Physiology Fundamentals of Anaesthesia and Acute Medicine

Page i
Fundamentals of Anaesthesia and Acute Medicine
Cardiovascular
Physiology
Second edition
Edited by
Hans-Joachim Priebe
Professor of Anaesthesia, University Hospital, Freiburg,
Germany
and
Karl Skarvan
Professor of Anaesthesia, University of Basel, Switzerland
Page ii
© BMJ Books 2000
BMJ Books is an imprint of the BMJ Publishing Group

www.bmjbooks.com
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, electronic, mechanical, photocopying,
recording and/or otherwise, without the prior written permission of the publishers.
First published in 1995
by the BMJ Publishing Group, BMA House, Tavistock Square,
London WC1H 9JR
Second edition published in 2000
British Library Cataloguing in Publication Data

A catalogue record for this book is available
from the British Library
ISBN 0-7279-1427-8
Typeset in Great Britain by

Apek Digital Imaging, Nailsea, North Somerset
Printed and bound in Great Britain by J.W.Arrowsmith Ltd, Bristol
Page iii
Contents
Page
Contributors v
Foreword vii
RONALD M JONES, ALAN R AITKENHEAD, PIERRE FOËX
Preface viii
HANS-JOACHIM PRIEBE, KARL SKARVAN
1 Cardiac cellular physiology and metabolism 1
HEINRICH TAEGTMEYER, ANNE B TAEGTMEYER
2 Ventricular performance 27
KARL SKARVAN
3 Cardiac electrophysiology 73
JOHN L ATLEE
4 Coronary physiology 119
HANS-JOACHIM PRIEBE
5 The pulmonary circulation 171
KEITH SYKES
6 Regulation of the cardiovascular system 213
NIRAJ NIJHAWAN, DAVID C WARLTIER
7 Cerebral circulation 240
DAVID K MENON
8 Renal, splanchnic, skin, and muscle circulations 278
NGUYEN D KIEN, JOHN A REITAN
9 Microcirculation 307
JAMES E BAUMGARDNER, ALEX L LOEB,
DAVID E LONGNECKER
10 Anaesthesia and the cardiovascular system 331
WOLFGANG BUHRE, ANDREAS HOEFT
Index 375
Page iv
FUNDAMENTALS OF ANAESTHESIA AND ACUTE MEDICINE
Series editors
Ronald M Jones, Professor of Anaesthetics, St Mary’s Hospital Medical School, London,

UK
Alan R Aitkenhead, Professor of Anaesthetics, University of Nottingham, UK
Pierre Foëx, Nuffield Professor of Anaesthetics, University of Oxford, UK
Titles already available:
Cardiovascular Physiology (second edition)

Edited by Hans Joachim Priebe and Karl Skarvan
Clinical Cardiovascular Medicine in Anaesthesia

Edited by Pierre Coriat
Intensive Care Medicine

Edited by Julian Bion
Management of Acute and Chronic Pain

Edited by Narinder Rawal
Neuro-Anaesthetic Practice
Edited by H Van Aken
Neuromuscular Transmission
Edited by Leo HDJ Booij
Paediatric Intensive Care

Edited by Alan Duncan
Forthcoming:
Pharmacology of the Critically Ill

Edited by Maire Shelly and Gilbert Park
Anaesthesia for Obstetrics and Gynaecology

Edited by Robin Russell
Page v
Contributors
John Atlee III, MD

Professor of Anesthesiology
Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, USA
James E Baumgardner, PhD, MD

Assistant Professor of Anesthesia and Bioengineering
Department of Anesthesia, University of Pennsylvania, Philadelphia, USA
Wolfgang Buhre, MD
Department of Anaesthesia, Georg-August Universität, Göttingen, Germany
Andreas Hoeft, PhD, MD

Professor of Anaesthesia and Chairman
Department of Anaesthesia, Rheinische Friedrich-Wilhelms Universität Bonn, Germany
Nguyen D Kien, PhD

Department of Anesthesiology, School of Medicine, University of California, Davis, USA
Alex L Loeb, PhD

Assistant Professor of Anesthesia and Pharmacology
David E Longnecker, MD
Robert Dunning Dripps Professor and Chairman
David K Menon, MD, FRCP, FRCA

Lecturer in Anaesthesia, University of Cambridge
Director, Neurosciences Critical Care Unit, Department of Anaesthesia, Addenbrooke’s
Hospital, Cambridge, UK
Page vi
Niraj Nijhawan, MD, MS

Assistant Professor of Anesthesiology
Zablocki Veterans Administration Medical Center, Milwaukee, Wisconsin, USA
Hans-Joachim Priebe, MD, FRCA

University Hospital, Freiburg, Germany
John A Reitan, MD
Department of Anesthesiology, School of Medicine, University of California, Davis, USA
Karl Skarvan, MD
Professor of Anaesthesia
Department of Anaesthesia, University Hospital, Basel, Switzerland
Keith Sykes, MB BChir, FRCA

Emeritus Professor, Nuffield Department of Anaesthetics, University of Oxford, UK
Anne B Taegtmeyer, BM BCh

Clinical Research Fellow
Cardiothoracic Surgery Division, National Heart and Lung Institute, Imperial College of
Science, Technology and Medicine, London, UK
Heinrich Taegtmeyer, MD, DPhil

Professor of Medicine
Department of Internal Medicine, Division of Cardiology, University of
Texas –Houston Medical School, Houston, USA
David C Warltier, PhD, MD

Professor of Anesthesiology, Cardiology and Medicine and Vice-Chairman of Research
Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, USA
Page vii
Foreword
The pace of change within the biological sciences continues to increase and nowhere is this
more apparent than in the specialties of anaesthesia, acute medicine, and intensive care.
Although many practitioners continue to rely on comprehensive but bulky texts for
reference, the accelerating rate of biomedical advances makes this source of information
increasingly likely to be dated, even if the latest edition is used. The series Fundamentals of
anaesthesia and acute medicine aims to bring to the reader up to date and authoritative
reviews of the principal clinical topics which make up the specialties. Each volume will
cover the fundamentals of the topic in a comprehensive manner but will also emphasise
recent developments or controversial issues.
International differences in the practice of anaesthesia and intensive care are now much
less than in the past, and the editors of each volume have commissioned chapters from
acknowledged authorities throughout the world to assemble contributions of the highest
possible calibre. Three volumes will appear annually and, as the pace and extent of
clinically significant advances varies among the individual topics, new editions will be
commissioned to ensure that practitioners will be in a position to keep abreast of the
important developments within the specialties.
Not only does the pace of advance in biomedical science serve to justify the appearance
of an international series of this nature but the current awareness of the need for more
formal continuing education also underlines the timeliness of its appearance. The editors
would welcome feedback from readers about the series, which is aimed at both established
practitioners and trainees preparing for degrees and diplomas in anaesthesia and intensive
care.
RONALD M JONES
ALAN R AITKENHEAD
PIERRE FOËX
Page viii
Preface
In the few years since the first edition of Cardiovascular Physiology, knowledge in this
field has again expanded tremendously. This second edition is therefore written to keep the
reader updated on progress in this area, and to further improve his/her understanding of
basic cardiovascular physiology.
The understanding of basic cardiovascular physiology is more important than ever. The
patient population is becoming progressively more elderly and infirm. Even the old and
sick undergo increasingly complex and stressful procedures.
More effective prehospital trauma care improves initial survival of previously fatal
injuries, but for subsequent long-term survival the cardiovascular system is challenged to
the maximum. New guidelines on perioperative cardiac evaluation restrict the extent of
preoperative testing. The concept of “same day surgery” limits the time available for
preoperative optimisation of cardiovascular performance. Minimally invasive surgical
techniques decrease tissue trauma, but they can impose an additional burden on the
cardiovascular system.
Furthermore, current medical practice encourages restriction of blood transfusion and
acceptance of low haemoglobin concentrations which often requires an increase in cardiac
work to maintain oxygen delivery. Early postoperative extubation and rapid hospital
discharge can pose additional stress on the cardiovascular system.
Not surprisingly, therefore, cardiovascular complications remain the dominant cause of
overall perioperative morbidity and mortality. For all of these and many others a solid
understanding of cardiovascular physiology remains a prerequisite for optimum patient
care.
All chapters have been completely revised and updated. We gratefully acknowledge the
contributions of all authors. Without their work and expertise, this monograph would not
have been possible.
HANS-JOACHIM PRIEBE
KARL SKARVAN
Page 1
1: Cardiac cellular physiology and

metabolism
HEINRICH TAEGTMEYER, ANNE B TAEGTMEYER
New horizons
Cardiac cellular physiology and metabolism have long been considered areas of interest to
only a small group of basic scientists. Although, in the past, clinicians have laid the
foundations for much of the work in this area,1 2 their research was descriptive and its
clinical use limited to the detection of coronary artery disease by the release of lactate or
alanine from the ischaemic myocardium.3-5 When, it was shown (by isotopic methods) that
lactate release occurred in the presence of net lactate extraction by the normal human
heart,6 it appeared that the potential for invasive assessment of cardiac metabolism to detect
coronary artery or other forms of heart disease could not be fulfilled.
A number of technical advances in the diagnosis and treatment of heart disease have
reawakened the interest in cardiac cell metabolism by clinical investigators and basic
scientists alike. The most dramatic recent examples include reports on enhanced myocardial
function in transgenic mice overexpressing the β2-adrenergic receptor,7 highly efficient
gene transfer into adult ventricular myocytes,8 the grafting of fetal myocytes into adult host
myocardium,9 and the transfer of autologous myoblasts into damaged myocardium.10
It seems, however, that the ultimate success of gene therapy for the failing heart
continues to be constrained by the inadequate understanding of the underlying
pathophysiological events. New insights into cellular physiology and pathophysiology have
come from the use of positron labelled metabolic tracers or tracer analogues, which are
used for non-invasive assessment of regional myocardial blood flow and metabolism as a
tool to differentiate between reversible and irreversible myocardial ischaemia.11–13 Other
examples also include the use of tomographic nuclear magnetic resonance (NMR)
spectroscopy for the early detection of contractile dysfunction in the pressure overloaded
left ventricle.14 Cellular mechanisms of adaptations to ischaemia, reperfusion, and
reperfusion injury have come into focus when it became possible to reverse the deleterious
effects of compromised
Page 2
Fig 1.1 Metabolism as link between gene expression and contractile function of the heart. The heart acutely
adapts to changes in function by changing its metabolic rates. Chronic changes in function alter the expression
of metabolic genes.16
or absent blood flow.15 Recently, we found that ventricular unloading reproduces the fetal
pattern of gene expression also found in the hypertrophied heart.16 The induction of a fetal
gene response provides a molecular basis for the functional improvement of the failing
heart after treatments such as the left ventricular assist device (LVAD) which allow the
heart to ‘‘rest”.
We now recognise that the precise and rapid regulation of energy substrate metabolism
allows the heart to adapt to changes in workload from one beat to the next. We also
recognise that metabolism forms the essential link between gene expression, on the one
hand, and contractile function of the heart, on the other (Fig 1.1).
The message is clear: modern management of patients with overt or latent cardiac
dysfunction includes physiological approaches that only a decade ago were unimaginable.
It is therefore appropriate to review some of the salient concepts of cellular function and
energy transfer in heart muscle and relate them to specific clinical situations, which are
addressed in the second part of this chapter. The interested reader may also wish to refer to
the recent monograph “Energy metabolism of the heart: from basic concepts to clinical
applications” for more detailed information.17
Principles of energy transfer in heart muscle

Heart muscle possesses a complex, yet very efficient system of energy transfer. At the
centre of this system is a network of enzyme catalysed reactions. Although bewildering at
first glance, the purpose of this network is easy to understand, and a number of simple
principles on function and
Page 3
Salient features of heart muscle

• Consumer and provider of energy
• High rate of energy turnover
• Metabolic omnivore
• Depends on O2 for energy production
• Only limited endogenous food reserves
• Wide range of adaptations
metabolism of the heart are worth remembering. These are listed in the box and discussed
below.
The heart, like any organ of the mammalian body, consists of a number of different
components, all of which are in a constant state of flux. These include:
z interactive proteins
z purine bases
z energy providing intermediates
z membranes
z ions
z signal molecules.
In addition, heart muscle has retained its ability to adapt to environmental changes by
altering the synthesis and/or degradation rates of specific proteins or, acutely, by changing
the flux through metabolic pathways in order to maintain its state of equilibrium. Only the
most severe environmental changes, such as those induced by an interruption of O2 supply,
result in a cessation of energy production and consequent heart muscle failure, which in
turn causes a collapse of the blood circulation.
Heart muscle is both a consumer and a provider of energy. The heart consumes energy
locked in the chemical bonds of fuels through their controlled combustion, and converts
chemical energy into physical energy. The predominant form of physical energy of the
heart consists of pump work. In this respect, the heart can be considered to be a transducer
(that is, a device that receives energy from one system and transmits it to another). As a
result of its ability to convert chemical energy into mechanical energy, the heart also
provides energy in the form of substrates and O2 both for itself and to the rest of the body.
In this context, two important concepts emerge:
1 The heart is a “hot spot” of metabolic activity because ATP (the chemical energy
available for conversion to mechanical energy at the contractile site) must be
continuously resynthesised from its breakdown products
Page 4
ADP and Pi (inorganic phosphate). The greater the work output, the higher the rate of ATP
turnover.
2 When the heart’s ability to convert chemical into mechanical energy is impaired (for
whatever reason), the consequences result in functional and metabolic abnormalities in
the rest of the body. These abnormalities are commonly referred to as “heart failure”.
There is no organ in the human body that is not affected by an impairment of energy
transfer in the heart.
The role of ATP as the main provider of chemical energy for various cell functions was
first postulated by Lipmann18 when he drew attention to the biological importance of the
ATP–ADP couple. The rate of ATP turnover in the heart is far greater than in other organs
of the mammalian body, and it is often underestimated. A simple calculation, based on
measurements of myocardial O2 consumption, indicates that in the course of 24 hours the
human heart produces (and uses) 5 kg ATP; in other words, more than 10 times its own
weight and more than 1000 times the amount of ATP stored in the heart and readily
available for hydrolysis at any one instant in time.19 Although the human heart accounts for
only 0.5% of the total body weight, it claims 10% of the body’s O2 consumption. Lastly, it
is important to remember that the rate of energy turnover, and not the tissue content of
ATP, is the determinant of myocardial energy metabolism.20–22
As the heart meets the bulk of its energy needs by oxidative phosphorylation of ADP, it
is not surprising that heart muscle cells are also richly endowed with mitochondria, the cell
organelles that contain the enzymes of oxidative metabolism. There is a close correlation of
mitochondrial volume fraction, heart rate, and total body O2 consumption.23 Not only are
cardiac mitochondria abundant in number, they also contain a far larger number of cristae
(the morphological sites of the respiratory chain enzymes) than mitochondria in other
organs such as the liver, brain, or skeletal muscle.24
Finally, energy metabolism of the heart must also be considered in the context of energy
transfer in biological systems in general. Knowledge of the vast array of metabolic
pathways in the cell is often regarded with apprehension by students of biochemistry. The
complexities of pathways become comprehensible when one considers the following three
general principles:
1 The first law of thermodynamics
2 The second law of thermodynamics
3 The principle of moiety conservation.
Understanding these three principles also makes it easier to comprehend the clinical
relevance of altered myocardial metabolism in the setting of myocardial ischaemia,
infarction, or reperfusion.
Page 5
Thermodynamic aspects of energy transfer in biological systems
Energy transfer in biological systems obeys the first and second law of thermodynamics.
These two laws state that, within a closed system, energy can be converted only from one
form into another, and that a process will occur spontaneously only if it is associated with
an increase in disorder (or entropy) of the system. In short: nothing comes from nothing.
Energy is captured through the process of photosynthesis. The captured energy is, in
turn, released through the reactions of intermediary metabolism to produce reducing
equivalents, which combine with molecular O2 to from H2O (see below). Most
dehydrogenase reactions are also linked to decarboxylation reactions (for example,
pyruvate dehydrogenase, isocitrate dehydrogenase, and 2-oxoglutarate dehydrogenase)
resulting in the liberation of carbon dioxide. Carbon dioxide and water are, in turn, the
substrate for photosynthesis. The description of this simple energy cycle emphasises the
fact that, in the biological environment, molecules are recycled.
Metabolic pathways and moiety conserved cycles
It is a characteristic property of all living cells, including heart muscle, to provide an

environment in which complex chemical reactions can proceed quickly at relatively low
temperatures and low substrate concentrations. The efficient transfer of energy occurs via
enzyme catalysed metabolic pathways, at the centre of which are moiety conserved cycles
(that is, a cycle in which the concentration of the participating intermediaries neither
increases nor decreases). Moiety conserved cycles permit multiple use of given resources
and are most likely to result from evolutionary selection.25 26
A metabolic pathway is defined as a series of enzyme catalysed reactions that starts with
a flux generating step, usually a reaction catalysed by a non-equilibrium reaction or
transport of the metabolite across a membrane, and ends with the removal of a product (for
details see the literature27 28).
Biochemists distinguish between the control and regulation of metabolism, because these
determine energy transfer in the cell.26 In heart muscle, control of a metabolic pathway
means that a change in the level of a control factor (for example, work load, hormones or
drugs, substrate concentration, coronary flow, or O2 availability) will change the rate of
substrate flux. In contrast, regulation of a metabolic pathway means that flux of a substrate
in a metabolic pathway is dictated by the activity of enzymes, regulators of enzymes,
cofactors, and signal transduction pathways.
Many, but not all, of the regulators of energy transfer pathways are part of moiety
conserved cycles. The largest moiety conserved cycle is in fact the circulation itself, where
erythrocytes and plasma serve as a vehicle for the transport of O2 and substrates and the
removal of CO2 and metabolic end
Page 6
Fig 1.2 Energy transfer in heart muscle: efficient energy transfer occurs in moiety conserved cycles. (See text
for further details.)
products. The hydrolysis of ATP through cross bridge cycling inside the myocardial cell
itself is also a moiety conserved cycle; it acts:
z to decrease the proton gradient

z to increase the oxidation of NADH
z to increase flux through the citric acid cycle
z to increase acetyl-CoA utilisation
z to increase substrate consumption.
A convenient way of viewing energy transfer in heart muscle is as a “three-ring-circus’’,

consisting of systemic circulation, cellular metabolism, and cross bridge cycling (Fig
1.2).The cycles interact, just like cog-wheels, in such a way that an increase in the rotation
rate of one cycle causes a concurrent increase in the rotation rate of the other two (Fig 1.3).
Thus an increase in cross bridge cycling leads to an increase in cellular metabolism and
systemic circulation, which in turn provides the larger substrate load required by the
metabolism cycle to maintain the new level of functioning. Conversely, it has been pointed
out that, in animals with long circulation times, less substrate and less oxygen are delivered
to the cell, and as both limit the rate of the metabolic reaction, the speed of the cycles is
slow. In animals with short circulation times, all the needed components of the
Fig 1.3 Interaction of cycles involved in energy transfer: feedback control of cycles is illustrated by the
example of increased contractile activity. (See text for further details.)
Page 7
reactions are delivered to the site in a continuous rapid stream and reactions occur almost
explosively.29
Another example of a moiety conserved cycle is one that governs intracellular calcium
homoeostasis. According to Barry and Bridge,30 calcium homoeostasis in cardiac myocytes
is of functional importance for at least three reasons:
1 The resting cytosolic calcium concentration ([Ca2+]) of less than 0.2 μmol/l necessary to
allow the contractile elements to relax during diastole must be maintained against a
5000-fold gradient across the sarcolemma ([Ca2+]>1 mmol/l).
2 Excitation–contraction coupling involves a complex interaction of membrane electric
elements mediated by specific ion channels. This results in Ca2+ influx, which triggers
the release of large amounts of Ca2+ from the sarcoplasmic reticulum via Ca2+ specific
release channels31 and subsequent extrusion of Ca2+. To maintain steady state
homoeostasis in this cycle the amount of Ca2+ entering the cell with each contraction
must be extruded before the next contraction. Likewise, the large amount of Ca2+
released from the sarcoplasmic reticulum must be pumped back into the storage
compartment. As a net result, only small amounts of Ca2+ enter and leave the cell with
each cardiac cycle.
3 The force of contraction in cardiac myocytes is modulated by variations in the magnitude
of the Ca2+ transient. Hormones or drugs that modify Ca2+ homoeostasis may
significantly alter the force of contraction. In addition, cytosolic Ca2+ may be taken up
into the mitochondria. Although the mitochondrial Ca2+ stores are only indirectly related
to the contraction–relaxation cycle, Ca2+ ions are regulators of a number of
intramitochondrial enzymes that are activated by Ca2+.32
Thus Ca2+ ions form a link between utilisation and production of ATP. Although this
hypothesis has not yet been proven,33 Denton and McCormack34 have proposed that the
main role of the Ca2+ transporting system within the inner mitochondrial membrane should
be viewed primarily as a means by which changes in the cytosolic [Ca2+] could be relayed
into mitochondria and hence influence the activity of the intramitochondrial Ca2+ sensitive
dehydrogenases. As Ca2+ is the only second messenger for hormones that is transferred
across the inner mitochondrial membrane, an important feature of this hypothesis is a
suggested mechanism by which mitochondrial oxidative metabolism and, hence, ATP
supply could be stimulated to meet increased demands for ATP that are associated with the
stimulation of the processes promoted by increases in cytosolic Ca2+.32 The same authors
suggest that, when the Ca2+ dependent mechanism for activating oxidative metabolism is
available (that is, via the Ca2+ sensitive enzymes in the mitochondrial matrix), then it is the
preferred mechanism for promoting the overall process of oxidative phosphorylation.32
Page 8
The sliding filament model of contraction represents another example of a moiety
conserved cycle. The molecular mechanisms involved in the sliding filament model of cross
bridge formation between actin and myosin have recently been elucidated.35 The essence of
the sliding filament model is that the myosin head binds to the actin filament in one
orientation, rotates to a second orientation, and then detaches. The cycle is driven in one
direction by coupling these transitions to the steps of ATP hydrolysis. Elucidation of the
structure of myosin and a model for the actomyosin complex during contraction, including
the description of an ATP binding pocket, have advanced our understanding of the
molecular design of muscle motors.
One important consequence of energy transfer through moiety conserved cycles is that
the loss of a moiety in any one of the cycles may lead to a loss of energy transfer within the
cell. The following are examples of such losses:
1 The loss of contractile proteins (such as when degradation exceeds synthesis), for
example, in certain forms of dilated cardiomyopathy or chronic myocardial ischaemia.
2 The loss of oxaloacetate from the citric acid cycle through either side reactions, such as
transamination, or the inhibition of one or more of the cycle enzymes.
Depletion of a moiety is recovered through its resynthesis from precursors via a series of
reactions termed ‘anaplerosis’. Hans Kornberg has defined anaplerosis as the replenishment
of a depleted cycle by an intermediate precursor.36 A case in point is the contractile
dysfunction of the isolated working rat heart perfused with acetoacetate, which is
completely reversed by the addition of glucose as a second substrate.37 The cause for the
contractile dysfunction is an inhibition of the enzyme 2-oxoglutarate dehydrogenase as a
result of sequestration of free coenzyme A,38 39 resulting in a shortage of oxaloacetate for
the citrate synthase reaction.40 The cause of normalisation of contractile function with the
addition of glucose is the carboxylation of pyruvate through the NADP dependent malic
enzyme reaction.41
As ischaemia also depletes the citric acid cycle of its intermediates, especially
succinate,42 it is tempting to speculate that the increased glucose and/or lactate requirement
in postischaemic myocardium43–45 may be a reflection of the increased need for
replenishment of the depleted cycle.
A second important consequence of energy transfer through moiety conserved cycles is
the effective recycling of moieties. These moieties not only involve larger carbon
molecules, such as glucose and fatty acids, but also the smaller organic acids of the citric
acid cycle, and especially recycling of CO2 and H2O. Without the recycling of H2O, ATP
production in the citric
Page 9
acid cycle would be 60% less than it is with its recycling of H2O (6 versus 15 moles ATP
per mole pyruvate oxidised). As Ephraim Racker wrote:46
Mitochondria cleave water without the drama of sunlight and chlorophyll. They perform this task,
unnoticed by textbooks, in the quiet and unobtrusive manner characteristic of Hans Krebs and his
cycle.
It stands to reason that, under certain circumstances, the excess of a moiety may also lead
to contractile dysfunction of the heart. This occurs for example in ischaemia, reperfusion,
and myocardial stunning, where the cells and their organelles may be flooded with Ca2+,47–
49 oxygen derived free radicals,50–53 and protons,54 55 resulting in cell swelling, osmotic
stress, and membrane damage.56
Catabolism of substrates
In the heart, the direction of most enzyme catalysed reactions is catabolic, that is,
substrates with high potential energy are broken down to products with low potential
energy. Synthetic, or anabolic, reactions such as those serving protein, glycogen, or
triglyceride synthesis are quantitatively of lesser importance, but ultimately they serve to
improve the efficiency of energy production in heart muscle. Thus, heart muscle is
endowed with an efficient system of energy transfer, which liberates energy locked in
chemical bonds through the generation of reducing equivalents and their reaction with
molecular O2 in the respiratory chain. It should be stated once more that the main purpose
of intermediary metabolism in normal heart muscle is the production of reducing
equivalents for ATP synthesis by oxidative phosphorylation of ADP.
As proposed by Lehninger,57 it is convenient to group the breakdown of substrates into
three stages:
1 The first stage consists of the breakdown of substrates to acetyl-CoA.
2 The second stage is the oxidation of acetyl-CoA in the citric acid cycle.
3 The third stage is the reaction of reducing equivalents with molecular O2 in the
respiratory chain, where electron transfer is coupled to rephosphorylation of ADP to
ATP.
As ATP production is tightly coupled to ATP utilisation, so is substrate oxidation tightly
coupled to cardiac work.37 58–62 It appears that, in the presence of adequate substrate
supply, the maximal rate of oxidation of substrate is determined by the capacity of the 2-
oxoglutarate dehydrogenase reaction in the citric acid cycle.63 The exact mechanism by
which respiration is coupled to energy expenditure in vivo is not, however, known.64 The
efficiency of oxidative phosphorylation for energy production is, however, well established
– 1 mole of glucose, when oxidised, yields 36
Page 10
moles of ATP, whereas the same amount of glucose yields only 2 moles of ATP when
metabolised to lactate under anaerobic conditions:
On a mole for mole basis, the energy yield from the oxidation of long chain fatty acids is
even greater than that from glucose or lactate.
Nutrition of the heart and myocardial protein turnover

The recent interest in healthy nutrition for the heart has almost exclusively focused on
cholesterol because of its role in the development of coronary artery disease. There is little
appreciation of the fact that, in terms of general descriptors of energy metabolism, heart
muscle functions not simply as a conformer in response to substrate availability,65 but that
substrate utilisation is controlled by the physiological demands on the system. Likewise,
there is little appreciation of the fact that the heart stores endogenous substrates such as
glycogen and triglycerides, and it does so in response to changes in the dietary state.66 67 In
contrast to skeletal muscle, starvation increases the tissue content of both glycogen and
triglycerides in heart muscle, an observation consistent with a biologist’s definition of true
“hibernation”.
Another fact that is not appreciated is that the heart continuously synthesises and
degrades its own constituent proteins,68 a process that is significantly slowed down by
myocardial ischaemia.69 70 Although protein turnover is perhaps the most difficult metabolic
process to study in the heart in vivo, and although it appears that each protein has its own
characteristic half life,71 recent estimates indicate that 4.8% of myocardial protein is
synthesised each day,72 that is, the mammalian heart regenerates itself completely over a
period of three weeks. As the net muscle mass is a function of both synthesis and
degradation, hypertrophy may be the result of either increased rates of protein synthesis or
decreased rates of protein degradation. Although acute volume overload of the myocardium
leads to an increase in synthesis, it appears that chronic volume overload leads to
suppression of protein degradation.72 The pathways linking mechanical signals to changes
in cardiac myocyte degradation rates are not known,73 and the unravelling of mechanisms
involved in myocardial protein degradation continue to pose a challenge to cellular and
molecular biologists.
Page 11
Substrate competition
As a result of the omnivorous nature of the heart, glucose, lactate, fatty acids, ketone
bodies, and, under certain circumstances, amino acids are all converted to acetyl-CoA and
so compete to be the fuel of respiration (Fig 1.4). The relative predominance of one fuel
over another depends on the arterial substrate concentration (which, in the case of fatty
acids, ketone bodies, and lactate, can vary greatly – Table 1.1), hormonal influences,
workload, and O2 supply. Likewise, the utilisation of specific substrates by the heart varies
with the physiological state of its environment. When Bing1 cannulated the coronary sinus
and measured aorta–coronary sinus differences in substrate concentrations across the heart,
he observed a proportional relationship between substrate concentration in the blood and
substrate uptake by the heart for all substrates investigated, that is, glucose,
Fig 1.4 Essential and non-essential fuels for cardiac energy production. Note that glucose, lactate, and
pyruvate provide both substrates for the citrate synthase reaction – acetyl-CoA and oxaloacetate.
Carboxylation of pyruvate leading to the formation of oxaloacetate is an anaplerotic pathway. The importance
of anaplerosis in the normal contractile function of the heart has recently been elucidated.41
Table 1.1 Metabolite concentrations in human plasma under various conditions
Glucose Lactate Free fatty acids Ketone bodies

(μmol/1) (μmol/1) (μmol/1) (μmol/1)
Rest 5.0 0.5 0.1 0.1

(postabsorptive)
Running (90 min) 5.0 5.0 1.7 1.8
Fasting 4.5 1.0 1.6 4.5
Diabetic 30.0 1.0 1.7 10.0
ketoacidosis
Reproduced by permission from Taegtmeyer H. Basic Res Cardiol 1984;79:322–36.

Page 12
lactate, fatty acids, ketone bodies, and amino acids. Subsequent work by Keul et al.2 has
established that the contribution of fuels to the fuel of respiration for the heart depends on
the physiological state of the whole body, which can vary greatly. The data from Keul’s
work are of interest, because they show glucose uptake to be relatively constant (16–31%),
whereas the uptake of fatty acids plus ketone bodies and the uptake of lactate vary
considerably (from 25% to 63%, and from 5% to 61%, respectively). This observation is of
relevance with respect to fatty acids and ketone bodies. Although fatty acid oxidation can
be almost completely suppressed when lactate and pyruvate are abundant, there is a
consistent rate of carbohydrate use. The need for glucose or lactate is most probably the
result of the need for pyruvate carboxylation and the anaplerosis of the citric acid cycle. In
keeping with this hypothesis we have shown that lactate (40 mmol/l) suppresses glucose
uptake by the isolated working rat heart by 90%, whereas β-hydroxybutyrate at the same
concentration suppresses glucose uptake by only 64%.74 Collectively these findings suggest
that the fuels for cardiac energy metabolism can be grouped (Fig 1.4) into essential fuels,
which provide both acetyl-CoA and oxaloacetate:
z glucose
z lactate
z pyruvate
z certain amino acids
and non-essential fuels, which provide only acetyl-CoA:
z fatty acids of all chain lengths

z ketone bodies
z leucine.
Fatty acids are the preferred fuel for respiration in the fasted state,75 but, even when fatty
acid or ketone body concentrations are high, a certain amount of glucose continues to be
oxidised.37 76 Conversely, high lactate concentrations, such as those observed with
strenuous exercise, can provide almost all77 or the bulk of the fuel for respiration.76
Even amino acids, when present in very high concentrations, can become a fuel for
respiration in heart muscle.1 In this respect, the heart is not different from the body as a
whole. When an omnivorous animal consumes a normal meal containing protein,
carbohydrate, and fat, the degradation of any excess amino acids (that is, amino acids not
needed for growth and replacement) takes precedence over the degradation of
carbohydrates and fats.78 This phenomenon results from strict control of amino acid
metabolism by their Km values (the Michaelis constant – the concentration of substrate
required for half maximal velocity of an enzyme catalysed reaction) and reveals an
important principle of metabolic control. As dietary protein or amino acids cannot be stored
in major quantities, the amino
Page 13
acids from intestinal digestion are distributed unchanged in blood plasma and tissues. By
contrast, products of carbohydrate and fat digestion can be stored rapidly as either glycogen
or triglycerides. As this storage process begins immediately, fluctuations in plasma glucose
and fatty acid levels are moderate and transient compared with fluctuations in amino acid
levels. The increased amino acid concentrations in blood and tissues after a meal
automatically cause an increased rate of amino acid degradation, because the Km values of
the enzymes initiating amino acid degradation are in general high and exceed the
concentration of amino acids in the tissues.
In summary, many factors contribute to the selection of energy providing fuels for the
heart. According to Krebs,78 they may be classified under three main categories:
1 Concentration of the direct fuel in the tissue.
2 The presence, in the tissue, of the enzymes required for the degradation.
3 The kinetic properties of the key enzymes, especially of those that initiate the release of
energy.
Each of these three main factors is, in turn, very complex and depends on a variety of
components. The entry of fuels into the cell, as well as synthesis and degradation of stored
fuel reserves, is controlled by hormones such as insulin and epinephrine (adrenaline) as
well as by other environmental factors, with cAMP (cyclic adenosine 3':5'-monophosphate)
and a cascade of intracellular signals acting as second messengers. Among the kinetic
properties of the key enzymes, the important ones are the Km values (see above) and the
inhibition and activation of enzymes by tissue constituents, which exercise either feedback
inhibition or allosteric control through allosteric effectors or covalent modification. When
the workload of the heart is raised acutely, these factors act together to trigger the
preferential oxidation of glycogen,79 80 thereby ensuring immediate availability of energy
for contraction. In short: the heart functions best when it oxidises several substrates
simultaneously.
Clinical relevance of myocardial metabolism

Altered energy metabolism is the cause of many clinical forms of heart disease
(summarised in the box). In a review on myocardial metabolism, Lionel Opie81 82 referred
to a “decline and resurgence of myocardial metabolism”. We have identified three areas of
clinical relevance:
1 Tracing of metabolic pathways for the diagnosis of ischaemia and other forms of heart
disease.
Page 14
Clinical forms of heart disease caused by altered energy metabolism
Myocardial ischaemia
Latent ischaemia
Hibernation
Myocardial infarction
Reperfusion/reperfusion injury
Stunning/postischaemic dysfunction
Preconditioning/stress response
Cardiomyopathies
Dilated cardiomyopathies
Systolic dysfunction
Hormonal or nutritional deficiencies
Hypertrophy
Adaptation
Maladaptation
Diastolic/systolic dysfunction
2 ‘‘Metabolic mechanisms of heart disease”, where contractile failure represents the end
result of profound metabolic derangements.
3 Metabolic support for the failing heart includes replenishment of cofactors or
intermediary metabolites in certain forms of dilated cardiomyopathy with resultant
improvement in contractile performance, and support for the failing heart after prolonged
periods of ischaemia as occurs in hypothermic ischaemic arrest.81 82
Tracing metabolic pathways in the intact heart
A detailed knowledge of the pathways of individual substrates for energy production is

usually not required by the clinician diagnosing or treating patients with heart disease.
Metabolism comes under scrutiny, however, when coronary arteries are not (or are no
longer) obstructed and yet the heart fails to contract, for example, as occurs in
cardiomyopathies or in reperfused myocardium after complete coronary occlusion. More
importantly, substrate metabolism has come into focus through the development of new,
non-destructive imaging techniques such as NMR spectroscopy and positron emission
tomography (PET), which permit the assessment of regional metabolic processes in the in
vitro and in vivo beating heart.13 21 83–91 Although NMR spectroscopy is able to detect
derangements in energy rich phosphate metabolism before the development of contractile
dysfunction,14 PET imaging allows us to detect reversibly ischaemic, viable myocardium.92–
94
Page 15
NMR spectroscopy
The basis of NMR spectroscopy is that, even though all nuclei of atoms have an overall
positive charge, some also have a “spin” which gives them a magnetic moment.95 A
powerful magnetic field orients the nuclear spins and, hence, establishes their different
energy states. Transitions between adjacent energy states are induced by the application of
an oscillating magnetic (or radiofrequency) field. The device that provides the
radiofrequency field is also used to detect the result and signal or resonance (hence, the
name radiofrequency coil or probe). Biologically important nuclei with a spin are 1H, 2H,
13C, 15N, 17O, 31P, 23Na, 39K, 87Rb, and 19F. Selective enrichment of low abundance
nuclei (for example, 13C) leads to an increase in their sensitivity.

Natural abundance NMR spectroscopy is most commonly used in the form of 31P NMR
spectroscopy, which yields distinct, quantitative resonance peaks for monophosphate esters,
inorganic phosphates, phosphocreatine, and ATP. Analysis of energy rich phosphates in the
beating heart in vivo by NMR spectroscopy of 31P supports the view that, over a relatively
wide range, the tissue content of ATP does not correlate with the rate of energy use as
measured by the rate of ATP turnover, that is, O2 consumption or contractile performance
of the heart.21 The recent introduction of a tomographic (spatial stacked plot) analysis of
31P NMR spectra has added a new dimension to the analysis of energy rich phosphates in
vivo. When this technique was applied to a group of patients with left ventricular
hypertrophy caused by aortic stenosis and/or insufficiency, heart failure was characterised
by a decline in the phosphocreatine:ATP ratio.14 The adaptation of isotopomer analysis of
13C natural abundance or labelled compounds permits the analysis of flux through specific
pathways, especially the citric acid cycle and glycogen turnover, through the acquisition of
serial spectra.84 89–91 A main advantage of NMR spectroscopy is the specificity of the
technique, which allows tracing of the flux of specific metabolites into and out of metabolic
pools. The isotopomeric enrichment of glutamate as an index for flux through the citric acid
cycle is an example.
Positron emission tomography

The tracing of metabolic pathways with short lived, positron emitting tracers has so far
been more successful in its clinical application than NMR spectroscopy, mainly because
there is technology that makes it possible to assess regional differences of metabolic
activity of the heart by visual inspection and quantitative analysis of radioactivity in
“regions of interest”.13 96 Two types of approaches can be distinguished:
1 Uptake and retention of a tracer analogue such as fluorodeoxyglucose (FDG).
Page 16
2 Uptake and clearance of tracers such as 11C-labelled fatty acids, where the rapid phase of
clearance from the tissue represents either β-oxidation and oxidation in the citric acid
cycle (in the case of long chain fatty acids), or oxidation in the citric acid cycle alone (in
the case of acetate).
Whereas the uptake and retention of FDG is linear with time and follows zero order
kinetics, the clearance of labelled fatty acids is biexponential,97–99 suggesting both rapid
and slow turnover pools for both long and short chain fatty acids. Relative size and slope of
each of the exponential components of the 11C time–activity curve relate to oxidation and
release from storage of the labelled compound. Both FDG and 11C-labelled fatty acids have
been used clinically to assess substrate metabolism in normal and ischaemic myocardium.
The argument of whether enhanced glucose uptake (assessed with FDG) or residual
oxidative capacity (assessed by the early, rapid clearance phase of [11C] acetate) constitutes
the gold standard for reversible tissue injury in ischaemic, reperfused, or “hibernating”
myocardium has not been settled. The clinical utility of a perfusion–metabolism mismatch
is, however, clear: preserved metabolic activity in the absence of significant coronary flow
(manifested by the retention of FDG and the absent uptake of the flow marker 13NH3,
respectively) is strongly suggestive of myocardium that has the potential to resume normal
contractile function (and hence oxidative metabolism) once blood flow and O2 supply have
been restored. It appears that the usefulness of imaging regional metabolic activity in heart
muscle is limited, because the same functional information can be obtained with less
expensive, more direct methods such as the assessment of contractile reserve.
Metabolic adaptation and deadaptation: the cellular consequences of

ischaemia and reperfusion
Heart muscle regulates its energy supply by regulating coronary blood flow in
accordance with the energy needs of the cell. For example, under resting conditions,
coronary flow is about 1 ml/min per g wet weight in humans, and it increases in proportion
to myocardial oxygen consumption; that is, when oxygen consumption doubles, coronary
flow doubles, and so on. Conversely, a reduction in coronary flow results in a reduction in
myocardial oxygen delivery and a consequent reduction in contractile force. In clinical
practice, this relationship manifests itself as stress induced asynergy or “hibernating
myocardium”.
The earliest forms of ischaemia, defined as lack of oxygen supply resulting from
inadequate blood flow, occur in patients who are unable to increase coronary flow in
response to increased energy demands. As resting coronary flow is normal in this setting,
this form of ischaemia is sometimes referred to as ‘‘normal flow” ischaemia. By contrast,
when coronary flow is
Page 17
reduced at rest, the term “low flow ischaemia” has been used. The extreme form of
ischaemia is, of course, reached by the complete occlusion of a coronary artery with
subsequent necrosis of the tissue supplied. Thus, there is a continuum of ischaemia, with
mild, “normal flow” ischaemia at one end of the spectrum and the extreme situation of
myocardial infarction at the other.
Ischaemia affects myocardial energy metabolism by slowing down aerobic metabolism
of substrates, reducing the tissue content of phosphocreatine and adenine nucleotides, and
first increasing and then slowing down anaerobic metabolism of substrates. Just as there is a
continuum of relative restriction of oxygen delivery, one might expect a continuum of
metabolic responses to ischaemia. With “normal flow” ischaemia, heart muscle is still
capable of oxidising fatty acids and glucose under resting conditions. As coronary blood
flow decreases, the relative contribution of glucose to the residual oxidative metabolism
increases, and oxidation of glucose may account for a greater percentage of aerobic ATP
production.100 Increased uptake of a glucose analogue by ischaemic myocardium has also
been found when the energy demand for the heart was increased by pacing or exercise.6 101
There is increased lactate release from the stressed myocardium6 102 and increased glucose
uptake, especially when fatty acid levels are low.103 Possible reasons for increased glucose
uptake with stress and ischaemia are as follows:
1 Glucose makes better use of the limited amount of O2 available to the myocyte. If blood
supply is mildly reduced, the heart switches from fatty acids to glucose as the preferred
fuel for respiration.
2 Glycolysis yields a small amount of ATP through substrate level phosphorylation in the
cytosol, independent of the availability of O2 (2 mol ATP/mol glucose, whereas 36 mol
ATP are produced per mol glucose oxidised).
3 Glucose transport is enhanced in oxygen deprived tissue. Thus, more glucose enters the
cell, and glucose is preferred over fatty acids as a substrate for energy production.
The regulation of intermediary metabolism of glucose, fatty acids, and amino acids
during ischaemia is complex and requires further discussion with respect to accumulation
of intermediary metabolites and reversibility of ischaemic tissue damage. When oxygen
becomes rate limiting for energy production, flux through the electron transport system of
the respiratory chain slows down and the ratio of the reduced form of nicotinamide adenine
dinucleotide (NADH) to the oxidised form (NAD+) (that is, [NADH]:[NAD+]) increases.
This reduced state reflects a lack of ATP production by oxidative phosphorylation, which is
accompanied by a loss of contractile function.
Page 18
The exact biochemical mechanisms responsible for the rapid loss of contractile function
are not yet known with certainty. There are those who implicate the loss of ATP104 and
others who implicate the accumulation of potentially toxic intermediary products such as
hydrogen ions (H+)105 106 or lactate.107 Kübler and Katz108 thought it unlikely that decreased
ATP supplies for energy consuming reactions in the myocardial cell cause the observed
decrease in myocardial contractility, because of the low Km for ATP at the substrate
binding sites of energy consuming reactions in the heart. In other words, at prevailing
concentrations of ATP in the ischaemic, non-contracting tissue, enzymes such as myosin
ATPase should still operate at near maximal velocity. Instead, Kübler and Katz108
speculated that small changes in ATP may already exert modulatory effects on ion fluxes,
and the large amount of inorganic phosphate may form insoluble precipitates of calcium
phosphate that trap calcium in the sarcoplasmic reticulum and mitochondria. Another
possible explanation for the discrepancy between ATP content and ATP conversion into
useful energy for the heart is the trapping, or “compartmentation”, of ATP in a
compartment that is not accessible to the enzymes of the contractile apparatus or ion pumps
(for example, mitochondria).
Examining the acute effects of ischaemia on phosphocreatine and ATP, Gudbjarnason et
al109 found that breakdown of phosphocreatine was more rapid than that of ATP. The
kinetic heterogeneity of ATP and phosphocreatine depletion seems to indicate an inhibition
of transfer of ATP from mitochondria to the cytosol, and it has been speculated that the
reduction in regeneration of cytosolic ATP causes the early cessation of contractile activity
in ischaemic myocardium. It is reasonable to state that the actual biochemical mechanism
for contractile failure in the ischaemic and infarcted myocardium continues to remain
elusive. Recent experimental work has emphasised the phenomenon of ischaemic
preconditioning110 and the role of stress proteins in myocardial protection.111 With the
exception of glutamate, glucose is the only substrate yielding ATP by anaerobic substrate
level phosphorylation.17 Glucose uptake is increased with low flow ischaemia both in
vitro112 and in vivo,113 as a result of translocation of glucose transporters to the plasma
membrane. Addition of insulin further enhances glucose uptake and glycogen.112 Thus, the
effects of ischaemia are additive.
Metabolic support of the acutely ischaemic myocardium
The use of glucose, insulin, and potassium (GIK) as inotropic metabolic support for the
acutely ischaemic, reperfused myocardium is controversial and has largely been abandoned
on the basis of theoretical114 and experimental107 argument. Likewise, the use of GIK in the
setting of acute myocardial infarction, first proposed by Sodi-Pallares and his co-workers in
Mexico (1962) and further developed by Rackley and his co-workers in the
Page 19
USA115–117 has not generally been accepted because of inconclusive evidence in earlier
clinical trials.118 In spite of substantial experimental evidence in support of beneficial
effects of substrate manipulation especially promoting glucose metabolism in myocardial
ischaemia,119–122 the concept of metabolic support for the failing ischaemic (or
postischaemic) myocardium was relegated to the antics of medical therapy.
Glycogen loading of rat hearts 90 min before hypothermic ischaemic arrest significantly
improves ischaemia tolerance, as evidenced by a return of normal left ventricular function
after 12 h of ischaemia (instead of 3 h in controls).123 In contrast, glycogen depletion before
ischaemia failed to improve left ventricular function of rabbit heart after hypothermic
ischaemic arrest.124 There is a correlation between glycogen content, on the one hand, and
the tissue content of energy rich phosphates and recovery of function with reperfusion, on
the other, although the mechanism for the protective effect of GIK is still unknown.
The effect of glycogen loading on recovery of function and associated biochemical
parameters after a brief (15 min) period of normothermic ischaemia and reperfusion in rat
hearts125 126 showed that glycogen loaded hearts recovered faster than their controls, used
more glucose, maintained normal energy rich phosphate levels, and lost a significantly
smaller amount of marker proteins (myoglobin, lactate dehydrogenase, citrate synthase)
with reperfusion. Although these studies are largely descriptive, they point to a
physiological role for glycogen, which complements its role as endogenous substrate but is
still elusive to a mechanistic analysis.
Although there are no prospective, controlled, clinical studies that examine the efficacy
of GIK in patients with refractory left ventricular failure after cardiopulmonary bypass and
hypothermic ischaemic arrest for aortocoronary bypass surgery, a small randomised clinical
trial on 22 patients examined the efficacy of GIK (for the protocol see Table 1.2) for up to
48 h.127 The results were so striking (a 50% increase in cardiac index, a 30% decrease in the
requirement for inotropic drugs, and a 75% decrease in 30 day mortality) that surgeons at
the Texas Heart Institute now use GIK
Table 1.2 Glucose–insulin–potassium (GIK) for metabolic support of the postischaemic
failing heart
D-Glucose 500 g
Regular insulin 80 U
KCl 100 mmol
In 1000 ml H20, infusion rate 1 ml/kg per h; requires indwelling catheter
Protocol: blood for glucose and K+ before, and at 1, 6, 12, 24, and 48 h after initiation of
treatment. Supplemental insulin only if blood glucose exceeds 300 mg% and/or K+
exceeds 5.3 mmol/l
Exclusion criteria: creatinine >3 mg/dl, bilirubin>3 mg/dl
From Gradinak et al.127

Page 20
routinely in the management of postoperative refractory left ventricular failure of different
aetiologies. In addition, protocols of preoperative glycogen loading are being developed for
the ex vivo preservation of a donor heart for cardiac transplantation and for high risk
patients with compromised left ventricular function (left ventricular ejection fraction <30%
before surgery). To a large extent, these protocols are now employed on an empirical basis
with good success, but unfortunately lack the benefit of rigorous scientific scrutiny.
An important, but little appreciated, intervention for reducing mortality in acute
myocardial infarction is the concept of metabolic support with intravenous GIK. A recent
meter analysis by Fath-Ordoubadi and Beatt128 revealed that GIK reduced in-hospital
mortality of myocardial infarction by 28–48%. This magnitude of reduction in mortality is
comparable to that achieved with thrombolytic therapy129 and supports the concept that
metabolic protection of ischaemic myocardium is as important as reperfusion itself.130
Conclusions
The heart is both a consumer and a provider of energy. Energy transfer in heart muscle is
highly efficient and occurs through a series of moiety conserved cycles. New methods
developed over the past decade have resulted in a better understanding of the physiology of
myocardial cell function, and gene therapy for the correction of cellular defects is looming
on the horizon. The ultimate success of new treatment modalities is, however, still
constrained by an inadequate understanding of the underlying pathophysiological events.
These recent developments point to a need for the re-examination of the concept and
application of metabolic treatment for the failing myocardium in defined clinical settings,
such as reperfusion after an acute ischaemic event, controlled hypothermic ischaemic arrest,
or acute myocardial infarction.
Acknowledgements
We thank Rachel Ralston for her help in preparing the manuscript for publication. The
authors’ laboratory is supported by grants from the US Public Health Service, National
Institutes of Health (R01-HL 43113) and the American Heart Association, National Center.
1 Bing RJ. The metabolism of the heart. Harvey Lect 1955;50:27–70.
2 Keul J, Doll E, Steim H, Homburger H, Kern H, Reindell H. Über den Stoffwechsel des menschlichen
Herzens I. Pflügers Arch Ges Physiol 1965;282:l–27.
Page 21
3 Gorlin R, Brachfeld N, Messer JV, Turner JD. Physiologic and biochemical aspects of disordered
coronary circulation. Ann Intern Med 1959;51:698–706.
4 Krasnow N, Neill WA, Messer JV, Gorlin R. Myocardial lactate and pyruvate metabolism. J Clin Invest
1962;41:2075–85.
5 Mudge GH, Mills RM, Taegtmeyer H, Gorlin R, Lesch M. Alterations of myocardial amino acid
metabolism in chronic ischaemic heart disease. J Clin Invest 1976;58:1185–92.
6 Gertz EW, Wisneski JA, Neese RA, Bristow JD, Searle GL, Hanlon JT. Myocardial lactate
metabolism:Evidence of lactate release during net chemical extraction in man. Circulation 1981;63:1273–
9.
7 Milano CA, Alien LF, Rockman HA, et al. Enhanced myocardial function in transgenic mice
overexpressing the β2-adrenergic receptor. Science 1994;264:582–6.
8 Kirshenbaum LS, MacLellan WR, Mazur W, French BA, Schneider MD. Highly efficient gene transfer
into adult ventricular myocytes by recombinant adenovirus. J Clin Invest 1993;92:381–7.
9 Soonpaa MH, Koh GY, Klug MG, Field LJ. Formation of nascent intercalated disks between grafted fetal
cardiomyocytes and host myocardium. Science 1994;264:98–101.
10 Taylor DA, Atkins BZ, Hungspreugs P, et al. Regenerating functional myocardium: improved
performance after skeletal myoblast transplantation. Nature Med 1998;4:929–33.
11 Schelbert HR, Schwaiger M. Positron emission tomography studies of the heart. In: Phelps M, Mazziota J,
Schelbert H, eds, Positron emission tomography and autoradiography: principles and applications for the
brain and the heart. New York: Raven Press, 1986:581–661.
12 Bergmann SR. Clinical applications of assessments of myocardial substrate utilization with positron
emission tomography. Mol Cell Biochem 1989;88:201–8.
13 Schwaiger M, Hicks R. The clinical role of metabolic imaging of the heart by positron emission
tomography. J Nucl Med 1991;32:565–78.
14 Conway MA, Allis J, Duwerkerk R, Niioua T, Rajagopalan B, Radda GK. Low phosphocreatine/ATP
ratio detected in vivo in the failing hypertrophied human myocardium using 31P magnetic resonance
spectroscopy. Lancet 1991;338:973–6.
15 Kloner RA, Przyklenk K. Understanding the jargon: a glossary of terms used (and misused) in the study
of ischaemia and reperfusion. Cardiovasc Res 1993;27:162–6.
16 Depre C, Shipley GL, Chen W, et al. Unloaded heart in vivo replicates fetal gene expression of cardiac
hypertrophy. Nature Med 1998;4:1269–75.
17 Taegtmeyer H. Energy metabolism of the heart: From basic concepts to clinical applications. Curr Prob
Cardiol 1994;19:57–116.
18 Lipmann F. Metabolic generation and utilization of phosphate bond energy. Adv Enzymol 1941;1:99–165.
19 Taegtmeyer H. Cardiac preconditioning does not require myocardial stunning. Ann Thorac Surg
1993;55:400.
20 Taegtmeyer H, Russell RR, Silvestain AL, Shafer D. Depressed function and energy metabolism of hearts
from spontaneously diabetic BB/W rats. J Mol Cell Cardiol 1985;17:44.
21 Balaban RS, Kontor HL, Katz LA, Briggs RW. Relation between work and phosphate metabolite in the in
vivo paced mammalian heart. Science 1986;232:1121–3.
22 Kupriyanov VV, Lakomkin VL, Kapelko VI, Steinschneider AY, Ruuge EK, Saks VA. Dissociation of
adenosine triphosphate levels and contractile function in isovolumic hearts perfused with 2-deoxyglucose.
J Mol Cell Cardiol 1987;19:729–40.
23 Barth E, Stämmler G, Speiser B, Schaper J. Ultrastructural quantitation of mitochondria and myofilaments
in cardiac muscle from 10 different animal species including man. J Mol Cell Cardiol 1992;24:669–81.
24 McNutt NS, Fawcett DW. Myocardial ultrastructure. In: Langer G, Brady A, eds, The mammalian
myocardium, New York: JohnWiley & Sons, 1974:1–49.
25 Baldwin JE, Krebs HA. The evolution of metabolic cycles. Nature 1981;291:381–2.
26 Brown GC. Control of respiration and ATP synthesis in mammalian mitochondria and cells. Biochem J
1992;284:1–13.
Page 22
27 Newsholme EA, Start C. Regulation in Metabolism. London: John Wiley & Sons, 1973: 349 PP.
28 Newsholme EA, Leech AR. Biochemistry for the medical sciences Chichester: John Wiley, 1983:952 PP.
29 Coulson RA, Hernandez T, Herbert JD. Metabolic rate, enzyme kinetics in vivo. Comp Biochem Physiol
1977;56A:251–62.
30 Barry WH, Bridge JHB. Intracellular calcium homeostasis in cardiac myocytes. Circulation
1993;87:1806–15.
31 Fabiato A. Calcium induced release of calcium from the cardiac sarcoplasmic reticulum. Am J Physiol
1983;245:C1-14.
32 McCormack JG, Halestrap AP, Denton RM. Role of calcium ions in reperfusion of mammalian
intramitochondrial metabolism. Physiol Rev 1990;70:391–425.
33 Lehninger AL, Reynafarie B, Vercesi A. Transport and accumulation of calcium in mitochondria. Ann
NYAcad Sci 1978;307:160–76.
34 Denton RM, McCormack JG. On the role of the calcium transport cycle in the heart and other mammalian
mitochondria. FEBS Lett 1980;119:l–8.
35 Rayment I, Holden HM, Whittacker M, et al. Structure of actin–myosin complex and its implications for
muscle contraction. Science 1993;261:58–65.
36 Kornberg HL. Anaplerotic sequences and their role in metabolism. Essays Biochem 1966;2:1–31.
37 Taegtmeyer H, Hems R, Krebs HA. Utilization of energy providing substrates in the isolated working rat
heart. Biochem J 1980;186:701–11.
38 Taegtmeyer H. On the inability of ketone bodies to serve as the only energy providing substrate for rat
heart at physiological work load. Basic Res Cardiol 1983;78:435-50.
39 Russell RR, Taegtmeyer H. Coenzyme A sequestration in rat hearts oxidizing ketone bodies. J Clin Invest
1992;89:968–73.
40 Russell RR, Taegtmeyer H. Changes in citric acid cycle flux and anaplerosis antedate the functional
decline in isolated rat hearts utilizing acetoacetate. J Clin Invest 1991;87:384–90.
41 Russell RR, Taegtmeyer H. Pyruvate carboxylation prevents the decline in contractile function of rat
hearts oxidizing acetoacetate. Am J Physiol 1991;261:H1756-62.
42 Taegtmeyer H. Metabolic responses to cardiac hypoxia:Increased production of succinate by rabbit
papillary muscles. Circ Res 1978;43:808–15.
43 Schwaiger M, Schelbert H, Ellison D, et al. Sustained regional abnormalities in cardiac metabolism after
transient ischemia in the chronic dog model. J Am Coll Cardiol 1985;6:337–47.
44 Schwaiger M, Neese RA, Araujo L, et al. Sustained nonoxidative glucose utilization and depletion of
glycogen in reperfused canine myocardium. J Am Coll Cardiol 1989;13:745–54.
45 Czernin J, Porenta G, Brunken R, et al. Regional blood flow, oxidative metabolism, and glucose
utilization in patients with recent myocardial infarction. Circulation 1993;88:884–95.
46 Racker E. Energy cycles in health and disease. Curr Top Cell Regul 1981;18:361–75.
47 Steenbergen C, Murphy E, Levy L, London RE. Elevation in cytosolic free calcium concentration early in
myocardial ischemia in perfused rat heart. Circ Res 1987;60:700–7.
48 Tani M, Neely JR. Role of intracellular Na+ in Ca2+ overload and depressed recovery of ventricular
function of reperfused ischemic rat hearts. Possible involvement of H+-Na+ and Na+-Ca2+. Circ Res
1989;65:1045–56.
49 Marban E, Kitakaze M, Koretsune Y, Yue DT, Chacko VP, Pike MM. Quantification of [Ca2+]1 in
perfused hearts. Critical evaluation of the 5F-BAPTA and nuclear magnetic response method as applied to
the study of ischemia and reperfusion. Circ Res 1990;66:1255–67.
50 Rao PS, Cohen MV, Mueller HS. Production of free radicals and lipid peroxides in early experimental
myocardial ischemia. J Mol Cell Cardiol 1983;15:713–16.
51 Kloner RA, Przyklenk K, Whittacker P. Deleterious effects of oxygen radicals in ischemia/reperfusion.
Resolved and unresolved issues. Circulation 1989;80:1115–27.
Page 23
52 Ferrari R, Alfieri O, Curello S, et al. Occurrence of oxidative stress during reperfusion in human heart.
Circulation 1990;81:201–11.
53 Bolli R. Mechanism of myocardial "stunning". Circulation 1990;82:723–38.
54 Klein HH, Puschmann S, Schaper J, Schaper W. The mechanism of the tetrazolium reaction in identifying
experimental infarction. Virchow’s Arch (A) 1983;393:287–97.
55 Dennis SC, Gevers W, Opie LH. Protons in ischemia:Where do they come from, where do they go to? J
Mol Cell Cardiol 1991;23:1077–86.
56 Jennings RB, Reimer KA, Steenbergen C. Myocardial ischemia revisited. The osmolar load, membrane
damage, and reperfusion. J Mol Cell Cardiol 1986;18:769–80.
57 Lehninger AL. Biochemistry, 1st edn. New York: Worth Publishers, 1970:1013 PP.
58 Winterstein H. Ueber die Sauerstoffatmung des isolierten Säugetierherzens. Z Allg Physiol 1904;4:333–
59.
59 Rohde E. Über den Einfluss der mechanischen Bedingungen auf die Tätigkeit und den
Sauerstoffverbrauch des Warmblüterherzens. Naunyn-Schmiedeberg’s Arch Ges Exp Path Pharmakol
1912;68:401–10.
60 Evans CL. The effect of glucose on the gaseous metabolism of the isolated mammalian heart. J Physiol
(Lond) 1914;47:407–18.
61 Neely JR, Liebermeister H, Battersby EJ, Morgan HE. Effect of pressure development on oxygen
consumption by isolated rat heart. Am J Physiol 1967;212:804–14.
62 Nguyêñ VTB, Mossberg KA, Tewson TJ, et al. Temporal analysis of myocardial glucose metabolism by
18F–2-deoxy–2-fluoro-D-glucose.
Am J Physiol 1990;259:H 1022-31.
63 Cooney GJ, Taegtmeyer H, Newsholme EA. Tricarboxylic acid cycle flux and enzyme activities in the
isolated working rat heart. Biochem J 1981;200:701–3.
64 Balaban RS. Regulation of oxidative phosphorylation in the mammalian cell. Am J Physiol
1990;258:C377-89.
65 Jones BP, Shan X, Park Y. Coordinated multisite regulation of cellular energy metabolism. Annu Rev Nutr
1992;12:327–43.
66 Evans, G. The glycogen content of the rat heart. J Physiol (Lond) 1934;82:468–80.
67 Denton RM, Randle PJ. Concentrations of glycerides and phospholipids in rat heart and gastrocnemius
muscles. Biochem J 1967;104:416–22.
68 Gevers W. Protein metabolism of the heart. J Mol Cell Cardiol 1984;16:3–32.
69 Lesch M, Taegtmeyer H, Peterson MB, Vernick R. Studies on the mechanism of the inhibition of
myocardial protein synthesis during oxygen deprivation. Am J Physiol 1976;230:120–6.
70 Taegtmeyer H, Lesch M. Altered protein and amino acid metabolism in myocardial hypoxia and
ischemia. Amsterdam: Elsevier/North Holland, 1980:347–60.
71 Morgan HE, Rannels DE, McKee EE. Protein metabolism of the heart. In: Berne, R, ed, Handbook of
physiology: the cardiovascular system: the heart. Washington, DC: American Physiology Society,
1979:845–71.
72 Magid NM, Borer JS, Young MS, Wallerson DC, Demonteiro C. Suppression of protein degradation in
progressive cardiac hypertrophy of chronic aortic regurgitation. Circulation 1993;87:1249–57.
73 Samarel AM. Hemodynamic overloaded the regulation of myofibrillar protein degradation. Circulation
1993;87:1418–20.
74 Taegtmeyer H, Doenst T, Mommessin JI, Guthrie PH, Williams CM. Further evidence for the importance
of anaplerosis in the isolated working rat heart: A tracer kinetic study with [18F] fluoro–2-deoxyglucose
(FDG). Circulation 1993;88:I–284 (Abstract).
75 Rothlin ME, Bing RJ. Extraction and release of individual free fatty acids by the heart and fat deposits. J
Clin Invest 1961 ;40:1380–5.
76 Keul J, Doll E, Keppler D. Energy metabolism of human muscle. Basel: S Karger, 1972:313 PP.
77 Drake AJ, Haines JR, Noble MM. Preferential uptake of lactate by the normal myocardium in dogs.
Cardiovasc Res 1980;14:65–72.
78 Krebs HA, Williamson DH, Bates MW, Page MA, Hawkins RA. The role of ketone bodies in caloric
homeostasis. In: Weber, G, ed, Advances in enzyme regulations, vol. 9. NewYork: Pergamon Press,
1971:387–409.
Page 24
79 Goodwin GW, Ahmad F, Doenst T, Taegtmeyer H. Energy provision from glycogen, glucose and fatty
acids upon adrenergic stimulation of isolated working rat heart. Am J Physiol 1998;274:H1239-47.
80 Goodwin GW, Taylor CS, Taegtmeyer H. Regulation of energy metabolism of the heart during acute
increase in heart work. J Biol Chem 1998;273:29530–9.
81 Opie LH. Cardiac metabolism–emergence, decline, and resurgence. Part I. Cardiovasc Res 1992;26:721–
33.
82 Opie LH. Cardiac metabolism – emergence, decline, and resurgence. Part II. Cardiovasc Res
1992;26:817–30.
83 Gadian DG, Hoult DI, Radda GK, Seeley PJ, Chance B, Barlow C. Phosphorous nuclear magnetic
resonance studies in normoxic and ischemic cardiac tissue. Proc Natl Acad Sci USA 1976;73:291–332.
84 Weiss ES, Hoffman EJ, Phelps ME, et al. External detection and visualization of myocardial ischemia
with 11C substrates in vitro and in vivo. Circ Res 1976;39:24–32.
85 Jacobus WE, Taylor G, Hollis DP, Nunnally RL. Phosphorous nuclear magnetic resonance of perfused
working rat hearts. Nature 1977;265:756–8.
86 Ingwall JS. Phosphorous nuclear magnetic resonance spectroscopy of cardiac and skeletal muscles. Am J
Physiol 1982;242:H729-44.
87 Bottomley PA. Noninvasive study of high energy phosphate metabolism in human heart by depth-
resolved 31P NMR spectroscopy. Science 1985;229:769–72.
88 Schelbert HR. Assessment of myocardial metabolism by PET: A sophisticated dream or clinical reality?
Eur J Nucl Med 1986;12:570–5.
89 McMillin-Wood JB. Biochemical approaches in metabolism: application to positron emission
tomography. Circulation 1985;72:IV145-50.
90 Taegtmeyer H, Mossberg KA, Nguyen VTB. Positron labelled tracers: A window for the assessment of
energy metabolism in heart and skeletal muscle. Acta Radiol 1991;376:40–44.
91 Lewandowski ED. Nuclear magnetic resonance evaluation of metabolic and respiratory support of work
load in intact rabbit hearts. Circ Res 1992;70:576–82.
92 Tillisch J, Brunken R, Marshall R, et al. Prediction of reversibility of cardiac wall motion abnormalities
predicted by positron tomography, 18fluoro-deoxyglucose, and 13NH3. N Engl J Med 1986;314:884–8.
93 Gould KL, Yoshida K, Haynie M, Hess MJ, Mullani NA, Smalling RW. Myocardial metabolism of
fluoro-deoxyglucose compared to cell membrane integrity for the potassium analogue R6–82 for assessing
viability and infarct size in man by PET. J Nucl Med 1991;32:l–9.
94 Yoshida K, Gould KL. Quantitative relation of myocardial infarct size and myocardial viability by
positron emission tomography of left ventricular ejection fraction and 3-year mortality with and without
revascularization. J Am Coll Cardiol 1993;22:984–97.
95 Radda GU. Control, bioenergetics, and adaptation in health and disease: noninvasive biochemistry from
nuclear magnetic resonance. FASEB J 1992;6:3032–8.
96 Bergman RN. Toward physiological understanding of glucose tolerance. Diabetes 1989;38:1512–27.
97 Schelbert H, Henze E, Sochor H. Effects of substrate availability on myocardial 11C palmitate kinetics by
positron emission tomography in normal subjects and patients with ventricular dysfunction. Am Heart J
1986;111:1055–65.
98 Brown MA, Marshall DR, Sobel BE, Bergmann SR. Delineation of myocardial oxygen utilization with
carbon–11 labelled acetate. Circulation 1987;76:687–96.
99 Buxton DB, Schwaiger M, Nguyen NA, Phelps ME, Schelbert HR. Radiolabelled acetate as a tracer of
myocardial tricarboxylic acid cycle flux. Circ Res 1988;63:628–34.
100 Opie LH, Owen P, Thomas M, Samson R. Coronary sinus lactate measurements in assessment of
myocardial ischemia: Comparison with changes in lactate/pyruvate and ß-hydroxybutyrate/acetoacetate
ratios and with release of hydrogen, phosphate, and potassium from the heart. Am J Cardiol 1973;32:295–
305.
101 Schelbert HR. The Heart. In: Ell P, Homan B, eds, Computed emission tomography. Oxford: Oxford
University Press, 1982:91–133.
102 Gertz EW, Wisneski JA, Neese R. Myocardial lactate extraction: Multidetermined metabolic function.
Page 25
103Wisneski JA, Gertz EW, Neese RA, Gruenke LD, Morris DL, Craig JC. Metabolic fate of extracted
glucose in normal human myocardium. J Clin Invest 1985;76:1819–27.
104 Hearse DJ. Myocardial enzyme leakage. J Mol Med 1977;2:185–200.
105 Katz AM, Hecht HH. The early "pump" failure of the ischemic heart. Am J Med 1969;47:497–502.
106Williamson JR, Shaffer SW, Ford C, Safer B. Contribution of tissue acidosis to ischemic injury in the
perfused rat heart. Circulation 1976;53:3–14.
107Neely JR, Grotyohann LW. Role of glycolytic products in damage to myocardium: Dissociation of
adenosine triphosphate levels and recovery of function of reperfused canine myocardium. Circ Res
1984;55:816–24.
108Kübler W, Katz AM. Mechanism of early "pump" failure of the ischemic heart: Possible role of adenosine
triphosphate depletion and inorganic phosphate accumulation. Am J Cardiol 1977;40:467–71.
109Gudbjarnason S, Mathes P, Ravens KG. Functional compartmentation of ATP and creatine phosphate in
heart muscle. J Mol Cell Cardiol 1970;l:325–39.
110Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in
ischemic myocardium. Circulation 1986;74:1124–36.
111 Marber MS. Stress proteins and myocardial protection. Clin Sci 1994;86:375–81.
112Chen TM, Goodwin GW, Guthrie PH, Taegtmeyer H. Effects of insulin on glucose uptake by rat hearts
during and after coronary flow reduction. Am J Physiol 1997;273:H2170-7.
113Young LH, Renfu Y, Russell R, et al. Low-flow ischemia leads to translocation of canine heart GLUT–4
and GLUT–1 glucose transporters to the sarcolemma in vivo. Circulation 1997;95:415–22.
114Neely JR, Morgan HE. Relationship between carbohydrate and lipid metabolism and the energy balance
of heart muscle. Annu Rev Physiol 1974;36:413–39.
115Rogers WJ, Stanley AW, Breing JB, et al. Reduction of hospital mortality rate of acute myocardial
infarction with glucose-insulin-potassium infusion. Am Heart J 1976;92:441–54.
116Rackley CE, Russell RO, Rogers WJ, Papapierto SE. Clinical experience with glucose-insulin-potassium
therapy in acute myocardial infarction. Am Heart J 1981;102:1038–49.
117Whitlow PL, Rogers WJ, Smith LR, et al. Enhancement of left ventricular function by glucose-insulin-
potassium infusion in acute myocardial infarction. Am J Cardiol 1982;49:811–20.
118Medical Research Council Working Party. Potassium, glucose, and insulin treatment for acute myocardial
infarction. Lancet 1968;ii:1355–60.
119Opie LH. The glucose hypothesis:Relation to acute myocardial ischemia. J Mol Cell Cardiol 1970;1:
107–15.
120Hearse DJ, Chain EB. The role of glucose in the survival and "recovery" of the anoxic isolated perfused
rat heart. Biochem J 1972;128:1125–33.
121Opie LH, Bruyneel K, Owen P. Effects of glucose, insulin, potassium infusion and tissue metabolic
changes within first hour of myocardial infarction in the baboon. Circulation 1975;52:49–57.
122Apstein CS, Gravino FN, Haudenschild CC. Determinants of a protective effect of glucose and insulin on
the ischemic myocardium. Effects on contractile function, diastolic compliance, metabolism, and
ultrastructure during ischemia and reperfusion. Circ Res 1983;52:515–26.
123McElroy DD, Walker WE, Taegtmeyer H. Glycogen loading improves left ventricular function of the
rabbit heart after hypothermic ischemic arrest. J Appl Cardiol 1989;4:455–65.
124Lagerstrom CF, Walker WE, Taegtmeyer H. Failure of glycogen depletion to improve left ventricular
function of the rabbit heart after hypothermic ischemic arrest. Circ Res 1988;63:81–6.
125Schneider CA, Nguyêñ VTB, Taegtmeyer H. Feeding and fasting determine postischemic glucose
utilization in isolated working rat hearts. Am J Physiol 1991;260:H542-8.
126Schneider CA, Taegtmeyer H. Fasting in vivo delays myocardial cell damage after brief periods of
ischemia in the isolated working rat heart. Circ Res 1991;68:1045–50.
Page 26
127Gradinak S, Coleman GM, Taegtmeyer H, Sweeney MS, Frazier OH. Improved cardiac function with
glucose-insulin-potassium after coronary bypass surgery. Ann Thorac Surg 1989;48:484–9.
128Fath-Ordoubadi F, Beatt KJ. Glucose-insulin-potassium therapy for treatment of acute myocardial
infarction. An overview of randomized placebo-controlled trials. Circulation 1997;96:1152–6.
129Apstein CS, Taegtmeyer H. Glucose-insulin-potassium in acute myocardial infarction The time has come
for a large prospective trial. Circulation 1997;96:1074–7.
130 Taegtmeyer H. Metabolic support for the postischaemic heart. Lancet 1995;345:1552–5
Page 27
2: Ventricular performance
KARL SKARVAN
Preservation of optimal cardiovascular function represents one of the foremost goals of

anaesthetic management during the perioperative period. Only rarely does the heart work
under such challenging and rapidly varying conditions as during this time. Acute changes in
ventricular loading, gas exchange, intrathoracic pressure, autonomic nervous tone, and
blood properties, as well as the effects of anaesthetic and other drugs, and various released
mediators, all put a formidable stress on the heart. This may give rise to perioperative
cardiac morbidity even in patients with normal hearts, although patients with limited
cardiac reserve are, of course, at much higher risk and require a great deal of the
anaesthetist’s attention. Hence, a proper understanding of ventricular function is a
prerequisite to correct evaluation and optimisation of cardiac function in the perioperative
period. With regard to the predominant role of the left ventricle in haemodynamic function,
this chapter focuses on the normal performance of the left ventricle and its determinants.
Ventricle as a muscle
The fundamental properties of the myocardium have been thoroughly studied in isolated
animal and human heart muscle preparations. The resting length of the unstressed muscle
strip represents the starting point for the following considerations (Fig 2.1). The resting
length can be increased by attaching a small weight to one end of the muscle strip. This
weight will stretch the muscle to a longer resting length in proportion to the attached
weight, as well as in proportion to the elastic properties of the muscle. This distending force
or weight (expressed in grams) is called the preload of the muscle. The preload, which is
related to the resting length of the sarcomeres of the myocardium, has an important impact
on the next contraction of the muscle.1 When the muscle is prevented from shortening
(isometric contraction), the active force developed during contraction is directly
proportional to its preload. When the muscle is allowed to shorten (isotonic contraction),
both the extent and velocity of shortening will increase in proportion to the preload. This
dependence of contraction
Page 28
Fig 2.1 Force–length–velocity relationships of an isolated strip of myocardium. (a) Passive force builds up
with increasing resting length; (b) active developed force increases with increasing resting length; (c)
shortening velocity decreases with increasing load; (d) extent of shortening decreases with increasing load.
Arrows indicate alterations caused by an increase in myocardial contractility.
characteristics on resting muscle length (preload) is a fundamental property of the

myocardium and is known as the force–length relationship.1 2 The ability of the
myocardium to develop progressively more force with increasing sarcomere length has also
been termed ‘‘length dependent activation” and appears to be related to an increase in the
number of active actin–myosin cross bridges, increased sensitivity to intracellular calcium,
and increased calcium release from sarcoplasmic reticulum.3
A second weight can be attached to the moving end of the isotonically contracting
muscle strip. This additional weight is engaged only during contraction when it is lifted by
the shortening muscle, and hence ensures a constant tension in the muscle during its
shortening. This second weight
Page 29
represents the afterload of the muscle. Similar to preload, afterload also has an important
influence on muscle contraction. Both the extent of muscle shortening and that of
shortening velocity are inversely related to afterload. Thus, an increase in afterload
decreases, whereas a reduction in afterload increases, the extent and velocity of shortening.1
The dependence of contractile performance on the shortening load is the second
fundamental property of the myocardium.4–6 The reduced shortening resulting from
increased afterload can be reversed up to a given limit by an appropriate augmentation of
the preload.7
When the preload and afterload are held constant, the developed tension (in isometric
contraction) or extent and velocity of shortening (in isotonic contraction) can be increased
by increasing the inotropic state of the muscle, for example, by adding calcium and
consequently increasing myocardial contractility.1 Thus, the contractile behaviour of the
isolated heart muscle can be exhaustively described within the framework of a force–length
relationship. It is determined by the interplay of preload, afterload, and contractility. This
physiological concept is also most useful for the understanding of the function of the intact
ventricle. Its too simplistic application to the human cardiovascular system may, however,
be misleading.
Isolated ventricle
The principles determining contractile behaviour of an isolated strip of myocardium can
also be applied to the intact isolated and perfused ventricle. The ventricle can contract
either isometrically against an infinite outflow resistance (aortic clamp) or isotonically
against a variable resistance. The first situation is called isovolumic contraction. The resting
muscle fibre length can be increased or decreased by changing the diastolic filling, and
consequently the volume of blood present in the ventricle at the end of diastole. In
isovolumic preparations, the volume of a balloon positioned in the ventricular cavity can
easily be changed. The passive distension of the ventricle induced by the increased filling
volume leads to a progressive increase in passive tension in the ventricular wall opposing
distension. Depending on the elastic properties of the myocardium and the ventricular
chamber the pressure within the cavity will increase with the increasing volume. The
relationship between resting pressure (P) and volume (V) is known as a diastolic P/V
relationship, and describes the elasticity of the relaxed ventricle. During contraction, force
is generated in the ventricular wall and transferred to the blood contained in the cavity,
causing intraventricular pressure to rise (Fig 2.2). If the ventricle were severed in two equal
parts along an imaginary plane, the intracavitary pressure would immediately tear both
halves apart. Consequently, a force
Page 30
Fig 2.2 Passive (diastolic) and active (systolic) pressure–volume relationships of the ventricle. Curves
illustrate filling and contractile behaviour of the ventricle for a given inotropic state.
of equal dimension but opposite direction must be operational in the virtual dividing plane
of the ventricle that holds both parts together. On the basis of this assumption,
mathematical models were developed that allow net wall forces to be calculated.4 5
Depending on the orientation of the imaginary plane, circumferential, meridional, and
radial forces can be estimated. When compared with the strip of myocardium in the isolated
ventricle, the resting volume corresponds to the resting fibre length, and the developed
pressure (or calculated wall force) corresponds to the directly measured force of the isolated
muscle.
During isovolumic contraction, the developed ventricular pressure and wall force are
directly proportional to the filling volume of the ventricle.1 4 8 At a given volume, the
maximal developed pressure will increase with positive inotropic stimulation. The
ventricle, which is allowed to eject, also contracts under isovolumic conditions until the
aortic (or pulmonic) valve opens. During this isovolumic contraction, the pressure increases
while the ventricular volume remains unchanged, although the fibre length may change
because the ventricle changes its geometry and assumes a more spherical shape. After the
opening of the semilunar valve, myocardial fibres begin to shorten and volume (stroke
volume) is expelled from the ventricle into the aorta or pulmonary artery. The fibre
shortening (and ejection) is terminated when the maximal wall force that can be sustained
by the myocardial fibres at the given level of contractility has been generated. This
Page 31
force is defined by the systolic or active P/V relationship of the isovolumically contracting
ventricle and is independent of preload. Thus, the ventricle operates within the boundaries
determined by the passive diastolic and active systolic P/V relationship.9 10 Similar to an
isolated muscle and isovolumic preparation, the contractile behaviour of the ejecting
ventricle is also controlled by resting fibre length (preload), instantaneous wall force
(afterload), and contractile state.1 5 The wall force during ejection is a function of
ventricular size and geometry, and of developed pressure. As ventricular size decreases in
the course of ejection, the instantaneous wall force and hence the ventricular afterload
decrease. Thus, a normally contracting ventricle unloads itself towards the end of ejection.4
Ventricle in situ
In clinical terms, the functions of the left and right ventricle are also described in terms
of pressure and volume. Both variables are used for the construction of the pressure–
volume diagram, which is an analogue of the force–length diagram of the strip of
myocardium. In the P/V diagram, the volume plotted on the x axis is related to the
myocardial fibre length, and the pressure plotted on the y axis corresponds to the generated
force.9 During one cardiac cycle of the ejecting heart, a P/V loop is inscribed (Fig 2.3). The
cycle starts at end diastole, characterised by end diastolic volume and the corresponding
end diastolic pressure (Fig 2.3, point 1). During isovolumic contraction, pressure increases
whereas volume remains constant until the aortic valve opens (Fig 2.3, point 2) and ejection
begins. Ejection continues until the end systolic P/V relationship line is reached (Fig 2.3,
point 3). Ventricular relaxation causes the pressure to fall and the aortic valve to close. The
intraventricular pressure falls further without changes in volume during the following
isovolumic relaxation period. When the ventricular pressure falls below the atrial pressure,
the mitral valve opens (Fig 2.3, point 4) and ventricular filling starts.9–11 In the context of
the P/V diagram, the effects of the three major determinants of the ventricular function
(preload, afterload, and contractile state) can be illustrated.
Preload
An increase in end diastolic volume shifts the starting point 4 of the loop to the right
along the passive P/V relationship, whereas the end systolic point remains unchanged. This
results in an increase in ejected volume (stroke volume). A decrease in end diastolic volume
shifts the loop leftwards and, because the end systolic P/V relationship line remains
constant, this results in a decrease in stroke volume. This dependence of the
Page 32
Fig 2.3 Left ventricular pressure–volume relationship. Points 1 to 4 demarcate the pressure–volume loop that
is explained in the text. ESPVR (end systolic pressure–volume relationship) and EDPVR (end diastolic
pressure–volume relationship) are curvilinear, but on this diagram, for simplification, they are approximated
by straight lines.
ventricular performance on preload is an expression of the Frank–Starling law of the heart

(see page 53).
Afterload
An increase in ejected stroke volume can also be achieved by a reduction of the
afterload. A reduction in the force opposing ejection and, consequently, in the force
developed by the contractile fibres (wall force) will result in a downward shift of the end
systolic point of the loop along the active and systolic P/V relationship, and in an increase
in stroke volume.12 This phenomenon is the basis of the afterload reduction, a therapeutic
principle applied to a failing ventricle. In contrast, an increase in afterload causes an
upward shift of the end systolic point. As a consequence, the ventricle will not be able to
empty completely and the stroke volume will decrease. A normal left ventricle responds to
an acute increase in afterload by an immediate increase in end diastolic volume, which
allows restoration of the stroke volume. The so called homoeometric regulation or Anrep
effect will subsequently restore the end diastolic volume, in spite of the persisting higher
afterload, by adjusting the intrinsic myocardial contractility.13 More recent studies have
shown that, during acute elevations in afterload, the end systolic P/V relationship does not
remain constant but
Page 33
shifts to the left, a change compatible with enhanced inotropic state.14 A failing ventricle is
deprived of these compensatory mechanisms and, consequently, becomes exquisitely
sensitive to any increase in afterload.
Contractility
The third major determinant of ventricular function is the contractile state of the
myocardium. A change in myocardial contractility will breech the confines of the diastolic
and end systolic P/V relationship. With an increase in contractility, the end systolic points
of the P/V loop will shift upwards and to the left of the original active end systolic P/V
relationship; consequently, stroke volume will increase despite unchanged preload and
afterload. The new end systolic P/V relationship, which now determines the extent of
ejection, is shifted to the left and its slope is steeper. Thus, the increase in contractility by
positive inotropic stimulation causes the ventricle to empty more completely to a smaller
end systolic volume.9–11 With limited ventricular filling (for example, during hypovolaemia
and reduced venous return), the end diastolic point of the P/V loop will also shift to the left,
neutralising the increase in stroke volume. Under such circumstances, the positive inotropic
stimulation will become evident by an increase in pressure rather than by an increase in
stroke volume. Some positive inotropic agents may also exert vasodilating and afterload
reducing effects, which further enhance the ventricular performance provided that a
decrease in preload can be prevented. In addition, inhibitors of the enzyme
phosphodiesterase exhibit a positive lusitropic effect. This effect, by virtue of improving
ventricular relaxation, shifts the diastolic P/V relationship downwards, which facilitates
ventricular filling and further enhances ventricular ejection.
Thus, P/V diagrams help the anaesthetist to have a better understanding of the changes in
ventricular function under rapidly changing conditions of ventricular loading and
contractility, and to predict the effects of his or her therapeutic interventions.
The heart as a pump

Functioning as a pump, the ventricle generates pressure and displaces volume. The
fundamental mechanical properties of the ventricular pump are elasticity, resistance, and
inertance.4 12 Elasticity reflects the rate independent relationship between pressure and
volume in the ventricle and determines the volume displaced from the ventricle during
systole (stroke volume). Ventricular resistance is related to viscous properties of the
ventricular pump that is operational during ejection, and is rate dependent because it
changes with flow velocity. Finally, inertance describes the force required to accelerate the
mass of the ventricle and the blood contained
Page 34
Fig 2.4 Pump function curve of the left ventricle. Ventricular pump output decreases with increasing
developed pressure and eventually ceases when the pressure necessary to generate flow exceeds the capacity
of the ventricular pump. Arrows indicate the effects of increased ventricular filling (left), contractility
(middle), and heart rate (right).14
within its cavity at end diastole. These three properties of the ventricular pump determine
the amount of pressure generated. In an isovolumically contracting ventricle, the maximum
pressure is determined solely by the elasticity. In contrast, the ejecting ventricle develops
less pressure because part of its contractile energy is used to overcome the resistive and
inertial forces.4
The pump function of the ventricle can be described by means of a pump function graph
plotting pressure against flow (Fig 2.4). The graph shows an inverse curvilinear parabolic
relationship between mean left ventricular pressure and flow. Increased ventricular filling
(preload) shifts the curve upwards and to the right; an increase in heart rate has the same
effect. An increase in contractility causes the curve to rotate clockwise around the flow axis
intercept.15 Power output of the ventricular pump can be calculated as the product of mean
ventricular pressure and flow. Knowing the myocardial oxygen consumption, which
represents the power input, the efficiency of the ventricular pump can be calculated. On the
basis of this pump model applied to a cat heart in situ, an efficiency of 20% was
estimated.16 It was also shown that the left ventricle tends to work at maximal power
output, while minimising its volume. As maximal power is highly preload dependent, it
must be normalised for end diastolic volume before it can be used as an index of
myocardial contractility.17
Clinical evaluation of global ventricular pump function

In the perioperative period, the blood pressure resulting from the interaction between the
left ventricle and the systemic vasculature often represents the only monitored index of
ventricular performance. Never-
Page 35
theless, when normal values of blood pressure are accompanied by normal heart rate and
clinical signs of an adequate peripheral circulation (pulse quality, capillary refill time, skin
temperature, urine output, pH), it can be assumed that ventricular performance is
satisfactory. If more information on the cardiovascular function is required, intracardiac
pressures and flows18 as well as blood gases must be measured (Table 2.1).
Cardiac output can be measured invasively (contrast ventriculography, Fick’s method,
thermodilution) or non-invasively by rebreathing, ultra sonography, or impedance
techniques. The output or work data obtained must be evaluated against the actual
pulmonary capillary pressure (Frank–Starling relationship) and the adequacy of O2 supply
to the tissues as reflected by mixed venous O2 tension or saturation. The most commonly
used index of global left ventricular function is the ejection fraction (EF):19
It can be measured non-invasively by radionuclide ventriculography or echocardiography.

During anaesthesia, transoesophageal echocardiography allows continuous monitoring of
the left ventricular cross sectional area and calculation of left ventricular fractional area
change (FAC):
Table 2.1 Clinical assessment of left ventricular performance18 113
Assessment Normal values
Symptoms
Signs
Haemodynamics
Heart rate 50–90 beats/min
Systolic/diastolic arterial pressure
Mean arterial pressure 70–105 mm Hg

Left ventricular end diastolic pressure 4–15 mm Hg
Pulmonary capillary wedge pressure 5–15 mm Hg
Left ventricular end diastolic volume 70 ml/m2
index
Left ventricular end systolic volume index 25 ml/m2
Cardiac index 2·6–4·2 l/min per m2
Left ventricular stroke volume index 45–50 ml/m2
Left ventricular ejection fraction 55–68%
Left ventricular mean systolic ejection rate 160 ml/s per m2
Left ventricular stroke work index 40–60 g·m/m2
Left ventricular stroke power index 170 g·m/m2 per s
Oxygen transport parameters
O2 consumption 110–150 ml/min per m2
Arteriovenous O2 difference 40–55 ml/1
O2 delivery 400–800 ml/min per m2
Mixed venous O2 saturation 65–75%
Mixed venous O2 partial pressure 40–50 mm Hg
Page 36
This has been shown to correlate closely with simultaneously measured radionuclide
ejection fraction.20–22 The monitoring of FAC became easier after the introduction of
automated border detection (Fig 2.5). The ejection fraction (or FAC) is the most useful and
widely used index of global pump function. As an index of left ventricular contractility,
however, it has the flaw of being load dependent. On the other hand, this dependency can
guide the anaesthetist towards optimising the left ventricular function by primarily
adjusting the left ventricular preload and afterload.
Determinants of ventricular function
What appears to be simple and easily understandable in an isolated papillary muscle

preparation becomes a complex and sometimes controversial issue in the clinic. On the one
hand, the indiscriminate use of the
Fig 2.5 Automated border detection. Top: two dimensional, echocardiographic, short axis view of the left
ventricle with left ventricular cavity area encircled by endocardial borders and the line drawn around the
region of interest. Bottom: electrocardiogram and left ventricular area waveform—EDA (end diastolic area),
ESA (end systolic area), and FAC (fractional area change).
Page 37
physiologically well defined terms, such as preload, afterload, and contractility, by
clinicians occasionally gives rise to ciriticism and confusion; on the other, many useful
concepts of perioperative haemodynamic management are based on the interplay of these
major determinants of ventricular function. Therefore, a detailed review of these
determinants and the methods used for their assessment is warranted.
Preload
As already stated, preload is the force that stretches the resting myocardium and
determines the resting length of its contractile fibres. In the intact ventricle, the fibre length
cannot be measured. However, the end diastolic volume of the ventricle is proportional to
the resting length of its contractile elements.1 4 6 9 The measurement of end diastolic volume
in patients undergoing surgery is difficult. Most data on left ventricular volume are based
on contrast ventriculography, which requires radiography equipment, left heart
catheterisation, and injection of contrast medium; it cannot be used for serial
measurements. The calculated volumes depend on ventricular geometry, model and the
mathematical formula used, and accurate identification of the endocardial borders.
Inclusion of the volume of papillary muscles and trabeculae into the ventricular volume
may lead to an overestimation of the volume. Furthermore, only a limited number of beats
can be evaluated. Nevertheless, results obtained from ventriculography represent the
standard to which the results from other less invasive and non-invasive methods have to be
compared. Recently, ultrafast computed tomography (CT) and magnetic resonance imaging
(MRI) have also been used for measurement of ventricular volumes.23 Neither of these
methods, although accurate, can be of use in the perioperative setting.
Equilibrium and first pass radionuclide ventriculography have been used in intensive care
units. Although accurate with regard to determination of ejection fraction, radionuclide
ventriculography can only approximate the absolute ventricular volume using either a
geometric or a count based method. Moreover, the radionuclide methods are not suitable for
intra-operative use. Echocardiography, in contrast, is much easier to use even in the
operating room and allows measurements of ventricular volumes. A real breakthrough was
the introduction of transoesophageal echocardiography, which permits continuous
monitoring of ventricular size and function without interfering with the surgical field.
Several studies have shown a good correlation between left ventricular volumes assessed by
transoesophageal echocardiography and those obtained by contrast or radionuclide
ventriculography or transthoracic echocardiography.20 22 24 25 In spite of the good
correlations found, transoesophageal echocardiography appears to underestimate absolute
left ventricular volumes systematically. This results, in part, from foreshortening of the
ventricular cavity in its long axis.
Page 38
For perioperative online estimation of left ventricular filling volume, however, the
calculation of end diastolic volume is not necessary. The area of the cross sectional view of
the left ventricle in the transgastric short axis view has been shown to correlate reasonably
well to left ventricular volume. The automated border detection method traces the changes
of the left ventricular area throughout the cardiac cycle and displays, in real time, the end
diastolic and end systolic area values together with the FAC (see Fig 2.5). The FAC
represents the echocardiographic analogue of the radio-nuclide ejection fraction.26 27 The
most recent development, three dimensional echocardiography, has not yet been widely
used in the perioperative setting.
Findings of a small ventricular size at end diastole, partial or complete obliteration of left
ventricular cavity at end systole, and the ‘‘kissing papillary muscles” sign, all of which can
be assessed usually by the naked eye, are most useful in the clinical online diagnosis of
inadequate filling volume (preload). Furthermore, the dimensional data obtained by trans
oesophageal echocardiography (cavity radius and wall thickness), together with measured
filling pressure (wedge pressure, pulmonary diastolic pressure, or mean left atrial pressure)
allow calculation of the end diastolic wall stress, which may correlate even better to the end
diastolic fibre length than the end diastolic volume.28
Without the possibility of monitoring left ventricular size, the filling pressure, usually
measured as the mean pulmonary capillary wedge pressure by using a pulmonary artery
catheter, remains the only means of estimating left ventricular filling volume during the
perioperative period. The limitations of the wedge pressure as a measure of preload must,
however, be kept in mind. The relationship between pressure and volume in diastole is
curvilinear and changes when the ventricle moves along the P/V relationship.9 10 Moreover,
acute changes in ventricular stiffness may also substantially alter the relationship between
filling pressure and volume (see section on “Diastolic performance”). Apart from
monitoring ventricular size, echocardiography also provides additional information on left
ventricular filling based on Doppler measurements of blood flow velocity across the mitral
valve and in the pulmonary veins.
Afterload
It is difficult to apply the concept of ventricular afterload defined in an isolated heart
muscle preparation to the patient’s heart in situ. In heart muscle in vitro, the afterload
remains constant during shortening and can be described by a single value of force. In the
intact ventricle in situ, afterload represents the sum of forces opposing the shortening of
myocardial fibres and the ejection of blood during systole. It is not constant during ejection
and depends on the complex interplay of factors both internal and external to the
myocardium. The internal factor (muscle load)
Page 39
refers to the instantaneous force or tension within the ventricular wall, which is related to
the size and shape of the chamber, and the pressure within it according to the law of
Laplace.4 6 12 The external factor (arterial load) refers to the physical properties of the
arterial system that the ventricle encounters during ejection.29
Another, although similar, concept of systolic load describes the total systolic load as
comprising intrinsic and extrinsic components: the intrinsic component corresponds to
intraventricular pressure gradients whereas the extrinsic component is related to the aortic
root pressure waveform resulting from the interaction between ventricular ejection and
aortic input impedance.30 This total systolic ventricular load and the changing ventricular
geometry determine the total muscle load that can be expressed in terms of systolic wall
stresses. These stresses, in turn, determine the contractile behaviour of the ventricle within
the framework of the force–velocity–length relationship. The novel aspect of this concept
of the left ventricular systolic load is the incorporation of transient intraventricular pressure
gradients related to both the blood inertia and impulsive flows in the early ejection and the
convective flow acceleration in the left ventricular outflow tract at peak ejection.
Instantaneous high fidelity measurements of pressure and flow did indeed demonstrate
significant inhomogeneity of intraventricular pressure during ejection with local gradients
up to 8 mm Hg at rest and 16 mm Hg at exercise.30 These pressure gradients increase with
diminishing chamber volume as a result of underfilling and/or concentric hypertrophy, as
well as with positive inotropic stimulation. These findings are the basis for understanding
the phenomenon of dynamic intraventricular obstruction associated with hypovolaemia and
high adrenergic tone, which occasionally develops in the perioperative period. The intrinsic
and extrinsic loads are complementary and competitive. Thus, a reduction in the extrinsic
load by a vasodilator can be offset by a compensatory increase in the intrinsic component of
the total load.
How can left ventricular afterload be assessed in the clinic? All of the following have
been proposed as a measure of left ventricular afterload:
z arterial and ventricular systolic pressures

z peripheral vascular resistance
z aortic input impedance
z systolic wall stresses
z effective arterial elastance.
The meaning and the clinical use of each of these parameters are discussed below.
Analogue parameters are valid for the right ventricle and the pulmonary vascular bed.
Arterial and ventricular systolic pressures The peripheral arterial pressure is an unreliable
index of left ventricular afterload. The arterial pressure
Page 40
waveform results from a complex interaction between ventricular and vascular factors.
Arterial pressure can remain unchanged in spite of relevant changes in the systolic load. Of
all the ventricular pressures that can be measured (instantaneous, peak, end systolic, and
mean), the mean left ventricular pressure reflects the afterload best.31 The left ventricular
pressure measurement is presently limited to the heart catheterisation laboratory.
Peripheral vascular resistance The peripheral resistance is calculated as the ratio of mean
arterial pressure divided by flow (cardiac output) and is inversely related to the fourth
power of the radius of the vascular bed. In a steady flow system, the peripheral resistance
indeed represents the total load imposed on the ventricular pump. In the presence of
intermittent ejection of blood into a distensible reservoir and of pulsatile blood flow,
however, the total external load also includes elastic properties of the arterial system and
wave reflections.29 31
Aortic input impedance The pulsatile characteristic of the cardiac pump flow is taken into
account when aortic input impedance is calculated as a measure of the total ventricular
load.31 32 In general, impedance is a measure of opposition to flow present within a
conducting system. In contrast to peripheral resistance, input impedance incorporates the
oscillatory motion of blood. In the arterial system, oscillatory waveforms are superimposed
on a mean non-pulsatile component, and the total opposition to flow is a sum of both. The
aortic impedance relates the oscillatory or sinusoidal pressure in the aorta to the oscillatory
flow and is affected by distensibility (compliance) of the arterial system and inertia of the
blood. The impedance modulus is calculated as the ratio of oscillatory pressure and flow,
and displayed graphically in the frequency domain by plotting the impedance moduli
against their respective frequencies. (Fig 2.6). The frequencies are expressed as multiples of
the frequency of the original pressure and flow waveform. In this input impedance
spectrum, the modulus of zero frequency (obtained by dividing mean pressure by mean
flow) represents the peripheral resistance. To calculate the other moduli, the original
pressure and flow waveforms have to be reduced to a finite number of constituent sine
waves (harmonics) by Fourier analysis. The oscillatory component of the load is expressed
as characteristic impedance Z0 (units dyn·s/cm5) and is the arithmetic mean of the
impedance moduli above the frequency of 2 Hz.31 33 Aortic impedance moduli oscillate
around the characteristic impedance line as a result of reflected waves (Fig 2.6). A decrease
in resistance or an increase in aortic compliance will decrease the amplitude of reflected
waves and moduli oscillation, whereas an increase in resistance or a decrease in aortic
compliance will have the opoposite effect. The reflected waves exert a prominent influence
on impedance moduli, particularly at low frequencies, and represent an additional
component of ventricular afterload.34 The
Page 41
overall contribution of the frequency dependent oscillatory impedance to flow in the normal
systemic circulation is, however, only approximately 10% of the total external load. As a
result of anatomical differences, it is significantly higher in the pulmonary vascular bed.
The oscillatory analysis can be completed by calculation of the phase of the harmonics
that fluctuates between +1 and -1. A negative phase value indicates that the flow harmonic
leads the pressure harmonic whereas a positive phase indicates that the flow harmonic lags
behind the pressure harmonic. The phase is less than 0 at low frequencies and crosses 0
close to the frequency of the minimal impedance modulus.
In a study in normal subjects, the impedance modulus fell from the high level at zero
(units dyn·s/cm5) frequency (peripheral resistance) to a minimum at 4 Hz and slightly
increased thereafter. The characteristic impedance was 90 (34) dyn·s/cm5 where this
represents the mean (SD), which corresponded to 8% of the total peripheral resistance.33 In
elegant
Fig 2.6 Aortic input impedance spectrum. The impedance moduli are plotted against their respective
frequencies. The impedance modulus at zero frequency represents the peripheral vascular resistance. The
impedance minimum for this example is 4 Hz.
Page 42
studies, the aortic input impedance was shown to be a major determinant of left ventricular
load and function. An increase in characteristic impedance with constant resistance clearly
reduced the extent and velocity of shortening and decreased cardiac output.32 35 However
attractive from a physiological point of view, the need for continuous pressure and flow
measurement, as well as its cumbersome calculation, renders aortic input impedance
unsuitable for assessment of ventricular afterload under clinical conditions.
Systolic wall stresses The measurement of ventricular size and wall thickness by
echocardiography allows assessment of the forces acting across the ventricular wall during
systole as systolic wall stress.28 36 Stress is defined as the force acting on a surface divided
by the cross sectional area over which the force acts. The total stress can usually be reduced
to component stresses acting perpendicular or parallel to the surface. The units of stress are
dyn/cm2 or g/cm2. The theoretical calculation of ventricular wall stress relies on complex
mathematical models based on Laplace’s law. In practice, assumptions for simplifications
are inevitable. The left ventricular geometry is usually approximated by a sphere or a
prolate ellipsoid, and the ellipsoid has a stress distribution between that of a sphere and that
of a cylinder. In the human ventricle, stress distribution at the equator more closely
resembles that of a cylinder. Consequently, the models based on an ellipsoid or a sphere
tend to underestimate the circumferential and overestimate the longitudinal (meridional)
stress. According to Laplace’s law, the force in the wall of a hollow, thin walled structure is
proportional to the transmural pressure and the principal radii of curvature. In a thin walled
structure, the bending and radial stresses can be neglected. The meridional and
circumferential stresses in the wall of a prolate, thin walled ellipsoid can be calculated
when systolic pressure, corresponding radii of curvature, and wall thickness are known.36
The basic assumption of the thin walled models is that the wall is thin relative to the
cavity diameter. In an attempt to obtain a better approximation to the real stresses and to
study the stress distribution across the wall, thick walled ventricular models were
developed. The earlier models, which assumed isotropic (homogeneous) properties of the
myocardium, showed that the stresses are apparently higher in the inner layers of the
ventricular wall than in the outer layers. Newer models incorporating anisotropic properties
of the myocardium, cylindrical geometry, and fibre direction revealed uniform distribution
of stress across the ventricular wall.37 The mechanism keeping the transmural distribution
of the stress uniform is related to the interplay of the field of deformation and regional fibre
orientation. When change in ventricular volume induces transmural gradients in fibre strain
(relative elongation), an appropriate amount of torsion redistributes the fibre strain from the
inner to the outer layers and equalises wall stresses. This uniformity of the mechanical load
in the
Page 43
myocardium is not restricted to the left ventricular wall alone, but is found also in the
papillary muscles and the free right ventricular wall.37
In animal experiments, systolic wall stress was compared with the aortic input impedance
with regard to the determination of left ventricular afterload. It was found that alterations in
ventricular stress more accurately predicted alterations in ventricular shortening than the
aortic input impedance. Thus, systolic wall stress, which represents the internal load
imposed on the contracting myocardium, also appears to reflect alterations in the external
load reliably.35
In echocardiographic studies, the meridional end systolic wall stress is usually calculated
according to the angiographically validated formula as follows:
where P is end systolic pressure, D is the end systolic left ventricular diameter (dyn/cm2),
and WT the end systolic wall thickness.24 Peak systolic and mean systolic wall stresses can
also be calculated.
Although it is rather difficult to calculate the systolic wall stress in the perioperative
setting, knowledge of at least directional changes of left ventricular afterload may help to
evaluate the effects of therapeutic interventions and to optimise left ventricular function.
For this purpose, the directional changes of the simple ratio [P×D]/[WT] obtained by echo-
cardiography can give us useful information on changes in left ventricular afterload. In the
absence of major alterations in arterial pressure, even the end systolic size (diameter or
cross sectional area) of the left ventricle alone will help to estimate online the directional
changes in afterload.
Effective arterial elastance Another measure of left ventricular afterload has been
introduced with the concept of the time varying elastance (see “Ventriculoarterial
coupling”).The effective arterial elastance (Ea) relates end systolic arterial pressure to
stroke volume. The concept of arterial elastance is a reflection of pressure that will be
generated when the stroke volume is ejected into the arterial system. The greater the stroke
volume ejected and the greater the opposition of the arterial system to the ejection, the
greater the pressure increase in the arterial system. By plotting the developed pressures at
end systole against varying stroke volumes, the arterial P/V relationship can be constructed.
The term for the slope of this linear relationship is “effective arterial elastance” (Ea), and it
is used in the end systolic P/V diagram as a measure of afterload to analyse the coupling of
the left ventricle with the arterial system.38
Contractility
As previously mentioned, contractility is one of the fundamental properties of the heart
muscle manifesting itself through the velocity and
Page 44
extent of shortening. To ascribe an observed change in shortening to a change in
contractility, the other determinants of pump function, that is, preload, afterload, and heart
rate, must be held constant. This requirement disqualifies the ejection phase parameters
such as stroke volume, ejection fraction, or the first derivative of the left ventricular
pressure, dP/dtmax, as reliable measures of myocardial contractility.39 The long search for a
quantitative and reliable single index of contractility (see box) will probably be futile
because it is most difficult to separate contractility from external factors and their feedback
interactions with internal factors.18 Future research will have to use a multivariate approach
to this complex and salient issue.
Currently, the preferred way of assessing myocardial contractility in human studies is the
determination of time varying elastance or of preload recruitable stroke work. Time varying
elastance can be calculated from the instantaneous P/V relationship38 40 41 (Fig 2.7). It
represents the slope of the P/V relationship and increases towards the end of systole to a
maximum: therefore, maximal elastance (Emax) or end systolic elastance (Ees) is an index
of myocardial contractility. In isolated ejecting ventricles, Emax and Ees are almost identical
over a wide range of varying preloads. When afterload is altered, they may differ
considerably. This is frequently observed in vivo
Indices of ventricular contractility
Pre-ejection phase parameters

Maximum rate of change of ventricular pressure dP/dtmax
(dP/dtmax)/IP (IP = developed pressure at dP/dtmax minus end diastolic
pressure
Left ventricular (dP/dtmax)/end diastolic pressure
Pre-ejection period (PEP)
Ejection phase parameters

Ejection fraction (EF%)
Velocity of circumferential fibre shortening (Vcf)
Ejection time (ET), PEP/ET
Peak systolic flow velocity, flow acceleration, time to peak flow (Doppler
techniques)
Methods minimising load dependence

End systolic pressure/volume relationship (left ventricular elastance Emax)
End systolic wall stress/volume relationship
Preload recruitable stroke work (SW/EDV relationship)
(dP/dtmax)/end diastolic volume relationship
Maximal ventricular power
Velocity of circumferential fibre shortening/end systolic wall stress
relationship
Ejection fraction/end systolic wall stress relationship
Page 45
and is probably caused by resistive and inertial influences on the end systolic pressures and
by the changes in time required to reach end systole as a function of load.42 In such a case,
Emax is steeper and does not fall on the end systolic (left upper corner) points of the P/V
relationship. As Ees can be determined more easily, it represents the preferred index of
myocardial contractility. To construct the P/V relationship of the left ventricle, several P/V
loops must be recorded starting at different (usually decreasing) end diastolic volumes. To
construct an end systolic P/V relationship in an isolated heart, the preload can be varied by
changing the volume of saline in the balloon placed in the ventricle. During open chest
conditions, preload can be changed by tightening a ligature around the inferior vena cava.
During closed chest conditions, venous return and preload can be changed by briefly
inflating a balloon at the tip of a catheter introduced into the inferior vena cava. Another
possibility is to vary preload pharmacologically, for instance with glyceryl trinitrate
(nitroglycerine).
It has been commonly assumed that the end systolic P/V relationship is linear and,
therefore, can be described by two parameters: slope Ees and volume axis intercept V0.43
Recent studies, however, demonstrated significant non-linearity of the relationship under
various conditions, for instance during myocardial ischaemia or a high contractile state.42 44
Fig 2.7 End systolic pressure–volume relationship and ventriculoarterial coupling. Ved (left ventricular end
diastolic volume), V0 (volume intercept of the end systolic pressure-volume relationship), Ees (end systolic
elastance), Ea (effective arterial elastance), grey area (area of the left ventricular pressure–volume loop). The
intersection of the Ees and Ea lines defines the effective stroke volume.
Page 46
Moreover, the end systolic P/V relationship is not completely load insensitive. Changes in
afterload induce parallel shifts of the relationship and may even alter the slope. The changes
in P/V relationships caused by load dependence have, however, been shown to be of minor
importance compared with the substantial changes brought about by variation in the
inotropic state.42
The development of the conductance catheter now allows determination of Ees in
patients.45 In the normal left ventricle, values of Ees of 3·5_4·5 mm Hg/ml were found,
decreasing to 2·5 mm Hg/ml in patients with mildly depressed left ventricular function and
to 1·5 mm Hg/ml in patients with severely depressed ventricles.46 The Ees was used for
evaluation of myocardial contractility in patients before and after open heart surgery. The
Ees varied between 0·9 and 5·6 mm Hg/ml with a mean of 2·5 and standard deviation (SD)
of 1·5, and showed individually variable changes after cardiopulmonary bypass.47 Ees is
very sensitive to positive inotropic stimulation, for example, to dobutamine, which clearly
increases the slope without changes in V0.48 It has recently been shown that the use of a
conductance catheter to measure ventricular volume can be replaced by the online
measurement of cross sectional left ventricular area using trans oesophageal
echocardiography and automated border detection (Fig 2.8). In an open chest study in dogs
using instantaneous echocardiographic left
Fig 2.8 Left ventricular time varying elastance (Emax) obtained by instantaneous pressure–area relationship.
Left: left ventricular pressure–area relationship demonstrating the increasing slope of the relationship (E) from
the onset to the end of ejection. Right: the change in elastance (E) plotted against time with the maximal value
of the elastance (Emax) occurring 250 ms after the onset of ejection. (Reproduced with permission from
Gorcsan et al.47
Page 47
ventricular area instead of volume, dobutamine markedly increased both Ees and Emax49
(Fig 2.9). In a study in patients undergoing open heart surgery, significant decreases in Ees
and Emax following cardiopulmonary bypass were observed. Such evidence of impaired left
ventricular contractility was not reflected by any other haemodynamic parameter such as
cardiac output, stroke work, or ejection fraction.50
Fig 2.9 Effect of positive inotropic stimulation on the left ventricular pressure–area relationship (Emax). (a)
Pressure–area loops and slope of the end systolic pressure–area relationship at control, (b) Increase in slope of
the end systolic pressure–area relationship (Emax) after dobutamine administration. (c) Increase in Emax and
shortening of the time to Emax after dobutamine administration (——) compared with control (—). (d)
Increase in the slope of the stroke force–end diastolic area relationship after dobutamine administration
compared with control . (Reproduced with permission from Gorcsan et al.46)
Page 48
In patients with heart disease, the end systolic pressure may be an inaccurate index of
ventricular afterload. In aortic stenosis or hypertensive cardiomyopathy, the end systolic
pressure is high, although the afterload may be normal as a result of the development of
ventricular hypertrophy, which is able to normalise the wall stress. Therefore, in these
patients the systolic (peak or end systolic) wall stress is a more reliable reflection of
ventricular afterload and should replace end systolic pressure in the P/V diagram. The end
systolic wall stress/area relationship was used for estimating myocardial contractility in
patients undergoing aortocoronary bypass surgery.51 The relationship between end systolic
pressure or wall stress and area or diameter of the left ventricle can be constructed from
non-invasive data using cuff blood presssure and echocardiography. The use of systolic cuff
blood pressure instead of end systolic pressure does not significantly alter the relationship.
When the pressure in the radial artery is recorded perioperatively, the left ventricular end
systolic pressure can be approximated with reasonable accuracy from the dicrotic notch
pressure by adding 8 mm Hg.52
In the daily practice of anaesthesia, as well as in cardiology, it has not been possible to
analyse the end systolic stress/volume relationship. However, it was suggested that the
simple ratio of end systolic stress to end systolic volume, obtained by echocardiography,
may be useful in evaluating left ventricular performance at the bedside. Although such a
simple ratio must not be used as a substitute for the slope of the pressure, stress/volume, or
area relationship, it can, however, provide simple but important information: a ventricle that
can contract to a small end systolic volume at a normal or even high end systolic pressure
or wall stress is in a better contractile state than one that remains large at end systole in
spite of a similar or even lower pressure.53 54 The maintenance of a low end systolic size (in
the presence of adequate filling volume) should be the goal of perioperative haemodynamic
management in patients with heart disease. Recently, a method for real time continuous
monitoring of ventricular and arterial elastance, and their coupling ratio based on single-
beat analysis, was developed.55 The method, based on echocardiographic automated border
detection and computation of isovolumic left ventricular pressure, obviates the need for
preload manipulation and has a potential for use in the clinical setting.
The second index of myocardial contractility validated in patients is the preload
recruitable stroke work. It is based on the linear relationship between left ventricular stroke
work (product of mean systolic pressure and stroke volume) and left ventricular end
diastolic volume.56 The left ventricular stroke work is calculated as the area of the P/V
loops, obtained in a similar fashion to elastance measurements by varying the end diastolic
volumes. The slope of the stroke work/end diastolic volume relationship (denoted as Mw) is
a measure of myocardial contractility. The relationship
Page 49
is highly linear and, in addition, the slopes and volume axis intercepts (Mo) are less variable
than those of the end systolic P/V relationships.57 In a study in patients, both preload
recruitable stroke work and end systolic P/V relationships showed high and comparable
linearity, and both responded to dobutamine by marked increases in Emax and Mw,
respectively.57 In contrast, following afterload reduction with captopril, the preload
recruitable stroke work remained unchanged whereas the end systolic P/V relationship
shifted to the right.58 With increases in afterload, the preload recruitable stroke work
relationship again does not change, but the end systolic P/V relationship shifts to the left.
Thus, the independence of the afterload is the advantage of the preload recruitable stroke
work relationship as a method of assessing myocardial contractility. Echocardiography
using automated border detection allows the measurement of preload recruitable stroke
force. End diastolic volume and stroke work are replaced by end diastolic area and stroke
force, which is the integral of the pressure–area loop. In patients, there was a marked
decrease in the slope of the preload recruitable stroke work relationship after
cardiopulmonary bypass, not reflected by standard haemodynamic indices.49 50
Other validated methods for clinical assessment of myocardial contractility make use of
the general principle of relating the parameter of left ventricular function to left ventricular
afterload. One of these methods, which has been also applied in surgical patients, is the
determination of the shortening velocity of circumferential fibres (Vcf) in relation to the
actual systolic wall stress. Vcf is obtained by echocardiography applying the formula:
When multiplied by the square root of the RR interval (where RR is the cardiac cycle), it
becomes independent of heart rate. The relationship between end systolic stress and Vcf is
inversely linear and reflects the force–velocity relationship. It shifts upwards with a positive
inotropic intervention and downwards with impaired contractility.59 A similar relationship
exists between end systolic wall stress and ejection fraction.
More recently, left ventricular power was proposed as an index of myocardial
contractility.17 Power is work per unit time and the left ventricular power represents the rate
of energy expenditure during the pressure–volume work of the ventricle. Maximal left
ventricular power (PWRmax) is the peak instantaneous product of chamber pressure and
rate of volume change obtained by simultaneous left ventricular pressure and volume
measurements with a conductance catheter technique, and has been validated as an index of
left ventricular contractility. It can be also measured non-invasively from the product of
peak arterial pressure and
Page 50
flow.60 The advantage of PWRmax is its afterload independence; on the other hand, it is
highly sensitive to preload. This preload dependency can be eliminated by dividing
PWRmax by EDV or EDV2.61
Heart rate
An isolated strip of heart muscle responds to changes in frequency of electrical
stimulation by changes in developed force. The ability of the heart muscle to develop force
and/or to shorten clearly depends on the interval between preceding contractions. This
dependence has been described as a force–interval, strength–interval, or a force–frequency
relationship. The physiological phenomena reflecting this fundamental relationship are
Bowditch’s positive staircase response, Woodforth’s negative staircase response,
postextrasystolic potentiation, and the response to paired pacing. Basically, an increase in
heart rate leads to an increase in myocardial contractility and vice versa. The short term
effect is much more pronounced than the chronic steady state effects. The existence of these
phenomena in intact, conscious animals and humans has been questioned.62 Many earlier
studies, however, were using unreliable, load dependent indices of contractility. When the
less load sensitive, end systolic P/V relationship was used to look at the effects of heart rate
on myocardial contractility in conscious dogs, a marked dependence of the contractility on
the pacing rate was found.63 The slope of the relationship, Emax, showed a positive
correlation with heart rate. In addition, with increasing heart rate, the volume intercept (V0)
of the P/V relationship shifted to the right on the x axis. This may be partly responsible for
the changes in stroke volume with changing heart rate. The increased slope of the
relationship and the rightward shift of the V0 point result in an increase in stroke volume at
higher end systolic pressures, and in a decrease in stroke volume at low end systolic
pressures.63 The first effect may be beneficial during exercise by facilitating ejection despite
an increase in afterload. Isotonic exercise has been shown to potentiate the effects of
increasing heart rate on myocardial contractility.64 As similar augmentation of the force–
frequency relationship was observed during dobutamine infusion, these effects can be
attributed to β-adrenergic stimulation.65 Besides contractility, myocardial relaxation is also
facilitated by increasing heart rate, and this effect is augmented by adrenergic stimulation.66
Except for patients with severe forms of bradycardia, it is not common practice to use
cardiac pacing to improve systolic performance. Although short term increases in heart rate
appear to have beneficial effects on ventricular performance, the opposite is true for chronic
tachycardia. The development of myocardial stunning, dilated cardiomyopathy, and heart
failure after prolonged fast pacing in animal experiments points to adverse
Page 51
effects of long term tachycardia.67 A transient ventricular dysfunction also occurs in
patients after successful conversion of a long lasting tachy-arrhythmia.
It is interesting that heart muscle strips recovered from failing hearts do not respond to
increases in stimulation rate; in contrast, a fall in developed force with increasing frequency
was described, manifesting itself as a descending limb of the force–frequency
relationship.68
Paired pacing that makes use of the phenomenon of postextrasystolic potentiation was,
however, successfully applied in patients with severe heart failure to enhance ventricular
performance. Postextrasystolic potentiation, as well as the positive staircase response, result
from increased availability of intracellular calcium for binding to contractile proteins.
Ventriculoarterial coupling
The coupling of the left ventricle with the peripheral vascular system has been intensely
studied in terms of the end systolic P/V relationship.69 In this model, the left ventricle and
the arterial system are regarded as two coupled elastic chambers. The distribution of blood
between these chambers is determined by their relative elastances. The elastance E is a
measure of respective chamber stiffness and is represented by the slope of the P/V
relationship (see Fig 2.7).The instantaneous elastance increases from a low value during
diastole to its maximum value (Emax) close to end systole. The pressure varies inversely
and linearly with the stroke volume according the following equation:
where Ees = end systolic elastance related to contractile properties of the ventricle, Pes =
end systolic ventricular pressure, Ved = end diastolic volume, SV = stroke volume and V0 =
volume-axis intercept of the P/V relationship.
The arterial system is characterised in this model by the relationship between end systolic
pressure and stroke volume. The slope of this relationship, the effective arterial elastance
(Ea=Pes/SV), serves as an index of total external load opposing ejection. Ea comprises
resistance, compliance, and characteristic impedance of the arterial vascular bed. During
the ejection phase, elastance varies with time; throughout ejection, ventricular elastance
progressively increases from the onset to the end of ejection whereas arterial elastance
progressively decreases (time varying elastance). The effective stroke volume resulting
from the ventriculoarterial coupling is determined by the intersection of the ventricular end
systolic P/V and arterial end systolic P/V relationships.48 69 Graphic analysis of the
ventriculoarterial coupling provides expeditious information on left ventricular function and
its determinants during acute changes in loading
Page 52
conditions.38 45 The left ventricle delivers maximal external work (stroke work) when the
Ea/Ees ratio approximates 1. The mechanical efficiency of the ventricle, relating work to
amount of energy consumed, is maximal when Ea is approximately half of Ees.70 In normal
subjects, Ees and Ea values of 3·5–7·0 and 1·6_4·0 mm Hg/ml, respectively, were found.46 71
The Ea/Ees ratio was 0·4_0·6 and increased significantly with increasing afterload as a result
of increased Ea. An afterload reduction caused Ea and the ratio Ea/Ees to decrease. Positive
inotropic stimulation also decreased Ea/Ees as a result of an increase in Ees.71
The pressure/volume area (PVA) encompassed by the end systolic and diastolic P/V
relationship and the systolic part of the P/V loop provides an interesting insight into
myocardial energetics.72–74 PVA is composed of the PE area (PE being the potential energy
of an isovolumic contraction) and the EW area (EW being the external work of ejecting
contraction) (Fig 2.10). PVA (= PE + EW) is a measure of the total mechanical energy of
contraction and was shown to correlate linearly with left ventricular myocardial oxygen
consumption . The slope of this relationship represents the oxygen cost of the PVA
and its reciprocal represents the mechanical efficiency. This linear /PVA
relationship remains constant during changes in preload and afterload, but is shifted
upwards in parallel manner by positive inotropic stimulation. In contrast, negative inotropic
Fig 2.10 Left ventricular pressure–volume relationship and pressure–volume area (PVA): ESPVR (end
systolic pressure–volume relationship), EDPVR (end diastolic pressure–volume relationship), Ees (end
systolic elastance), Ea (effective arterial elastance), EW (left ventricular external work (stroke work)), PE (left
ventricular potential energy). The PVA is the sum of EW and PE areas.
Page 53
interventions shift the relationship downwards, again without changing its slope.75 In
patients with normal left ventricular function, a contractile efficiency of 40% was found.76
A decrease in efficiency from 46% to 35% was demonstrated in healthy volunteers during
positive inotropic stimulation with dobutamine.77 The ratio of external work (EW or stroke
work) to PVA can be used to express the mechanical efficiency of the whole left ventricle.
In patients with normal left ventricular function, this ratio was 0·6. It decreased with an
increase in afterload and increased when afterload was reduced.71
Ventricular function curves and the Frank–Starling law

of the heart
Clinical evaluation of the overall pump function of the heart is usually based on the
ventricular function curve, which relates a measure of ventricular performance (cardiac
output, stroke volume, or stroke work) as a dependent variable to a measure of ventricular
preload, such as ventricular filling pressure, end diastolic volume, end diastolic diameter,
cross sectional area, or end diastolic wall stress.9 10 78 The ventricular function curve
describes the fundamental dependence of ventricular performance on ventricular preload
and represents the expression of the Frank–Starling law of the heart: ‘‘the energy of
contraction is a function of the length of the muscle fibre”.79 (Fig 2.11). This law is based
on the length–tension relationship of myocardial sarcomeres, which at rest operate below
the maximum length (Lmax) on the ascending limb of their length–tension relationship.
Increasing stretch on the sarcomeres towards Lmax increases the number of activated cross
bridges and the developed tension.
The dependence of ventricular performance on afterload and contractile state is reflected
by displacement of the ventricular function curve with changing contractility and/or
afterload. For example, with decreasing afterload and/or increasing contractility, the curve
is shifted upwards and to the left; an opposite shift is observed with increasing afterload
and/or deteriorating contractility.78 Hence, the function of the ventricle cannot be described
by one function curve; instead, a family of ventricular function curves, reflecting the
changing load and inotropic states, gives a better characterisation of the ventricular
function.78 79 Incremental pacing also shifts the cardiac output curve upwards; the
magnitude of the shift is, however, less than during exercise at a comparable heart rate, as a
result of an additional positive inotropic effect of increased sympathetic tone during
exercise.80 At higher preloads the ventricular function curve tends to plateau and further
augmentation in ventricular filling no longer enhances ventricular performance. The so
called “descending limb of the Starling curve”, which has given rise to considerable
controversy in the past, is now
Page 54
Fig 2.11 Ventricular function curves. The curves describe the relationship between resting length of
contractile fibres (preload) and ventricular performance (cardiac output, stroke volume, stroke work), known
as the Frank-Starling law. An enhanced inotropic state shifts the relationship leftward and upward; depressed
inotropy has an opposite effect.
considered to be an artefact caused by non-physiological experimental conditions.79
Internal control mechanisms
Cardiac pump function or its output (cardiac output) is regulated by both internal and
external factors. The Frank–Starling mechanism represents the internal control mechanism
of the pump. It plays an eminent role in the maintenance of balance between right and left
ventricular outputs, and the distribution of blood volume between the systemic and
pulmonary circulation.10 78 81 The Frank–Starling mechanism is also activated during early
stages of exercise, possibly because of the delay in sympathetic activation, but it does not
play a major role during later stages of vigorous exercise.82 In normal supine subjects and in
the presence of normal filling pressures (10–12 mm Hg), the left ventricle operates close to
the maximum of its function curve. Attempts to increase the filling volume lead to an
increase in filling pressure, but only to modest improvement in ventricular performance.83 84
In contrast, in the upright position and in the presence of filling pressures that are lower
than normal, the ventricle clearly operates on the ascending limb of its function curve and,
consequently, fluid administration can markedly enhance pump function. In the presence of
Page 55
myocardial depression, for example, resulting from anaesthesia and obtunded baroreflex
function, the role played by the Frank–Starling mechanism becomes more important.13
External control mechanisms
In contrast to the internal control of the heart pump provided by the Frank–Starling
mechanism, the external control factors interacting with ventricular performance reside
within the systemic circulation.78 During the steady state, the heart cannot pump more
blood than it receives from the periphery. The flow of blood returning from the periphery is
known as venous return and is equivalent to cardiac output except for very short periods of
time, as during transient compression of the vena cava. Systemic venous return is described
by the venous return curve, relating flow to right atrial pressure. In the negative filling
pressure range, the venous return becomes maximal but soon levels out because of the
limiting effect of venous collapse at the level of the thoracic inlet. In the positive pressure
range, there is a linear decrease in venous return with increasing atrial pressure until a point
is reached when the flow ceases. The pulmonary venous return curve, reflecting the
relationship between left atrial pressure and cardiac output, is analogous to the systemic
venous return curve.
When the heart pump stops and blood flow ceases, the pressures within the heart and
circulation will equalise at the so called mean systemic pressure, which can be estimated by
extrapolation during long diastoles. This mean systemic pressure depends on blood volume
present in the circulation, vascular, notably venous tone, and external forces compressing
the vessels. Venous return to the heart is directly proportional to this mean systemic
pressure and to the difference between mean systemic pressure and right (or left) atrial
pressure. The complex interplay of all major factors involved in cardiac output regulation,
for instance during intraoperative changes in blood volume and intrathoracic pressure, can
be studied on Guyton’s graph where both ventricular function curves and venous return
curves are combined. The points at which the curves cross represent respective equilibria
and determine the actual value of cardiac output (Fig 2.12).
Afterload mismatch
As a result of the immediate upregulation in force generation in response to acute

increases in afterload, the steady state relationship between stroke work and end diastolic
volume remains insensitive to wide alterations in afterload.85 This is true as long as there is
a sufficient preload reserve; only then does the ventricular performance (generated stroke
volume or stroke
Page 56
Fig 2.12 Guyton’s diagram showing right and left ventricular function curves. At steady state, the output of
both ventricles is identical and corresponds to cardiac output (CO). The outputs of the right and left ventricles
are determined by the intersections of ventricular output curves with the corresponding systematic and
pulmonary venous (PV) return curves.68
work) not decrease with increasing afterload, because the ventricle can maintain its stroke
volume by mobilising its preload reserve and making use of the Frank–Starling mechanism.
In contrast, when preload reserve is exhausted – for example, by overtransfusion or
inadequate venous return – stroke volume becomes dependent on systolic pressure and
linearly declines with any farther increase in afterload. This can produce an apparent
descending limb of the ventricular function curve.84 The condition has been termed
“afterload mismatch” and can be described as the inability of the ventricle at a given level
of myocardial contractility to maintain a normal stroke volume against the prevailing
systolic load.86 A change in myocardial contractility will alter the afterload sensitivity of the
ventricle, and the afterload mismatch will occur at different levels of systolic pressure and
stroke volume. The concept of afterload mismatch is useful in the haemodynamic
management of the perioperative phase even in patients with normal ventricular function.
These patients may have a normal cardiac preload reserve, but peripheral factors may
generate a venous return that is inadequate to maintain the end diastolic volume needed for
adequate ventricular performance.84
Page 57
Ventricular diastolic performance

During systole, the ventricle can pump only the amount of blood into the aorta that it has
received during the preceding diastole. Furthermore, the ventricle, which fails to fill
properly, is deprived of its intrinsic ability to increase the strength of its contraction by
increasing the length of its contractile fibres. Therefore, the diastolic function of the
ventricle is as important as the systolic one.87 88 A normal diastolic function can be defined
as the amount of filling of the ventricle that is necessary to produce a cardiac output
commensurate with the body needs at normal pulmonary venous pressure.89 Traditionally,
diastole is defined as the part of the cardiac cycle between the closure of the aortic valve
and the closure of the mitral valve. It comprises:
z isovolumic relaxation period

z rapid filling
z slow filling (diastasis)
z atrial systole.
An alternative concept assigns ventricular relaxation, which is an active, energy

consuming process of myocardial inactivation, to the systole which, in turn, extends into
the rapid filling phase.90 The major determinants of diastolic function are:
z myocardial relaxation
z elastic recoil
z passive filling characteristics of the ventricle
z atrial function
z heart rate.
Isovolumic relaxation period
During relaxation, ventricular pressure is determined by two overlapping processes: the

decay of the pressure actively developed during the preceding systole and the build up of
the passive filling pressure. Together, these two components result in the effective pressure
of the isovolumic relaxation period. The rapid fall in pressure can be approximated by a
mono-exponential curve and described by its time constant τ.88 Maximal negative dP/dt and
the duration of the isovolumic relaxation period have been also used for the evaluation of
ventricular relaxation. The rate of relaxation is modulated by sympathetic tone and
circulating catecholamines. It increases with positive inotropic stimulation and increasing
heart rate. Interventions causing an increase in the rate and extent of relaxation are called
positive lusitropic effects.
A delay of ventricular relaxation can be caused by pressure or volume overload, which
primarily prolongs the contraction. The delayed relaxation
Page 58
can affect the early filling but does not cause diastolic failure. In contrast, the real
impairment in the rate and extent of relaxation will compromise diastolic function. Such
impairment becomes evident as an upward shift of the entire diastolic P/V relationship,
reflecting increased resistance to ventricular filling.91 Diastolic dysfunction and failure may
result from disturbances of the mechanisms that control myocardial relaxation. They
include intracellular calcium homoeostasis, function of sarcoplasmic reticulum and
contractile proteins, loading conditions, and uniformity of relaxation.92 93 The most
common cause of impaired relaxation in the perioperative period is myocardial ischaemia.
Rapid filling
The driving force of early filling is the atrioventricular pressure gradient. The gradient is
enhanced by the ventricular elastic recoil or suction. During contraction, potential energy is
stored at end systole in the form of a longitudinal gradient of circumferential rotation
(twist) and released during early diastole as elastic recoil.94 Additional, so called restoring
forces include myocardial compression and stretching of the mitral valve apparatus during
contraction. The left ventricular twist is a special form of systolic deformation (torsion) and
manifests itself as a counterclockwise rotation of the apex relative to the base of the left
ventricle. At end systole the fast untwisting starts with a half of the twist already dissipated
during the isovolumic relaxation period, followed by a slower untwisting during the early
filling.95 Consequently, 60–80% of the stroke volume enters the ventricle during the first
third of diastole.89 Restoring forces become more important when the end systolic volume
is small, such as during exercise, tachycardia, and hypovolaemia.
Slow filling
The following phase of slow filling, diastasis, is characterised by only modest increases
in pressure and volume in the ventricle (in accordance with the passive diastolic P/V
relationship) and is caused by ongoing venous return.
Atrial contraction
At the end of diastole, atrial contraction raises the atrial pressure and hence the filling
pzressure gradient, and completes the filling. Atrial contraction represents a diastolic
function reserve (atrial booster pump) that becomes increasingly important during
shortening in filling time (such as during tachycardia) as well as in all situations of
impaired ventricular filling, hence the importance of sinus rhythm for normal ventricular
performance and the usually detrimental effect of its loss in patients with heart disease.
Page 59
The contribution of the atrium to ventricular filling can be easily assessed by Doppler
measurements of blood flow velocity across the mitral valve. The late diastolic flow
velocity wave caused by atrial contraction (A wave) is normally lower than the early
velocity (E wave).
The final amount of blood accommodated by the ventricle at end diastole depends on
passive characteristics of the chamber, which are determined by the following:
z elastic properties of the myocardium

z geometry of the chamber
z thickness of the ventricular wall
z viscoelastic (flow velocity dependent) effects
z external constraints (pericardium, lungs).
These properties, taken together, can be expressed by chamber stiffness, which relates the
instantaneous change in filling pressure to instantaneous change in filling volume (dP/dV)
and reflects the pressure change induced by a unit change in volume (Fig 2.13). Chamber
compliance (dV/dP) is the reciprocal of chamber stiffness.96 The diastolic P/V relationship
is exponential in shape and its slope increases as the end diastolic pressure increases more
steeply at high filling volumes. The slope of the relationship at any part of the curve
correlates linearly with the end diastolic pressure, allowing the calculation of the modulus
of chamber stiffness as the slope of the (dP/dV)/P relationship.96 Changes in chamber
stiffness may occur as a consequence of changes in operating end diastolic pressure
(preload). With
Fig 2.13 Diastolic (passive) pressure–volume relationship. Left: the slope of the pressure–volume relationship
(dP/dV) represents the chamber stiffness and increases with increasing diastolic volume. Right: the
relationship between chamber stiffness (dP/dV) and diastolic presure is linear and its slope represents the
chamber stiffness constant kc.85
Page 60
progressive increase in end diastolic volume and a rightward and upward displacement of
the ventricular P/V loop along the diastolic P/V line, the chamber becomes stiffer. In
contrast, with a real alteration of chamber stiffness, the whole diastolic P/V relationship and
its slope (modulus of chamber stiffness) are altered. A decrease in chamber stiffness is
associated with a downward and rightward shift of the diastolic P/V relationship and vice
versa.
Ventricular hypertrophy and myocardial infarction with ensuing remodelling represent
the most common causes of increased chamber stiffness and diastolic dysfunction. In
contrast to chamber stiffness, which is a measure of the ability of the ventricle to oppose
distension, myocardial stiffness is a measure of the resistance to stretching of the
myocardium itself.88 97 The ventricular diastolic stress (σ), similar to systolic stress, is
defined as force per unit of cross sectional area of the ventricular wall and is related to
cavity pressure, radius, and wall thickness. Strain (ε) is the change in length with respect to
a reference length and is expressed as a percentage. The stress/strain relationship (σ/ε) of
the myocardium is nonlinear and its shape resembles that of the diastolic P/V curve. The
slope of any part of this exponential curve is termed the “elastic stiffness of the
myocardium” (dσ/dε). When this elastic stiffness is plotted against the diastolic wall stress,
the relationship becomes linear and its slope can be quantified by a single stiffness constant.
An increase in the stiffness constant represents a true increase in myocardial stiffness,
which is mostly caused by myocardial fibrosis.28 88
Heart rate
Heart rate is also a major determinant of diastolic performance. With progressively

increasing heart rate and decreasing diastolic filling time, the diastolic reserve based on
increased rate of relaxation and atrial booster becomes exhausted and cardiac output falls,
even in patients with normal ventricular function. In patients with impaired diastolic
function at rest, cardiac output may fall even with modest tachycardia.
Right ventricular performance

The right ventricle receives systemic and coronary venous return and pumps it into the
left ventricle across the pulmonary vascular bed. The right and left ventricles can be
described as two pumps in series coupled by the lungs and operating as one functional unit.
Both parts of this unit have a common blood supply, a common muscular septum separating
both cavities, and common intertwining myocardial bundles. Moreover, both ventricles are
confined within the common pericardium and exposed to the same changes in intrathoracic
pressure and lung volume. There are,
Page 61
however, important differences between both ventricles. Anatomically, the right ventricle
consists of the free wall and septum arranged into the inflow tract (sinus) and the outflow
tract (conus); the conus represents the phylogenic relic of the bulbus cordis.98 The right
ventricular outflow tract contracts after the inflow tract with a delay of up to 25–50 ms; this
asynchrony causes the outflow tract to expand during the contraction of the inflow tract and
to maintain right ventricular ejection by the time the inflow tract is beginning to relax.
Under intense sympathetic stimulation, significant pressure gradients between inflow and
outflow tracts can occur.99–101
Wall thickness and myocardial mass of the right ventricle are considerably less than
those of the left ventricle, reflecting the lower external work of the former. Although the
right ventricle pumps the same amount of blood (cardiac output) as the left ventricle, it does
so into the low pressure pulmonary vascular bed with low resistance to flow. Pulmonary
arteries are much more distensible than systemic arteries, and the pulse wave velocity in the
pulmonary arteries is lower than in the systemic arterial tree. This causes the reflected
pressure waves to return after the closure of the pulmonic valve. The pulmonary input
impedance, representing the external load of the right ventricle, differs from the aortic
impedance in that the oscillatory component (characteristic impedance) is relatively greater
and makes up to 25–30% of the total pulmonary resistance.102 The right ventricular ejection
into the low impedance pulmonary vascular bed, together with the delayed return of the
reflected waves, affect the right ventricular pressure waveform. There is a very short
isovolumic contraction period, lower right ventricular dP/dt, and systolic peak pressure
occurs early during the ejection. Furthermore, the ejection continues despite the rapid and
marked decline in pressure.103
As a result of the low intraventricular and, consequently, low intramural pressure, the
coronary blood flow in the right ventricle is continuous throughout the cardiac cycle. In
spite of these differences, the mechanical behaviour of the right ventricle closely resembles
that of its left companion. An increase in right ventricular filling (reflected by increases in
end diastolic volume and pressure, and related to an increase in resting fibre length) leads to
an augmentation of right ventricular stroke volume and stroke work according to the Frank–
Starling law. Similar to the left ventricle, there is an inverse relationship between the
impedance opposing ejection and the right ventricular function expressed as either ejection
fraction or stroke volume.
An augmentation of right ventricular preload and positive inotropic stimulation will
improve right ventricular performance.99 101 The right ventricle is more sensitive to
increases in afterload than the left one. With increasing resistance to ejection (for example,
in pulmonary hypertension), the right ventricle readily uses up its preload reserve and
dilates. As a result of the high chamber compliance of the thin walled right ventricle, the
Page 62
increase in end diastolic volume can be more pronounced than the increase in filling
pressure. Thus, the right ventricle is able to maintain normal pulmonary blood flow and left
ventricular filling without an undue increase in central venous pressure. The right
ventricular ejection fraction will, however, exhibit a linear decrease with increasing
afterload.99 101 A prerequisite to the maintenance of flow is an adequate preload reserve.
Yet, there is a limitation to right ventricular performance, as a normal, non-hypertrophied
right ventricle will not sustain an acute increase in peak systolic pressure of more than 70-
80 mm Hg without failing.101
Similar to the left ventricle and the systemic circulation, the coupling of the right
ventricle and the pulmonary vasculature was studied by means of the time varying elastance
model, which revealed an optimal matching between the right ventricle and its load under
physiological conditions.102 The right ventricular P/V loop reflects its ejection
characteristics. Isovolumic contraction is almost absent, and there is a continuous decrease
in ventricular volume after the end systolic point. The slope of the end systolic P/V
relationship (right ventricular Emax) is lower and the volume intercept (V0) is higher in the
right than in the left ventricle.103
The right ventricle appears to have better tolerance of acute decreases in pulmonary
compliance (for example, occlusion of a central pulmonary artery branch) than of increases
in resistance (for example, peripheral pulmonary embolisation or lung hyperinflation).104 105
Considering the close anatomical relationship between both ventricles, it is not surprising
that changes in geometry, pressure, and volume of one ventricle directly affect the function
of the other. This is known as ventricular cross talk or interdependence, and occurs in both
diastole and systole.106 107 An increase in filling volume of one ventricle causes an upward
shift of the diastolic P/V relationship, a decrease in diastolic chamber compliance, and
impaired filling of the other ventricle. For instance, an increase in right ventricular diastolic
volume and pressure leads to an inversion of the diastolic trans-septal pressure gradient,
flattening of the ventricular septum curvature, and a leftward shift of the septum during
diastole; these effects increase the stiffness of the left ventricle and limit its filling. In
patients with an overloaded and failing right ventricle, right ventricular filling pressure
(right atrial pressure) exceeds left sided filling pressure (left atrial pressure or wedge
pressure). Diastolic ventricular interdependence is more pronounced with the pericardium
intact.108 109
The term “systolic interdependence’’ describes the observation that an increase in
pressure in one ventricle leads to an immediate pressure increase in the other ventricle. The
contribution of the left ventricle to pressure generation in the right ventricle exceeds the
contribution of the right to the left ventricle and is estimated to be about one third of the
right ventricular systolic pressure.110 Although ventricular interdependence operates
primarily through the interventricular septum, the free walls of both ventricles
Page 63
are also involved: in the case of the right ventricle, this is accomplished by pulling the right
ventricular free wall against the septum during contraction of the intertwining muscle
bundles shared by both ventricles.108 111 This mechanism of “left ventricular assistance”,
combined with the force acting from behind (“vis a tergo”) imparted by left ventricular
contraction, explains why there is only a modest depression of haemodynamic function
after total exclusion of the right ventricular free wall, provided that there is low pulmonary
vascular resistance.112
The diastolic and systolic interdependence plays an important role in clinical conditions
such as right ventricular volume or pressure overload and right ventricular ischaemia. In
these conditions, both diastolic filling and systolic performance of the left ventricle are
compromised and left ventricular support of the right ventricle reduced. Therapy aimed at
improved left ventricular function and developed pressure will result in enhanced right
ventricular function. That will result partly from the interdependence mechanism, and
partly from the increase in right ventricular coronary perfusion pressure.
The right ventricle plays an important role in the perioperative period and critical care
medicine where acute changes in loading conditions, gas exchange, ventilatory patterns,
and coronary blood flow can often eventually lead to right ventricular failure. To assess the
right ventricular function properly, to detect its dysfunction in time, and to treat it correctly,
the measurement of pressures and flows in the right heart and lesser circulation is necessary
(Table 2.2).18 113
The use of fast thermistor pulmonary catheters and the thermodilution method for
assessment of right ventricular ejection fraction and volumes can provide additional useful
information. The most valuable information on the structure and function of the right
ventricle at the bedside is now obtained by transthoracic and transoesophageal
echocardiography. With the help of echocardiography, and the end diastolic and end
systolic size of the right ventricle, its ejection fraction, regional wall motion, septum shifts,
and presence of tricuspid and pulmonic valve regurgitation can be evaluated. By means of
Doppler measurements of forward and regurgitant
Table 2.2 Right ventricular pump function18 113
Normal values
Right atrial pressure 5 mm Hg

Right ventricular pressure (systolic/diastolic) 25/5 mm Hg
Pulmonary artery pressure (systolic/diastolic) 25/9 mm Hg
Pulmonary artery pressure (mean) 15 (10–20) mm Hg
Right ventricular end diastolic volume index 65–100 ml/m2
Right ventricular ejection fraction 48–66%
Right ventricular stroke work index 5–10 g·m/m2
Page 64
blood flow velocities across the tricuspid and pulmonic valve, estimates of right ventricular
and pulmonary artery pressures can be made.
Ventricular performance during exercise

Exercise testing has proven useful as a method of preoperative assessment of cardiac
reserve and cardiac risk stratification before anaesthesia and surgery. The ability of the
heart to raise its output in response to the stress of trauma and surgery is a prerequisite to
uneventful recovery and survival.
During exercise, the heart must increase its output in order to meet the increasing need of
the body for oxygen and removal of metabolic end products. The metabolic demands of the
exercising muscles are supported by the increased cardiac output and increased oxygen
extraction in the tissues. During dynamic (isotonic) exercise, cardiac output increases
linearly with increasing work performance and oxygen consumption . The
relationship between and cardiac output as a dependent variable has an average y axis
intercept of about 6 and a slope of 5–6.113 For instance, a threefold increase in from
0·25 to 0·75 1/min is associated with an increase in cardiac output from 7·5 to 10·5 1/min.
In healthy individuals, the change in cardiac output divided by the change in is more
than 6. In this example, it is 3:0·5 = 6. Lower values indicate an inadequate augmentation of
cardiac output in relation to oxygen demand.
Maximal oxygen consumption is reached when levels off despite a

further increase in workload and represents the measure of the cardiac reserve (30–85
ml/min per kg depending on the level of physical fitness). Cardiac output increases three to
four times to 20–25 1/min during peak exercise in young healthy subjects. In highly trained
athletes, values up to 35 1/min were measured.113 The increased oxygen extraction is
reflected by the falling mixed venous oxygen saturation and a two- to threefold increase in
the arteriovenous O2 difference at peak exercise.114 115
The left ventricular and haemodynamic function during exercise is dependent on the
position of the body.116 In the supine position, the resting values of end diastolic, end
systolic, and stroke volumes are higher than in the upright position, although the heart rate
tends to be lower. During exercise in the supine position, there is only a modest initial
increase (up to + 20%) in stroke volume, and the increased cardiac output results
predominantly from tachycardia. In contrast, during upright exercise the Starling
mechanism is more important. This is reflected by a marked increase in the initially lower
end diastolic volume and a more pronounced increase (up to 50%) in stroke volume. There
is a consistent reduction in end systolic volume and, consequently, an increase in ejection
fraction.117 118 The stroke volume levels off at about 40% of the maximum
Page 65
workload and, therefore, it is the progressive increase in heart rate to its maximum (180–
200 beats/min) that is responsible for the attainment of the maximum cardiac output. The
final maximal heart rate and cardiac output in both positions are the same, and ventricular
volumes also become comparable.116 117 Thus, the increase in preload (brought about by
enhanced venous return from working muscles and mobilisation of blood volume, mostly
from the splanchnic reservoir) and the increase in heart rate, together with sympathetically
mediated enhancement of myocardial contractility, result in the required increase in cardiac
output. Systemic vascular resistance progressively falls with an increasing level of isotonic
exercise, reflecting vasodilation in the exercising muscles. Systolic blood pressure rises
while diastolic pressure remains unchanged or falls.113 Consequently, there is only a modest
increase in left ventricular afterload during isotonic exercise. However, during isometric
exercise (weight lifting, hand grip) systemic vascular resistance increases together with
systolic and diastolic blood pressure, and this is associated with a marked augmentation of
left ventricular afterload and myocardial oxygen consumption. There is also a difference
between leg and arm exercise with the latter producing a more pronounced rate and
pressure response despite a lower .
Endurance training based on isotonic exercise of adequate intensity and repetition
frequency leads to increases in resting blood volume and oxygen carrying capacity, end
diastolic and end systolic volumes, and ejection fraction. The heart rate is low at rest. The
higher cardiac output achieved by trained individuals results primarily from a higher stroke
volume, whereas the peak heart rates are similar in athletes and sedentary subjects.
Prolonged severe exercise produces alterations in ventricular function that can persist for
up to 24 hours. In tests on athletes immediately after finishing a triathlon race, a decrease in
left ventricular end diastolic dimension and shortening fraction associated with a depression
of the wall stress/shortening velocity relationship was found, suggestive of decreased
preload and impaired contractility.119 In tests on runners after a high altitude mountain run,
an isolated and marked right ventricular dilatation and depressed pump function associated
with pulmonary hypertension were described in one third of them.120 One can only
speculate on the cardiac consequences of prolonged intensive therapy when accompanied
by tachycardia comparable to that observed during strenuous exercise.
Changes in ventricular performance related to ageing

As age itself is rarely a contraindication to surgery, anaesthetists are increasingly
encountering very old patients undergoing major surgery that puts formidable stress on the
cardiovascular system. It is therefore
Page 66
important to understand the changes in cardiac function associated with ageing.121
On consideration of the high prevalence of clinically latent coronary artery disease and
hypertension in elderly people, it is not easy to separate the effects of ageing from other
pathological processes, and this may well explain some of the variations in the published
data.122 Ageing is associated with increasing stiffness of the great arteries, which in turn
leads to increases in arterial pulse wave velocity and pulse wave amplitude. The ejection of
stroke volume into a stiffer arterial tree and the accelerated reflected pulse waves returning
before the aortic valve closure result in higher systolic pressures.123 Thus, there is an
increase in left ventricular systolic load manifesting itself as alterations in aortic input
impedance. The characteristic impedance increases, there is increased fluctuation of the
impedance moduli about the mean, and both the minimum modulus and the pressure–flow
phase crossover shift to higher frequencies. The zero frequency impedance modulus,
representing resistance to steady flow (systemic vascular resistance), was found to increase
with age in some but not in all studies. The left ventricle responds to the augmented
hydraulic load and wall stress by a modest increase in left ventricular wall thickness and
mass, which, according to Laplace’s law, will normalise the systolic wall stress. In
experimental studies, changes in excitation–contraction coupling were found in senescent
myocardium resulting in prolonged contraction.122 At rest, ejection time, stroke volume, and
ejection fraction are well preserved even in very old individuals. On the other hand, there
are marked alterations in left ventricular filling.124 Ageing prolongs the isovolumic
relaxation time, reduces peak filling rate, and increases inhomogeneity in regional
lengthening. Surprisingly, end diastolic volume remains well preserved despite these filling
abnormalities and was even shown to increase with age. This may be related to the small
increase in left atrial (wedge) pressure and increased atrioventricular pressure gradient
and/or increased filling in the late diastole resulting from enhanced atrial contraction. The
resting heart rate decreases with advancing age, accompanied by reduced heart rate
variability and diminished intrinsic rate of the sinus node. Reports on changes in cardiac
output with age vary, with both reductions and no change in cardiac output being
described.122 In elderly populations with maintained cardiac output, an increase in stroke
volume associated with the reported increase in end diastolic volume presumably made up
for the reduction in heart rate.
Ageing is associated with a progressive decline in both basal and maximal oxygen uptake
and maximal heart rate, whereas the relationship between and heart rate during
increasing workload remains linear and has the same slope in both old and young subjects.
At peak exercise, increases in heart rate, cardiac output, and ejection fraction are lower in
old compared with young subjects. Furthermore, the mechanism by which
Page 67
cardiac output rises during exercise is altered in old subjects: there is more reliance on the
Frank–Starling mechanism manifesting itself as an increase in end diastolic volume.
Furthermore, compared with younger subjects, end systolic volume increases at higher
levels of exercise; together with the diminished heart rate response, these represent the
principal age related alterations, resulting in blunted or even absent increase in ejection
fraction and yet a well preserved stroke volume. In men at maximal workloads, end
diastolic and end systolic volumes and systemic vascular resistance increased with
increasing age, whereas inverse correlations between age and heart rate, cardiac index,
ejection fraction, and contractility index were described.125 126 There is also an inverse linear
relationship between age and peak filling rate during exercise.124 Thus, there is unequivocal
evidence of a progressively diminished cardiac reserve in healthy ageing individuals.
1 Braunwald E, Ross J Jr, Sonnenblick EH. Mechanisms of contraction of the normal and failing heart.
Boston: Little, Brown & Company, 1968:31–76.
2 Brady AJ. Mechanical properties of cardiac fibers. In: Berne RM, ed, Handbook of physiology, Section 2,
Cardiovascular system. The heart, Vol 1. Bethesda, MD: American Physiological Society, 1979:461–74.
3 Lew WYW. Mechanisms of volume-induced increase in left ventricular contractility. Am J Physiol
1993;265:H 1778–86.
4 Weber KT, Janicki JS, Hunter WC, Shroff S, Pearlman ES, Fishman AP. The contractile behavior of the
heart and its functional coupling to the circulation. Prog Cardiovasc Dis 1982;XXIV:375–400.
5 Weber KT, Janicki JS. The dynamics of ventricular contraction: force, length, and shortening. Fed Proc
1980;39:188–95.
6 Weber KT, Hawthorne EW. Descriptors and determinants of cardiac shape: an overview. Fed Proc
1981;40:2005–10.
7 Strobeck JE, Krueger J, Sonnenblick EH. Load and time considerations in the force–length relation of
cardiac muscle. Fed Proc 1980;39:175–82.
8 Weber KT, Janicki JS. The metabolic demand and oxygen supply of the heart: physiologic and clinical
considerations. Am J Cardiol 1979;44:722–9.
9 Parmley WW, Talbot L. Heart as a pump. In: Berne RM, ed, Handbook of physiology, Section 2,
Cardiovascular system. The heart, Vol 1. Bethesda, MD: American Physiological Society, 1979:429–60.
10 Braunwald E, Ross J Jr. Control of cardiac performance. In: Berne RM, ed, Handbook of physiology,
Section 2, Cardiovascular system. The heart, Vol 1. Bethesda, MD: American Physiological Society,
1979:533–80.
11 Katz AM. Influence of altered inotropy and lusitropy on ventricular pressure–volume loops. J Am Coll
Cardiol 1988;11:438–45.
12 Weber KT, Janicki JS. The heart as a muscle-pump system, and the concept of heart failure. Am Heart J
1979;98:371–84.
13 Vainer SFE, Braunwald E. Cardiovascular control mechanisms in the conscious state. N Engl J Med
1975;293:970–6.
14 Baan J, van der Velde EJ. Sensitivity of left ventricular end systolic pressure–volume relation to type of
loading interventions in dogs. Circ Res 1988;62:1247–58.
15 Westerhof N, Elzinga G. Cardiac pump function. In: Strackee J, Westerhof N, eds, The physics of heart
and circulation. Bristol: Institute of Physics, 1993:207–21.
16 Toorop GP, Van Den Horn GJ, Elzinga G, Westerhof N. Matching between feline left ventricle and
arterial load: optimal power or efficiency? Am J Physiol 1988;254:H279–85.
Page 68
17 Kass DA, Beyar R. Evaluation of contractile state by maximal ventricular power divided by the square of
end-diastolic volume. Circulation 1991;84:1698–708.
18 Yang SS, Bentivoglio LB, Maranhao V, Goldberg H. From cardiac catheterization data to hemodynamic
parameters, 2nd edn. Philadelphia: FA Davis Co., 1978:233–325.
19 Robotham JL, Takata M, Berman M, HarasawaY. Ejection fraction revisited. Anesthesiology
1991;74:172–83.
20 Clements FM, Harpole DH, Quill T, Jones RH, McCann RL. Estimation of left ventricular volume and
ejection fraction by two-dimensional transoesophageal echo-cardiography: comparison of short axis
imaging and simultaneous radionuclide angiography. Br J Anaesth 1990;64:331–6.
21 Urbanowitz JH, Cahalan MK, Chatterjee KL, Schiller NB. Comparison of transoesophageal
echocardiographic and scintigraphic estimates of left ventricular end-diastolic volume index and ejection
fraction in patients following coronary artery bypass grafting. Anesthesiology 1990;72:607–12.
22 Smith MD, MacPhail B, Harrison MR, Lenhoff SJ, DeMaria AN. Value and limitations of
transesophageal echocardiography in determination of left ventricular volumes and ejection fraction. J Am
Coll Cardiol 1992;19:1213–22.
23 Pohost GM, O’Rourke RA. Principles and practice of cardiovascular imaging. Boston: Little, Brown &
Co, 1991:383–503.
24 Leung JM, Schiller NB, Mangano D. Assessment of left ventricular function using two-dimensional
transesophageal echocardiography. In: de Bruijn NP, Clements FM, eds, Intraoperative use of
echocardiography. A Society of Cardiovascular Anesthesiologists Monograph. Philadelphia: JB
Lippincott Co., 1991:59–75.
25 Nessly ML, Bashein G, Detmer PR, Graham, MM, Kao R, Martin KW. Left ventricular ejection fraction:
single-plane and multiplanar transesophageal echocardiography versus equilibrium gated-pool
scintigraphy. J Cardiothorac Vasc Anesth 1991;5:40–5.
26 Cahalan MK, Ionescu, P, Melton HE Jr, Adler S, Kee LL, Schiller NB. Automated real-time analysis of
intraoperative transesophageal echocardiograms. Anesthesiology 1993;78:477–85.
27 Lindower PD, Rath L, Preslar J, Burns TL, Rezai K, Vandenberg BF. Quantification of left ventricular
function with an automated border detection system and comparison with radionuclide ventriculography.
Am J Cardiol 1994;73:195–9.
28 Mirsky I. Elastic properties of the myocardium: a quantitative approach with physiological and clinical
applications. In: Berne RM, ed, Handbook of physiology, Section 2, Cardiovascular system. The Heart,
Vol 1. Bethesda, MD: American Physiological Society, 1979:407–32.
29 Nichols WW, Pepine CJ. Left ventricular afterload and aortic input impedance: implications of pulsatile
blood flow. Prog Cardiovasc Dis 1982;XXIV:293–305.
30 Pasipoularides A. Clinical assessment of ventricular ejection dynamics with and without outflow
obstruction. J Am Coll Cardiol 1990;15:859–82.
31 Noble MIM. Left ventricular load, arterial impedance and their interrelationship. Cardiovasc Res
1979;13:183–98.
32 Covell JW, Pouleur H, Ross J Jr. Left ventricular wall stress and aortic input impedance. FedProc
1980;39:202–7.
33 Merillon JP, Fontenier GJ, Lerallut JF, et al. Aortic input impedance in normal man and arterial
hypertension: its modification during changes in aortic pressure. Cardiovasc Res 1982;16:646–56.
34 O’Rourke MF, Kelly RP. Wave reflection in the systemic circulation and its implications in ventricular
function. J Hypertens 1993;11:327–37.
35 Pouleur H, Covell JW, Ross J Jr. Effects of alterations in aortic input impedance on the force–velocity–
length relationship in the intact canine heart. Circ Res 1979;45:126–36.
36 Yin FC-P. Ventricular wall stress. Circ Res 1981;49:829–42.
37 Arts T, Prinzen FW, Reneman RS. Mechanics of the wall of the left ventricle. In: Strackee J, Westerhof
N, eds, The physics of heart and circulation. Bristol: Institute of Physics, 1993:153–74.
38 Sunagawa K, Sagawa K, Maughan ML. Ventricular interaction with the vascular system in terms of
pressure–volume relationships. In: Yin FCP, ed, Ventricular/Vascular coupling. New York: Springer
Verlag, 1987:210–39.
Page 69
39 Kass DA, Maughan WL, Guo ZM, Kono A, Sunagawa K, Sagawa K. Comparative influence of load
versus inotropic states on indexes of ventricular contractility: experimental and theoretical analysis based
on pressure–volume relationships. Circulation 1987;76:1422–36.
40 Suga H, Sagawa K, Shoukas AA. Load independence of the instantaneous pressure–volume relation of the
canine left ventricle and the effects of epinephrine and heart rate on the ratio. Circ Res 1973;32:314–22.
41 Grossman W, Braunwald E, Mann T, McLaurin LP, Green LH. Contractile state of the left ventricle in
man as evaluated from end-systolic pressure–volume relations. Circulation 1977;56:845–52.
42 Kass DA, Maughan WL. From “Emax” to pressure–volume relations: a broader view. Circulation
1988;77:1203–12.
43 Mehmel HC, Stockins B, Ruffmann K, van Olshausen K, Schuler G, Kübler W. The linearity of the end-
systolic pressure–volume relationship in man and its sensitivity for assessment of left ventricular function.
Circulation 1981;63:1216–22.
44 Van Der Velde ET, Van Dijk AD, Steendijk P, et al. Nonlinearity and afterload sensitivity of the end-
systolic pressure–volume relation of the canine left ventricle in vivo. Circulation 1991;83:315–27.
45 Kass DA. Clinical evaluation of left heart function by conductance catheter technique. Eur Heart J
1992;13(suppl E):57–64.
46 Asanoi H, Shigetake S, Kameyama T. Ventriculoarterial coupling in normal and failing heart in humans.
Circ Res 1989;65:483–93.
47 Schreuder JJ, Biervliet JD, van der Velde ET, et al. Systolic and diastolic pressure–volume relationships
during cardiac surgery. J Cardiothorac Vasc Anesth 1991;5:539–45.
48 Sasayama S. Matching of ventricular properties with arterial load under normal and variably depressed
cardiac states. In: Lewis BS, Kirnchi A, eds, Heart failure mechanisms and management. Berlin:
Springer-Verlag, 1991:59–67.
49 Gorcsan J III, Romand JA, Mandarino WA, Deneault LG, Pinsky MR. Assessment of left ventricular
performance by on-line pressure–area relations using echocardiographic automated border detection. J Am
Coll Cardiol 1994;23:242–52.
50 Gorcsan J III, Gasior TA, Mandarino WA, Deneault LG, Hattler BG, Pinsky MR. Assessment of the
immediate effects of cardiopulmonary bypass on left ventricular performance by on-line pressure–area
relations. Circulation 1994;89:180–90.
51 O’Kelly BF, Tubau JF, Knight AA, et al. Measurement of left ventricular contractility using
transesophageal echocardiography in patients undergoing coronary artery bypass grafting. Am Heart J
1991;122:1041–9.
52 Dahlgren G, Veintemilla F, Settergren G, Liska J. Left ventricular end-systolic pressure estimated from
measurements in a peripheral artery. J Cardiothorac Vasc Anesth 1991;5:551–3.
53 Carabello BA, Spann JF. The uses and limitations of end-systolic indexes of left ventricular function.
Circulation 1984;69:1058–64.
54 Carabello BA. Ratio of end-systolic stress to end-systolic volume: is it a useful clinical tool? J Am Coll
Cardiol 1989;14:496–8.
55 Shih H, Hillel Z, Declerck C, Anagnostopoulos C, Kuroda M, Thys D. An algorithm for real time
continuous evaluation of left ventricular mechanics by single-beat estimation of arterial and ventricular
elastance. J Clin Monit 1997;13:157–70.
56 Glower DD, Spratt JA, Snow ND, et al. Linearity of the Frank–Starling relationship in the intact heart: the
concept of preload recruitable stroke work. Circulation 1985;71:994–1009.
57 Feneley MP, Skelton TN, Kisslo KB, Davis JW, Bashore TM, Rankin JS. Comparison of preload
recruitable stroke work, end-systolic pressure–volume and dP/dtmax–end-diastolic volume relations as
indexes of left ventricular contractile performance in patients undergoing routine cardiac catheterization. J
Am Coll Cardiol 1992;19:1522–30.
58 Takeuchi M, Odake M, Takaoka H, Hayashi Y, Yokoyama M. Comparison between preload recruitable
stroke work and the end-systolic pressure–volume relationship in man. Eur Heart J 1992;13(suppl E):80–
4.
Page 70
59 Lang RM, Briller RA, Neumann A, Borow KM. Assessment of global and regional left ventricular
mechanics: applications to myocardial ischemia. In: Kerber RE, ed, Echocardiography in coronary artery
disease. Mount Kisco, NY: Futura, 1988:221–57.
60 Sharir T, Haber H, Feldman M, Marmor A, Kass DA. Ventricular systolic assessment in patients with
dilated cardiomyopathy by preload-adjusted maximal power: validation and non-invasive application.
Circulation 1994;89:2049–53.
61 Nakayama M, Chen C-H, Nevo E, Fetics B, Wong E, Kass DA. Optimal preload adjustment of maximal
ventricular power index varies with cardiac chamber size. Am Heart J 1998;136:281–8.
62 Seed WA, Walker JM. Relation between beat interval and force of the heartbeat and its clinical
implications. Cardiovasc Res 1988;22:303–14.
63 Freeman GL, Little WC, O’Rourke RA. Influence of heart rate on left ventricular performance in
conscious dogs. Circ Res 1987;61:455–64.
64 Miura T, Miazaki S, Guth BD, Kambayashi M, Ross J Jr. Influence of the force–frequency relation on left
ventricular function during exercise in conscious dogs. Circulation 1992;86:563–71.
65 Ross J Jr, Miura T, Kambayashi M, Eising GP, Ryu KH. Adrenergic control of the force–frequency
relation. Circulation 1995;92:2327–32.
66 Kambayashi M, Miura T, Oh B-H, Rockman HA, Murata K, Ross J Jr. Enhancement of the force
frequency effect on myocardial contractility by adrenergic stimulation in conscious dogs. Circulation
1992, 86:572–80.
67 DePauw M, Bao SM, Heyndrikx GR. Reversible left ventricular dysfunction induced by short term (48
hrs) rapid pacing in conscious dogs: non-ischemic myocardial stunning (abstract). Circulation 1993;88
(part 2):I29.
68 Erdmann E, Beuckelmann D, Boehm M, Schwinger HG. Klinische Gesichtspunkte der medikamentoesen
Differentialtherapie der chronischen Herzinsuffizienz. Z Kardiol 1992;81(suppl 4):97–103.
69 Sunagawa K, Maughan WL, Burkhoff D, Sagawa K. Left ventricular interaction with arterial load studied
in isolated canine ventricle. Am J Physiol 1983;245:H773–80.
70 Burkhoff D, Sagawa K. Ventricular efficiency predicted by an analytical model. Am J Physiol
1986;250:R1021–7.
71 Starling MR. Left ventricular–arterial coupling relations in the normal human heart. Am Heart J
1993;125:1659–66.
72 Suga H. Ventricular energetics. Am J Physiol 1990;70:247–77.
73 Nozawa T, Yasumura Y, Futaki S, et al. Relation between oxygen consumption and pressure–volume area
of in situ dog heart. Am J Physiol 1987;253:H31–40.
74 Nozawa T, Cheng C-P, Noda T, Little WC. Relation between left ventricular oxygen consumption and
pressure–volume area in conscious dogs. Circulation 1994;89:810–17.
75 Suga H. How we view systolic function of the heart: E max and PVA. 1994 CSDS Konrad Witzig
Lecture. In: Ingels NB Jr, Daughters GT, Baan J, Covel JW, Reneman, RS, Yin FC-P, eds, Systolic and
diastolic function of the heart. Amsterdam: IOS Press, 1995:215–25.
76 Takaoka H, Takeuchi M, Odake M, Yokoyama M. Assessment of myocardial oxygen consumption (Vo2)
and systolic pressure–volume area (PVA) in human hearts. Eur Heart J 1992;13 (suppl E):85–90.
77 Vanoverschelde J-LJ, Wijns W, Essamri B, et al. Hemodynamic and mechanical determinants of
myocardial O2 consumption in normal heart: effects of dobutamine. Am J Physiol 1993;265:H1884–92.
78 Guyton AC, Jones CE, Coleman TG. Circulatory physiology: cardiac output and its regulation.
Philadelphia: WB Saunders, 1973:137–252.
79 Elzinga G. ‘‘Starling’s law of the heart” a historical misinterpretation. Basic Res Cardiol 1989;84:1–4.
80 Melbin J, Detweiler DK, Riffle RA, Noordergraaf A. Coherence of cardiac output with rate changes. Am J
Physiol 1982;243:H499–504.
81 Jacob R, Dierberger B, Kissling G. Functional significance of the Frank–Starling mechanism under
physiological and pathophysiological conditions. Eur Heart J 1992;13 (suppl E):7–14.
Page 71
82 Plotnick GD, Becker LC, Fisher ML, et al. Use of the Frank–Starling mechanism during submaximal
versus maximal upright exercise. Am J Physiol 1986;251:H1101–5.
83 Parker JO, Case RB. Normal left ventricular function. Circulation 1979;60:4–11.
84 Lee J-D, Tajimi T, Patritti J, Ross J Jr. Preload reserve and mechanisms of afterload mismatch in normal
conscious dog. Am J Physiol 1986;250:H464–73.
85 Feneley MP, Karunanithi M, Michniewicz J, Young J, Kalnins W. Rapid contractile upregulation
rematches stroke work to increased afterload independent of ventricular geometry, afterload-related
coronary perfusion pressure fluctuations and baseline contractile state. In: Ingels NB Jr, Daughters GT,
Baan J, Covel JW, Reneman RS, Yin F C-P, eds, Systolic and diastolic function of the heart. Amsterdam:
IOS Press, 1995:325–44.
86 Ross J Jr. Afterload mismatch and preload reserve: a conceptual framework for the analysis of ventricular
function. Prog Cardiovasc Dis 1976;28:255–63.
87 Nishimura RA, Abel MD, Hade LK, Tajik AJ. Assessment of diastolic function of the heart: background
and current applications of Doppler echocardiography. Part II. Clinical studies. Mayo Clin Proc
1989;64:181–204.
88 Mirsky I, Pasipoularides A. Clinical assessment of diastolic function. Prog Cardiovasc Dis
1990;XXXII:291–318.
89 Little WC, Downes TR. Clinical evaluation of left ventricular diastolic performance. Prog Cardiovasc Dis
1990;XXXII:273–90.
90 Brutsaert DL, Sys SU. Relaxation and diastole of the heart. Physiol Rev 1989;69:1228–315.
91 Brutsaert DL, Sys SU, Gillebert TC. Diastolic dysfunction in post-cardiac surgical management. J
Cardiothorac Vase Anesth 1993;7(suppl 1):18–20.
92 Gillebert TC, Sys SU. Physiologic control of relaxation in isolated cardiac muscle and intact left ventricle.
In: Gaasch WH, LeWinter MM, eds, Left ventricular diastolic dysfunction and heart failure. Philadelphia:
Lea & Febiger, 1994:25–42.
93 Brutsaert DL, Sys SU, Gillebert TC. Diastolic failure: pathophysiology and therapeutic implications. J Am
Coll Cardiol 1993;22:318–25.
94 Moon MR, Ingels NB Jr, Daughters GT, Stinson EB, Hansen DE, Miller DC. Alterations in left
ventricular twist mechanics with inotropic stimulation and volume loading in human subjects. Circulation
1994;89:142–50.
95 Beyar R, Yin F C-P, Hausknecht M, Weisfeldt MC, Kass DA. Dependence of left ventricular twist–radial
shortening relations on cardiac cycle phase. Am J Physiol 1989;257:H1119–26.
96 Gaasch WH, Quinones MA, Waisser E, Thiel HG, Alexander JK. Diastolic compliance of the left
ventricle in man. Am J Cardiol 1975;36:193–201.
97 Gaasch WH. Passive elastic properties of the left ventricle. In: Gaasch WH, LeWinter MM, eds, Left
ventricular diastolic dysfunction and heart failure. Philadelphia: Lea &Febiger, 1994:143–9.
98 March HW, Ross JK, Lower RR. Observations on the behavior of the right ventricular outflow tract, with
reference to its developmental origins. Am J Med 1962;32:835–45.
99 Morris JJ III, Wechsler AS. Right ventricular function: the assessment of contractile performance. In: Fisk
RL, ed, The right heart. Philadelphia: FA Davis, 1987:3–18.
100 Furey SA III, Zieske HA, Levy MN. The essential function of the right ventricle. Am Heart J
1984;107:404–10.
101 Hurford WE, Zapol WM. The right ventricle and critical illness: a review of anatomy, physiology, and
clinical evaluation of its function. Intensive Care Med 1988;14:448–57.
102 Piene H. Matching between right ventricle and pulmonary bed. In: Yin FC-P, ed, Ventricular/Vascular
coupling. NewYork: Springer-Verlag, 1987:180–202.
103 Dell’Italia LJ, Santamore WP. Can indices of left ventricular function be applied to the right ventricle?
Prog Cardiovasc Dis, 1998;40:309–24.
104 Pouleur H, Lefevre J, van Eyll C, Jaumin PM, Charlier AA. Significance of pulmonary input impedance
in right ventricular performance. Cardiovasc Res 1978;12:617–29.
105 Ghignone M, Girling L, Prewitt RM. Effect of increased pulmonary vascular resistance on right
ventricular systolic performance in dogs. Am J Physiol 1984;246:H339–43.
106 Bove A, Santamore WP. Ventricular interdependence. Prog Cardiovasc Dis 1981; 23:365–88.
Page 72
107Olsen CO, Tyson GS, Maier GW, Spratt JA, Davis JW, Rankin JS. Dynamic ventricular interaction in the
conscious dog. Circ Res 1983;52:85–104.
108Weber KT, Janicki JS, Schroff S, Fishman AP. Contractile mechanics and interaction of the right and left
ventricles. Am J Cardiol 1981;47:686–95.
Beyar R, Dong S-J, Smith ER, Belenkie I, Tyberg JV. Ventricular interaction and septal deformation: a
109
model compared with experimental data. Am J Physiol 1993;265: H2044–56.
110Santamore WP, Dell’Italia LJ. Ventricular interdependence: significant left ventricular contributions to
right ventricular systolic function. Prog Cardiovasc Dis 1998;40:289–308.
Clyne CA, Alpert JS, Benotti JR. Interdependence of the left and right ventricles in health and disease.
111
Am Heart J 1989;117:1366–73.
112Jones DL, Guiraudon GM, Klein GJ. Total disconnection of the right ventricular free wall: physiological
consequences in the dog. Am Heart J 1984;107:1169–77.
113 Lentner C. Geigy scientific tables, the heart and circulation, Vol 5. Basel: Ciba-Geigy Ltd, 1990:9–219.
114Fleg JL, Schulman SP, O’Connor FC, et al. Cardiovascular responses to exhaustive upright cycle exercise
in highly trained older men. J Appl Physiol 1994;77:1500–6.
115Hammond K, Froelicher VF. Normal and abnormal heart rate responses to exercise. Prog Cardiovasc Dis
1985;27:271–96.
116Poliner LR, Dehmer GJ, Lewis SE, Parkey RW, Blomquist CG, Willerson JT. Left ventricular
performance in normal subjects: a comparison of the responses to exercise in the upright and supine
positions. Circulation 1980;62:528–34.
117Cotsamire DL, Sullivan MJ, Bashore TM, Leier CV. Position as a variable for cardiovascular responses
during exercise. Clin Cardiol 1987;10:137–12.
118Iskandrian AS, Hakki A-U, DePace NL, Manno B, Segal BL. Evaluation of left ventricular function by
radionuclide angiography during exercise in normal subjects and in patients with chronic coronary heart
disease. J Am Coll Cardiol 1983;l:1518–29.
ML Weisfeldt, ed, The aging heart. Its function and response to stress. Aging, vol 12, New York: Raven
119
Press, 1980:1–6.
Douglas PS, O’Toole ML, Hiller WD, Hackney K, Reichek N. Cardiac fatigue after prolonged exercise.
120
Circulation 1987;76:1206–13.
121Davila-Roman VG, Guest TM, Tuteur PG, Rowe WJ, Ladenson JH, Jaffe AS. Transient right but not left
ventricular dysfunction after strenuous exercise at high altitude. J Am Coll Cardiol 1997;30:468–73.
122 Lakatta EG. Cardiovascular regulatory mechanisms in advanced age. Physiol Rev 1993;73:413–53.
123Fleg JL, Gerstenblith G, Lakatta EG. Pathophysiology of the aging heart and circulation. In: Messerli FH,
ed, Cardiovascular disease in the elderly, 2nd edn, Boston: Martinus Nijhoff Publishing, 1988:9–35.
124Schulman SP, Lakatta EG, Fleg JL, Lakatta L, Becker LC, Gerstenblith G. Age-related decline in left
ventricular filling at rest and exercise. Am J Physiol 1992;263:H1932–8.
125Rodeheffer RJ, Gerstenblith G, Becker LC, Fleg JL, Weisfeldt ML, Lakatta EG. Exercise cardiac output
is maintained with advancing age in healthy human subjects: cardiac dilatation and increased stroke
volume compensate for a diminished heart rate. Circulation 1984;69:203–12.
126Fleg JL, O’Connor F, Gerstenblith G, et al. Impact of age on the cardiovascular response to dynamic
upright exercise in healthy men and women. J Appl Physiol 1995;78:890–900.
Page 73
3: Cardiac electrophysiology
JOHN LATLEE
The heart’s primary function is to generate contractile forces for the distribution of
adequate amounts of blood to the lungs and systemic tissues. For contraction to occur, cells
must first be excited – the function of the heart’s specialised conducting system. Normally,
excitation is initiated by spontaneous depolarisation (automaticity) of pacemaker cells
located in the sinoatrial (SA) node. In turn, SA automaticity brings surrounding cells to the
threshold for transient depolarisation – termed the “action potential” (AP). Action
potentials propagate rapidly to excite atrial and ventricular muscle in preferential atrial
pathways and the His–Purkinje conducting system, respectively. The atrioventricular (AV)
node functions to slow AV impulse transmission, in order to provide sufficient time for
ventricular filling.
The study of normal and abnormal mechanisms for generation and propagation of
cardiac APs is termed “cardiac electrophysiology”. It is the basis for understanding normal
sinus rhythm and cardiac arrhythmias, and also antiarrhythmic drug or device actions.
There are normal and abnormal cardiac electrophysiological processes; the latter can result
from either an abnormal cardiac substrate (for example, accessory pathways or scar tissue)
or a pathophysiological disease process (for example, ischaemia, adverse drug actions, or
electrolyte imbalance).
Normal cardiac electrophysiology

Cardiac electrophysiology can be resolved at the macroscopic and cellular levels.1 2 At
the macroscopic level, the surface electrocardiogram (ECG) records electrical activity from
an aggregate of myocardial cells. Slow depolarisation and delayed repolarisation are seen as
QRS widening and prolongation of the QT interval, respectively. At the microscopic level,
a single cardiac cell is impaled by a microelectrode, and an AP is recorded after a suitable
excitatory stimulus. The rising AP upstroke reflects the rate of depolarisation, whereas final
rapid AP repolarisation reflects the rate of repolarisation. Even higher resolution is
provided by testing the effects of
Page 74
rapid changes in membrane voltage (voltage clamp) in single cardiac cells, and then
recording the associated current – whole cell or macroscopic current.3 Finally, most
resolution is provided by the application of a highly polished microelectrode to a single cell
to isolate electrically a small patch of the cell membrane (patch clamp).4 Patch clamp
permits resolution of the opening and closing transitions of individual ion channels.
Properties of cardiac ion channels
The basis for normal excitability (pacemaker potential) and conduction (AP) is the
passage of Na+, K+, Ca2+, and Cl- ions across the cardiac plasma membrane (sarcolemma)
through specialised voltage and ligand operated ion channels.1 2 5–7 Ion channels are large
membrane spanning proteins. They provide relatively low resistance pathways for the
passive (not requiring energy) movement of ions across the lipid bilayer. In contrast, active
ion transport mechanisms – such as the membrane Na+/K+ exchange pump – do require
energy from the hydrolysis of adenosine triphosphate (ATP). Transmembrane ion fluxes
probably occur through aqueous pores within the channel protein. It is hypothesised that
ions move in single file through the channel pores in a discontinuous fashion, “hopping’’
between discrete selective binding sites. The variable size of hydrated ions and preferential
binding possibly explain ion channel selectivity.1
Gating of cardiac ion channels
Based on the process that accounts for opening and closing (gating) of cardiac ion
channels, the channels fall into three classes: (1) voltage gated ion channels; (2) ligand or
receptor gated ion channels; and (3) stretch activated ion channels.1 2 These mechanisms are
not mutually exclusive. Neurotransmitters and hormones may modulate voltage gated
channels, whereas ligand gated channels may show voltage dependence.
Voltage gated ion channels

Voltage gated channels are the largest group of cardiac ion channels. They have charges
(dipoles) that sense changes in membrane potential, causing a conformational change in
charged ion channel regions, generating small “gating” currents.8 These precede voltage
dependent changes in ion conductance through the ion specific channel. Voltage gated ion
channels occupy open or closed conductance states; these are termed the resting state. Most
resting channels open in response to depolarisation. The open channel permits ion
movement across the cell membrane, with the direction determined by a particular ion’s
electrochemical gradient (Table 3.1).
The number of closed states preceding channel opening determines the activation
kinetics. With a single closed state, the rate of ion conductance
Page 75
Table 3.1 Ion concentration and equilibrium potentials in cardiac fibres
Ion Extracellular Intracellular Ratio Ei

(mmol/l) (mmol/1) (E/I) (mV)
Na+ 145 15 9.7 +60
K+ 4 150 0.027 –94
Cl- 120 5 24 –83
Ca2+ 2 0.0001 20000 +129
increase is maximal at the onset of depolarisation (Fig 3.1a). With multiple closed states,
there is a delay in the rise in conductance (Fig 3.1b). Some ion channels maintain their high
state of conductance for the duration of depolarisation. With others, conductance declines
despite persistent depolarisation as a result of channel inactivation (Fig 3.1c).
The inactivated state is a second non-conducting state, different from the resting state,
because inactivated channels do not usually open in response to membrane depolarisation.1
Some resting channels bypass the open state, and pass directly into the inactivated state. To
reverse inactivation, the membrane must be repolarised to highly negative values – termed
“recovery from inactivation”. This process is highly voltage dependent, and hastened by
membrane hyperpolarisation (that is, repolarisation to a level of resting membrane potential
that is more negative than normal). Further, the nature of the voltage dependence of
inactivation is characteristic for each channel type. Factors such as temperature, pH, and the
concentration of divalent cations may influence the voltage dependence of inactivation.
Hodgkin–Huxley model of Na+ channel gating

Nearly 50 years ago, Hodgkin and Huxley proposed a model for the Na+ channel to
account for the transient nature of Na+ current in response to depolarisation.9 Despite
extensive studies using more complex methods, the basic tenets of their proposal still
account for more than 95% of Na+ channel behaviour.1 Hodgkin and Huxley suggested that
ion movement through the Na+ channel is controlled by activation (m) and inactivation (h)
gates. The behaviour of the channel in response to depolarisation is illustrated in Fig 3.2. A
similar model may hold true for other ion channels.1
Ligand gated ion channels

Neurotransmitters, ATP or Ca2+ ions are ligands for membrane bound receptors and gate
ion channels.1 2 The K+ selective ion channel opened by muscarinic (M2) receptor
activation is a well studied example.10 Acetylcholine (ACh) interacts with the muscarinic
receptor initiating a sequence of events that result in the dissociation of the α and β–γ
subunits of a guanine nucleotide, regulatory G protein – Gi. The α and β–γ subunits of
Page 76
Fig 3.1 Depiction of membrane current (IM) waveforms generated in response to a voltage step form the
holding potential (VH) to test potential (VT) in voltage gated ion channels. (a) Non-inactivating channel with a
single closed state (C) preceding opening (O). Ion conductance (g) is a function of a single activation variable
(n) and maximal from the outset, (b) Non-inactivating channel with several closed states (C1, C2, C3). Ion
conductance remains a function of a single activation variable; however, there is a delay in the rise in ion
conductance. (c) Inactivating channel with multiple closed states and a single inactivated state (I). Activation
(m) and inactivation (h) variables are required for a full description of conductance changes. As in (b), there is
a delay in the rise in ion conductance to a maximal value. Current then declines to zero at I, despite
maintenance of voltage step depolarisation.
Page 77
Gi open a K+ channel.10 This process is rapid, being limited by hydrolysis of ACh in the
synaptic cleft and desensitisation – a sequence of intracellular processes that collectively
result in a diminution of the effect of ACh receptor interaction, despite the continued
presence of ACh.
Other ligands bind to receptors to close an ion channel. The K+ ATP sensitive channel is
closed by normal intracellular levels of ATP; however, a reduction in intracellular ATP
concentration to below 0.5 mmol/1 results in channel opening.11 Intracellular pH and the
ADP/ATP ratio modulate the sensitivity of the K+ ATP channel to ATP. Activation of this
channel by
Fig 3.2 Hodgkin-Huxley model of Na+ channel gating. Depiction of membrane current (IM) and single
channel gating behaviour in response to a step in cell membrane potential (VM). The Na+ channel is shown as
a pore spanning the sarcolemma bilayer in resting, open, and two inactivated states. (1) Resting state: the
activation or m gate is closed, and the inactivation or h gate open. (2) Open state: upon depolarisation, the h
gate remains open and the m gate opens to allow movement of Na+ into the cell. (3) First inactivated state: the
h gate then closes to block further Na+ movement into the cell. (4) Second inactivated state: when the
membrane potential has returned to its resting level, the m gate also closes. After a variable time interval, the
h gate moves into the open position – a process termed “recovery from inactivation” – returning the Na+
channel to the resting state.
Page 78
low ATP during myocardial ischaemia shortens the duration of the action potential (APD)
of ischaemic fibres, thereby increasing disparity of APD among myocardial fibres. Finally,
physical factors such as stretch can activate ion channels,1 and second messenger systems
are involved in signal transduction.
Structure–function relationships of ion channels
The simplest ion channel carries the slow component of the delayed rectifier current (IK)
in cardiac muscle.12 It consists of a single chain of 130 amino acids spanning the
sarcolemmal membrane. The larger family of voltage gated K+ channels provides the next
higher order of structural complexity.1 Each channel has four subunits, each consisting of
500–1000 amino acids that form six membrane spanning helices. The four subunits may be
the same – a homotetramer – or different – a heterotetramer.13
The voltage gated Na+ and Ca2+ channels have the highest order of complexity.1 6 14–16
They consist of the α, β1, and β2 subunits. The α subunit is, however, sufficient to form a
functional channel. It consists of four homologous domains (I–IV). Each domain has six
sarcolemma spanning segments (S1–S6). The structure of each segment is highly conserved
between domains; however, peptide chains connecting the segments show more variability.
The β subunit of the Na+ channel modulates the level of expression and kinetics of
inactivation of the α subunit,17 the structure of which is shown in Fig 3.3. The fourth
sarcolemma spanning segment has a positively charged amino acid in every fourth position
(Fig 3.3), and acts as the voltage sensor (‘‘m” gate) for the activation process.5 15 The loop
between S5 and S6 extends for some distance in the membrane, and is believed to form the
channel pore. The loop between the third and fourth homologous domains plays a central
role in normal inactivation.5 15
Terminology
Three types of cardiac transmembrane potentials (TMPs) can be recorded by inserting a
microelectrode (tip diameter=0.5 μm) into a single cell:
Resting membrane potential (RMP) – The membrane potential during electrical quiescence
(diastole) in most fast response fibres.
Action potential (AP) – The changes in transmembrane potential (TMP) that occur when the
cell has been brought to threshold for regenerative excitation.
Automatic TMP – Spontaneous, diastolic depolarisation brings some specialised conducting
system cells to threshold for an AP. Instead of an RMP, these fibres have a maximum
diastolic potential (MDP).
APs for quiescent and automatic cardiac fibre types are shown in Fig 3.4, and a
comparison of TMP characteristics for these fibre types in Table 3.2.
Page 79
Fig 3.3 Structure of portion of the α subunit of a Na+ channel. Voltage gated ion channels (Na+, K+, and Ca2+)
are glycosylated proteins. Each subunit of these contains four covalently linked domains (except for K+
channels) designated I – IV – shown are domain III and a portion of domain IV Each domain contains six α-
helical trans-sarcolemmal segments (S1–S6). S4 is thought to represent the voltage sensor (“m” gate) for the
activation process, and the cytoplasmic peptide chain between domains III–S6 and IV–S1 the inactivation or
“h” gate.
TMPs may be depolarised, repolarised, or hyperpolarised with respect to a reference or

immediately preceding level of TMP (Fig 3.5).
Cardiac action potential
The action potential of a quiescent Purkinje fibre has five phases:

1 Phase 0: upstroke or rapid depolarisation phase.
2 Phase 1: early rapid repolarisation.
3 Phase 2: plateau phase.
4 Phase 3: final rapid repolarisation.
5 Phase 4: resting or diastolic membrane potential.
That portion of phase 0 when the membrane potential is positive to 0 mV is termed the
AP overshoot. Cardiac AP phases, major ionic currents
Cellular cardiac electrophysiology

AP action potential LMP loss of membrane potential
AV atrioventricular MDP maximum diastolic potential
DAD delayed after RMP resting membrane potential
depolarisation
DFR depressed fat response SA sinoatrial
EAD early after depolarisation TMP transmembrane potential
Page 80
Terms
Action potential: change in TMP after excitation of cell
Automaticity: spontaneous diastolic depolarisation
Fast response fibre: atrial, ventricular, or Purkinje fibres with AP upstroke

dependent on fast Na+ current
Maximum diastolic most negative diastolic TMP (automatic fibres)

potential:
Resting membrane stable TMP during diastole (most fast response fibres)
potential:
Slow response fibre: SA and AV node cells with AP upstroke mostly

dependent on slow Ca2+ current
Transmembrane potential: electrical potential across cell membrane
responsible for generation of the APs, and ion exchange pumps that help to restore ion
gradients during phase 4 are depicted in Fig 3.6.1 2 19 20
Resting membrane potential (phase 4)

K+ is the major ion determining RMP, because the cell membrane is quite permeable to
K+ during phase 4, but relatively impermeable to other ions. A membrane bound, Na+/K+
exchange pump (Fig 3.6), fuelled by the hydrolysis of ATP, transports three Na+ out of the
cell for two K+ into the cell during phase 4. As this pump generates a net outward
movement of positive charges, it is “electrogenic’’. As a result, [Na+]i and [K+]i are kept
low and high, respectively, with the inside of the cell being negative with respect to the
outside. Indeed, in most cardiac cell types, were it not for a small leak of Na+ ions (Fig
3.6), RMP would approach the equilibrium potential for a K+ electrode ( – 96 mV). Also,
depending on TMP, and Na+ and Ca2+ concentrations inside and outside the cell, a passive
Na+/Ca2+ exchanger can run in the forward or reverse modes to move Na+ or Ca2+ into or
out of cells (Fig 3.6). This exchanger depends partly on maintenance of the Na+
concentration gradient by the Na+/K+ exchange pump. Under normal conditions, one
intracellular Ca2+ is exchanged for three or more external Na+ ions. If [Na+]1 is abnormally
high – for example, with digitalis toxicity – external Ca2+ may be exchanged for internal
Na+.
Action potential
The cardiac AP is a propagating wave of transient depolarisation, which begins when an
excitatory stimulus (propagating AP, an external stimulus) depolarises the cell membrane
beyond threshold potential. The AP is inscribed by the movement of Na+, K+, Ca2+, and Cl-
ions through at
Page 81
least nine distinct voltage or ligand operated ion channels.1 The Na+/K+ pump and
Na+/Ca2+ exchanger also contribute to the genesis of the AP by: (1) maintaining membrane
ion gradients essential for excitability; (2) generating small currents as the result of net ion
movements; and
Fig 3.4 Schematic representation of action potentials (APs) from various cardiac fibre types. Approximate
timing of APs in relationship to events of surface ECG lead II and His bundle ECG (HBE) for one cardiac
cycle is shown. Note that SA and AV nodal activation is electrically silent in both ECG II and HBE.
Page 82
Table 3.2 Comparison of transmembrane potentials and other action potential
characteristics from various cardiac fibre typesa
SA node Atrial AV node Purkinje Ventricular
RMP or MDP –50 to – –80 to – –60 to – –90 to – –80 to –90

(mV) 60 90 70 95
Amplitude (mV) 60–70 110–120 70–80 120 110–120
Overshoot (mV) 0–10 100–200 5–15 30 30
Duration (ms) 100–300 100–300 100–300 300–500 200–300
Upstroke (V/s) 1–10 100–200 5–15 500–700 100–200
Conduction (m/s) <0·05 0·3–0·4 0·1 2–3 0·3–0·4
aData from Sperelakis.18 SA, sinoatrial; AV, atrioventricular; RMP or MDP, resting
membrane potential (quiescent fibres) or maximum diastolic potential (automatic fibres).
Upstroke, maximum upstroke velocity.
(3) restoring normal intracellular ion concentration once the AP is inscribed.
By convention, inward (depolarising) currents during the AP reflect movement of
positive charges into the myocyte. INa and ICa (below) are the major physiological inward
currents. Outward currents reflect movement of positive charges out of the cell, and
repolarise the cell during AP phase 3. K+ ions are the major charge carrier for outward
current in heart.
Fig 3.5 Terminology used to describe changes in transmembrane potential (TMP) during cardiac action
potential (AP). Depolarisation is towards a more positive (low) level of TMP, and repolarisation towards a
more negative (high) level of TMP. Note that action potential “B” arises from a depolarised membrane
potential compared with action potential “A”. As a result, some Na+ channels are inactivated, less Na+ current
can flow, and action potential ‘‘B” has a slower rate of rise and less overshoot compared with action potential
“A”. See text for further discussion.
Page 83
Fig 3.6 Transmembrane currents, ion exchange pumps, and refractory periods in a typical fast response fibre.
Inward currents (black) include the fast Na+ current (INa); transient or tiny Ca2+ current (ICa,T); long-lasting or
large Ca2+ current (ICa,L); and a non-selective cation current (INS) carried by a channel activated by
intracellular Ca2+ loading. Outward currents (grey) include the inward rectifier current (IK1); delayed rectifier
current (IK – shown as the sum of its two components); a rapidly activated, time dependent, non-Ca2+
dependent K+ current (Ito,1); a transient current that is dependent on [Ca2+]i, and carried by C1- in some
species (Ito2); and a hyperpolarising K+ current activated by acetylcholine (IACh). A chloride conductance (ICl)
dependent protein kinase A can be activated in some species. Typical intra- and extracellular ion
concentrations during phase 4 are shown (top right). The predominant intracellular anions (A-) are large,
impermeable proteins. The ATP dependent Na+/K+ pump maintains the steep outwardly and inwardly directed
gradients for K+ and Na+, respectively, and also generates a small net outward current. The passive Na+/Ca2+
exchanger generates a small net inward current. The resting membrane potential (Vr) results from high
membrane permeability to K+ relative to other ions and the transmembrane concentration gradient for K+. A
small inward “leak” of Na+ ions keeps Vr slightly positive to the K+ equilibrium potential. Finally, the fibre
cannot be excited during the absolute refractory period (ARP). During the relative refractory period (RRP),
excitation produces action potentials with reduced amplitude, slower upstrokes, and no overshoot (AP “a”).
These either fail to propagate or do so slowly. Following the RRP, the threshold potential (TP) returns to
normal, so that a fully regenerative AP (AP “b”) occurs with excitation. See the text for further discussion.
Page 84
Finally, we must define the term “rectification’’ before describing ionic mechanisms for
the AP. Rectification describes the voltage dependence of resistance to ion flow through
some ion channels. For example, the delayed rectifier K+ current (IK) and transient outward
currents (Ito1 and Ito2) (Fig 3.6) exhibit outward going rectification, that is, more current
flows with increasing depolarisation. In contrast, current flow progressively decreases with
increasing depolarisation for the inward rectifier K+ current (IK1).
Phase 0: AP upstroke (rapid depolarisation)

There are two types of fibres, depending on the primary mechanism for generation of the
AP upstroke – “fast” (Na+) and slow (Ca2+) response fibres. The first includes atrial,
Purkinje, and ventricular muscle fibres, and the second SA and AV nodal cells.
In fast response fibres, depolarisation increases the probability of Na+ channel activation.
It must be sufficient to bring the membrane to threshold potential (TP), where enough
inward Na+ current is activated to overcome the repolarising influence of the outward K+
conductances. TP ranges from – 70 to – 65 mV in normal Purkinje fibres. Once reached, the
escalating influx of Na+ ions causes regenerative depolarisation – whereby movement of a
little Na+ into the cell further depolarises the cell membrane, allowing more and more Na+
to enter the cell. Depolarisation approaches, but never actually reaches, the Na+ equilibrium
potential (+ 70 mV). Purkinje fibres, with the highest AP upstroke velocities (see Table
3.2), have a significantly higher density of Na+ channels than atrial or ventricular muscle
fibres. AP upstroke velocity and amplitude are major determinants of myocardial
conduction velocity. Finally, smaller depolarising stimuli that do not bring fast response
fibres to threshold for regenerative excitation can result in non-propagated APs (local or
“electrotonic effects”). Non-propagated APs, however, may impair conduction of
subsequent propagating APs.
Slow response fibres have a lower RMP, slower upstrokes, and little or no overshoot (see
Table 3.2 and Fig 3.4). Depolarisation during phase 0 is dependent primarily on the slow
inward current (Isi) carried predominantly by Ca2+ (ICa). Its threshold for activation is
about –30 to –40 mV. In fast response fibres, this current is activated during phase 0 by
regenerative depolarisation caused by INa. Although current flows through both the fast
(INa) and slow channels (Isi) during the second half of phase 0, Isi is much smaller than
peak INa, and therefore contributes little to the AP until after inactivation of INa (following
completion of phase 0). In addition, Isi can be activated and may play a prominent role in
incompletely depolarised fast response fibres (“depressed fast response” – see below) in
which INa has been inactivated, provided that conditions are appropriate for Isi activation.
Isi passes through protein membrane channels that are selective for Ca2+. Two types of
Ca2+ inward current exist in cardiac fibres:
Page 85
1 A slowly inactivating, high threshold, dihydropyridine sensitive current (long lasting,
large or “L type” Ca2+ current – ICa,L). ICa,L contributes to depolarisation and impulse
propagation in slow response fibres. In fast response fibres, ICa,L contributes to the AP
plateau and triggering the release of Ca2+ from the sarcoplasmic reticulum.2 Ca2+
channel blockers block ICa,L, but it is stimulated by drugs that increase cyclic AMP
levels (β-adrenergic agonists and phosphodiesterase inhibitors).
2 A fast inactivating, low threshold, dihydropyridine insensitive current (transient, tiny, or
“T type” – ICa,T). ICa,T is activated at thresholds intermediate between those for INa and
ICa,L. It probably contributes inward current to the later stages of phase 4 depolarisation
in SA node cells and Purkinje fibres.2
Phase 1: early rapid repolarisation

In fast response fibres, the AP upstroke peak is followed by rapid, early repolarisation
(phase 1). This brings the membrane potential back to + 10 ± 10mV.1 2 Phase 1 results from
inactivation of INa along with activation of Ito. Ito consists of at least two components:21
1 Ito1 is a voltage dependent, rapidly activating K+ current with outward going rectification.
It undergoes voltage and time dependent inactivation, and its channels are notably slow
to recover from inactivation – diminishing its importance to phase 1 at fast heart rates.
2 Ito2 is believed to be a Ca2+ dependent Cl- current.22 Ito2 activation correlates with Ca2+
release from the sarcoplasmic reticulum, triggered by Ca2+ influx from L type Ca2+
channels.
There is marked regional variation in the prominence of Ito2. For example, AP of
ventricular epicardial cells have prominent Ito2 and phase 1 repolarisation. In contrast, Ito2
is small and phase 1 negligible in endocardial cells.23 Cl- ions, along with Ito2, may
contribute to early rapid repolarisation during adrenergic stimulation via rapidly activating–
non-inactivating Cl-current.24
Phase 2: AP plateau
The AP plateau phase (phase 2) may last several hundred milliseconds. Membrane
conductance for all ions falls to rather low levels.1 2 Decreased Na+ conductance (caused by
inactivation of INa) along with decreased outward K+ conductance (consequent to IK1
inward going rectification – so that little K+ leaves the cell) are primarily responsible.
Minor currents during phase 2 include:
1 Ca2+ current (ICa,T, ICa,I): current through T type channels, activated at cell membrane
potentials positive to – 70 mV, is negligible, but may contribute to the pacemaker
potential (see below). Current through the
Page 86
L type channels, activated at cell membrane potentials positive to – 40 mV, inactivates
slowly. Its primary role is to trigger the release of Ca2+ from the sarcoplasmic reticulum
to initiate contraction.
2 ATP dependent Na+/K+ pump: this does not turn on and off with each AP, but instead
restores the ionic gradient over a cumulative time period. It pumps 3 Na+ ions out for 2
K+ ions into the cell.
3 Cl- current: Ca2+ dependent Cl- current (Ito2) also contributes to phase 2.22
4 Slowly inactivating or “late” Na+ current: a small, persistent inward Na+ current that
continues to flow during phase 2 in ventricular muscle and Purkinje fibres.25 It is
unknown whether it results from delayed or failed inactivation of some Na+ channels
(INa), or is carried by a distinct subpopulation of Na+ channels that open with long
latencies.1 Inhibition of this “late’’ Na+ current is thought to be responsible for the
shortening of the ventricular AP duration by lidocaine (lignocaine) and other class IB
antiarrhythmic drugs.26
5 Passive Na+/Ca2+ exchanger: this exchanger generates a small membrane current by
virtue of its 3:1 Na+:Ca2+ transport ratio.27 The direction of the current flow is
determined by the relationship of the membrane potential to the Na+/Ca2+ equilibrium
potential. Inward current (Na+ influx/Ca2+ efflux) flows at rest because RMP is negative
to ENa–Ca (– 30 to –40mV). Outward current flows (Na+ efflux/Ca2+ influx) towards the
end of phase 0 and during phase 1 because RMP is positive to ENa-Ca-As [Ca2+]i
increases (phase 2), ENa-Ca becomes more positive and the exchanger reverts to its Na+
influx/Ca2+ efflux mode.
Phase 3: final rapid repolarisation

Progressive decay of ICa,L with increased outward K+ current terminates phase 2 and
initiates phase 3.1 2 Outward current is generated by IK (early phase 3) and IK1, along with a
small contribution from the ATP dependent Na+/K+ pump. IK1 channels, closed during
phase 2, progressively reactivate during phase 3 to cause regenerative repolarisation back to
the RMP.
Automaticity: phase 4 (diastolic) depolarisation
Pacemaker fibres in the SA node and latent (also subsidiary or secondary) pacemakers
spontaneously depolarise during phase 4 towards threshold for regenerative excitation –
termed “automaticity”.
Sinus node automaticity

The SA nodal action potential with underlying currents is shown in Fig 3.7.20 From a
maximum diastolic potential of – 50 to – 60 mV, the cell undergoes slow diastolic (phase 4)
depolarisation. This phase merges smoothly with the AP upstroke (phase 0), and there is no
distinct AP overshoot, phase 1, or phase 2 (Fig 3.7).
Page 87
SA node automaticity must be the result of a net gain in intracellular positive charges
during diastole.1 2 Contributing to this change is a voltage dependent channel activated at
membrane potentials negative to –50 to –60 mV, with a reversal potential of around –20
mV.This pacemaker current (If) is carried mostly by monovalent cations (Na+, K+).
Hyperpolarisation increases the If rate of activation, so that it carries only about 20% of
phase 4 depolarising current in SA nodal cells. Therefore, automaticity is primarily
dependent on IK, ICa,L, ICa,T, and an unidentified background current (Ib). Repolarisation
results in the closure or deactivation of the K+ channels (IK) opened during the preceding
AP. This causes K+
Fig 3.7 Membrane currents underlying SA node automaticity and the modulation of automaticity. Inward
currents (black) include the pacemaker current (If), background current (Ib), transient Ca2+ current (ICa,T), and
long lasting Ca2+ current (ICa,L). Outward currents include two inward rectifying K+ currents – the delayed
rectifier (IK) and a hyperpolarising K+ current that is activated by vagal stimulation (IACh). (1) The rate of SA
node discharge can be increased by increasing the slope of phase 4 depolarisation. (2) The rate can be slowed
by reducing the slope of phase 4 depolarisation, (3) reducing threshold potential (TP), or (4) by increasing
maximum diastolic potential (MDP).
Page 88
conductance to decline during phase 4, contributing to net depolarising current. The
proposed Na+ background current (Ib) is carried by a voltage independent channel in SA
nodal cells – one that is different from the one responsible for INa. Finally, ICa,L and ICa,T
contribute to the SA node AP upstroke.
Purkinje fibre automaticity

The MDP of automatic Purkinje fibres is hyperpolarised ( – 70 to – 90 mV) compared
with that of SA nodal cells.1 2 Consequently, If is the major pacemaker current in automatic
Purkinje fibres, and INa is also the primary carrier for the AP upstroke.
Modulation of automaticity
The intrinsic rate of SA nodal pacemaker discharge is determined by: (1) maximum
diastolic potential (MDP); (2) threshold potential (TP); and (3) the slope of phase 4
depolarisation (Fig 3.7). Vagal stimulation slows SA node automaticity by: (1) activation of
inward rectifying, hyperpolarising K+ current (IKACh), (2) reduction of ICa,L, and (3)
reduction of If.28 IKACh hyperpolarises MDP and shortens AP duration. This, together with
the reduction in If, slows the rate of phase 4 depolarisation, causing it to take longer to
reach TP for a regenerative AP (Fig 3.7). Reduced ICa,L slows the rate of phase 0
depolarisation. Catecholamines accelerate automaticity by moving the potential for If
activation to a more positive value, and by increasing IK and ICa – both increase the slope
of phase 4 depolarisation.
Subsidiary (latent) pacemakers

In addition to Purkinje fibres, cells in certain portions of the atria (sulcus terminalis,
Bachman’s bundle, coronary sinus ostia), muscle of the tricuspid and mitral annuli, and
distal AV node cells may undergo gradual phase 4 depolarisation, The SA node discharge
normally exceeds that of these pacemakers, so that they are kept from reaching threshold
potential. This is termed “overdrive suppression” of automaticity. The mechanisms for
automaticity in latent pacemakers are probably similar to those for SA node cells and
Purkinje fibres, with the actual mechanism most dependent on fibre MDP. Compared with
the SA node, latent pacemakers are not as subject to the influence of neurotransmitters and
the autonomic nervous system.
Impulse propagation
The SA node impulse must be propagated to excite the rest of the heart. AP propagation
is dependent on excitability, conduction, and refractoriness.
Page 89
Excitability
The term excitability describes the ease with which a regenerative AP capable of
propagation to the rest of the heart is initiated in response to a depolarising stimulus.29
Excitability depends upon the complex interplay of membrane active (“generator”) and
passive (“resistive”) properties.30 The former include: (1) the kinetics and amplitude of
ionic currents responsible for the AP; (2) threshold potential for activation of INa (fast
response fibres) or ICa (slow response fibres); and (3) the difference between RMP (or
MDP) and TP, which determines the amount of current needed to evoke an AP. Passive
properties refer to the impedance elements or cable properties of the myocyte: (1)
membrane resistance, (2) intra- and extracellular resistance, and (3) membrane
capacitance.31 32 The last comprises the myoplasmic and gap junctional channel resistances.
Gap junctions connect adjacent myocytes, and provide low resistance electrical coupling
between cells by establishing aqueous pores that directly link their respective cytoplasms.2
Gap junctions permit the multicellular heart to function electrically like an orderly,
synchronised, interconnected unit. Electrolyte abnormalities (especially K+ and Mg2+
imbalance), antiarrhythmic drugs, general and local anaesthetics, and ischaemia may
influence excitability by affecting one or more of its determinants.1 2 29 30 33 34
Conduction
AP propagation by local circuit currents is best understood by assuming that electrical
behaviour of a small segment of cell membrane – the unit membrane – is equivalent to
resistance (Rm) and capacitance (Cm) arranged in parallel.35 Resistive properties arise from
the presence of ion channels, and capacitance properties from the lipid bilayer – as a result
of its hydrophobic core and polar cytoplasmic and extracellular surfaces. The entire cell is
then considered as a cable, consisting of many unit membranes – each coupled to adjacent
units by the resistances of cytoplasm (Ri) and the extracellular fluid (Ro).
Local circuit current is initiated by INa or ICa, depending on cell type. The resulting
influx of positive charges moves longitudinally through the cytoplasmic resistance Ri and
discharges the capacitance Cm of the adjacent non-excited membrane segment This brings
it to threshold for activation of Na+ or Ca2+ inward current. The local circuit is completed
when positive charges (released to the extracellular space by discharge of Cm) and flow
through gap junctions (see above) return to the initiating active membrane segment through
Ro. Local currents within the cell are carried by K+ ions, which are by far the most
predominant intracellular cation.
Propagation between adjacent cells occurs by specialised, low resistance, gap junctional
channels (see above), which are located most densely at the ends of ventricular muscle and
Purkinje fibres.1 2 Gap junctions are probably partly responsible for the fact that conduction
in the heart is
Page 90
anisotropic – namely, its anatomical and biophysical properties vary according to the
direction in which they are measured. Usually, conduction is two to three times faster
longitudinally than transversely (perpendicular to long axis). However, the safety factor for
conduction is greater transversely, and conduction delay or block occurs more commonly in
the longitudinal direction. As a result of anisotropy, propagation is discontinuous and can
be a cause of re-entry.
The velocity of impulse propagation is determined by several factors.1 2 Active
membrane properties include: (1) the amplitude and kinetics of INa and ICa; and (2) the
difference between RMP and TP. Passive properties include: (1) the excitability threshold;
(2) Ri and gap junctional resistance; and (3) cross-sectional area of the cell. Conduction
velocity may be increased by factors that enhance the magnitude of inward current during
phase 0: (1) hyperpolarisation of RMP or MDP; (2) increase in Rm – for example, with
hypokalemia; or (3) decrease in Ri – for example, with catecholamines. Finally, direction of
impulse propagation is crucial to the influence of anisotropy – as mentioned earlier.
Refractoriness
Refractoriness describes the ability to re-excite a cell (or tissue) after previous excitation.
A cell may be absolutely (no response to stimulation), relatively (partial response), or no
longer refractory (normal response). The state of refractoriness depends on Na+ or Ca2+
channel availability in fast or slow response fibres, respectively, as well as the ability of the
depolarising stimulus to bring the membrane potential to threshold for activation of the
available Na+ or Ca2+ channels. In fast response fibres, almost all Na+ channels are
inactivated beginning with the AP upstroke and continuing through AP phase 2 to phase 3.
Consequently, a regenerative AP cannot be initiated regardless of stimulus strength. Once
the cell has repolarised during phase 3 to about –50 mV, however, an increasing proportion
of Na+ channels recovers from inactivation as a function of time and voltage. This defines
the end of the absolute refractory period (ARP), and the beginning of the relative refractory
period (RRP) (see Fig 3.6). During the RRP, a regenerative AP may occur, but this requires
a suprathreshold stimulus. Also, APs initiated during the RRP have a reduced AP upstroke
velocity, amplitude, and overshoot, and are conducted more slowly (see Fig 3.6).
In fast response fibres, the time course for recovery of excitability closely follows that
for membrane repolarisation – usually full recovery within approximately 50 ms of
achievement of the normal RMP. This reflects the rapid time course for Na+ channel
recovery from inactivation. Deactivation of K+ repolarisation currents may also, however,
affect Na+ channel recovery from inactivation and excitability. Delmar36 has shown that
dynamic changes in IK and IK1, by altering Rm, exert an important role in determining the
ease with which the membrane depolarises to threshold.
Page 91
Further, IK inactivates slowly during the AP, so that complete deactivation can extend
beyond restoration of the normal RMP. If so, current through still open IK channels may
oppose inward stimulating current, despite full recovery of the Na+ channels from
inactivation.36
The time course for recovery from inactivation is much slower for Ca2+ channels of slow
response fibres (SA and AV nodes), so that relative refractoriness outlasts the return to
MDP by more than 100 ms. Fast response fibres that have been partially depolarised by
ischaemia or other factors that slow recovery of Na+ channels (for example, long acting
local anaesthetics and class 1C antiarrhythmic drugs) may also exhibit prolonged post-
repolarisation refractoriness.
Abnormal cardiac electrophysiology

Abnormal cardiac electrophysiological phenomena include impaired conduction and
uneven refractoriness associated with the depressed fast response, and repetitive impulse
formation arising de novo in fibres that do not normally exhibit automaticity (that is,
abnormal automaticity) or triggered by early or delayed after depolarisations, and re-entry
of excitation. Such abnormal phenomena result from the disruption of normal cardiac
electrophysiological processes, as a result of either altered physiological states or
myocardial disease. The first is considered to be “imbalance’’, implying that the cause is at
least potentially reversible (see box). The second is considered a “substrate” – a more or
less fixed defect that is conducive to arrhythmogenesis (see box).
Loss of membrane potential and depressed fast response
Loss of membrane potential (LMP) is the partial depolarisation of fast response fibres
resulting from a metabolic abnormality (for example, ischaemia, hyperkalaemia). LMP
reduces Na+ channel availability.2 LMP reduces AP upstroke velocity, amplitude, and
overshoot, and prolongs conduction time of propagating AP – even to the point of block.
APs with their upstrokes dependent on INa flowing through partially inactivated Na+
channels are called depressed fast responses (DFR). DFR AP contours resemble those of
slow response fibres (see Fig 3.4). AP changes with LMP are likely to be heterogeneous, as
a result of variable inactivation of the INa. As a result, there is uneven conduction and
refractoriness – conditions especially favourable for re-entry arrhythmias. LMP may also
contribute to the genesis of abnormal automaticity and triggered activity (see below).
Cellular mechanisms for arrhythmias
Arrhythmias are disorders of impulse initiation (automaticity, triggered activity – from

early or delayed after depolarisations), propagation, or both
Page 92
Myocardial imbalance
Autonomic imbalance
Electrolyte imbalance
Metabolic imbalance
Acid–base imbalance
Temperature extremes
Adverse drug effects
Adverse PCD effects
Drug interactions
Myocardial ischaemia
Too light anaesthesia
Too deep anaesthesia
Hypo- or hypercapnia
Cerebral hypoxia
Micro shock
Myocardial substrates
Healed myocardial infarction
Hypertrophic cardiomyopathy
Dilated cardiomyopathy
Restrictive cardiomyopathy
Sick sinus syndrome
Ventricular pre-excitation (WPW)
Pericarditis, myocarditis
Congenital long QT syndrome
PCD, pacemakers and cardioverter defibrillators; WPW, Wolff–Parkinson–White

syndrome.
(for example, a focus of abnormal automaticity protected by entrance block – parasystole).
Automaticity may be normal (SA node, latent pacemakers), altered normal, or abnormal.
“Altered normal” simply means that the ionic mechanisms for automaticity are normal;
only the rate of automaticity has changed as a result of an increase or decrease in current
magnitude. Reentry is the only disorder of impulse propagation, although some authorities
also include deceleration dependent block (block at slow heart rates – possibly caused by
reduced AP amplitude or excitability), tachycardia dependent block (resulting from
refractoriness caused by incomplete recovery of excitability), and decremental conduction
(whereby properties of a fibre change along its length so that the propagating AP loses its
efficacy as a stimulus to excite the fibre ahead of it).2
Abnormal automaticity
Abnormal automaticity is different from normal automaticity because: (1) it occurs in

fibres that do not normally exhibit automaticity; or (2) ionic
Page 93
mechanisms for abnormal automaticity are dissimilar to those for normal automaticity in
the same fibre type (for example, Purkinje fibres). It occurs in fibres with reduced MDPs –
often at potentials positive to –50 mV, when ICa,L and IK may be operative.2 Electrotonic
effects from surrounding, more depolarised or normally polarised myocardium will
influence the development of automaticity. Application of constant depolarising current can
produce abnormal automaticity in atrial and ventricular muscle, and Purkinje fibres. If so,
loss of membrane potential, along with a background inward current provided by declining
IK and activating ICa, possibly explain abnormal spontaneous phase 4 depolarisation.1 2
This mechanism has been observed in experimental canine infarction, rat myocardium
damaged by epinephrine (adrenaline), diseased human atrium, and ventricular myocardium
from patients undergoing ventricular aneurysectomy or endocardial resection for
tachyarrhythmias.2
Triggered activity
Triggered activity is repetitive impulse formation that is initiated by depolarising

oscillations in membrane potential induced by one or more APs. These oscillatory
potentials are also referred to as after potentials or after depolarisations. After
depolarisations can occur before or after full repolarisation of the fibre (Fig 3.8).Those
arising during phase 2 or 3 of the AP are termed “early after depolarisations” (EADs),
whereas those that arise after full repolarisation (phase 4) are termed “delayed after
depolarisations’’ (DADs).37 Not all after depolarisations reach the threshold for
Fig 3.8 Depiction of linear re-entry in a fibre bundle with a local region of depressed excitability (M1), similar
to that first proposed by Schmitt and Erlanger.61 Region M1 is also the zone of unidirectional conduction
block. The impulse propagates in fibre A from left to right, and blocks at x – the proximal border of M1. In
adjacent fibre B, however, excitability is less depressed (M2); therefore, the impulse conducts slowly beyond
the more depressed region (M1) in fibre A. It continues to propagate in fibre B, and also crosses back to distal
fibre A via lateral connections. In fibre A it propagates distally as well as slowly back through the former site
of block – with excitability restored (that is, y to x in M1) to re-excite proximal fibre A, before it is excited by
another impulse arriving from above.
Page 94
regenerative depolarisation, but, if they do, they may trigger another AP and repetitive
impulse formation. Such repetitive impulse formation is termed “triggered activity”
because, in contrast to automaticity, it is critically dependent on prior impulses (EAD,
DAD) or stimulation.2
Early after depolarisations

EADs arise from incompletely repolarised membrane during phase 2 (type 1 EAD) and 3
of the AP (type 2 EAD).1 2 38 EADs and triggered activity are associated with bradycardia
and drugs that prolong the QT interval – acquired long QT syndrome (A-LQTS) –
particularly when associated with hypokalaemia and hypomagnesaemia.38 Quinidine, the
prototype class 1A antiarrhythmic, is well-known to cause QT prolongation and torsade de
pointes ventricular tachycardia attributed to EADs and triggered activity.2 38 39 Other drugs
that have caused torsade de pointes in association with a long QT interval and torsade
include other type 1A and some class III antiarrhythmics, non-sedating antihistamines
(terfenadine, astemizole), and erythromycin.38 40
Not all drugs that prolong the QT interval 38 are associated with torsade de pointes
ventricular tachycardia caused by EADs.39 Class 1C antiarrhythmics also cause QT interval
prolongation and may be associated with ventricular proarrhythmia (the term used to
describe paradoxical ventricular arrhythmia formation with antiarrhythmic drugs 38 39 41–43).
However, with 1C drugs, the polymorphic ventricular tachycardia resembling torsade de
pointes is more likely as a result of re-entry – attributed to depression and delayed recovery
from inactivation of INa, as well as prolonged refractoriness.39 Either facilitates re-entry
(see below), especially in patients with poor left ventricular function, acute myocardial
ischaemia, or pre-existing ventricular arrhythmias, 39 41–43
The mechanism for EADs and triggered activity is uncertain, especially given that EADs
may arise from two ranges of membrane potential (type 1 EAD, potentials < –40 mV; type
2 EAD, potentials between –50 and –70 mV). At least type 1 EAD must involve activation
(or reactivation) of ICa,L in conjunction with delayed repolarisation as a result of block of
IK.2 39 At least with quinidine, however, EADs and triggered activity are enhanced at slower
heart rates.40 44 Quinidine blocks both INa and IK at sites within the channel pores; only
receptor occupancy within the K+ channel is greater at slow rates 44 – possibly as a result of
reduced extracellular K+ accumulation.40 45 This phenomenon – greater block at slower
heart rates – is termed “reverse use dependence”, because this denotes greater block at
faster heart rates.
Finally, EADs (also delayed after depolarisations – see below) and triggered activity, and
torsade de pointes ventricular tachycardia, are associated with QT interval prolongation in
the congenital (idiopathic) long QT syndrome (C-LQTS). QT prolongation is attributed to
defects in the
Page 95
genes encoding K+ and Na+ channels, left sided sympathetic predominance, or both.46–50 β-
Adrenergic stimulation facilitates EADs and triggered activity, and torsade de pointes with
C-LQTS, hence the use of β-blockers for the management of patients with this syndrome.
β-Adrenergic stimulation is expected to increase ICa,L (and amplitude of EAD), and
possibly also inhomogeneity of repolarisation.47
Delayed after depolarisations

In contrast to EADs, DADs are secondary depolarisations that arise from a fully
repolarised membrane.1 2 Triggered activity resulting from DADs has been observed in
atrial and ventricular muscle and Purkinje fibres under a wide variety of experimental
conditions (for example, digitalis intoxication, one day after experimental myocardial
infarction, exposure to catecholamines). It has also been observed in vitro in diseased
human atrial and ventricular fibres. DADs and triggered activity are believed to be
responsible for at least some tachyarrhythmias precipitated by digitalis toxicity, and
possibly accelerated junctional or idioventricular rhythms with acute myocardial infarction.
The mechanism believed to be responsible for DADs involves an increase in [Ca2+]i,
which activates a transient inward current (Iti) mediated either by the Na+/Ca2+ exchanger
or a Ca2+ dependent non-specific ion channel.51 52 This current (Iti) is small or absent
normally, but activated by an increase in intracellular Ca2+ (“Ca2+ overload”),53 54 such as
might occur with myocardial ischaemia or injury, catecholamines, or digitalis excess. Na+
is the primary carrier for Iti. Drugs that (1) reduce diastolic Ca2+ transient, by reducing
Ca2+ overload (Ca2+ channel blockers, β-blockers), (2) inhibit Ca2+ release from the
sarcoplasmic reticulum (caffeine, ryanodine), or (3) reduce INa, and consequently [Na+]i,
inhibit DADs.2
In contrast to the behaviour of EAD triggered activity, pacing at a rate faster than the rate
of triggered activity or premature stimulation increases the amplitude and prematurity of
subsequent DADs.2 This is in contrast to normal automaticity, which is temporarily
suppressed (for example – sinus pause or bradycardia after termination of paroxysmal
supraventricular tachycardia). Further, because a single premature stimulus can both initiate
and terminate DAD triggered activity, the differentiation of this mechanism from re-entry
may be difficult.
Re-entry of excitation
The excitatory wavefront emanating from the SA node continues until all of the heart has
been activated and become completely refractory. If for some reason a group of fibres is not
activated by the propagating impulse, and it returns by another pathway to excite them, this
process is termed ‘‘re-entry of excitation”. It is also termed “circus movement or
reciprocation”,
Page 96
and extrasystoles caused by re-entry reciprocal or echo beats. Three basic criteria for
ascribing abnormal beats or rhythm to re-entry were first formulated by Mines:55 56
1 There must be an area of unidirectional conduction block.
2 The re-entrant pathway must be defined – namely, the movement of the excitatory
wavefront should be observed to progress through the pathway, return to its point of
origin, and then return to re-excite the same pathway.
3 It must be possible to terminate re-entry by interrupting the circuit at some point to rule
out a focal origin (that is, automatic, triggered).
These criteria have been satisfied for many clinical tachyarrhythmias, because it is now
possible to map tachycardia circuits (electrophysiological studies) and interrupt the circuits
by surgical or radiofrequency catheter ablation techniques. Discussed here are the basic
requirements for re-entry (slow conduction, unidirectional block), anatomical and
functional reentry, re-entry involving the SA and AV nodes, and atrioventricular re-entry
with accessory pathways.
Slow conduction
It has long been known that a basic requirement for re-entry is sufficient delay of the
propagating impulse in an alternative pathway to permit proximal tissue a site of
unidirectional block to recover from refractoriness. If so, re-entry would be facilitated by
conduction that was slower than normal. In fast response fibres, the speed of conduction is
dependent on the magnitude of INa during phase 0, and the rapidity with which INa reaches
its maximum. This will depend on the number of available Na+ channels, which in turn is
dependent on the level of TMP. For example, membrane depolarisation in a fast response
fibre to levels of –60 to –70 mV may inactivate half the Na+ channels, whereas
depolarisation to –50 mV or less may inactivate all the Na+ channels.2 Further, time is
required for Na+ channels to regain excitability after inactivation. If so, conduction of a
premature AP can be slowed in distal tissue as a result of decreased Na+ channel
availability and/or incomplete recovery from inactivation.
The amplitude and upstroke velocities of premature AP initiated before full
repolarisation are reduced (see Fig 3.6), as is their speed of propagation. Re-entry may
occur during propagation of premature AP in regions with different AP durations, as a
result of slowed conduction and unidirectional block (see below). Re-entry may also occur
in fibres with persistent low levels of membrane potential (namely slow response or
depressed fast response fibres), caused by slow conduction (reduced Na+ channel
availability) and refractoriness that extends beyond full repolarisation. Finally, coupling
resistance between adjacent cells is another factor
Page 97
that may influence the speed of conduction. As coupling resistance is increased, the speed
of conduction decreases. Increased [Ca2+]i and acidosis are two factors that may increase
coupling resistance.57
Unidirectional conduction block

Unidirectional block of conduction may occur as the result of non-uniform recovery of
excitability, geometric factors, or asymmetrical depression of conduction and excitability.57
Non-uniform recovery of excitability –When an impulse propagates through tissue with
regional differences in refractoriness, conduction may fail in the regions with the longest
refractory periods. These regions will, however, be available for re-excitation, provided the
propagating impulse can somehow return to the site of former block. This is unlikely in
normal working myocardium, even at fast heart rates, because there is still enough time
during diastole when tissue is non-refractory. Re-entry induced by premature extrasystoles
may, however, be facilitated because refractory periods shorten with short cycle lengths
resulting from prematurity or fast heart rates. It follows that the pathways over which the
impulse propagates are also shortened. The amount of refractory period non-uniformity
needed for unidirectional conduction block after a premature impulse may be quite small –
with minimal differences of 10–40 ms reported for atrial and ventricular fibres,
respectively.57
Geometric factors – When a thin bundle of fibres inserts into a larger muscle mass – for
example, the Purkinje fibre–ventricular muscle junction (“junction”) – the insertion point
can be the site for unidirectional block of conduction.57 Such block has not been observed
in normal fibres, even though the anterograde junctional conduction delay may be longer
than in the retrograde direction. With reduced Na+ channel availability, however,
anterograde block may occur at the junction whereas retrograde conduction remains
possible.58 More recent studies indicate that the Purkinje fibre–muscle junction is better
represented by a three dimensional model of overlying two dimensional sheets of fibres,
rather than terminal Purkinje fibres inserting into a three dimensional ventricular muscle
mass.59 If so, activation of the ventricular muscle layer occurs only at specific junctional
sites; otherwise, a considerable resistive barrier exists between the two cell layers.57 It
follows that reasons for unidirectional block between Purkinje and ventricular fibres
include differences in excitability and fibre thickness between the two layers, as well as
increased coupling resistance at sites of unidirectional conduction block.60 Sites where the
cross sectional area of interconnected cells suddenly increases may also be sites for
unidirectional conduction block (for example, ventricular insertion points for accessory
pathways in patients with the Wolff–Parkinson–White syndrome, as well as sites of fibre
branching and junctions of separate muscle bundles).57 Finally, the way in which
myocardial cells are connected to one another influences
Page 98
the speed of conduction, with longitudinal conduction velocity being about three times
transverse conduction velocity.57
Asymmetrical depression of excitability and conduction – There may be an asymmetrical
region of depressed excitability and conduction, which can also result in unidirectional
conduction block (see Fig 3.8), first suggested by Schmitt and Erlanger.61
Anatomical re-entry
The simplest model of re-entry involving a fixed anatomical obstacle (Fig 3.9a) was
proposed by Mines in 1913.55 He suggested that re-entry could occur in an anatomically
defined circuit, provided that conduction was sufficiently slowed along with shortened
refractoriness. An important
Fig 3.9 Schematic depiction of variations on anatomical and functional re-entry. The arrows represent the
crest of the circulating wavefront, and in its wake are areas of absolute (black) or relative refractoriness
(stippled). White areas represent excitable tissue, that is, the “excitable gap” (see also fig 3.10). (a) Circus
movement around a gross anatomical obstacle, as envisioned by Mines.55 (b) Circus movement around the
caval orifices, separated by a zone of functional block, (c) Circus movement in a loop composed of bundles of
fibres having a greater conduction velocity than surrounding tissue. (d, e) Circus movement based on a
combination of an anatomical obstacle (circle) and an adjacent area of diseased tissue (hatched area) with
depressed conduction, (f) Circus movement around a relatively small anatomical obstacle becomes possible as
a result of a shortened refractoriness and reduced conduction velocity – thereby shortening the wavelength of
the circulating impulse. (g) Leading circle concept of re-entry, with no need for an anatomical obstacle. See
the text for further discussion. (Reproduced from Allessie et al,62 with permission.)
Page 99
feature of Mines’ model was the presence of an excitable gap (Figs 3.9 and 3.10). The
excitable gap implies that an impulse originating outside the reentry circuit can penetrate it
and influence the rhythm.57 But, normal pacemakers of the heart are suppressed (that is,
overdriven) during reentrant tachycardia. Consequently, the impulse that can penetrate the
circuit must come from another source (extrastimulation, burst pacing, etc).
Fig 3.10 The excitable gap, and how a single premature stimulus can reset or terminate re-entry tachycardia.
The arrows represent the crest of the circulating wavefront, and the black tails the zones of absolute
refractoriness. Relative refractoriness within the excitable gap (stippling) decreases with increasing distance
from the tail of absolute refractoriness. (a) A premature stimulus reaching the circuit relatively late, when
tissue within the excitable gap has almost regained full excitability. (b) A moment later, the stimulus
wavefront has penetrated the circuit in both directions, and blocks in the retrograde (left) limb as it encounters
the advancing tachycardia wavefront, but advances in the anterograde (right) limb to change the phase or reset
the tachycardia. (c) The premature stimulus reaches the reentry circuit earlier, when tissue behind the tail of
absolute refractoriness is only partially excitable (relatively refractory). (d) This stimulus blocks in both the
retrograde and anterograde limbs as a result, respectively, of the advancing tachycardia wavefront and relative
refractoriness – terminating the tachycardia. (Reproduced from Janse et al,63 with permission.)
Page 100
There are several variations on re-entry involving a fixed anatomical obstacle.57 62 Lewis
and co-workers used rapid atrial pacing or alternating current to initiate re-entrant
tachycardia (that is, atrial flutter) around the caval orifices in dogs (Fig 3.9b), but the
question remained whether such natural obstacles were large enough for sustained re-entry
to occur. Moe’s group suggested that anisotropic atrial conduction would remove the need
for a large anatomical obstacle (Fig 3.9c).64 For example, anatomically distinct, interatrial
or intranodal, preferential conducting pathways, with faster conduction velocities than
surrounding atrial myocardium, could form the loops required for re-entry. It is now
believed that such anatomically distinct pathways do not exist; rather, preferential
conduction is the result of different electrophysiological properties or geometrical
arrangements among the fibres. Allessie and co-workers 62 proposed that re-entry would be
facilitated by an anatomical obstacle with an adjacent area of depressed conduction (Fig
3.9d,e). Finally, it has been suggested that altered electrophysiological properties of fast
response fibres (namely, loss of membrane potential – see above) might reduce the size of
an anatomical obstacle needed for re-entry.57
Anatomically defined re-entry circuits might occur in fibrotic regions of the atria or
ventricles, or in surviving muscle fibres of healed infarctions.65 Critical slowing of
conduction and unidirectional block for re-entry may be caused by loss of membrane
potential (above) with atrial cardio-myopathies,66 or by increased effective axial resistivity
with healed infarcts.67 Defined anatomical circuits are also involved in ventricular
tachycardia as a result of bundle branch re-entry,68 supraventricular tachycardia caused by
atrioventricular re-entry involving accessory pathways,69 and atrial flutter resulting from re-
entry confined to the right atrium.70
Functional re-entry
Functional re-entry lacks restrictive anatomical boundaries, and occurs in contiguous
fibres with heterogeneous electrophysiological properties caused by local differences in
APs.2 65 Dispersed excitability and/or refractoriness, and anisotropic distribution of
intercellular resistances, allow the initiation and maintenance of re-entry.71–73
Leading circle re-entry – Leading circle re-entry (Fig 3.9g) was first described by
Allessie and co-workers in experiments on atrial muscle,74–76 and is an important
mechanism in clinical atrial fibrillation.2 64 Re-entry is initiated by a precisely timed
premature beat in regions that are activated normally at regular spontaneous or paced rates.
Sustained re-entry is made possible by different refractory periods of atrial fibres in close
proximity to one another.75 The premature beat initiating re-entry blocks in fibres with long
refractory periods, although propagating in those with shorter ones. It eventually returns to
the former site of block after excitability has recovered
Page 101
there. The impulse may continue to circulate around a central core that is kept refractory
because it is constantly bombarded by impulses propagating towards it from all sides of the
circuit (Fig 3.9g). The circumference of the smallest leading circle is as small as 6-8 mm,
and is a pathway in which the circulating wavefront is just able to excite relatively
refractory tissue ahead of it.2 64 No, or only a very short, excitable gap exists, which
distinguishes this from other mechanisms for re-entry.2 Therefore, impulses propagating
outside the circuit cannot easily enter the circuit to terminate or reset the tachycardia.77
Theoretically, drugs that prolong refractoriness but do not delay conduction (for example,
dofetilide, ibutilide, bretylium, and sotolol) would slow tachycardia as a result of the
leading circle mechanism, although not affecting re-entry tachycardia with an excitable
gap.2 Conversely, an antiarrhythmic drug that primarily slows conduction with little effect
on refractoriness (for example, class 1 drugs – Na+ channel blockers) would have major
effects on tachycardia with an excitable gap, but little effect on leading circle re-entry.2
These examples illustrate how an antiarrhythmic drug may be targeted against specific
mechanisms for clinical arrhythmias (see below) – the approach advanced by the Sicilian
Gambit.20 78
Random re-entry – Random re-entry is also an important mechanism in atrial
fibrillation.2 64 It occurs when the impulse propagates randomly and continuously, re-
exciting areas that were excited shortly before by another wavelet. Examples of random re-
entry include anisotropic and spiral wave re-entry. Anisotropic re-entry is the result of the
structural features responsible for variations in conduction velocity and the time course of
repolarisation – such as concentration of gap junctions at the ends rather than sides of cells
– which can result in slowed conduction, unidirectional conduction block, and re-entry.
Therefore, in contrast to leading circle reentry, the functional characteristic that leads to re-
entry is not spatial differences in refractory period but rather spatial differences in effective
axial resistivity caused by non-uniform anisotropy.64 Anisotropic re-entry has been shown
in atrial and ventricular muscle, and an excitable gap may be present.2 64 In contrast to
anisotropic re-entry, spiral wave re-entry (scroll wave re-entry in three dimensions) does
not require any permanent heterogeneities for its initiation or maintenance.79–82 Instead,
transient heterogeneities can be created in normally uniform tissue. Similar to anisotropic or
leading circle re-entry, however, electrical activity with spiral waves is organised around a
central fulcrum (the core). But, in contrast to leading circle re-entry, in which the central
core is inexcitable, the central core during spiral wave re-entry is excitable but not excited.
The mechanism of the core of the spiral is the pronounced curvature of the wavefront at the
tip of the spiral, which is higher than the critical curvature for successful propagation. As a
result, the rotating wave cannot invade the core because of its inability to stimulate it, not
because the core is
Page 102
refractory. Another important difference between leading circle re-entry and spiral waves is
the presence of a fully excitable gap with the spiral waves. In the heart these have three
distinct dynamics: (1) stationary, whereby the core remains in the same position, giving rise
to a rhythmic pattern of activation; (2) drifting, whereby the core drifts away from the site
of origin, giving rise to complex and irregular patterns of activation; and (3) anchored,
whereby a drifting core becomes stationary by anchoring to small areas of discontinuity in
cardiac muscle (for example, blood vessel, connective tissue). Finally, “anchoring’’ of a
drifting spiral possibly explains the transition between polymorphic and monomorphic
ventricular tachycardia.80
SA and AV node re-entry

As the result of slow conduction and prolonged refractoriness, the SA node has the
potential for dissociation of conduction – whereby an impulse can propagate in some but
not other fibres, permitting re-entry to occur.2 57 83 The re-entrant circuit may be located
entirely within the SA node or involve both the SA node and the atrium. Supraventricular
tachycardia (SVT) caused by SA node re-entry is generally tolerated better than other types
of SVT resulting from slower tachycardia rates. SVT caused by SA node re-entry accounts
for 3–16% of paroxysmal SVT referred for electrophysiological study.83
Although longitudinal electrophysiological dissociation of the AV node into two or more
pathways has been demonstrated in animal investigations,84 whether such dissociation
accounts for paroxysmal SVT resulting from AV nodal re-entry in humans is more
dubious.2 The presence of dual AV nodal pathways is not disputed,85 86 but whether they are
intranodal and undergo functional longitudinal dissociation, or are extranodal – involving
separate inputs to the AV node – has been the question.2 Results of clinical
electrophysiological investigation, and radiofrequency catheter and surgical ablation, are
now conclusive - the slow and fast AV nodal pathways have their origins well outside the
limits of the compact AV node, and consist of ordinary atrial muscle fibres.2 87–89 Thus,
these pathways are atrial to nodal approaches or connections, not distinct intranodal
pathways. Target sites for ablation of the fast pathway in AV nodal re-entry tachycardia
(AVNRT) are located along the anterosuperior portion of the interatrial septum near the
tricuspid annulus, just proximal to the compact AV node (that is, the anterosuperior atrial
approach – Fig 3.10a).88 89 Target sites for ablation of the slow pathway in AVNRT are
found more posteriorly along the tricuspid annulus, close to the ostium of the coronary
sinus (that is, the posteroinferior atrial approach – Fig. 3.10a). During the common form of
AVNRT, atrial to AV nodal (anterograde) conduction occurs via the posteroinferior (slow)
atrial approach, with AV nodal to atrial (retrograde) conduction over the anterosuperior
(fast) atrial approach. With the uncommon form of
Page 103
AVNRT, the reverse occurs. (The distinction between common and uncommon forms of
AVNRT is moot as far as recognition – P waves are usually non-apparent with either, and
both are narrow QRS tachycardias, or treatment – which is the same for both forms). Either
or both atrial approaches can be selectively ablated by catheter or surgical methods to cure
AVNRT.88 89 In patients with dual AV nodal physiology, anterograde conduction probably
occurs over the anterosuperior atrial approaches during sinus rhythm, and the premature
impulse that initiates AVNRT blocks in this pathway (with longer refractoriness) as well.2
Finally, paroxysmal SVT caused by AV nodal re-entry accounts for 28% to 50% of
paraoxysmal SVT referred for electrophysiological study.90
Atrioventricular re-entry with accessory pathways (see Fig 3.11)

Accessory atrioventricular connections (accessory pathways – AP) are the anatomical
substrate for ventricular pre-excitation and AV reciprocating tachycardia (AVRT) in the
Wolff–Parkinson–White (WPW) syndrome – that is, short PR interval, slurred QRS
complex upstroke (δ wave), prolonged QRS duration, and associated paroxysmal
tachyarrhythmias.2 91–96 APs consist of small fibres that resemble ordinary atrial tissue, and
can bridge the AV groove at any location along the mitral or the tricuspid annulus, except
for the region where the mitral annulus is contiguous with the aorta. Thus, APs provide
abnormal electrical continuity between the atrium and ventricle. AP conduction is normally
rate independent – a contributing factor to the extremely rapid ventricular rates sometimes
seen during atrial flutter and fibrillation. Such rates can be the cause for sudden death, and
the ECG may be difficult to differentiate from polymorphic ventricular tachycardia or
coarse fibrillation. In this regard, however, it should be noted that primary ventricular
arrhythmias are unusual in WPW patients, unless they have other heart disease – often
coronary artery disease.94
In patients who present to the emergency room with paroxysmal SVT, but with no
evidence of ventricular pre-excitation in sinus rhythm, about 30% have AVRT, 60% have
AVNRT, and the remainder have SVT caused by SA or intra-atrial re-entry.92 APs may be
capable of anterograde and retrograde conduction. Alternatively, they may be incapable of
anterograde conduction, and only participate in antidromic AVRT or not at all (see below),
or they may conduct only in the retrograde direction during orthodromic AVRT (see
below). If so, there could be no ventricular pre-excitation, and the AP re-entry would be
“concealed”. As already noted, paroxysmal SVT resulting from concealed AV re-entry is
about half as common as that caused by AV node re-entry.
Finally, AV reciprocating tachycardia can be orthodromic or antidromic, depending on
the direction of impulse propagation in the AP. With the former (Fig 3.11b), activation of
the ventricles during tachycardia is by the
Page 104
Fig 3.11 Depiction of the usual mechanisms for AV node re-entry tachycardia: (a) AVNRT and (b) AV
reciprocating tachycardia AVRT. With the common form of AVNRT, the re-entry loop involves the AV node,
possibly the atria (still debated2), and the anterosuperior and posteroinferior atrial approaches. The latter are
the fast and slow pathways in AVNRT, respectively. With orthodromic AVRT, the anterosuperior and
posteroinferior atrial approaches may or may not participate in tachycardia. Anterograde conduction is from
the atrium (and/or AV nodal approaches) to the AV node, His bundle, left and right bundle branches (LBB,
RBB), and the ventricles – resulting in a narrow QRS tachycardia. See the text for further discussion.
Page 105
normal pathway (AV node to His bundle, etc). The circulating impulse returns to the atria
via the AP. This gives rise to a narrow QRS tachycardia. With antidromic tachycardia (≤
10% of AVRT91), the ventricle is activated by anterograde impulses in the AP, giving rise
to a wide QRS (pre-excited) tachycardia, which can be difficult to distinguish from
ventricular tachycardia. This, however, is quite unusual in the WPW patient without other
heart disease,94 which is worth remembering when confronted with a regular wide QRS
tachycardia in a patient known to have the WPW syndrome.
Conduction block
There can be block of conduction anywhere within the specialised AV conducting

system. SA, atrial, AV nodal, bundle branch, fascicular, and intraventricular block can be
diagnosed or inferred – primary or secondary SA block only – by inspection of the surface
ECG. Diagnosis of tertiary SA block, intra-atrial block, or block within the His bundle
requires catheter electrode recording techniques. SA and AV heart block is classified as:
primary, delayed conduction; secondary, some but not all beats are conducted with (type 1)
or without delay (type 2); tertiary, no beats are conducted to the atria (SA block) or
ventricles (AV block). Advanced secondary SA or AV block is two or more successive
non-conducted beats to the atria or ventricles, respectively. Bundle branch block is block in
the right or left bundle branches. Block in the left anterior or posterior fascicle of the left
bundle branch is fascicular block. Bifascicular block is a block in either both fascicular
branches (that is, left bundle branch block) or one branch and the right bundle branch.
The site of conduction block has clinical and prognostic significance. The ratio of
conducted beats with SA or AV nodal block may be increased with antimuscarinic drugs or
β1-agonists. Intra-atrial and infranodal (AV node) block is not as likely to respond
favourably. Block within or below the His bundle carries a worse long term prognosis than
block within the AV node or higher. Further discussion of conduction block and literature
citations can be found elsewhere.40
Mechanisms for clinical arrhythmias

Clinical arrhythmias most certainly result from the abnormal cellular
electrophysiological phenomena discussed above. Involvement of a particular mechanism
in specific arrhythmias – with some exceptions – can, however, never be certain because in
vitro cardiac electrophysiological studies cannot model circumstances of the intact heart.
Nowhere is this more true than in perioperative and critical or emergent care situations,
where drugs and imbalance – metabolic, electrolyte, autonomic – combine to affect the
genesis of arrhythmias.33 34 38Thus, a patient with hypertension
Page 106
and healed myocardial infarction from coronary artery disease, who develops monomorphic
ventricular tachycardia (VT) after a difficult tracheal intubation, could have several
mechanisms involved in the genesis and maintenance of VT:
1 Re-entry of excitation: islands of normal myocardium interspersed with regions of
fibrosis provide an anatomical substrate for re-entry, likely to be the mechanism which
sustains monomorphic VT in this setting.
2 Loss of membrane potential and depressed fast response: increased heart rate and blood
pressure act in concert to increase myocardial O2 demand and reduce supply (reduced
diastolic time). This and occlusive coronary disease cause acute myocardial ischaemia
and, in turn, loss of membrane potential and the depressed fast response.
3 Abnormal automaticity: depolarisation induced (abnormal) automaticity in ischaemic
Purkinje or ventricular muscle fibres could be the cause for slow monomorphic VT.
Catecholamines released in response to endotracheal intubation can increase the rate of
abnormal automaticity.
4 Triggered activity: increased [Ca2+]i with ischaemia and catecholamines might underlie
DAD or EAD triggered activity.
5 Re-entry initiated by automatic or triggered beats: finally, automatic or triggered
premature beats might initiate re-entry of excitation involving anatomical (fibrous scar
tissue) or functional re-entry loops (caused by varying depression of the fast response
and inhomogeneous conduction and refractoriness). Ideal drug treatment for VT in this
example is not at all obvious – although treatment may be successful, it can just as easily
produce proarrhythmia (see below).
In contrast to this complexity, the genesis and treatment of paroxysmal SVT caused by
AV node or AV re-entry are far more clear. Re-entry pathways are defined by
electrophysiological mapping, and subsequently ablated by catheter or surgical means to
“cure” SVT. Acute paroxysms of SVT can be readily terminated by drugs that increase AV
node conduction time and refractoriness. So, mindful that cellular mechanisms may vary
for the same disturbance, depending on the setting, postulated mechanisms for specific
clinical arrhythmias are given in the box.
Antiarrhythmic drug action

There is no question that the emphasis on therapy for arrhythmias, especially ventricular
tachyarrhythmias with coronary artery disease, has moved from drugs to
“electricity” (implanted pacemakers, cardioverter defibrillators), or catheter-surgical
ablation. When the heart is structurally normal, antiarrhythmic drugs have proved safe and
effective. However, in the structurally abnormal heart (for example, myocardial infarction
and
Page 107
Postulated mechanisms for specific clinical arrhythmias

Mechanism Arrhythmias
Altered normal automaticity Sinus bradycardia/tachycardia; sinus arrhythmia; AV

junctional/idioventricular escape rhythms; wandering
atrial pacemaker
Abnormal automaticity Slow monomorphic VT; accelerated AV junctional or
idioventricular rhythm with acute myocardial
infarction; some ectopic atrial tachycardia
Triggered activity (DAD) Some VT in first 24 hours after infarction;
atrial/ventricular tachycardias with digitalis toxicity;
catecholamine mediated VT
Triggered activity (EAD) Polymorphic VT insetting of QT interval prolongation
(torsade de pointes)
Re-entry (anatomical) SVT to SA, AV node, or AV re-entry; VT with healed
infarction; atrial flutter (possibly anatomical and
functional)
Re-entry (functional) Atrial fibrillation; monomorphic and polymorphic VT
with acute myocardial infarction; ventricular
fibrillation
congestive heart failure), drug efficacy has been modest. Indeed, their use has been
associated with increased mortality.98–100 Nevertheless, drugs are an important adjuvant to
electrical therapy for management of ventricular tachyarrhythmias, and the mainstay of
treatment for atrial fibrillation.101 In this brief overview of antiarrhythmic drugs, the
decided emphasis is on ion channels as targets for antiarrhythmic drug action.
Antiarrhythmic treatment: role of structure of the myocardium
Antiarrhythmic drug therapy is least effective in structurally abnormal heart.39 41–43 101
The myocardium is terminally differentiated. Replacement microfibrosis and a reduction in
the number of gap junctions follow cell loss between surviving fibres (“myocardial
remodelling”). The implications of changes in the normal distribution of gap junctions for
the genesis of arrhythmias – namely, emergence of discontinuous conduction phenomena
resulting from increased cellular electrical loading – are discussed by Spach and Boineau.102
So, if structural heart disease is a primary culprit in the genesis of arrhythmias, perhaps one
goal of pharmacological treatment should be the prevention of myocardial remodelling
secondary to ischaemia and infarction, hypertrophy, and probably heart failure.101 At least
atrial remodelling seems to be a part of ageing, and may be important in the
Page 108
increased prevalence of atrial fibrillation in elderly people.103 Although development of
means to reduce myocardial remodelling with disease (for example, drugs that influence
gap junction expression 101) has been neglected as part of the antiarrhythmic strategy, and
this is a promising new area for investigation.
Antiarrhythmic treatment: desirable drug properties
Ion channels and neurohormonal receptors that modulate their function, and not
electrophysiological phenomena themselves,98 99 have now become the primary targets for
antiarrhythmic drug action.20 42 78 100 101 104 The goal of drug therapy is, as far as possible, to
restore normal impulse initiation and propagation – or to slow the effective ventricular rate
(chronic atrial fibrillation). The following are desirable properties of an antiarrhythmic
drug:
z specificity: the drug should interact with the specific ion channel (s) involved in the
genesis of the arrhythmia
z pharmacodynamics: the kinetics of association and dissociation should be such that
effects (slowing of conduction, prolongation of refractoriness) are maximal during
tachycardia
z pharmacokinetics: brief onset and duration of action (parenteral administration); high
bioavailability, rapid onset of action, and infrequent dosing (oral administration).101
Specific ion channels as targets for antiarrhythmic drug action
Pacemaker currents
The pacemaker current (If) helps initiate automaticity in SA node cells and subsidiary
pacemaker fibres. Although specific blockers of this current have been identified,105 their
use is quite limited because sinus tachycardia is usually a compensatory rhythm
disturbance. Both T type and L type Ca2+ current also contribute to the pacemaker potential
and conduction in SA node and subsidiary pacemakers, but clinically effective antagonists
(verapamil, diltiazem) block only L type current. Verapamil and diltiazem have low lipid
solubility, and access their intracellular Ca2+ channel binding site predominantly via the
open channel.104 Either is effective for slowing AV conduction with atrial fibrillation-
flutter, and terminating SA and AV node re-entry. Dihydropyridine Ca2+ antagonists
(nifedipine, nicardipine) are neutral compounds that are highly lipid soluble, and ineffective
blockers of L type current in the heart. They occupy receptors on the extracellular surface
of the Ca2+ channel, and bind preferentially to the channel in depolarised tissue (that is, in
its inactivated state).104 As they are neutral and lipid soluble, they also dissociate rapidly,
especially from more polarised tissue. Hence, dihydropyridine Ca2+ blockers are primarily
effective for blocking Ca2+ channels in more depolarised vascular smooth muscle.
Page 109
Na+ channel: fast inward current

Block of the fast Na+ inward current (INa) is useful against re-entrant arrhythmias
involving atrial, ventricular, and Purkinje fibres – with AP upstrokes largely dependent on
INa. There is, however, recognised potential for pro arrhythmia with Na+ channel blockers,
which appears to be greatest with class 1A (quinidine-like) and 1C (flecanide-like) drugs.
These drugs have intermediate or slow dissociation kinetics – consequently, lesser block at
relatively slow heart rates may be substantially enhanced at faster rates (that is, use-
dependent block). This might promote re-entrant tachycardia, and in part explain increased
mortality with 1C drugs in the Cardiac Arrhythmia Suppression Trial (CAST) studies.98 99
The antiarrhythmic efficacy of Na+ channel blockers is determined mainly by how they
interact with the multiple states of the Na+ channel (see Fig 3.2). This has a number of
clinical ramifications. For one, high affinity binding states may be accessible only
phasically, so that drug binding occurs primarily during depolarisation. If so, a fast binding
drug may compete with and displace one with slower kinetics – as shown in a patient with
propoxyphene overdose in whom marked QRS widening was reversed by lidocaine.106
Also, ischaemia or stretch with consequent partial depolarisation may incompletely
inactivate Na+ channels. Compared with primarily open state blockers (disopyramide and
quinidine), drugs that block both open and inactivated Na+ channels (lidocaine and
mexiletine) depress AP phase 0 and conduction more in partially depolarised fibres.105
Consequently, inactivated state blockers may be more effective against reentrant ventricular
tachyarrhythmias with acute myocardial infarction and ischaemia.107 Open channel Na+
blockers penetrate the Na+ channel pore a significant distance to produce block, and Na+
ions bind to specific sites as they transit the pore.104 Competition between the open channel
blocker and the permeating Na+ ions may result. If so, administration of Na+ lactate or
bicarbonate solutions may decrease the rate of association of open but not inactivated state
Na+ channel blockers with the Na+ channel.
In addition to producing open state Na+ channel blockade, class 1A drugs (quinidine and
disopyramide) prolong AP duration. This increases the period of inactivation of Na+
channels. If so, open state blockers may enhance the blocking action of inactivated state
blockers (lidocaine and mexiletine),104 which is the rationale behind combining class 1A
and 1B drugs.108 109
Action potential plateau currents

Both the Na+ and Ca2+ channels contribute inward current during the AP plateau.
Increasing either current component will prolong AP duration, and directly (Ca2+) or
indirectly (Na+) cause positive inotropy.101 This might occur in a patient receiving a
positive inotrope for ventricular dysfunction and lidocaine for arrhythmia suppression after
cardiopulmon-
Page 110
ary bypass. There is, however a downside to this. If Na+ or Ca2+ activators bind during the
AP plateau, AP prolongation could become regenerative and EAD with triggered activity
result.101
K+ channel blockers
The emphasis of antiarrhythmic drug research and development has shifted away from
drugs that are primarily Na+ channel blockers to those that prolong AP duration and
refractoriness by blocking voltage gated K+ channels (class 3 effect). The basis of this shift
is multifactorial:104
Class 1 proarrhythmia – There is accumulating evidence of relative inefficacy and
increased proarrhythmia with class 1 agents.98 99
Class 3 efficacy – Prolongation of AP duration and refractoriness is particularly
efficacious for arrhythmias with an excitable gap (atrial flutter, monomorphic VT).104 110
Class 3 drugs appear to be more effective than class 1 drugs in preventing death and
arrhythmia recurrence in patients with ventricular tachyarrhythmias,111 and in experimental
models of atrial flutter and lethal ventricular arrhythmias.104
Contractility – In contrast to class 1A and 1C drugs, which depress myocardial
contractility, class 3 drugs either do not affect contractility or slightly increase it.104 This
effect is attributed to AP prolongation, which increases Ca2+ current during the AP plateau,
thereby enhancing Ca2+ release from the sarcoplasmic reticulum.
Defibrillation threshold (DFT) – Class 3 drugs have been reported to decrease DFT in
canine models, thereby facilitating defibrillation.104
The K+ channel blockers are a diverse group of compounds that share the property of
prolonging AP duration and refractoriness in fast response fibres. They include drugs
traditionally grouped as class 1A, 1C, and 3 agents in the modified Vaughan Williams
classification.112 113 The degree of AP duration prolongation is quite variable among class
1A and 1C drugs, despite the fact that all block the delayed rectifier (Ik), and some
(quinidine and disopyramide) block the inward rectifier (IK1) and transient outward
currents (Ito1) as well.104 This is attributed to the concomitant influence of these drugs on
INa, particularly the slowly inactivating or late component.104 Potent block of this
component by flecainide shortens AP duration, which counters AP duration prolongation
by block of IK.104 Increased Ca2+ or Na+ current during the AP plateau can also prolong AP
duration. Indeed, part of the effect of the class 3 drug ibutilide – a drug recently approved
for chemical conversion of atrial flutter and fibrillation – is mediated by block of the late
component of INa.104 114 Ibutilide may also prolong AP duration by blocking IK.
Quinidine and disopyramide (class 1A) are non-specific blockers of K+ current, blocking
all three components (IK, IK1, and Ito1).104 Procainamide (class 1A) and flecainide (class
1C) are selective blockers of IK. The class 3
Page 111
drugs, sotolol, amiodarone and bretylium are relatively non-selective K+ blockers. Sotolol
is also a non-selective β- blocker (β1, β2), and bretylium causes initial release of
norepinephrine (noradrenaline). Amiodarone exhibits all four class actions. In addition to
blocking IK and IK1, amiodarone also blocks Na+ and Ca2+ channels, and has non-
competitive α- and β- adrenergic blocking effects. This non-specificity may in part explain
the lower proarrhythmic potential of amiodarone compared with other antiarrhythmic
agents.104 110
Most Na+ channel blockers, especially in higher concentrations, cause use-dependent
block, that is, depression of Na+ current and AP upstroke velocity (hence, AP propagation
or conduction) is greater at faster heart rates. The ‘‘ideal” K+ channel blocker should also
display use-dependent block. Most available ones (except, amiodarone and flecainide) do
just the opposite, however, produce maximal prolongation of AP duration and
refractoriness at slow heart rates, with progressively diminishing effects at faster heart
rates.104 This has been termed “reverse use dependence”.115 In fact, it really isn’t “reverse
use dependence”.104 Most K+ channel blocking drugs (including amiodarone and flecainide)
block IK in a use-dependent fashion. Furthermore, block is increased at depolarised
potentials, when channels are in the open state. Accordingly, drug block (what occurs at the
K+ channel level, the conventional meaning for “use dependence”116) shows normal use
dependence.104 If so, the term “reverse rate dependence”117 may be the more appropriate
way to describe K+ channel blocker behaviour.104
Reverse rate dependence is expected to limit the efficacy of most K+ blockers for
terminating sustained tachyarrhythmias. It may also help predispose to development of
bradycardia dependent prolongation of the QT interval and torsade de pointes.104 107 110 115
Nevertheless, as a group the K+ channel blockers are moderately effective antiarrhythmics,
possibly because prolongation of refractoriness decreases the diastolic window of
excitability for initiation of tachycardia.115 Finally, the fact that amiodarone does not exhibit
reverse rate dependence over a wide range of heart rates may explain its greater efficacy
and lower proarrhythmic potential compared with other antiarrhythmic drugs.104
Finally, and of possible relevance to further development of cardiac K+ channel blockers,
several mechanisms have been proposed to explain the disparity between use dependent
block of IK and reverse rate dependent prolongation of AP duration and refractoriness.104
First, at faster heart rates, the relative contribution of other ionic processes (for example, the
inactivation of ICa,L) may exceed that of IK. Second, at fast heart rates, as a result of
incomplete activation, the slowly activating component of IK (Iks) assumes more
importance than its faster component (Ikr) in mediating AP repolarisation.104 As most
selective IK blockers target Ikr and have little affinity for Iks, they should be less effective at
fast heart rates.104 118 119 In this
Page 112
regard, it is noteworthy that amiodarone primarily blocks IKs. This may partly explain why
the drug prolongs AP duration over a range of heart rates.104 Third, preferential block of
open K+ channels does not necessarily result in prominent use dependent block, especially
if recovery from block is slow (dofetilide) or onset of block is very rapid (quinidine).104
This results because, as for the Na+ channel blockers, the kinetics of onset and offset of K+
channel block are key determinants of use dependent block. Thus, if the onset of block is
rapid, steady state block could be achieved during a single AP, and there would be little
increase in block at faster rates. Similarly, with slow offset of block, there would be little
dissipation of the block between beats, and steady state block would be achieved at
relatively slow heart rates. It has been suggested that the “ideal” K+ channel blocker should
block the open channel with depolarisation, thereby permitting rapid recovery from
inactivation with repolarisation.119 Kinetic properties such as these are expected to result in
a normal use dependent (not reverse rate dependent) pattern of APD prolongation. This has
the potential for greater antiarrhythmic efficacy with less proarrhythmia.120
Anaesthetic arrhythmic potential

Anaesthetic drugs can have both pro- and antiarrhythmic actions.33 34 38 Proarrhythmic
actions include the facilitation of catecholamine mediated ventricular arrhythmias,
suppression of normal automaticity in primary and secondary pacemaker fibres, and
depression of AV conduction. Examples of antiarrhythmic actions include suppression of
abnormal automaticity and DAD triggered activity, and ventricular tachyarrhythmias in
canine models of myocardial ischaemia and infarction. Discussion of these in more detail is
beyond the scope of this chapter. Suffice it to say, no anaesthetic drug, inhalational or
otherwise, could be introduced into clinical practice today were it discovered in preclinical
trials to cause de novo arrhythmias in normal heart. Nor would it enjoy a very long clinical
life if it were proarrhythmic in subsets of patients with cardiovascular disease, especially
with myocardial remodelling. In this vein, it is disheartening that very little is really known
about anaesthetic drug effects in patients with arrhythmias or susceptibility to arrhythmias.
Finally, one cannot presume that an anaesthetic drug will be proarrhythmic in a patient with
heart disease, especially the patient with coronary disease and a history of ventricular
arrhythmias, just because it is sensitising (for example, halothane).121
1 Whalley DW, Wendt DJ, Grant AO. Basic concepts in cellular cardiac electrophysiology: Part I: Ion
channels, membrane currents, and the action potential. Pace 1995;18:1556–74.
Page 113
2 Zipes DP. Genesis of cardiac arrhythmias: Electrophysiological considerations. In: Braunwald E, ed,
Heart disease, 5th edn. Philadelphia, WB Saunders, 1997:548–92.
3 Brown AM, Lee KS, Powell T. Voltage clamp and internal perfusion of single rat heart muscle cells. J
Physiol (Lond) 1981;318:455–77.
4 Hamill OP, Marty A, Neher E, Sachmann B, Sigworth FJ. Improved patch-clamp techniques for high-
resolution current recording from cells and cell-free membrane patches. Pflügers Arch 1981;39:85–100.
5 Katz AM. Cardiac ion channels. N Engl J Med 1993; 328:1244–51.
6 Catterall WA. Cellular and molecular biology of voltage-gated sodium channels. Physiol Rev
1992;72:515–48.
7 Hille B. Ionic channels of excitable membranes. Sunderland Sinauer Association, Inc., 1984:1–19.
8 Armstrong CM. Sodium channels and gating currents. Physiol Rev 1981;61:544–683.
9 Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to
conduction and excitation in nerve. J Physiol (Lond) 1952;117:500–44.
10 Brown AM. Ion channels as g protein effectors. News Physiol Sci 1991;6:158–61.
11 Nichols CG, Lederer WJ. Adenosine triphosphate-sensitive potassium channels in the cardiovascular
system. Am J Physiol 1991;261: H1675–86.
12 Freeman LC, Kass RC. Expression of a minimal K+ channel protein in mammalian cells and immuno-
localization in guinea pig heart. Circ Res 1993;73:968–73.
13 Po S, Roberds S, Snyders DJ, et al. Heteromultimetric assembly of human potassium channels. Molecular
basis of a transient outward current? Circ Res 1993;72:1326–36.
14 Cohen SA, Barchi RL.Voltage-dependent sodium channels. Int Rev Cytol 1993;137C:55– 103.
15 Stuhemer W, Conti F, Suzuki H, et al. Structural parts involved in activation and inactivation of the
sodium channel. Nature 1993;339:597–603.
16 Backx PH, Yue DT, Lawrence JH, et al. Molecular localization of an ion-binding site within the pore of
mammalian sodium channels. Science 1992;257:258–51.
17 Isom LL, De Jongh KS, Patton DE, et al. Primary structure and functional expression of the β1 subunit of
the rat brain sodium channel. Science 1992;256:839–42.
18 Sperelakis N. Origin of the cardiac resting potential. In: Berne RM, ed, Handbook of physiology, Section
2, The cardiovascular system. The heart, Vol 1. Bethesda, MD: American Physiological Society,
1979:187–267.
19 Boyden PA. Cellular electrophysiologic basis of cardiac arrhythmias. Am J Cardiol 1996;78(suppl 4A):4–
11.
20 Task Force of the Working Group on Arrhythmias of the European Society of Cardiology. The Sicilian
Gambit. A new approach to the classification of antiarrhythmic drugs based on their actions on
mechanisms. Circulation 1991;84:1831–51.
21 Coraboeuf E, Carmeleit E. Existence of two transient outward currents in sheep cardiac Purkinje fibres.
Pflügers Arch 1982;392;352–9.
22 Zygmunt AC, Gibbons WR. Calcium-activated chloride current in rabbit ventricular myocytes. Circ Res
1991;68:424–37.
23 Litovsky SH, Antzelvitch C. Transient outward current prominent in canine ventricular epicardium but
not endocardium. Circ Res 1988;62:116–26.
24 Harvey RD, Clark CO, Hume JR. A chloride current in mammalian cardiac myocytes: Novel mechanisms
for autonomic regulation of action potential duration and resting membrane potential. J Gen Physiol
1990;95:1077–102.
25 Patlak JB, Ortiz M, . Slow current through single Na+ channels of adult rat heart, J Gen Physiol
1988;86:89–104.
26 Wasserstrom JA, Salata JJ. Basis for tetrodotoxin and lidocaine effects on action potentials in dog and
ventricular myocytes. Am J Physiol 1988;254:H1157–66.
27 Eisner DA, Lederer WT. Na-Ca exchange: stoichiometry and electrogenecity. Am J Physiol
1985,248:C189-202.
28 Irisawa H, Brown HF, Giles W. Cardiac pacemaking in the sinoatrial node. Physiol Rev 1993;73:197–227.
29 Arnsdorf MF. Arnsdorf’s paradox. J Cardiavasc Electrophysiol 1990;1:42–52.
Page 114
30 Arnsdorf MF. Basic understanding of electrophysiological actions of antiarrhythmic drugs: Sources, sinks
and matrices of information. Med Clin North Am 1988;68:1247–80.
31 Kleber AG, Riegger CB. Electrical constants of arterially perfused rabbit papillary muscle. J Physiol
(Lond) 1987;385:307–24.
32 Weidmann S. Passive properties of cardiac fibres. In: Rosen MR, Janse MJ, Wit AL, eds, Cardiac
electrophysiology: A textbook. Mount Kisco, NY: Futura Publishing Co., 1990:415–25.
33 Atlee JL III, Bosnjak ZJ. Mechanisms for cardiac dysrhythmias during anesthesia. Anesthesiology
1990;72:347–94.
34 Turner LA, Bosnjak ZJ. Autonomic and anesthetic modulation of cardiac conduction and arrhythmias. In:
Lynch C III ed, Clinical cardiac electrophysiology. Perioperative considerations. Philadelphia: JB
Lippincott Co., 1994:53–84.
35 Fozzard HA, Arnsdorf MF. Cardiac electrophysiology. In: Fozzard HA, Jennings RB, Haber E, Katz AM,
Morgan HE, eds, The heart and cardiovascular system, 2nd edn. New York: Raven Press, 1992:63–98.
36 Delmar M. Role of potassium currents on cell excitability in cardiac ventricular myocytes. J Cardiovasc
Electrophysiol 1992;3:474–86.
37 Cranefield PF. Action potentials, afterpotentials, and arrhythmias. Circ Res 1977;41:415–23.
38 Patterson E, Szabo B, Scherlag BJ, Lazzara R. Arrhythrnogenic effects of antiarrhythmic drugs. In: Zipes
DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside. 2nd edn. Philadelphia: WB Saunders,
1995: 496–511.
39 Grace AA, Camm AJ. Quinidine. N Engl J Med 1998;338:35–44.
40 Atlee JL. Perioperative cardiac dysrhythmias. Anesthesiology 1997;86:1397–424
41 Morganroth J. Proarrhythmic effects of antiarrhythmic drugs: evolving concepts. Am Heart J
1992;123:1137–9.
42 Kerin N, Somberg J. Proarrhythmia: Definition, risk factors, causes, treatment, and controversies. Am
Heart J 1994;128:575–86.
43 De Luna A, Guindo J, Vinolas X, et al. Proarrhythmic effects of anti-arrhythmic agents. Postgrad Med J
1994;70:S84–8.
44 Colatsky TJ, Follmer CH, Starmer CF. Channel specificity in antiarrhythmic drug action: mechanism of
potassium channel block and its role in suppressing and aggravating cardiac arrhythmias. Circulation
1990;82:2235–42.
45 Yang T, Roden DM. Extracellular K+ modulation of drug block by Ikr: implications for torsade de pointes
and reverse use-dependence. Circulation 1996;93:407–11.
46 Schwartz PJ, Srramba-Badiale M, Segantini A, et al. Prolongation of the QT interval and the sudden
infant death syndrome. N Engl J Med 1998;338:1709–14.
47 Schwartz PJ, Locati EH, Napolitano C, Priori SG. The long QT syndrome. In: Zipes DP, Jalife J, eds,
Cardiac electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB Saunders, 1995:788–811.
48 Schwartz P, Priori S, Locati, EH, et al. Long QT syndrome patients with mutations of the SCN5A and
HERG genes have differential responses to Na+ channel blockade and to increases in heart rate.
49 Rosen M. Long QT syndrome patients with gene mutations. Circulation 1995;92:3373–5.
50 Moss A, Zareba W, Benhorin J, et al. ECGT-wave patterns in genetically distinct forms of the hereditary
long QT syndrome. Circulation 1995;92:2929–34.
51 Kass RL, Lederer WJ, Tsien RW, Weingurt R. Role of calcium ions in transient inward currents and
aftercontractions induced by acetylstrophanthidin in cardiac Purkinje fibres. J Physiol 1978;281:187–208.
52 Kass RL, Tsien RW, Weingurt R. Ionic basis of transient inward current induced by strophanthidin in
cardiac Purkinje fibres. J Physiol 1978;281:209–26.
53 Luo CH, Rudy Y. A dynamic model of the cardiac ventricular action potential: I. Simulations of ionic
currents and concentration changes. Circ Res 1994;74:1071–96
54 Luo CH, Rudy Y. A dynamic model of the cardiac ventricular action potential: II. Afterdepolarizarions,
triggered activity and potentiations. Circ Res 1994;74:1097–113
55 Mines GR. On dynamic equilibrium in the heart. J Physiol (Lond) 1913;46:349–82.
Page 115
56 Mines GR. On circulating excitations in heart muscles and their possible relation to tachycardia and
fibrillation. Trans R Soc Canada, Section IV, 1914:43–52.
57 Fozzard HA, Haber E, Jennings RB, et al, eds. The heart and cardiovascular system: Scientific
foundations, 2nd edn. New York: Raven Press, 1991:63–98, 863–1169, 2021–193.
58 Mendez C, Mueller WJ, Urguiaga X. Propagation of impulses across the Purkinje fibre-muscle junctions
in the dog heart. Circ Res 1970;36:135–50.
59 Veenstra RD, Joyner RW, Raveling DA. Purkinje and ventricular activation sequences of canine papillary
muscle: Effects of quinidine and calcium on Purkinje–ventricular conduction delay. Circ Res
1984;54:500–15.
60 Joyner RW, Overhold VD, Ramza B, Veenstra RD. Propagation through electrically coupled cells: Two
inhomgeneously coupled cardiac tissue layers. Am J Physiol 1984;247:H596–609.
61 Schmidt FO, Erlanger J. Directional differences in the conduction of the impulse through heart muscle
and their possible relation to extrasystoles and fibrillary contractions. Am J Physiol 1929;87:326–47.
62 Allessie MA, Lammers WJ, Bonke IM, Hollen J. Intra-atrial reentry as a mechanism for atrial flutter
induced by acetylcholine and rapid atrial pacing in the dog. Circulation 1984;70:123–35.
63 Janse MJ, Van Cappele FJ, Freud GE, Durrer D. Circus movement within the AV node as a basis for
supraventricular tachycardia as shown by multiple microelectrode recording in the isolated rabbit heart.
Circ Res 1971;28:403–14.
64 Pastelin G, Mendez R, Moe GK. Participation of atrial specialized conduction pathways in atrial flutter.
Circ Res 1978;42:386–93.
65 Cabo C, Wit AL. Cellular electrophysiologic mechanisms of cardiac arrhythmias. Cardiol clin
l997;15:517–38
66 Boyden PA, Tilley LP, Albala A, Liu SK, Fenoglio JJ Jr, Wit AL. Mechanisms for atrial arrhythmias
associated with cardiomyopathy: A study of feline hearts with primary myocardial disease. Circulation
1984;69:1036–47.
67 Ursell PC, Gardner PI, Albala A, Fenoglio JJ Jr, Wit AL. Structural and electrophysiological changes in
the epicardial border zone of canine myocardial infarcts during infarct healing. Circ Res 1985;56:436–51.
68 Blanck Z, Sra J, Dhala A, Deshpande S, Jazayeri MR, Akhtar M. Bundle branch reentry: Mechanisms,
diagnosis, treatment. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd edn.
69 Miles WM, Klein LS, Rardon DP, Mitrani RD, Zipes DP. Atrioventricular reentry and variants:
Mechanisms, clinical features, and management. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology:
from cell to bedside, 2nd edn. Philadelphia: WB Saunders, 1995: 638–1, 642–55.
70 Waldo AL. Atrial flutter: Mechanisms, clinical features, and management. In: Zipes DP, Jalife J, eds,
71 El-Sherif N. Reentrant mechanisms in ventricular arrhythmias. In: Zipes DP, Jalife J, eds, Cardiac
electrophysiology: from cell to bedside. 2nd edn. Philadelphia: WB Saunders, 1995: 567–82.
72 Janse MJ, OpthofT. Mechanisms of ischemia induced arrhythmias. In: Zipes DP, Jalife J, eds, Cardiac
electrophysiology: from cell to bedside 2nd edn. Philadelphia: WB Saunders, 1995:489–96.
73 Wit AL, Dillon SM, Coromilas J. Anisotropic reentry as a cause of ventricular tachyarrhythmias in
myocardial infarction. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd
edn. Philadelphia: WB Saunders, 1995:511–26.
74 Allessie MA, Bonke FIM, Schopman FJG. Circus movement in rabbit atrial muscle as a mechanism of
tachycardia. Circ Res 1973;33:54–62.
tachycardia: II. The role of nonuniform recovery of excitability in the occurrence of unidirectional block,
as studied with multiple microelectrodes. Circ Res 1976;39:168–77.
Page 116
tachycardia: III The leading circle concept: A new model of circus movement in cardiac tissue without the
involvement of an anatomical obstacle. Circ Res 1977;41:9–18.
77 Allessie MA. Re-entrant mechanism underlying atrial fibrillation. In: Zipes DP, Jalife J, eds, Cardiac
electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB Saunders, 1995: 562–6.
78 The Sicilian Gambit. Antiarrhythmic therapy: a pathophysiologic approach. Armonk, NY: Futura
Publishing, 1994:3–337
79 Davidenko JM, Pertsov AM, Salomonsz R, Baxter W, Jalife J. Stationary and drifting spiral waves of
excitation in isolated cardiac muscle. Nature 1992;335:349–51.
80 Davidenko JM. Spiral waves in the heart: Experimental demonstration of a theory. In: Zipes DP, Jalife J,
eds, Cardiac electrophysiology: from cell to bedside 2nd edn. Philadelphia: WB Saunders, 1995:478–88.
81 Davidenko JM. Spiral wave activity: A possible common mechanism for polymorphic and monomorphic
ventricular tachycardias. J Cardiovasc Electrophysiol 1993;4:730–46.
82 Chen P-S, Garfinkel A, Weiss JN, Karagueuzian HS. Spirals, chaos, and new mechanisms of wave
propagation. PACE 1997;20:414–21.
83 Naccarelli GV, Shih HT, Jalal S. Sinus node reentry and atrial tachycardias. In: Zipes DP, Jalife J, eds,
84 Watanabe Y, Watanabe M. Impulse formation and conduction of excitation in the atrioventricular node. J
Cardiovasc Electrophysiol 1994;5:517–31.
85 Janse MJ, Anderson RH, McGuire MA, Ho SV. “AV nodal” reentry: 1. “AV nodal” reentry revisited. J
Cardiovasc Electrophysiol 1993;4:561–72.
86 McGuire MA, Janse MJ, Ross DL. ‘‘AV nodal” reentry: 2. AV nodal, AV junctional or atrial nodal
reentry? J Cardiovasc Electrophysiol 1993;4:573–86.
87 Jackman WM, Nakagawa H, Heidbuuchel H, et al. Three forms of atrioventricular nodal (junctional)
reentrant tachycardia: Differential diagnosis, electrophysiological characteristics, and implications for
anatomy of the reentrant circuit. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to
bedside, 2nd edn. Philadelphia: WB Saunders, 1995: 620–37.
88 Kalbfleisch SJ, Morady F. Catheter ablation of atrioventricular nodal reentrant tachycardia. In: Zipes DP,
Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB Saunders,
1995:1477–99.
89 Mahomed Y. Surgery for atrioventricular nodal reentrant tachycardia. In: Zipes DP, Jalife J, eds, Cardiac
electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB Saunders, 1995:1577–83.
90 Wellens HJJ. Supraventricular tachycardia with reentry in the AV node. In: Kastor JA, ed, Arrhythmias.
91 Gallagher JJ. Supraventricular tachycardia with reentry in accessory pathways. In: Kastor JA, ed,
Arrhythmias. Philadelphia: WB Saunders, 1994:262–96.
92 Miles WM, Klein LS, Rardon DP, Mitrani RD, Zipes DP. Atrioventricular reentry and its variants:
Mechanisms, clinical features, and management. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology:
from cell to bedside, 2nd edn. Philadelphia: WB Saunders, 1995: 638–55.
93 Packer DL, Zipes DP. Anatomical and physiological substrates for antidromic reciprocating tachycardia.
In: Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB
Saunders, 1995:655–66.
94 Yee R, Klein GJ, Guiraudon GM. The Wolff-Parkinson-White syndrome. In: Zipes DP, Jalife J, eds,
95 Haissaguerre M, Clémenty J, Warin J-F. Catheter ablation of atrioventricular reentrant tachycardia. In:
Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd edn. Philadelphia: WB
Saunders, 1995:1487–99.
96 Guiraudon GM, Klein GJ, Yee R, Thakur RK, Guiraudon CM. Surgery for the Wolff-Parkinson-White
syndrome. In: Zipes DP, Jalife J, eds, Cardiac electrophysiology: from cell to bedside, 2nd edn.
Page 117
97 Wit AL, Janse MJ. The ventricular arrhythmias of ischemia and infarction: Electrophysiological
mechanisms. Alt Kisco, NY: Futura, 1993.
98 CAST Investigators. Preliminary Report. Effect of encainide and flecainide on mortality in a randomized
trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321:406–12.
99 CAST-II. Effect of the antiarrhythmic agent moricizine on survival after myocardial infarction. N Engl J
Med 1992:327:227–33.
100 Flaker GC, Blackshear JL, McBride R, Kronmal RA, Hart RG. Antiarrhythmic drug therapy and cardiac
mortality in atrial fibrillation. J Am Coll Cardiol 1992;20:527–32.
101 Grant AO. Mechanisms of action of antiarrhythmic drugs: From ion channel blockage to arrhythmia
termination. PACE 1997;20; 432–4.
102 Spach MS, Boineau JP. Micofibrosis produces electrical load variations due to loss of side-to-side cell
connections. PACE 1997;20:397–413.
103 Weber KT, Spach MS, Starmer ME Chaos in the hall of mirrors. Cardiovasc Res 1995;30:336–44.
104 Whalley DW, Wendt DJ, Grant AO. Basic concepts in cellular cardiac electrophysiology: Part II: Block
of ion channels by antiarrhythmic drugs. Pace 1995;18:1686–704.
105 Van Bogaert P-P, Goethals M. Pharmacological influence of specific bradycardic agents on the
pacemaker current of sheep cardiac Purkinje fibres. A comparison between three different molecules. Eur
Heart J 1987;8(suppl L):35–42.
106 Whitcomb DC, Gilliam FR 3rd, Starmer CF, Grant AO. Marked QRS complex abnormalities and sodium
channel blockade propoxyphene reversed with lidocaine. J Clin Invest 1989;84:1629–43.
107 Zipes DP. Management of cardiac arrhythmias: Pharmacological, electrical, and surgical techniques. In:
Braunwald E, ed, Heart disease, 5th edn. Philadelphia: WB Saunders, 1997:593–639.
108 Duff HJ, Roden D, Primm RK, Oates JA, Woosley RL. Mexiletine in the treatment of resistant ventricular
arrhythmias: Enhancement of efficacy and reduction of dose related side effects of combination with
quinidine. Circulation 1983;67:1124–8.
109 Kim SG, Mercondo AD, Tom S. Combination of disopyramide and mexiletine for better tolerance and
additive effects for treatment of ventricular arrhythmias. J Am Coll Cardiol 1989;13:659–64.
110 Singh B. Controlling cardiac arrhythmias by lengthening repolarization: Historical overview. Am Heart J
1993;72:18F–24F.
111 Mason JW. A comparison of seven antiarrhythmic drugs in patients with ventricular tachyarrhythmias. N
Engl J Med 1993;329:452–8.
112 Vaughan Williams E. Classification of antiarrhythmic drugs. In: Sandoe E, Flensted-Jensen E, Olesen K,
eds, Cardiac arrhythmias. Sodertaljie, Sweden: AB Astra, 1971:449–72.
113 Harrison D, Winkle R, Sami M, Mason J. Encainide: a new and potent antiarrhythmic agent. In: Harrison
D, ed, Cardiac arrhythmias: a decade of progress. Boston: GK Hall, 1981:315–30.
114 Stambler BS, Wood MA, Ellenbogen KA, Perry KT, Wakefield LK, VanderLugt JT, and the Ibutilide
repeat dose study Investigators. Efficacy and safety of repeated intravenous doses of ibutilide for rapid
conversion of atrial flutter or fibrillation. Circulation 1996;94:1613–21.
115 Hondeghem LM, Snyders DJ. Class III antiarrhythmic agents have a lot of potential but a long way to go.
116 Carmeliet E. K+ channels and control of ventricular repolarization in the heart. Fundam Clin Pharmacol
1993;7:19–28.
117 Jurkiewicz NK, Sanguinetti MC. Rate-dependent prolongation of cardiac action potentials by a
methanesulfonanilide Class III antiarrhythmic agent. Circ Res 1993;73:75–83.
118 Colatsky TJ. Potassium channel blockers: Synthetic agents and their antiarrhythmic potential. In: Weston
AH, Hamilton TC, eds, Potassium channel modulators: pharmacological, molecular and clinical aspects.
London: Blackwell Scientific Ltd, 1992:304–10.
119 Carmeliet E. Use-dependent block and use-dependent unblock of the delayed rectifier K+ current by
almokalant in rabbit ventricular myocytes. Circ Res 1993;73:857–68.
Page 118
120Hondeghem LM. Development of Class III antiarrhythmic agents. J Cardiovasc Pharmacol 1992;20
(suppl 2):S17-22.
121 Atlee JL. Halothane: cause or cure for arrhythmias? Anesthesiology 1987;67:617–18.
Page 119
4: Coronary physiology
HANS-JOACHIM PRIEBE
A thorough knowledge of normal coronary physiology is required to understand the

mechanisms involved in coronary pathophysiology and, thus, for the optimal care of the
patient with coronary artery disease. The heart, unlike any other organ, not only provides
flow to the entire organism, but also has to generate its own perfusion pressure. This poses
an extraordinary metabolic load on the coronary circulation. Complicating matters further,
unlike any other regional circulation, the coronary circulation is subjected to marked
variations in extravascular compressive forces related to the cardiac cycle.
Such unique physiology requires a highly specialised circulation if metabolic
requirements are to be met. Consequently, the coronary circulation is composed of different
kinds of vessels, each with distinct physiological, anatomical, and pharmacological
characteristics.1 2
The myocardium is almost entirely dependent on aerobic metabolism. Thus, there is little
capacity to accumulate an oxygen (O2) debt, and O2 demand must be met on a beat-to-beat
basis. As myocardial O2 extraction is about 70% even at rest, increased myocardial O2
consumption is principally met by increases in coronary blood flow (CBF). This
results in a linear relationship between CBF and (Fig 4.1). As a consequence,
coronary sinus O2 tension (18–20 mm Hg or 2·4–2·7 kPa) and saturation (25–30%) remain
remarkably constant, O2 delivery in excess of demand (so called “luxury” perfusion) is not
usually observed, and inadequate O2 supply in relation to demand will rapidly cause
myocardial ischaemia.
Obviously, matching of O2 delivery to O2 demand is the essential task of the coronary
circulation. The anatomical, physical, and biochemical factors responsible for this precise
matching are discussed below. This provides the basis for understanding the
pathophysiology of the diseased coronary circulation.
Coronary anatomy
The right and left coronary arteries provide the entire blood supply to the myocardium.
They arise from the coronary ostia in the sinuses of Valsalva,
Page 120
Fig 4.1 Relationship between coronary blood flow and myocardial oxygen consumption in conscious dogs.
Each dog is represented by a separate symbol. A close correlation between the two variables over a wide
range is evident. (Reproduced with permission from Khouri EM, Gregg DE, Rayford CR. Effect of exercise
on cardiac output, left coronary flow and myocardial metabolism in the unanesthetized dog. Circ Res
1965;17:427–37.)
located at the aortic root above the anterior and left posterior cusps of the aortic valve. The
anatomical arrangement between valve leaflets and coronary ostia ensures CBF even during
systole. The two main coronary arteries run along their respective atrioventricular grooves
where they branch repeatedly.
The left coronary artery divides into the anterior interventricular (also called the left
anterior descending) and the circumflex artery. Branches of the anterior interventricular
artery supply the anterior walls of the left and right ventricles, part of the left lateral wall,
and most of the interventricular
Page 121
septum. The circumflex artery supplies the left atrium, parts of the lateral and posterior left
ventricular wall, and, in about 45% of hearts, the sinus node. It also perfuses the anterior
papillary muscle of the left ventricle, whereas the posterior papillary muscle is perfused by
branches of both right and left coronary arteries.
The right coronary artery supplies the remainder of the right ventricle and, in 90% of
hearts, the atrioventricular node through its posterior interventricular branch. In about 55%
of hearts, the sinus node artery originates from the right coronary artery. An extensive
network of collateral and communicating vessels between branches of right and left
coronary arteries encircles the epicardial surface of the heart. These vessels play an
important role in maintaining some degree of regional perfusion when coronary occlusion
develops.
The term “dominant” refers to that particular coronary artery from which the posterior
interventricular artery originates. This branch supplies the lower (apical) part of the
interventricular septum and the diaphragmatic surface of the left ventricle. The vast
majority of normal human hearts (80–85%) have a dominant right coronary system. Of the
remaining 15–20% of individuals, half demonstrate a left dominant coronary circulation
and the other half a balanced circulation in which right and left coronary arteries contribute
to septal and left ventricular diaphragmatic perfusion.3 When comparing results of different
studies, it is important to realise that significant species differences exist in regard to the
dominant coronary system. Porcine hearts – like human hearts – have a dominant right
coronary system, whereas the canine heart has a dominant left coronary system.
The myocardial wall is supplied by branches that arise from the large epicardial coronary
arteries. These branches penetrate the ventricular wall in an almost perpendicular fashion.
Within the myocardium there is further division of these branches into extensive networks
of small arteries and arterioles, which give rise to the myocardial capillary bed. Certain
branches from the epicardial vessels arborise in the subepicardium whereas others plunge
deep into the myocardial wall to form a dense subendocardial network of vessels. Control
of vasomotor tone differs between these two types of vessels, which enables changes in
transmural flow distribution.1 Coronary angiography can only visualise arterial segments
that are larger than 500 μm in diameter. Visualisation of coronary microvessels between 15
and 500 μm requires highly specialised microvascular imaging techniques.
Most of the coronary vascular bed lies within the myocardial wall. As a result of easier
accessibility, however, much of our knowledge of the coronary circulation is based on
studies of the large epicardial vessels. Although they are the most frequent site of
atherosclerotic alterations in coronary artery disease, they are mostly not representative of
coronary physiology in general.
Page 122
Morphologically, the major coronary vascular resistance (CVR) lies in arterioles smaller
than 450 μm in diameter.4 Almost half of the total CVR resides in vessels 100–450 μm in
diameter, and the rest in arterioles of less than l00μm. Whereas the small resistance vessels
constitute the major resistance to flow, the large conductance vessels determine the quantity
of blood arriving at the resistance vessels.
After passage through the capillary beds, most of the venous blood returns to the right
atrium through the coronary sinus. Of coronary sinus outflow, 90–95% is derived from the
left coronary artery. Most of the venous return from the right ventricle drains into the
anterior cardiac veins that empty directly into the right atrium. A small amount of venous
drainage of the heart drains via Thebesian veins directly into the left atrium, and the right
and left ventricles, so contributing to the physiological arteriovenous shunt. A striking
feature of the coronary venous system is the abundance of large anastomoses between all
major veins,5 which may result in considerable fluctuations in coronary sinus outflow.
The outer layer (adventitia) of small coronary arteries contains nerves. Although they do
not penetrate the media, release of transmitter substances from nerve varicosities close to
the smooth muscle layer is likely. Circumferentially arranged smooth muscle cells
constitute the media. Layers of smooth muscle cells number between six in vessels that are
300 μm in diameter, and one in arterioles that are 30–50 μm in diameter. The smooth
muscle cells behave electrically like a syncytium. The luminal surface of the coronary
vessels is lined by endothelial cells which penetrate into the media. Through this contact
with the smooth muscle cells, the vascular endothelium plays a key role in the control of
coronary vasomotor tone (see below).
Coronary microcirculation
The microcirculation regulates terminal arterioles and capillaries that are responsible for
the exchange of O2 from blood to myocardial tissue. Capillary flow is determined less by
the morphological properties of the capillary itself than by the tone of the feeder arteriole.
Of the more than 2000 capillaries/mm3, usually only 60–80% are open. The normal
intercapillary distance is about 17| μm (Table 4.1). During hypoxaemia, the intercapillary
distance decreases to 14·5 μm as a result of recruitment of additional capillaries.
Recruitment occurs by relaxation of precapillary sphincter tone. During prolonged
hypoxaemia, the intercapillary distance decreases even further to 11 μm (Table 4.1).
Recruitment of coronary capillaries with subsequent reduction in intercapillary distance
is an important compensatory mechanism to meet
Page 123
Table 4.1 Recruitment of coronary capillaries
Distances (μm)
Intercapillary Diffusion
Normal 17 8·5
Exercise (estimated) 14 7·0
Hypoxia 14·5 7·3
Prolonged anoxia 11 5·5
Maximal recruitment 6·5 3·3
Adapted with permission from Opie LH. The heart, 3rd edn. Philadelphia: Lippincott-
Raven, 1998:270.
increased . Augmentation of CBF alone at times of elevated O2 demands would be
insufficient compensation.
Coronary collateral circulation

Collateral vessels are accessory vascular channels which provide perfusion distal to an
obstructed native coronary artery. Even in the normal myocardium, inter- and intracoronary
collaterals exist at all levels of vessel sizes except the capillaries.6 In certain species (for
example, humans and dogs), they are present even in the neonatal heart. In the unstimulated
state, collaterals are 40 μm in diameter and often only one cell layer thick. In the stimulated
state, they may have diameters as large as 1 mm, and histologically they are similar to a
myocardial arteriole.
There are considerable species differences in localisation and extent of intercoronary
collaterals, which have to be taken into account when comparing results of different studies.
The dog and the guinea pig have a relatively well developed collateral circulation. Pigs
(similar to humans without coronary artery disease) have practically no anatomically
demonstrable collaterals. Consequently, acute coronary occlusion leads very quickly to
transmural infarction.
There are species differences not only in native (pre-existent) collaterals but also in
transformed (developed) collaterals. In the human and porcine heart, collaterals develop
predominantly in the subendocardium, and these look histologically like abnormally thin-
walled arteries. By contrast, collaterals in canine hearts develop only in a narrow
subepicardial zone around the edge of the zone that is potentially ischaemic. These
collaterals generally exhibit significantly higher resistance than the vessels that they have
replaced, although they are not nearly as flow limiting as the collaterals that develop in
humans with ischaemia.
The factors that induce the transformation from the unstimulated to the stimulated state
are largely unknown. Ischaemia seems to be a very
Page 124
powerful stimulus for the development of coronary collaterals.7 Several experimental
findings would, however, suggest that ischaemia of the cardiac myocyte may not
necessarily be a prerequisite for the stimulation of collateral growth:7 8 (1) growing
segments of collaterals that are relatively far removed from areas of tissue hypoxia; (2)
progressive ameroid induced coronary artery stenosis leading to the development of
collaterals in the surrounding subepicardium even though subepicardial perfusion is not
necessarily impaired; and (3) lack of defined time or spatial relationship between presumed
ischaemia and collateral growth.
There is evidence that an inflammatory response may be the primary stimulus for
collateral growth.8 As a result of the lack of preformed arteriolar connections between
superficial and deeper vascular territories in humans, a progressive stenosis of large
epicardial vessels is likely to cause ischaemia-induced local micronecroses. The subsequent
inflammatory response produces angiogenic mitogens mediated by
macrophages/monocytes that trigger collateral growth. Thus, in humans (unlike the dog and
like the pig) development of subendocardial collateral plexus appears to be a result of true
angiogenesis. Reduction in arterial inflow at rest may be the predominant trigger
mechanism. Other factors such as coronary occlusion, anaemia, and hypoxia have also been
shown to stimulate collateral growth.9
Well developed canine collateral vessels do not vasoconstrict in response to α-adrenergic
agonists. Sympathetic activation is therefore unlikely to cause collateral vasoconstriction.
There is, however, thromboxane A2 mediated vasoconstriction induced by platelet
activating factor (PAF) in mature canine coronary collaterals.10 As myocardial ischaemia
can rapidly increase lyso-PAF (the precursor of PAF) by as much as 50%,11 collateral
dependent myocardial areas may well be susceptible to PAF-induced vasoconstriction and
decrease in blood flow.
Although the protective role of coronary collaterals in humans has been debated in the
past, prospectively collected data provide evidence that the collateral circulation limits the
degree of ischaemia and improves survival during controlled intracoronary balloon
occlusion, and after coronary artery spasm, acute myocardial infarction and post-coronary
artery bypass graft closure.9 Thus, in humans the coronary collateral circulation appears to
constitute an important alternative source of blood supply to “jeopardised” myocardium.
Regulation of coronary blood flow

As with flow in any other vascular bed, CBF varies with perfusion pressure and
vasomotor tone (that is, vascular resistance). Three major
Page 125
factors regulate coronary vascular tone: (1) the vascular endothelium, (2) local metabolism,
and (3) the neurohumoral system.
Endothelial control of coronary vascular tone
The coronary vascular endothelium plays a major role in regulating vasomotor tone in
health and disease.12 It modulates the contractile activity of the underlying smooth muscle
through the secretion and synthesis of substances with different biological activities in
response to a variety of pharmacological agents (for example, acetylcholine, substance P,
catecholamines) and physical stimuli (for example, blood flow, pulsatile flow, shear stress).
Endothelium-derived relaxing factors

Since the initial observation that an intact endothelium is necessary for acetylcholine-
induced vasodilation,13 many physiological stimuli have been shown to cause vasodilation
by stimulating the release of powerful vasodilators such as endothelium-derived relaxing
factor (EDRF), nitric oxide, prostacyclin, and endothelium-derived hyperpolarising factor
(EDHF) (Fig 4.2). (By contrast, adenosine and the nitrovasodilators nitroglycerine and
nitroprusside are endothelium-independent vasodilators.)
Nitric oxide – For practical purposes, EDRF is identical to nitric oxide (NO). NO is
formed in endothelial cells from L-arginine by oxidation of its guanidine–nitrogen terminal.
The catalysing enzyme involved is endothelial NO synthase (eNOS). It is calcium
sensitive,14 constitutively expressed, and exists in several isoforms in endothelial cells,
vascular smooth muscle cells, platelets, macrophages, and the brain.
An inducible form of NO synthase (iNOS) exists in vascular smooth muscle,
endothelium, and macrophages. This enzyme is calcium-independent, and can produce
large amounts of NO. It is induced by cytokines such as endotoxin, interleukin-1β, and
tumour necrosis factor (TNF).
Nitric oxide formation can be pharmacologically inhibited by the administration of L-
arginine analogues such as L-NG-monomethyl arginine (L-NMMA) or L-nitroarginine
methyl ester (L-NAME). These compounds compete with the natural precursor L-arginine
at the catalytic site of NOS.
Nitric oxide travels from the endothelium to the vascular smooth muscle where it
stimulates guanylate cyclase to produce vasodilatory cGMP. NO has a very short half life
(in the order of seconds or smaller), and thus just acts at the site where it is formed.
Physiological stimuli that cause release of NO include acetylcholine, increased flow (via
shear force), bradykinin,15 hypoxia, and (somewhat surprisingly) endothelin.16 Damage to
the endothelium (for example, mechanical insult, formation of free radicals, early
Page 126
atheroma) results in inhibition of NO release. In addition, in the presence of a non-intact
endothelium, acetylcholine causes vasoconstriction (by a direct effect on vascular smooth
muscle), and angiotensin II, platelet-derived growth factor, free radicals, and thrombin
stimulate the release of vasoconstrictory endothelin.
Coronary vasodilation after intracoronary infusion of substances known to stimulate
endothelial NO production (for example, acetylcholine, substance P, and serotonin) have
provided indirect evidence of a role for NO in the human coronary circulation. The role of
NO in the regulation of basal coronary vasomotor tone, however, remains controversial.
Although several studies seem to indicate that basal formation of NO exerts a tonic dilator
influence on the resting coronary circulation,17 18 inhibition of NO synthesis does not
consistently decrease basal CBF.19 20 When compared with wild type rats, basal CBF (as
well as reactive hyperaemic response to
Fig 4.2 Endothelium-derived vasoactive substances (left contracting factors; right relaxing factors). AI =
angiotensin I; AII = angiotensin II; ACE = angiotensin converting enzyme; Ach = acetylcholine; ADP =
adenosine diphosphate; Bk = bradykinin; cAMP/cGMP = cyclic adenosine/guanosine monophosphate; E =
oestrogen; ECE = endothelin converting enzyme; EDHF = endothelium-derived hyperpolarising factor;
bET-l=big endothelin-1; ET-1 = endothelin-1; 5-HT = 5-hydroxytryptamine (serotonin); L-Arg = L-arginine;
NO = nitric oxide; NOS = nitric oxide synthase; O2 = superoxide; PGH2 = prostaglandin H2; PGI2 =
prostacyclin; TGFβ1 = transforming growth factor β1; Thr = thrombin; TXA-2 =
thromboxane A2. Circles represent receptors: AT1 = angiotensinergic; B2 = bradykinergic; ETA/B =
endothelin-A/B receptor; M = muscarinic; P = purinergic; S1 = serotoninergic; T = thrombin receptor; TX =
thromboxane/prostaglandin H2 receptor. (Reproduced with permission from Wight E, Noll G, Lüscher TF.
Regulation of vascular tone and endothelial function and its alterations in cardiovascular disease. Baillière’s
Clin Anaesthesiol 1997;11:531–60.)
Page 127
short term ischaemia, and responses to intracoronary administration of adenosine and
acetylcholine) was not different in endothelial NO synthase knockout mice.21 Even
combined NO synthase inhibition and adenosine receptor blockade failed to decrease
resting CBF.22 The combined evidence would suggest that under resting conditions NO is
not essential to maintain basal CBF.
By contrast, NO formation is apparently the major pathway involved in the dilatation of
large epicardial coronary arteries that is caused by elevated shear stress and by
acetylcholine.23 Residual dilatation of coronary conductance arteries to acetylcholine is,
however, commonly observed even after blockade of NO formation by arginine analogues.
The relative contribution of NO may well depend on the underlying stimulus. Whereas in
shear stress-induced (that is, flow-dependent) vasodilation NO is the major mediator, in
acetylcholine-induced (that is, receptor-operated) dilatation mediators other than NO seem
to be involved. One such intermediate may be EDHF.24 This EDHF could be a cytochrome
P450 metabolite of arachidonic acid.25 26
In contrast to conductance vessels, dilatation to acetylcholine that is resistant to arginine
analogues does not occur in coronary resistance vessels.26 This would suggest that NO is
the principal mediator of both flow-dependent and receptor-operated vasorelaxation in the
resistance vessels of the coronary circulation.
During high , deficiencies in NO release by the coronary endothelium may be
balanced by powerful metabolic stimuli. Adenosine release apparently compensates for
diminished NO release.27 Only blockade of both NO synthase and adenosine receptors
prevents an adequate increase in CBF during stress.27 Activation of ATP-sensitive K+
channels channels) may also compensate for diminished NO release.22 Thus,
although NO contributes to coronary vasodilation during exercise, a redundancy of
vasodilatator mechanisms apparently compensates for a loss of NO release in the otherwise
normal heart.
Nitric oxide synthase inhibition decreases total reactive hyperaemic flow.28 Combined
blockade of channels, adenosine receptors, and NO synthase does not, however,
fully block reactive hyperaemia.22 Other vasodilators such as prostacyclin, EDHF, or
metabolic alterations (that is, decreased pH and increased CO2 tension or PCO2) could
contribute to the residual vasodilation. In general, the respective response to any NO
inhibition will vary with species, baseline vasomotor tone, state of consciousness (awake
versus anaesthetised), and experimental model.
Prostaglandins – The first vasoactive endothelium-derived substance discovered was
prostacyclin, or prostaglandin I2. Prostacyclin (PGI2) is the major breakdown product of
cyclo-oxygenase and the major vasodilatory prostaglandin. It is released from the
endothelium in response to shear stress, pulsatile flow, hypoxia, and several other
substances that also release
Page 128
NO, such as adenosine diphosphate (ADP), adenosine triphosphate (ATP), serotonin, and
thrombin. In most blood vessels, however, its platelet inhibitory effects are probably more
important than its vasodilatory ones.
Bradykinin – Increased shear forces as a result of an increased blood flow stimulate the
formation of bradykinin in vascular endothelium.15 Bradykinin binds to two types of
endothelial receptors: the B1- and B2-receptor. B2-receptors are more sensitive to
bradykinin than B1-receptors. Their stimulation results in the release of the two powerful
vasodilators EDRF/ NO and PGI2. Local formation of bradykinin may, therefore, be
important in the mechanism of flow-induced vasodilation.
Endothelium-derived contracting factors

Endogenous (for example, arachidonic acid, noradrenaline, thrombin) and
pharmacological substances (for example, calcium, nicotine, high potassium), and
physicochemical stimuli (shear stress, mechanical stress, hypoxia) can stimulate
endothelium-dependent vasoconstriction.15 16 Major endothelium-derived vasoconstrictors
are endothelin, thromboxane A2, and prostaglandin H2 (see Fig 4.2)
Endothelin – Endothelin is a peptide composed of 21 amino acids. It is a potent
vasoconstrictor in disease states. When physiologically released in small amounts, however,
it may function as a vasodilator, possibly by stimulating endothelium-dependent release of
NO and PGI2 (which in turn inhibit endothelin production via a negative feedback
mechanism).
Of the three isoforms endothelin-1, endothelin-2, and endothelin-3, endothelial cells
produce exclusively endothelin-1. It is derived from pre-pro-endothelin and pro (big)-
endothelin, which is converted to the biologically active peptide endothelin-1 by the
endothelin converting enzymes (ECE-1 and ECE-2). The main pathophysiological stimuli
for the release of endothelin-1 are thrombin, platelet-derived transforming growth factor β1
(TGF β1), angiotensin II, noradrenaline, adrenaline, vasopressin, interleukin-1, hypoxia,
ischaemia, O2-derived free radicals, and shear stress.29
All three isoforms of endothelin bind to two types of endothelin receptors: the ETA and
the ETB receptors. ETA receptors are expressed on vascular smooth muscle cells and
mostly mediate the constrictor effects. ETB receptors are predominantly located on the
endothelial cell surface and mostly mediate vasodilation by facilitating the release of NO
and PGI2.
Whereas endothelin-1 causes vasodilation at very low concentrations, it becomes the
most potent endogenous vasoconstrictor at higher concentrations, eventually leading to
myocardial ischaemia, cardiac arrhythmias, and death.30 The exact role of endothelin in the
regulation of coronary vasomotor tone is not fully understood.
Prostanoids – The major endothelium-derived vasoconstrictor prostanoids are
thromboxane A2 (TXA2) and prostaglandin H2 (PGH2). Both promote
Page 129
platelet aggregation. They activate thromboxane receptors on vascular smooth muscle cells
and platelets, this way counteracting the protective effects of NO and PGI2 on both cell
types.
Mechanical forces (for example, shear stress) or pharmacological agonists (for example,
acetylcholine) cause the endothelial co-release of vasodilatory NO and PGI2 on the one
hand, and of vasoconstrictory TXA2 and PGH2 on the other. The final vascular response
will depend on the net balance between released vasodilating and vasoconstricting factors.
Decreased release of relaxing factors combined with an increased formation of contracting
factors (as in ageing or hypertension) will attenuate endothelium-dependent relaxation.
Vasoconstriction after platelet aggregation and damage is the result of smooth muscle cell
activation by platelet-derived TXA2 (and serotonin).
EDRFs and EDCFs in disease states

Normal coronary arteries exhibit endothelium-dependent dilatation in response to both
local acetylcholine and increased flow. These physiological responses are lost in humans
with advanced coronary artery disease.31 Oxidised low density lipoproteins and hypoxia or
anoxia inhibit the release of EDRF. Ischaemia and reperfusion induce impairment in
endothelium-dependent relaxation to most EDRF/NO agonists.32 Although such studies
underscore the potential clinical relevance of impaired release of EDRF/ NO in ischaemic
syndromes, a potentially important role for endothelin is also possible. Hypoxia,33 and
decreased flow and shear stress34 may induce endothelin gene expression and secretion.
After global myocardial ischaemia and reperfusion during cardiopulmonary bypass, not
only is there impaired release of EDRF in the coronary artery, but the ischaemic event may
also increase the production of an endothelium-derived contracting factor (EDCF).35
The impaired release of EDRF will suppress coronary vasodilation in areas of
jeopardised myocardium, and will facilitate platelet adhesion and aggregation as well as
platelet-induced coronary vasoconstriction. The stimulated release of endothelin (induced
by vascular injury, impaired release of EDRF, ischaemia, and decreased shear stress) will,
in turn, also promote vasoconstriction. The finding of augmented vasoconstrictor actions of
endothelin in the presence of simultaneous inhibition of EDRF synthesis36 supports the
hypothesis that the endogenous EDRF system serves as a functionally important modulator
of the vasoconstrictor actions of endothelin. When the delicate balance between
endothelium-mediated vasodilation and vasoconstriction is disturbed in disease states
associated with injured endothelium (for example, coronary atherosclerosis), the attenuated
release of EDRF and stimulated secretion of endothelin will facilitate vasoconstriction,
thrombosis, and smooth muscle cell proliferation. An intact endothelium protects the
vasculature from a variety of
Page 130
potentially vasoconstrictory stimuli. In turn, endothelial damage can convert usually
vasodilatory stimuli into vasoconstrictory ones.
All of the presented data indicate that the endothelial system cannot be viewed as simply
an inert lining of blood vessels, but rather it must be regarded as a highly active endocrine
organ that serves a wide variety of biological functions including synthesis, metabolism,
and binding of various vasoactive and non-vasoactive substances. Impaired endothelial
function is likely to contribute significantly to the pathophysiology of coronary artery
disease and myocardial ischaemic syndromes. Endothelial dysfunction of the coronary
arteries is found across a broad spectrum of conditions in patients free of angiographic
evidence of coronary atherosclerosis, including advanced age,37 hypertension,38 left
ventricular hypertrophy,38 hyperlipidaemia,39 diabetes mellitus,40 and chronic tobacco use.41
Metabolic control of coronary vascular tone
Relationship between CBF and

CBF and are closely coupled because the myocardium depends almost
completely on aerobic metabolism (see Fig 4.1) As baseline coronary venous O2 saturation
is only 25–30%, additional O2 extraction is limited. Coronary vascular resistance therefore
responds within a second to changes in . The precise mediators responsible for the
tight coupling between myocardial demand and supply still need to be identified.42 As
matching of O2 supply to metabolic demands remains almost unaffected when
neurohumoral factors are eliminated,43 there is general agreement that CBF is metabolically
regulated.44 Alterations in the level of myocardial energy use,45 or in the balance of O2
supply and demand cause production of vasodilator substances that restore the
supply/demand balance and, thus, maintain myocardial function.
Various substances (for example, adenosine, NO, prostaglandins, CO2, H+) have been
investigated in this context.46 Changes in their concentrations in the perivascular space are
small, however, and their significance in controlling steady state coronary vascular
resistance has been questioned.47
Adenosine
When ATP utilisation exceeds resynthesis by myocardial cells, adenosine
monophosphate (AMP) will be produced. Adenosine is formed from AMP by the enzyme
5′-nucleotidase. Adenosine (and its metabolites inosine and hypoxanthine) can enter the
interstitial space and appear in the coronary sinus effluent.
As a purine compound, adenosine acts on purinergic receptors. These are subdivided into
adenosine-sensitive P1-receptors and ATP-sensitive P2-receptors. The P1-receptors can, in
turn, be further subdivided into
Page 131
myocardial A1- and vascular A2-receptors. Vascular A2-receptors are located on vascular
smooth muscle cells, and promote coronary vasodilation by stimulating production of
cAMP.
Adenosine is a potent coronary vasodilator.48 In addition, the coronary circulation
appears to be more sensitive to adenosine than the peripheral circulation.49 The exact role of
adenosine in this context, however, remains inconclusive. Adenosine has been hypothesised
to be the major metabolite that regulates CBF.50 According to this hypothesis, adenosine is
released from myocardial cells as cellular O2 tension decreases. It traverses the interstitial
space and acts on coronary vascular smooth muscle to cause vasodilation. The subsequent
increase in CBF restores myocardial O2 tension. Data exist both to support51 and to reject
the hypothesis52 that adenosine is the principal metabolic regulator of CBF.
For example, the production of adenosine increases during myocardial O2
supply/demand mismatch, and the interstitial concentration correlates with CBF.
Furthermore, uncoupling of increases in CBF from increases in myocardial metabolism has
been demonstrated. Adenosine receptor blockade almost abolished the hyperaemia
associated with dobutamine-induced increases in cardiac work, and it dramatically
exaggerated the reduction in CBF associated with intracoronary vasopressin infusion. These
findings would suggest a role for adenosine as a mediator of CBF changes.
On the other hand, physiological and pathological conditions as diverse as resting CBF,
exercise-induced coronary dilatation, reactive hyperaemia, and rapid atrial pacing are
largely unrelated to the release of adenosine. Overall evidence would suggest that
adenosine is probably involved to some extent in the metabolic control of myocardial
perfusion. It does not, however, appear to be the primary mediator of the close coupling
between flow and metabolism.
ATP-sensitive K+ channel
There is considerable evidence that opening of ATP-sensitive K+ channels

channels) is the principal mechanism of metabolic coronary vasodilation. When this
channel opens, K+ leaves the cell resulting in hyperpolarisation of the cell membrane. Such
hyperpolarisation closes Ca2+ channels. Consequently, cytosolic Ca2+ decreases and
vascular smooth muscle relaxes. channels are opened either by receptor activation
(for example, by adenosine or acetylcholine) or by metabolic factors (for example, by a
decrease in ATP as in hypoxia or severe ischaemia). This constitutes an important
mechanism for metabolic coronary vasodilation.53
Blockade of channels lowers CBF both at rest and during exercise.22 Infusion of
pinacidil, a KATP channel opener, induces coronary vasodilation comparable to that
observed during the peak reactive flow response.54 With
Page 132
channels intact, NO and adenosine are not essential for maintaining resting CBF or
for coronary vasodilation during exercise.22 Blockade of adenosine receptors after
channel blockade does not further decrease resting CBF but reduces the exercise induced
increase in CBF by more than half.55 This finding emphasises the importance of adenosine
in causing metabolic vasodilation after channel blockade. Additional NOS
inhibition (after blockade of channels and adenosine receptors) entirely abolishes
coronary vasodilation during exercise.55 This finding implies that NO contributes to
coronary vasodilation during exercise when other vasodilator systems are blocked.
Blockade of any one of the three vasodilator mechanisms alone does not blunt the increase
in CBF during exercise. Hypoxia may cause vasodilation by opening channels.56
Neurohumoral control of coronary vascular tone
Autonomic control
The coronary arteries are richly innervated by adrenergic and parasympathetic nerves.57
Superior, middle, and inferior cervical and the first four sympathetic ganglia supply the
heart with sympathetic, and the vagi with parasympathetic innervation. Sympathetic
innervation is present at all coronary microvascular segments, but the density of innervation
may differ among various vascular levels.58 Rather than increasing concentrations of
circulating levels of catecholamines, cardiac efferent sympathetic signals seem to account
for the increase in CBF in response to activation of the sympathetic nervous system.59 The
increase in CBF appears to correlate with the magnitude of regional stores of noradrenaline
(norepinephrine) in cardiac sympathetic nerve terminals.59
The role of cholinergic nerves in the regulation of CBF is controversial.60 Although
parasympathetic stimulation seems to dilate small coronary arteries,57 the response is weak
and transitory. Coronary vasodilation that normally follows parasysmpathetic stimulation
may be dependent on the release of EDRF.61
α-Adrenergic control
α1 and α2-adrenergic receptors exist throughout the coronary circulation. They are not,
however, uniformly distributed.2 The α1receptors seem to predominate in large epicardial
vessels, whereas α2-receptors predominate in vessels that are less than 100 μm in diameter.1
2
Activation of both α1- and α2-adrenergic receptors produces coronary vasoconstriction.2

In accordance with the non-uniform distribution of the receptor subtypes, the α-adrenergic
responses of the coronary circulation to various physiological and pharmacological
interventions seem to vary
Page 133
between vessel segments of different morphology and location. For example, α1- and α2-
receptors are both involved in the vasoconstriction of resistance vessels. In contrast, larger
coronary arteries mainly respond to α1-receptor activation, and mature coronary collaterals
do not respond to α-receptor stimulation.62
Paradoxically, in vitro coronary arterioles are refractory to α-adrenergic agonists.63 This
is puzzling, because isolated coronary venules63 and large coronary arteries (> 500 μm in
diameter)64 do constrict on α1- and α2-adrenergic stimulation. It has recently been shown
that α1-adrenergic stimulation causes in vivo coronary arteriolar constriction by stimulating
cardiac myocytes to release a vasoconstrictor, possibly endothelin-1.65 It thus appears that
cardiac myocytes have a requisite role in mediating α1-adrenergic coronary arteriolar
constriction.
There is evidence for baseline α-adrenoceptor-mediated coronary vasoconstriction in
humans.66 Cardiac denervation decreases coronary vascular resistance, lowers arteriovenous
O2 extraction, and raises coronary venous O2 content, indicating direct vasodilation. α-
Adrenoceptor blockade also reduces coronary vascular resistance.
Under physiological conditions, increased sympathetic activity causes dilatation of
coronary resistance vessels and, subsequently, increases myocardial blood flow. This
vasodilator response appears to be partly mediated by endothelial function.67 It may result
from the direct stimulation of α2-adrenergic receptors in intact endothelial cells, and from
the release of NO, possibly through activation of local kinin synthesis.68 In fact, removing
the vascular endothelium enhances the vasoconstrictor response to noradrenaline,69 and
inhibition of NO synthesis in coronary arteries potentiates α-adrenergic vasoconstriction.70
During neurohumoral adrenergic stimulation, α-adrenergic stimulation competes with
metabolic vasodilation.66 Although under physiological conditions the net result of α-
adrenergic stimulation is always an increase in CBF, such stimulation nevertheless
attenuates the increase in CBF by 20–30%. Such attenuation of CBF in the presence of
increased myocardial O2 demand seems paradoxical. α-Adrenoceptor-mediated
vasoconstriction may, however, have beneficial effects on the transmural blood flow
distribution.71 An α-blockade during myocardial hypoperfusion results in redistribution of
CBF from the subendocardium to subepicardium.72 This would suggest that α-receptor-
mediated vasoconstriction during hypoperfusion exhibits an ‘‘anti-steal” effect by limiting
perfusion to the subepicardium and improving that to the subendocardium. The postulated
mechanism of this beneficial effect is that α-adrenoceptor-mediated vasoconstriction
preferentially “stiffens” coronary vessels that are larger than 100 μm in diameter. This will
reduce intramyocardial vascular capacity and, subsequently, to-and-fro oscillation of CBF
during the cardiac cycle.73
Page 134
Although there is little evidence for a physiological role of α-adrenergic coronary
vasoconstriction under normal conditions, controversy continues about the importance of α-
adrenergic receptor stimulation during myocardial hypoperfusion. No doubt, sympathetic
stimulation activates vasoconstricting α-adrenoceptors in coronary arteries. The
physiological role of such α-adrenoceptor-mediated vasoconstriction still has, however, to
be determined.71
β-Adrenergic control
Contrary to what might be expected from other vascular beds, coronary β-adrenoceptors
are not only of the non-cardiac β2 subtype. Where as β2-receptors may regulate coronary
vascular resistance (CVR) in small vessels, β1-receptors dominate in large human coronary
arteries.74 There appears to be a segmental distribution of β-adrenergic receptors, with a
higher density of β-receptors in resistance vessels than in large vessels. The resistance
vessels may contain both β1- and β2-receptor subtypes, whereas large epicardial vessels
have predominantly β1-receptors. Coronary vasodilation mediated by β1- and β2-
adrenoceptors of both large and small coronary vessels,43 75 76 and of mature canine
collaterals77 has been demonstrated.
At first it would seem paradoxical that administration of noradrenaline causes
simultaneous α-adrenoceptor-mediated coronary vasoconstriction and β-adrenoceptor-
mediated vasodilation.76 These seemingly opposing effects may, however, be of benefit. α-
adrenoceptor-mediated vasoconstriction of large and medium sized coronary arteries
possibly lessens the to-and-fro oscillation of flow from subendocardium to subepicardium
during systole and diastole, thus preserving flow.78 On the other hand, direct β-
adrenoceptor-mediated vasodilation probably occurs mostly in small arteries, and serves to
match blood flow and myocardial metabolism. It may thus be postulated that α-
adrenoceptor-mediated vasoconstriction serves to adjust phasic coronary impedance, which
preserves left ventricular subendocardial perfusion, whereas β-adrenoceptor-mediated
vasodilation prevents large decreases in myocardial tissue PO2.76
Inhibition of NO synthesis by L-NAME antagonises isoproterenol-induced coronary
vasodilation.79 This would suggest that β-adrenergic dilatation of resistance vessels
involves an endothelium-dependent mechanism that is linked to the L-arginine/NO
pathway.
Cholinergic control
Acetylcholine is a powerful coronary vasodilator when given intravenously. Cholinergic
stimulation consistently results in coronary dilatation.80 Whereas in patients with
angiographically normal coronary arteries intracoronary acetylcholine causes dilatation,81 in
atherosclerotic segments it elicits constriction.82 Intracoronary acetylcholine binds to
endothelial muscarinic receptors and stimulates the release of EDRF.
Page 135
Peptidergic control
It is generally thought that neural control depends primarily on the release of
noradrenaline and acetylcholine from sympathetic and parasympathetic nerve terminals,
respectively. It is now recognised, however, that, in addition to the classic
neurotransmitters, other putative transmitters (including several vasoactive peptides) may
also be involved in regulating coronary vasomotor tone. Such peptides identified in nerves
associated with coronary vessels include neuropeptide Y (NPY), vasoactive intestinal
polypeptide (VIP), calcitonin gene-related peptide (CGRP), and tachykinins such as
substance P and neuropeptide K.83
Human epicardial coronary arteries are supplied by numerous, peptide-containing,
perivascular nerve populations.83 The number of peptide-containing nerve fibres varies with
vessel size, with the distal segments of epicardial coronary arteries being more densely
innervated than the proximal segments. NPY immunoreactive nerve fibres seem to be the
most abundant of the peptide-containing nerve populations identified in human epicardial
coronary arteries. They appear to have a distribution pattern similar to that of nerves
containing the catecholamine-synthesising enzyme tyrosine hydroxylase.83 Most NPY-
containing nerves in the heart represent postganglionic sympathetic neurons originating in
the stellate and other paravertebral ganglia. This supports the extracardiac origin of this
neuropeptide.
Neuropeptide Y has generally been regarded as a vasoconstrictor peptide.84 Its vasomotor
activity appears, however, to vary with vessel size and location. NPY does not seem to
elicit a vasoconstrictor response in epicardial coronary arteries, but it induces some
constriction of intramyocardial resistance vessels.85 Its vasomotor action is possibly
mediated by specific Y1 and Y2 receptors.86 NPY is co-stored and co-released with
noradrenaline from sympathetic nerve terminals. Sympathetic stimulation increases the
ratio of NPY to noradrenaline release. NPY is concentrated around coronary arteries and is
a potent vasoconstrictor; it possibly contributes to coronary vasoconstriction during
profound sympathetic stimulation. Although the functional significance of NPY in the
regulation of coronary vasomotor tone still has to be defined, it may well play a role in
modulating the effect of other vasoactive substances.87
Substance P and CGRP immunoreactive nerve fibres are rare in the proximal region of
epicardial coronary arteries and increase in number distally.83 Substance P and CGRP
produce a marked relaxation of epicardial coronary arteries88 89 but exert only a weak
vasodilatory effect on intramyocardial resistance vessels.89 90
Vasoactive intestinal peptide is present in post-ganglionic parasympathetic (vagal) and
intrinsic cardiac nerve fibres of humans.83 Cardiac VIP nerve fibres are predominantly
found in coronary arteries, sinoatrial and atrioventricular nodes, the atria, and the right
ventricle.91 Within the
Page 136
coronary arteries, VIP nerve fibres are present in the epicardial vessels and to a lesser
extent in the arterioles.
Various findings suggest that VIP has a direct vasodilator effect on the coronary
arteries.92 VIP appears to have a more potent effect on large epicardial vessels than on
resistance vessels. The increase in CBF during administration or release of VIP is not the
result of an increase in and cardiac metabolism.92
Although the exact physiological role of VIP in the control of the coronary circulation is
still uncertain, VIP may well be important in the regulation of CBF. Vagal nerve
stimulation releases VIP which, in turn, directly dilates coronary arteries and increases CBF
to the left ventricle.92 The phenomenon of post vagal tachycardia (although not directly
reflecting changes in CBF) is thought to be the result of VIP release from cardiac vagal
nerves after intense vagal stimulation.93
In summary, a network of neuropeptide-containing nerve fibres located in the adventitia
and at the adventitial–medial border supplies human coronary arteries.83 Neuropeptides
appear to modify CBF either directly or indirectly by modifying the effects of other
vasoactive substances.
Opposing vasodilatory and vasoconstrictory mechanisms

Final vasomotor tone is the net result of opposing vasodilatory and vasoconstrictory
mechanisms operating at various levels and modifying each other (Table 4.2). As a result of
such opposing and interacting mechanisms, the final effect of each individual mechanism
on vasomotor
Table 4.2 Opposing vasodilatory and vasoconstrictory mechanisms in the regulation of vascular
smooth muscle tone
Site Vasodilation Vasoconstriction
Autonomic nervous system Cholinergic neurons (ACh) Adrenergic neurons (NA)

Adrenal medulla (A)
Nerve terminal receptor α1-Adrenoceptors β2-Adrenoceptors
Muscarinic receptors Angiotensin II receptors
Vascular receptor β2 α1 and α2
Adenosine Angiotensin II
Endothelin B (ETB) Endothelin A (ETA)
Endothelium EDRF/NO Endothelin (high
Prostacyclin (PGI2) concentration)
Endothelin (low
concentration)
Vascular ion channel
channel Ca2+ channels
(ROC,VOC)
channel
ACh, acetylcholine; NA, noradrenaline; A, adrenaline; EDRF, endothelium-derived relaxing

factor; NO, nitric oxide; K+ATP channel, ATP-sensitive potassium channel; K+Ca channel,
calcium-activated potassium channel; ROC, receptor-operated channel; VOC, voltage-operated
channel.
(Adapted from Opie LH. The heart. 3rd edn. Philadelphia: Lippincott-Raven, 1998:255.)
Page 137
tone is difficult to predict in vivo. Whereas the in vitro effect can be described clearly, the
in vivo effect will vary with baseline conditions (for example, baseline autonomic tone,
baseline concentrations of vasoactive substances, endothelial responsiveness), species, and
vascular bed. In addition, activation of one mechanism normally results in immediate
activation of counterbalancing mechanisms to maintain homoeostasis.
There is complex interaction of the autonomic nervous system, vascular smooth muscle
cell receptor, and endothelium (Fig 4.3). Release of noradrenaline from the nerve terminal
(as a result of sympathetic activation) stimulates postsynaptic vasoconstrictory α1- and α2-
receptors. A variety of hormones and autonomic signals modulates vasomotor tone by
modifying the release of noradrenaline. The parasympathetic transmitter acetylcholine
decreases noradrenaline release by stimulation of presynaptic muscarinic receptors.
Circulating adrenaline (epinephrine) can exert dual action: it may indirectly cause
vasoconstriction by increasing the release of noradrenaline through stimulation of
presynaptic β2-receptors; or it may cause direct vasodilation by stimulation of postsynaptic
β2-receptors. Angiotensin II promotes vasoconstriction, either directly by acting on
postsynaptic angiotensin II receptors, or indirectly by increasing the release of
noradrenaline via stimulation of presynaptic angiotensin II receptors.
A couple of interactions at the endothelial level may serve as examples of opposing and
modifying mechanisms. Endothelin may stimulate NO release via ETB receptors, thereby
limiting its vasoconstrictory effect.29 Part of NO-induced vasodilation may result from
concomitant inhibition of endothelin production by NO. Like NO, PGI2 may exert a
negative feedback inhibition on endothelin production. In porcine coronary arteries, PGI2
potentiates the relaxing activity of NO by stimulating its release. Certain stimuli (for
example, shear stress, acetylcholine) cause the concomitant release of vasodilatory NO and
cyclo-oxygenase-derived contracting factors, such as TXA2, PGH2, and superoxide radical,
which all attenuate the relaxing effects of NO.94 The final vascular response will depend on
the type and degree of interactions (which are related to amounts and potency of factors
released); these are likely to differ between vascular beds. Vasomotor tone is thus under the
triple control of autonomic tone, activity of vascular smooth muscle cell receptors, and
endothelial function.
Autoregulation
Autoregulation has been defined as “the intrinsic tendency of an organ to maintain

constant blood flow despite changes in arterial perfusion pressure”.95 Using a more
operational definition, autoregulation is a proportional change in vascular resistance in
response to a change in perfusion pressure. This active change in vascular resistance
constitutes an
Page 138
Fig 4.3 Regulation of vasomotor tone by neuromodulation, vascular receptors, and endothelium. NA =
noradrenaline. A II = angiotensin II. TGF-β = transforming growth factor β. Thr = thrombin. Ach =
acetylcholine. Bk = bradykinin. ET-1 =endothelin-l. EDRF = endothelium-derived relaxing factor. NO =
nitric oxide. GC = guanylyl cyclase. Circles represent receptors: ETA = endothelin-A receptor. M =
muscarinic. P = purinergic. = Excitatory effect. = Inhibitory effect.
(Modified from Opie LH. The heart, 3rd edn. Philadelphia: Lippincott-Raven, 1998: 270.)
Page 139
intrinsic mechanism that is independent of extrinsic neurohumoral factors.
Perfusion is autoregulated mainly by arterioles that are larger than 150 μm in diameter.
As perfusion pressure continues to decline, however, smaller arterioles are recruited.96
There are lower and upper limits of autoregulation beyond which CBF will (pressure-
dependently) decrease or increase, respectively.
Such a mechanism requires immediate adjustment of vasomotor tone in response to
alterations in perfusion pressure. Autoregulatory changes in coronary vasomotor tone
behave in a dynamic fashion. If metabolic demand is kept constant, a sudden change in
coronary perfusion pressure results in an immediate directionally identical change in CBF
which returns to normal over 10–30 s46 (Fig 4.4). As, at rest, the coronary vasculature
appears to be under greater vasoconstrictor tone than other vascular beds, this greater
vasodilatory reserve provides the capacity to increase flow remarkably.
The definition of autoregulation assumes that organ metabolism and venous pressure do
not change as arterial perfusion pressure is altered. As aortic pressure is a major
determinant of left ventricular afterload, and as developed left ventricular systolic pressure
correlates with left ventricular ,97 it is impossible to study coronary autoregulation
simply by
Fig 4.4 Dynamic coronary flow response to a sudden change in perfusion pressure in the left circumflex artery
in the dog. (a) Flow response to a step decrease in pressure; (b) response to a step increase in pressure.
(Reproduced with permission from Dole WP. Autoregulation of the coronary circulation. Prog Cardiovasc
Dis 1987;29:293–323.)
Page 140
changing aortic pressure because this would lead to marked changes in myocardial
metabolism. This problem can be overcome by cannulating a coronary artery and perfusing
it with a pump. Nevertheless, low perfusion pressures may cause myocardial ischaemia
with subsequent decreases in myocardial metabolism and increases in venous pressure. On
the other hand, higher than normal perfusion may increase cardiac contractility and
myocardial metabolism, secondary to the increase in CBF. Doubling of the resting CBF
may increase the strength of cardiac contraction by 15%.98 This phenomenon is referred to
as the “Gregg effect”, after its discoverer99 (see below). Either increasing CBF by
increasing cannulated coronary artery pressure or increasing it by pharmacological
vasodilation at constant coronary perfusion pressure will elicit this effect, thus changing
myocardial metabolism simultaneously. As autoregulation constitutes a flow response to
changes in perfusion pressure, a Gregg effect will always be involved in autoregulatory
investigations. Such simultaneous changes in will complicate the interpretation of
acquired data on autoregulation.
Adenosine
The precise mechanism(s) responsible for maintaining CBF in the presence of decreasing
coronary perfusion pressure remain (s) controversial.46 100 Autoregulation is largely
unrelated to the release of adenosine.101 Furthermore, intracoronary infusion of adenosine
deaminase or adenosine receptor antagonists that blunt reactive hyperaemia does not affect
coronary autoregulation,47 101 and interstitial levels of adenosine do not change with
decreases in coronary perfusion pressure within the autoregulatory range.47 These data
would suggest that adenosine plays at best a minor role in coronary autoregulation.
EDRF/NO
Endothelium-dependent production of NO in coronary vessels appears to be an important
mechanism in the regulation of myocardial perfusion only during hypoperfusion.102
Inhibiting NO synthase with L-NAME increased the critical pressure at which myocardial
ischaemia began (lower autoregulatory break point) from 45±3 mm Hg under control
conditions to 61 ± 2 mm Hg after L-NAME (Fig 4.5). In addition, both the slope of the
coronary pressure–flow relation below the autoregulatory point, and the peak reactive
hyperaemic flow response were reduced, reflecting impaired capability to minimise
coronary vascular resistance. On the other hand, flow recruitment in response to increased
metabolic demand (that is, a twofold increase in heart rate) was not affected by L-NAME.
These findings would suggest that both initial autoregulatory adjustments to decreases in
Page 141
Fig 4.5 Plots summarising pressure-flow relationships under control conditions (open circles) and following
inhibition of nitric oxide synthesis by Nω-nitro-L-arginine (L-NAME) (hatched triangles). L-NAME had no
significant effect on flow regulation over the autoregulatory plateau. The lower autoregulatory break point
(arrows) as well as the pressure–flow relationship during ischaemia were, however, shifted to the right after
inhibition of nitric oxide production. (Reproduced with permission from Smith TP, Canty JM Jr. Modulation
of coronary autoregulatory responses by nitric oxide. Circ Res 1993;73:232–40.)
coronary perfusion pressure and flow recruitment in response to increased metabolic

demand are probably mediated by metabolic factors independent of NO production. During
hypoperfusion, however, endothelium-dependent production of NO is importantly involved
in minimising coronary vascular resistance.
O2 and CO2 tension

Changes in myocardial O2 and CO2 tensions may mediate coronary autoregulation.103 104
There appears to be a strong inverse relationship between coronary venous O2 tension and
coronary autoregulation. Good autoregulation was observed when coronary venous O2
tension was 25 mm Hg(3·3 kPa) and autoregulation was lost when venous O2 tension was
more than 32 mm Hg(4·3 kPa).104 This, again, would indicate that the dominant mechanism
of coronary autoregulation is metabolic. The normal resting coronary sinus PO2 of 18–25
mm Hg(2·4–3.3 kPa) indicates a tight coupling between and myocardial O2
delivery.
Changes in PaO2 (arterial O2 tension) itself induce variations in coronary vasomotor
tone, independent of O2 content and metabolic regulation of CBF. A high PaO2 constricts
coronary arteries, possibly mediated by closure of channels.105

Page 142
Although possibly involved in the phenomenon of autoregulation, the effect of CO2
tension is probably small at venous O2 tension (PVO2) of more than 20 mm Hg(2·7 kPa).103
It remains unclear whether coronary auto-regulation is mediated directly by changes in
tissue O2 tension or by some mediating factors, such as channels.
ATP sensitive K+ channels
Glibenclamide, a putative blocker of channels, abolishes auto-regulation in the

canine heart perfused with blood in situ.106 channels open when intracellular ATP
concentration falls in myocardial and vascular smooth muscle cells.53 56 107 It has been
suggested that a decrease in myocardial tissue O2 tension may be sensed by the vascular
smooth muscle cell. By regulation of the generation of ATP, this decrease may
subsequently lead to an opening of channels in vascular smooth muscle cells.53 This
is consistent with the previous finding that coronary autoregulation is coupled strongly to
tissue O2 tension rather than to .104 It is, however, also conceivable that other tissue
factors modify coronary autoregulation because channel activity is also affected by

tissue concentrations of ADP, lactate, and extracellular cations.108 It is interesting to note
that adenosine-induced coronary vasodilation is in part mediated by channels.56 109
Myogenic control
Vascular smooth muscle contracts in response to increased distending force (referred to
as ‘‘stretch-induced contraction”). This mechanism may form the basis for the myogenic
control of perfusion. It might contribute to coronary autoregulation, particularly in
arterioles that are less than 100 μm in diameter.42 How this mechanism works is unknown;
stretch activated channels may be involved. Increased intraluminal pressure causes
arteriolar smooth muscle to contract. The subsequent increase in resistance tends to
normalise blood flow despite an elevated perfusion pressure. This mechanism is called
myogenic control. It seems to be more important in arterioles smaller than 100 μm,
compared with larger arterioles, and in subepicardial compared with subendocardial
arterioles.110
Studies in isolated coronary resistance arterioles that are uncoupled from metabolic
mediators of vascular tone have shown that both myogenic and flow-mediated,
endothelium-dependent mechanisms may be involved in local blood flow regulation
throughout much of the coronary microcirculation.111 There seem to be transmural
differences in the potential for myogenic dilatation. In subendocardial arterioles, myogenic
dilatation was maximal at pressures of 60 cm H2O, whereas in subepicardial arterioles
myogenic dilatation was present up to pressures as low as 40 cm H2O.112 The changes in
vessel diameter over the pressure range studied were,
Page 143
however, only modest (< 10% of resting values). In contrast, flow-dependent influences on
coronary vasomotor tone increased arteriolar diameter by up to 25% of resting value.111
Thus, although myogenic responses have been observed in isolated coronary vessels,
their demonstration in vivo has been difficult, probably because of the predominance of the
metabolic control of CBF. Although some argument can be made for a myogenic
component in coronary reactive hyperaemia,113 overall there is little experimental evidence
for myogenic effects in the coronary circulation.
Left ventricular autoregulation

Subendocardial flow of the left coronary artery (LCA) begins to decline at 37 mm Hg,
and subepicardial flow at 25 mm Hg.114 Down to a LCA perfusion pressure of 40 mm Hg
(that is, almost a 50% decrease), LCA autoregulation remains preserved. In anaesthetised
animals with stable myocardial metabolism, CBF stays rather constant over a pressure
range of 60–160 mm Hg.103 Increasing myocardial metabolism results in an upward shift of
the autoregulatory curves.104
Autoregulatory capacity of the subendocardium is considerably less than that of the
subepicardium. Whereas in the awake animal subepicardial autoregulation was preserved
until the perfusion pressure reached 25 mm Hg, subendocardial autoregulation was
impaired at 40 mm Hg.114 Such reduced subendocardial vasodilator reserve explains the
well known greater vulnerability of the left ventricular subendocardium to ischaemia when
compared with the subepicardium.115
Experimental conditions and extent of cardiac work influence the lower limits of
autoregulation. At a heart rate of 100 beats/min, perfusion pressure can decrease to 38 mm
Hg before subendocardial ischaemia develops, but ischaemia develops at 61 mm Hg if the
heart rate is 200 beats/min.116 Even during tachycardia and at pressures as low as 33 mm
Hg, subepicardial flow was maintained.
What could be the explanation for the lower subendocardial vasodilator reserve? The
pressure–flow relationship during a long diastole would suggest that pressure at zero flow
(Pzf) is higher in the subendocardium than in the subepicardium.117 Pzf is, however,
unlikely to be high in any myocardial layer, and unlikely to be higher in the
subendocardium by more than 2–3 mm Hg.118 This excludes the possibility that different
diastolic myocardial compressive forces account for the differences in regional vasodilator
reserve.
During systole, compressive forces are greatest in the subendocardium, resulting in
higher vascular resistance. With the start of diastole, flow follows first the path of lowest
vascular resistance (that is, the large subepicardial vessels) before reaching the higher
resistance subendocardial vessels. This hypothesis of an interaction between systole and
diastole has
Page 144
been proposed as a possible explanation for the lower subendocardial vasodilator reserve.118
Right ventricular autoregulation

Absence of effective right coronary artery (RCA) pressure–flow autoregulation (that is,
variation of RCA flow with RCA perfusion pressure) has been a consistent finding.119
However, when the change in right ventricular (accompanying changes in RCA
flow) is taken into account, the “corrected” RCA autoregulatory gain is comparable to the
LCA autoregulatory gain (gain being a measure of autoregulatory potency).120 RCA
autoregulation in conscious dogs fails at approximately 42 mm Hg,120 a pressure somewhat
higher than for failure of LCA autoregulation.114 Anaesthesia does not seem to depress
RCA autoregulatory gain, but attenuates the range.120 Right ventricular myocardial
performance is preserved in both conscious and anaesthetised animals until RCA perfusion
pressures fall below 50 mm Hg.120 Above 50 mm Hg, right ventricular function does not
seem to be coupled to RCA blood flow. Probably related to low right ventricular systolic
pressures, inner and outer layers of the right ventricle have similar lower limits of
autoregulation.
Conclusion
The precise activating signals and mediators of autoregulation still have to be defined. At
the lower range of autoregulation, NO formation102 and opening of channels106 may
be involved. It is possible that O2 tensions within a critical range may be the initial
metabolic stimulus for coronary autoregulation. Although adenosine does not appear to be
essential for autoregulation, it is probably reasonable to speculate that compensatory
mechanisms other than adenosine exist in the myocardium, which gain importance once the
action of adenosine is prevented. It is rather unlikely that a highly oxidative organ such as
the heart depends on just one “host defence” mechanism to preserve the balance between
supply and demand. Furthermore, the mechanisms involved in the control of coronary
vasomotor tone may well be distributed in a way that metabolic, flow-dependent, or
myogenic control mechanisms predominate in different classes of arteriolar vessel.
Furthermore, each of the mediators may have differing physiological relevance under
differing conditions.
Variations in myocardial perfusion
Blood supply to the heart is affected by ventricular contraction and relaxation. Any
myocardial stress is expected to alter underlying myocardial geometry and, in turn,
geometry of intramyocardial vessels. This may affect
Page 145
vascular resistance and flow. Forces acting on a myocardial segment may include
interactions between myofibres and adjacent vessels, intramyocardial tissue (fluid)
pressure, cavity pressure transmitted as radial stress, myofibre force transmitted
tangentially, and pericardial pressure. Intramyocardial tissue pressure possibly constitutes a
major component of coronary vascular resistance. It differs between right and left
ventricular wall, atrial and ventricular chambers, and epicardial and endocardial layers, and
it varies with systole and diastole.
During systole, the myocardial fibre bands that encircle both ventricles exert lateral
shearing forces on the perpendicularly penetrating intramyocardial branches of the large
epicardial vessels. This may entirely abolish flow to certain regions of the myocardium.121
At the same time, however, those intramyocardial vessels running parallel to the muscle
fibres are compressed during systole which propagates blood further downstream. Coronary
venous blood is drained almost entirely in systole partly as a result of this squeezing effect
of myocardial contraction. This in itself promotes coronary arterial inflow.122
Resting myocardial perfusion

Although mean resting myocardial blood flow in all larger mammals is consistently 0·6–
1·0 ml/min per g,58 local flow distribution is remarkably heterogeneous varying between
20% and 200% of the mean value.123 Such flow heterogeneity most probably reflects local
differences in aerobic metabolism.124 Thus, high flow areas do not represent a state of
luxury perfusion, but rather reflect higher local O2 demands. They should be as susceptible
to hypoperfusion as low flow areas.
Phasic myocardial perfusion

Intermittently high and low extravascular resistances during systole and diastole are
responsible for the phasic pattern of coronary perfusion (Fig 4.6). In the left ventricle,
extravascular compression during systole is so great that normally only 20–30% of left
coronary artery flow occurs during systole.125 As a result of considerably lower systolic
and, thus, intramyocardial pressure generated by the thin-walled right ventricle, right
coronary arterial systolic flow constitutes a much greater proportion of total coronary
inflow (30–50%)(Fig 4.6). Systolic myocardial compression increases with rises in heart
rate, afterload, preload, and contractility. The relative contribution of each of these factors
to the regulation of myocardial perfusion remains controversial.126 When coronary
perfusion pressure is increased experimentally, systolic and diastolic components of
epicardial flow increase, with diastolic flow dominating.127
The effects of contraction on phasic flow pattern seem to vary with the contractile state
of the myocardium. During uniform global contraction, local intramyocardial forces rather
than transmitted forces (that is, cavity
Page 146
Fig 4.6 Phasic blood flows of right and left coronary arteries in relation to aortic pressure. Whereas right
coronary artery flow exists throughout the cardiac cycle, left coronary flow is largely confined to diastole.
(Reproduced with permission from Berne RM, Levy MR Cardiovascular physiology, 5th edn. St Louis: CV
Mosby, 1986:200.)
pressure) are primarily responsible for the phasic flow variation.128 In a non-contracting
region of myocardium, however, left ventricular pressure becomes a major determinant of
phasic flow pattern.128
This dependence of flow characteristics on the contractile state may be explained by
differences in myocardial behaviour during systole. During normal contraction, the
myocardium stiffens and then becomes resistant to deformation by externally transmitted
stress such as cavitary pressure. When contraction is absent, however, the respective
myocardial segment fails to stiffen, and the intramyocardial vessels are now prone to
deformation by externally applied forces.
Normal intramyocardial and peripheral epicardial coronary arteries exhibit almost
exclusive forward flow during diastole. Reverse flow is frequently observed during
systole.129 With coronary artery stenosis,
Page 147
systolic reverse flow increases whereas diastolic forward flow decreases.130 Reduced back
pressure to systolic reverse flow as a result of decreased poststenotic distal pressure, and
increased coronary arterial capacitance resulting from a pressure-dependent capacitance
change131 may both serve as explanation.
Transmural myocardial perfusion

During the cardiac cycle, transmural flow distribution is non-uniform. Normally,
subendocardial flow exceeds subepicardial flow by about 10%, resulting in an endo-
/epicardial perfusion ratio of 1·1. As systolic intramyocardial compressive forces are
greatest in the subendocardium, but low in the subepicardium, it was postulated that the
subepicardium was perfused throughout the cardiac cycle, whereas the subendocardium
received blood only during diastole. However, findings of possibly very little flow to the
subepicardium even during systole argue against such a mechanism.118
During normal contraction, primarily subendocardial vessels are compressed,132 and a
steep transmural gradient of intramyocardial pressure persists in an empty beating heart.
When a beating heart is arrested, subendocardial flow has been shown to increase but
subepicardial flow to decrease.78 Thus, contraction appears to augment subepicardial
perfusion. It has been proposed that during cardiac contraction blood is squeezed out of
subendocardial vessels and translocated in a retrograde fashion to superficial layers of the
myocardium. In this way, the subepicardial vessels represent a low pressure and low
resistance “sink” for any translocation of blood from deep to superficial layers. It is to be
expected that the amount of retrograde flow will depend on the transmural pressure
gradient. Consequently, absent global78 or regional contraction128 causes marked changes in
phasic inflow to the myocardium and in transmural flow distribution, most probably by
altering the transmural pressure gradient. When contraction is absent, left ventricular
pressure becomes a powerful determinant of transmural flow distribution and,
subsequently, the subendocardial/subepicardial flow ratio more than doubles, indicating
favoured subendocardial perfusion.128
The left ventricular subendocardium is more susceptible to hypoperfusion than the
epicardium, for two major reasons: (1) coronary vessels penetrate the ventricular wall from
outside to inside, so that the inner layers are further away from the epicardial conduit
arteries; and (2) systolic compressive force is greater in the subendocardium than in the
epicardium.133 Whereas subendocardial arterioles narrow by about 20% during systole,
subepicardial arteriolar diameter changes little during the cardiac cycle.134 It has been
postulated that systole and diastole create a to-and-fro oscillation of blood flow in the
coronary vessels that penetrates the myocardial wall from outside to inside.78 135 The
oscillating flow only fills
Page 148
and empties intramyocardial arteries and arterioles, without providing nutritive flow to the
subendocardial capillaries. At the end of diastole, blood flows from the aorta through the
coronary vascular tree to all layers of the myocardium. At the start of systole, cardiac
contraction produces retrograde flow in penetrating coronary arteries. Part of the retrograde
flow originates at the subendocardial layers, and is diverted to the outer epicardial layers
that are less compressed than the inner layers.78 134 At the onset of diastole, forward flow
refills the intramyocardial vessels that had been compressed and emptied during the
preceding systole. Only then does nutritive flow through the capillaries of the
subendocardium begin. As a result of the oscillatory coronary flow pattern, the outer layer
of the myocardium (the epicardium) is perfused throughout the cardiac cycle, but the inner
layer (the subendocardium) is perfused only during diastole.
Tachycardia and intense adrenergic activation of the heart (as during maximal exercise)
pose a particular threat to the subendocardium because: (1) the duration of diastole becomes
very short, whereas systole hardly shortens; (2) increases; and (3) intramyocardial
blood volume increases as a result of metabolically-induced coronary vasodilation.
Increased intramyocardial capacitance will contribute to a proportionally greater oscillatory
flow, which encroaches on nutritive flow to the subendocardium. In other words, when
intense cardiac stress results in high and CBF, in tachycardia, and in compromise of
local metabolic vasodilator reserve, the mechanical effects of systolic compression and
oscillating transmural flow may lead to subendocardial hypoperfusion.
Thus, opposing factors such as wall stiffness, regional contractility, and cavity pressure
influence phasic inflow and transmural flow distribution. The net result on myocardial
perfusion will depend on their interactions during specific conditions. With changes in
baseline conditions, the relative importance of each of the factors will also change. During
uniform global contraction, local tissue–vessel interactions and/or intramyocardial fluid
pressure caused by active contraction appear to play a dominant role in determining
myocardial perfusion. When regional contraction is abolished, left ventricular pressure
assumes a more prominent role. This, however, does not imply that the effects of
intramyocardial forces caused by contraction are simply removed when myocardial
stiffness is low. Rather, changes in underlying conditions will result in complex changes in
the spatial and temporal course of forces that act on intramyocardial vessels. Such
considerations may help us to understand mechanisms governing myocardial perfusion
under clinical conditions of regional myocardial dysfunction (such as myocardial ischaemia
and “stunned” myocardium).
Coronary vascular resistance

Vascular tone of coronary arterioles larger than 100 μm in diameter is regulated
primarily by myogenic and local metabolic factors.136 Adenosine
Page 149
and dipyridamole produce their vasodilator effects mainly in vessels of this size. Up to 40%
of total coronary resistance resides, however, in small arteries that are 100–400 μm in
diameter. 137 Vasomotor tone in these small coronary arteries is controlled by endothelial,
humoral, and neural autonomic influences.138 Although these small vessels are not under
direct metabolic control, they may, nevertheless, influence maximal coronary blood flow
importantly. When vasodilation of the arterioles causes an increase in flow, the resultant
increase in endothelial shear stress will augment NO production and, subsequently, cause
vasodilation of the small arteries.22 136 Thus, arteriolar vasodilation leads to dilatation of the
small resistance arteries by way of endothelium-dependent flow/shear-mediated NO
release. In hyperlipidaemia, such endothelium-dependent, flow-mediated vasodilation of
the small arteries may be impaired or abolished.139
The principal functions of the resistive vessels are: (1) to match myocardial blood flow to
when metabolic demand varies; and (2) to maintain myocardial perfusion
(proportionate to metabolic demand) when perfusion pressure varies. Under pathological
conditions, dilatation of the resistance vessels compensates for the increase in resistance of
the conduit arteries caused by stenoses. In experimental animals, as well as in humans,
normal epicardial arteries vasodilate in response to increases in blood flow. This response is
endothelium-dependent, because removal of the endothelium abolishes the flow-mediated
vasodilation. Metabolic vasodilation of the coronary resistance vessels during exercise or
pacing also causes flow-mediated vasodilation of the epicardial arteries.18
For a better understanding of vasomotor dysregulation which may lead to myocardial
ischaemia, it is helpful to distinguish functionally between two components within the
resistive coronary vasculature, which are arranged in series (Fig 4.7): (1) more proximal,
‘‘prearteriolar” vessels in which vasomotor tone and flow are not metabolically regulated
because of the epicardial position or thickness of the vessel wall; and (2) more distal,
“arteriolar” vessels in which major pressure reductions occur, and in which tone and flow
are metabolically regulated.140 As no anatomical differentiation is possible between the two
segments, functional differentiation is expected to be vague.141
The continuous matching of flow to demand by the resistive vessels is so precise that
myocardial O2 extraction remains practically constant over a wide range of metabolic
demand and coronary perfusion pressure. The site of action and the mechanisms involved
may differ, however, when flow is varied either in response to changes in metabolic
demands at constant aortic pressure, or when flow is maintained in the presence of changes
in aortic pressure at constant metabolic demand.
Reduction of coronary perfusion pressure to 40 mm Hg causes dilatation of vessels
smaller than 100 μm in diameter, but constriction of larger
Page 150
vessels.96 Two possible explanations exist for such a heterogeneous response: (1) large
reductions in coronary perfusion pressure cause metabolically-induced dilatation of
arteriolar vessels but a passive, low, distending, pressure-induced reduction in prearteriolar
vessel diameter; or (2) reduced flow-mediated release of EDRF/NO causes prearteriolar
constriction.
When CBF increases in response to metabolically-mediated arteriolar vasodilation, one
would expect a proportionate increase in pressure drop across the prearteriolar vessels
unless they respond with compensatory flow-mediated vasodilation. If this does not occur
(as with endothelial dysfunction in coronary artery disease), pressure at the origin of the
maximally dilated arterioles may decrease to an extent that impairs
Fig 4.7 Schematic representation of pressure reduction from the aorta to capillaries. Three components of the
coronary circulation can be identified: (1) conductive vessels with negligible pressure drop; (2) prearteriolar
vessels with moderate pressure drop; and (3) arteriolar vessels with greatest pressure drop. (Reproduced with
permission from Maseri A, Crea F, Cianflone D. Myocardial ischemia caused by distal coronary
vasoconstriction. Am J Cardiol 1992;70: 1602-5.)
Page 151
subendocardial perfusion.135 Thus, during metabolically-induced arteriolar vasodilation
adequate perfusion pressure at the origin of the arterioles can be maintained only if there is
an appropriate change in vasomotor tone at the prearteriolar level. Flow-mediated
vasodilation on the basis of a tonic release of endothelial NO and/or other endothelium-
derived relaxing factors may be primarily involved in the adaptation of prearteriolar vessel
size to changes in flow.
Coronary blood flow varies little over a wide range of aortic pressures (see
Autoregulation). This requires constant changes in coronary vasomotor tone and vascular
resistance. Depending on the initiating stimulus, the adaptive mechanisms maintaining CBF
may differ. An increase in aortic pressure may primarily trigger prearteriolar constriction,
thus maintaining optimal pressure at the origin of the arterioles. On the other hand, when
aortic pressure decreases metabolically-induced arteriolar dilatation, in addition to flow-
mediated prearteriolar dilatation may be required to maintain CBF.
It is the traditional view that myocardial ischaemia in coronary artery disease is caused
by a fixed or dynamic obstruction of large conduit coronary arteries resulting in a critical
reduction in perfusion pressure at the origin of fully dilated arterioles. Clinical studies
suggest, however, that myocardial ischaemia can also be the result of constriction of small,
distal, resistive coronary vessels.142 Such small vessel constriction can result from
dysfunction of either prearteriolar or arteriolar vessels. The mechanisms of the abnormal
behaviour of the distal resistive vessels resulting in myocardial ischaemia can be multiple,
and may involve different sites.143 144
The concept that coronary vascular resistance resides in large conduit arteries and,
primarily, in small resistance vessels practically ignores the contribution of the venous
system to total coronary vascular resistance. Whereas under control conditions only 7% of
the total coronary vascular resistance resides in veins that are more than 150 μm in
diameter, under conditions of vasodilation with dipyridamole the total contribution of the
venous component increases to 31%.145 Thus, during coronary vasodilation the coronary
venous system may considerably modify myocardial perfusion. Furthermore, by affecting
ventricular distensibility related to changes in myocardial blood volume,146 alterations in
venous reactivity may also have an impact on diastolic cardiac function.
Isolated coronary venules dilate in response to an increase in flow.147 This flow-induced
vasodilation is endothelium-dependent and mediated by the release of a nitrovasodilator.
Endothelial disruption results in flow-induced constriction, suggesting that shear stress may
directly act on the vascular smooth muscle.147 Whereas the additive effects of flow-induced
dilatation and possibly myogenic relaxation of arterioles can maximise myocardial oxygen
delivery during elevated , the flow-induced venular dilatation may possibly
contribute to a reduction in postcapillary resistance.
Page 152
Pressure-flow relationships
Flow (F) across a resistance (R) is determined by the difference between pressures
upstream and downstream of the resistance (P), according to the equation:
The driving pressure, P, is the coronary perfusion pressure. Whereas upstream pressure can
clearly be defined as the pressure at the aortic root, definition of what constitutes
downstream pressure is still somewhat controversial.148 149 For Fig 4.8, the downstream
pressure is taken as the coronary sinus pressure or left ventricular end diastolic pressure,
and the
Fig 4.8 Schematic presentation of coronary pressure–flow relationships in the normal left ventricle during
autoregulated flow (A), and during maximal vasodilation (D). R1 and R2 denote coronary vascular reserves at
mean coronary perfusion pressures of 100 (R1) and 75 mm Hg (R2) (Reproduced with permission from
Hoffman JIE. Maximal coronary flow and the concept of coronary vascular reserve. Circulation 1984;70:153–
9.)
Page 153
(apparent) coronary perfusion pressure is the difference between downstream and mean
aortic diastolic pressure.
The normal pressure–flow relationship (A, Fig 4.8) has three separate regions of interest:
(1) a high pressure region, where flow increases with coronary perfusion pressure; (2) an
intermediate pressure range, where flow changes little with changes in coronary perfusion
pressure (referred to as “autoregulation”); and (3) a low pressure region, where flow
decreases with decreasing coronary perfusion pressure.
The controversy surrounding the downstream value for coronary perfusion pressure
centres around the extrapolation of the low pressure region to zero flow (Fig 4.9). It is
accepted that the coronary pressure–flow relationship always has a positive intercept.148 150
l51 It had been suggested that myocardial perfusion stopped at pressures considerably higher
than coronary sinus pressure.117 The pressure at which flow stopped was termed “critical
closing pressure” or ‘‘zero flow pressure” (Pzf or Pf=0) (Fig 4.9). This would imply that the
effective downstream pressure for the calculation of coronary vascular resistance would be
Pzf rather than the much lower coronary sinus or left ventricular end diastolic pressure.
Results were, however, obtained on the basis of measurements on large proximal epicardial
vessels, and it now seems that flow through intramural coronary vessels actually continues
after large superficial vessel flow has already ceased. Furthermore, it has been shown that
forward movement of red blood cells in arterioles that are 20 μm in diameter continues until
perfusion pressure is only a few millimetres of mercury higher than
Fig 4.9 Graphic presentation of the concept of critical closing pressure (Pzf). During a long diastole, pressure
in the aortic root or coronary artery, and flow through an epicardial coronary artery are recorded
simultaneously (solid line in pressure–flow plot). The pressure at which flow through the (epicardial)
coronary artery is zero (Pzf) is either determined directly or derived by linear extrapolation (dashed line).
(Reproduced with permission from Sethna DH, Moffitt EA. An appreciation of the coronary circulation.
Anesth Analg 1986;65:294–305.)
Page 154
coronary sinus pressure.152 Thus, the initial findings of a considerably higher Pzf than
coronary sinus pressure can probably be explained on the basis of arterial collateral flow153
and coronary capacitance.154
An increase in coronary sinus pressure can shift the entire pressure–flow relationship to a
higher Pzf without affecting its slope.150 151 This way, coronary sinus pressure may become
a determinant of CBF and transmural flow distribution. There appear to be regional
differences in the response of the pressure–flow relationship to an increase in coronary
sinus pressure: whereas the subepicardial pressure–flow relationship behaves like the total
relationship, in the subendocardium neither the pressure–flow relationship nor
intramyocardial tissue pressures changed in response to high coronary sinus pressure.151
Such transmural differences in intramyocardial pressures and Pzf support the existence of a
vascular waterfall mechanism in the myocardium.155
The concept of Pzf remains controversial, and the clinical relevance of changes in Pzf in
response to interventions remains questionable. For clinical purposes, coronary perfusion
pressure in normal coronary arteries can be defined as the difference between (mean) aortic
diastolic and coronary sinus or left ventricular end diastolic (or even right atrial) pressures.
Flow-function relationship
Tennant and Wiggers156 first reported in 1935 that myocardial performance is closely
coupled to perfusion. The degree of transmural flow reduction required to impair contractile
function, however, varies considerably. As the right ventricle differs importantly from the
left ventricle in O2 extraction, , and transmural pressure, differences in the flow–
function relationship between the two ventricles are to be expected.
Right and left ventricular functions are well maintained as perfusion pressures are
reduced to 50 mm Hg.114 120 Whereas left ventricular perfusion remained unchanged,114
RCA flow decreased by 34% as perfusion pressure declined.120 As the right ventricle has a
greater O2 extraction reserve than the left,157 moderate right ventricular hypoperfusion
might be compensated for by increasing O2 extraction. Alternatively, right ventricular
internal cardiac work and, thus, right ventricular decreases as the reduction in
coronary perfusion pressure reduces intravascular volume and, thus, right ventricular
systolic stiffness.158 For these reasons, a highly effective autoregulation is less important for
the right than for the left ventricle to maintain myocardial performance above a perfusion
pressure of 50 mm Hg.
Gregg was the first to note that coronary perfusion pressure independently affects
and contractile performance.99 The exact mechanism of this so called “Gregg phenomenon”
is still unknown. It is possible that the two aspects of the Greggs’ phenomenon are based on
different mecha-
Page 155
nisms: the increase in may be related to an augmentation in CBF, and the increase
in contractility may be secondary to changes in vascular shear stress induced by perfusion
pressure,159 possibly at the capillary level.160
The increase in correlates with an increase in coronary vascular volume.161 Such
expanded volume could result in a more rigid coronary hydraulic “skeleton”, which, in turn,
requires more energy (that is, greater for systolic contraction.158 This is in
agreement with the finding that perfusion-induced changes in and contractility are
greatest during poor coronary autoregulation, when coronary vascular volume is also
greatest.158 161
When the increase in CBF is caused by an elevation in perfusion pressure, this
phenomenon is termed the “garden hose effect’’, because the resultant coronary distension
is thought to cause an increase in myocyte sarcomere length, thereby augmenting cardiac
performance by the Frank–Starling mechanism. Increases in CBF without a change in
perfusion pressure have a similar effect.158
Coronary flow reserve
The pressure–flow relationship during maximum coronary vasodilation is almost straight

(D, Fig 4.8). Maximum flow at any given pressure is determined primarily by the cross–
sectional area of the resistance vessels. The difference between autoregulated (A, Fig 4.8)
and maximally dilated (D, Fig 4.8) flow is the reserve capacity for vasodilation. Coronary
flow reserve (CFR) is defined as the ratio of CBF during maximal coronary vasodilation to
CBF under resting conditions.162 Depending on the technique used (that is, nuclear
techniques, transoesophageal echocardiography, coronary catheterisation, intracoronary
Doppler wire), normal values for CFR have been reported from as low as 1·8 to as high as
5·5.163 A value of 3·0 or more is, however, generally considered to be normal.163
Three principal factors determine CFR: (1) a vascular factor (reflecting coronary vascular
resistance); (2) an extravascular factor (reflecting myocardial compressive forces); and (3) a
rheological component (reflecting blood composition). As CFR presents a ratio, its value
will depend substantially on the baseline coronary flow rate. Age, myocardial hypertrophy,
and elevated heart rate and blood pressure may increase resting coronary flow, thus
“falsely” lowering the CFR.163
In general, CFR values of less than 3·0 suggest microvascular disease on the basis of
functional or structural anomalies. In routine clinical practice, maximal vasodilation of the
resistance vessels is best achieved pharmacologically (for example, by adenosine). Physical
exercise or cardiac pacing is less suitable.
Page 156
Coronary flow reserve may be reduced principally by elevation of the autoregulatory line
or depression of the line for maximally attainable flow. Anaemia, left ventricular
hypertrophy, or increased lead to higher resting CBF, thus elevating the
autoregulation line (A2, Fig 4.10). If the maximally attainable flow at any given pressure
remains unchanged, coronary flow reserve will decrease (R1–R2, Fig 4.10). Concomitant
hypertension will, however, preserve CRF (R3, Fig 4.10).
Coronary artery disease, tachycardia, polycythaemia, and marked increases in left
ventricular end diastolic pressure or contractility all
Fig 4.10 Diagram of pressure–flow relationships during autoregulation in the normal (A1) and in the left
ventricle with anaemia or increased contractility or hypertrophy (A2). As pressure–flow relationships during
maximal vasodilation remain about the same (D), coronary vascular reserve will be lower in the abnormal
(R2) than in the normal (R1) left ventricle. Elevation of coronary perfusion pressure (as in hypertension) may
restore or even increase absolute coronary vascular reserve (R3). (Reproduced with permission from Hoffman
JIE. Maximal coronary flow and the concept of coronary vascular reserve. Circulation 1984;70:153–9.)
Page 157
decrease the slope of the line for maximally attainable flow (D, Fig 4.10), so reducing the
CFR (R1-R2, Fig 4.10). In the case of coronary artery disease, this reflects a decrease in
total cross sectional area of the coronary vascular bed.
Many studies have estimated CFR from the reactive myocardial hyperaemia that follows
transient total coronary occlusion. If there is no hyperaemia, coronary vascular reserve is
deemed to have been exhausted by having to compensate for a stenosis in the supply vessel,
and the supply vessel is deemed to be “critically constricted”.164 Reactive hyperaemia is,
however, the result of complex interactions between vasodilator metabolites, myogenic
relaxation, and coronary capacitance.165 It is difficult to standardise because it varies with
the duration of occlusion, basal , coronary perfusion pressure, sympathetic tone, and
165
reactivity of adenosine receptors. Pharmacological vasodilators such as adenosine or
dipyridamole may lower resistance much more than ischaemic stimuli166 (Fig 4.11). During
maximal exercise in humans, CBF may increase by two or four times the control values, but
with dipyridamole, increases of three to
Fig 4.11 Autoregulation and coronary vasodilator reserve. Traditionally, it was thought that the coronary
pressure–flow relationship below the autoregulatory range (line a) was identical with the pressure–flow
relationship in the maximally dilated bed (line b), indicating exhausted coronary vasodilator reserve.
Considerable flow reserve well below the autoregulatory range has, however, been demonstrated in response
to intracoronary application of vasodilation (line c). Thus, pharmacological vasodilator reserve (c–a) may be
preserved at times of exhausted (physiological) autoregulatory vasodilator reserve. (Reproduced with
permission from Dole WP. Autoregulation of the coronary circulation. Prog Cardiovasc Dis 1987;29:293–
323.)
Page 158
five times the control have been described.167 Thus, endogenous (ischaemic) stimuli may
not be as effective in revealing true (maximal) coronary flow reserve as potent
pharmacological vasodilators (Fig 4.11).
The overall evidence seems to indicate that the degree of CFR provides potentially valid
information on what might happen during maximal stress, as long as other factors that
might modify maximally attainable and autoregulated flows are controlled for (or
accounted for if changes do occur during the study). Determination of CFR using maximal
pharmacological vasodilation has been used to assess the physiological significance of
coronary stenoses, and the results of coronary angioplasty and coronary bypass surgery.
Methods to determine coronary/myocardial blood flow

Correct interpretation of coronary circulatory behaviour in response to a given
intervention is critically dependent on reliable determination of perfusion. Contradictory
findings may at times reflect differences in methodology employed to measure perfusion.
All methods of measuring blood flow have technical limitations.168
Radioactive microsphere technique
This has become the standard technique to evaluate transmural myocardial blood flow
distribution in the experimental animal.169 It is an extraction method that rests on the
assumptions that microspheres injected into the bloodstream are distributed like red blood
cells, and that basically all microspheres are trapped in tissues on their first passage. Under
such conditions, blood flow should be proportional to radioactivity per tissue mass.
Microspheres are denser and larger, and stream more centrally than red cells. They are,
therefore, not distributed in quite the same way as native blood. As subendocardial shunting
(non-entrapment) of 9 μm spheres is somewhat greater than shunting of 15 μm spheres,
endo-/epicardial flow ratios determined with 9 μm spheres tend to be lower than those
determined with 15 μm spheres. In general, endo-/epicardial values are lower in open chest
preparations than in conscious dogs.58 The necessity of obtaining multiple myocardial tissue
samples requires sacrifice of the animal and prohibits its use in humans.
Inert gas clearance technique
By determining the arteriocoronary sinus difference in gas saturation, CBF can be

calculated from equations originally developed by Kety and Schmidt.170 For this technique
to be accurate, several requirements must be
Page 159
fulfilled:171 (1) the indicator must be physiologically inert; (2) partition coefficients for the
gas in the myocardium, fat, red cells, and plasma must be known; (3) venous blood from
the entire myocardial region of interest must be sampled; (4) during the period of saturation
and desaturation (5–20 min), the CBF must be stable; and (5) the entire washout curve must
be analysed.
The principal disadvantages of the inert gas clearance technique (particularly when using
non-radioactive tracers) include: (1) poor spatial resolution; (2) length of time required for
making measurements; (3) variations in venous drainage patterns distorting results; and (4)
the inability to determine right ventricular and atrial flows because venous blood from those
areas cannot be sampled. Although the introduction of radioactive xenon-133(133Xe) has
added spatial resolution to the inert gas clearance technique,172 transmural blood
distribution cannot be determined, the temporal resolution remains limited (requiring stable
flow during measurements), only a limited number (two to five) of flow determinations is
possible (as a result of xenon’s eight times greater solubility in fat than in myocardium),
and only flow rates below 200 ml/min per 100 g can be measured with reasonable
accuracy.173 Therefore, determination of flow rates during maximal coronary vasodilation is
probably not possible.
Electromagnetic flow meters
The principle of this technique is based on Faraday’s induction law which states that a
conductor moving in a magnetic field generates electric current. When blood flows through
the vessel, electrical voltage proportional to the rate of the flow is generated, and this
voltage is recorded by an appropriate flow meter apparatus. Use of accurate, non-occlusive
(electronic) zero that can be used at any time and compared with occlusive (mechanical)
zero, automatic flow ranging (autoranging) that provides an instantaneous readout of blood
volume flow, and precalibration during flow probe production that eliminates the need for
time consuming flow calibration have made it possible to provide immediate, reliable
determinations of blood flow.
The principal drawbacks of this technique include the relatively large sized flow probes
(limiting its application to major epicardial vessels), its inability to determine transmural
blood flow distribution, and its invasiveness (requiring thoracotomy and coronary artery
dissection). Variability in contact between flow probe and vessel may result in major errors
in measurement.
Ultrasonic Doppler techniques
This technique is based on the Doppler principle that states that, when sound waves are
reflected from a moving structure, the frequency of the
Page 160
reflected wave is shifted to a higher or lower frequency (Doppler shift). In the case of blood
vessels, the Doppler shift is caused by ultrasonic waves that reflect off moving red blood
cells. Coronary flow velocity can be measured by several Doppler techniques: Doppler flow
meter, Doppler catheter,174 Doppler guidewire,175 176 an epicardial probe,177 a
transoesophageal probe,178 and a transthoracic Doppler probe.176 Guide wires as small as
0·36 mm in diameter, tipped with a 15 MHz piezoelectric ultrasound transducer, allow
reliable velocity measurements even in poststenotic areas of coronary arteries.176
Intracoronary Doppler measurements can be affected by positioning of the wire, tortuous
segments, and areas of varying luminal dimensions or configurations.179 Non-invasive flow
velocity measurement using transthoracic Doppler echocardiography, under the guidance of
colour Doppler flow mapping correlates well with measurements obtained invasively by an
intracoronary Doppler catheter.176
The temporal resolution of this technique is ideal, allowing continuous online
measurements of changes in velocity. The principal drawback of this technique is the fact
that velocity rather than flow is measured. For the correct interpretation of results, it is
crucial to have precise knowledge of the magnitude and direction of any change in coronary
cross sectional area in response to a given intervention. Otherwise, statements regarding
changes in absolute flow are not valid.
Thermodilution technique
This technique is based on the principle that a change in temperature of a downstream

fluid–blood mixture is proportional to blood flow. A fluid indicator (saline) with a known
temperature (lower than that of blood) is infused (upstream) into the coronary sinus or great
cardiac vein. Although coronary sinus thermodilution is still being used for the estimation
of myocardial blood flow,180 it has almost been forgotten that the initial studies validating
this technique were performed in animals under very rigid conditions, which cannot be
reproduced in humans (catheter tied into position in coronary sinus; assurance of adequate
mixing of blood and tracer; normal coronary arteries and myocardium).
The technique has very crude spatial and only modest temporal resolution. Wide
variations in coronary venous drainage patterns make direct comparisons of flow between
patients impossible. Even slight variations in catheter position in the coronary sinus may
have major effects on flow determination. This practical limitation will make it difficult to
interpret results obtained during interventions that are likely to affect cardiac size, and it
will make comparison of data gained on different occasions almost impossible (even in the
same patient).
As a result of both fundamental and practical deficiencies, thermodilution should be used
only in very special circumstances. At present, only
Page 161
large changes (>30%) in coronary sinus or great cardiac vein flow, in patients with no
evidence of coronary artery disease, should be accepted as semiquantitative indices of
directional changes in CBF.171
Positron emission tomography
With this technique the essential features of myocardial perfusion, function, metabolism,
and viability can be examined.181 Positron emission tomography (PET) in humans uses
either [13N] ammonia7 or 15O-labelled water182 as indicators of myocardial blood flow.
Dynamic imaging assesses uptake and retention of the tracer 13NH3 or H215O. The uptake
phase depends on blood flow, and the later (myocardial) retention phase depends on the
conversion of NH3 into glutamine. The conversion rate is considered to be constant, and
can thus be dealt with mathematically. As positrons emitted per unit mass constitute the
mass factor that is incorporated into the rate constant for the delivery of the tracer into the
myocardium, myocardial blood flow is reported as millilitres/100 grams per minute (ml/100
g per min).
At present, PET is the only method that allows non-invasive quantification of myocardial
blood flow in humans.183 The correlation with the reference technique (that is, microsphere
technique) is excellent (r = 0·97) over a wide flow range.184 PET measures actual
myocardial nutritive tissue perfusion. Its major disadvantages are great expense, and limited
spatial and transmural resolution.185
Thallium-201 scintigraphy
Thallium-201(201T1) is an analogue of potassium. Its myocardial distribution is the result

of two processes. The initial distribution phase reflects myocardial blood flow distribution.
The following (redistribution) phase reflects the uptake of 201T1 by the myocardial cell.
Uptake of the tracer depends on the same Na+/K+ pump (Na+/K+ ATPase) that transports
potassium. Tl binds faster, however, and to more sites than potassium. In hypoperfused but
viable myocardium, both the uptake and washout of 201T1 are delayed. Whereas initially
there is a difference in radioactivity between normally and hypoperfused areas of
myocardium (referred to as a “filling defect” when comparing the intensity of radioactivity
of the hypoperfused with that of the normally perfused area), levels of radioactivity in both
areas tend to equalise in time (referred to as “redistribution”, an actual misnomer because
there is no redistribution in the sense of the word). In infarcted areas of myocardium, there
is no uptake of 201T1, and no equalisation of radioactivity between various areas can take
place, that is, the ‘‘filling defect” persists. Thallium scintigraphy cannot assess absolute
blood flow, acute changes in flow, and transmural flow distribution.
Page 162
Myocardial contrast echocardiography
This technique involves intracoronary injection of microbubbles with a mean size of 10

μm or less under echocardiographic monitoring.186 Contrast enhancement indicates
adequate perfusion, whereas contrast defects are consistent with areas of myocardial
hypoperfusion.186 187 Myocardial contrast echocardiography enables assessment of the
spatial distribution of microvascular perfusion. It requires coronary artery catheterisation,
and does not allow quantification of blood flow.
Conclusion
All methods (with the exception of PET) are invasive. They require either venous access
or surgical dissection of vessels (electromagnetic flow probes), or they involve catheter
manipulation in coronary arteries (intracoronary Doppler ultrasonography, xenon-133
clearance). In addition, most methods expose the patient to ionising radiation. At present,
no technique is available for clinical use that would allow determination of absolute
myocardial perfusion with high temporal and spatial resolution.
In general, methods that use tracers have a poor temporal resolution because they require
at least one circulation time for tracer distribution, or 5–20 minutes for inert gas saturation
or desaturation to occur. In contrast, methods that assess arterial blood velocity or venous
drainage have good temporal but poor spatial resolution. Newer techniques such as
magnetic resonance imaging, contrast echocardiography, and ultrafast X-ray computed
tomography have theoretically a sufficent spatial resolution to distinguish variation in
transmural perfusion.168
1 Chilian WM. Adrenergic vasomotion in the coronary microcirculation. Basic Res Cardiol 1990;85(suppl
I):111–20.
2 Chilian WM. Functional distribution of α1- and α2-adrenergic receptors in the coronary microcirculation.
Circulation 1991;84:2108–122.
3 Bittl JA, Levin DC. Coronary arteriography. In: Braunwald E, ed, Heart disease, 5th edn. Philadelphia:
WB Saunders, 1997:240–72.
4 Sonntag M, Deussen A, Schultz J, Loncar R, Hort W, Schrader J. Spatial heterogeneity of blood flow in
the dog heart. I. Glucose uptake, free adenosine and oxidative/glycolytic enzyme activity. Pflügers Arch
1996;432:439–50.
5 Hutchins GM, Moore GW, Hatton EV. Arterial–venous relationships in the human left ventricular
myocardium. Anatomic basis for countercurrent regulation of blood flow. Circulation 1986;74:1195–202.
6 Schaper W, Görge G, Winkler B, Schaper J. The collateral circulation of the heart. Prog Cardiovasc Dis
1988;31:57–77.
7 Cohen MV. Myocardial ischemia is not a prerequisite for the stimulation of coronary collateral
development. Am Heart J 1993;126:847–55.
8 Schaper W. New paradigms for collateral vessel growth (Editorial). Basic Res Cardiol 1993;88:193–8.
9 Charney R, Cohen M. The role of the coronary circulation in limiting myocardial ischemia and infarct
size. Am Heart J 1993;126:937–45.
10 Kinn JW, Bache RJ. Effect of platelet activation on coronary collateral blood flow. Circulation
1998;98:1431–7.
Page 163
11 Leong L, Sturm M, Taylor R. The lyso-precursor of platelet activating factor (lyso-PAF) in ischemic
myocardium. J Lipid Mediat 1991;4:277–88.
12 Pearson PJ, Vanhoutte PM. Vasodilator and vasoconstrictor substances produced by the endothelium. Rev
Physiol Biochem Pharmacol 1993;122:l–67.
13 Furchgott RF, Zawadzky JV. The obligatory role of endothelial cells in the relaxation of arterial smooth
muscle by acetylcholine. Nature 1980;288:373–6.
14 de Belder AJ. Radomsky MW. Nitric oxide in the clinical arena (Editoral). J Hypertens 1994;12:617–24.
15 Lüscher TF, Boulanger CM, Yang Z. Interactions between endothelium-derived relaxing and contracting
factors in health and cardiovascular diseases. Circulation 1993;87(suppl V):V36–44.
16 Yanagisawa M. The endothelin system. A new target for therapeutic intervention. Circulation
1994;89:1320–22.
17 Losano G, Pagliaro P, Gatullo D, Marsh NA. Control of coronary blood flow by endothelial release of
nitric oxide. Clin Exp Pharmacol Physiol 1994;21:783–9.
18 Quyyumi AA, Dakak N, Andrews NP, Gilligan DM, Panza JA, Cannon RO. Contribution of nitric oxide
to metabolic coronary vasodilation in the human heart. Circulation 1995;92:320–6.
19 Duncker DJ, Bache RJ. Inhibition of nitric oxide production aggravates myocardial hypoperfusion during
exercise in the presence of a coronary artery stenosis. Circ Res 1994;74:629–40.
20 Bernstein RD, Ochoa FY, Xu X, et al. Function and production of nitric oxide in the coronary circulation
of the conscious dog during exercise. Circ Res 1996;79:840–8.
21 Gödecke A, Decking UKM, Ding Z, et al. Coronary hemodynamics in endothelial NO synthase knockout
mice. Circ Res 1998;82:186–94.
22 Ishibashi Y, Duncker DJ, Zhang J, Bache RJ. ATP-sensitive K+-channels, adenosine, and nitric oxide-
mediated mechanisms account for coronary vasodilation during exercise. Circ Res 1998;82:346–59.
23 Bassenge E. Endothelium-mediated regulation of coronary tone. Basic Res Cardiol 1991;86(suppl 2):69–
76.
24 Cohen RA, Vanhoutte PM. Endothelium-dependent hyperpolarization: beyond nitric oxide and cyclic
GMP. Circulation 1995;92:3337–49.
25 Campbell WB, Gebremedhin D, Pratt PF, Harder DR. Identification of epoxyeicosatrienoic acids as
endothelium-derived hyperpolarizing factors. Circ Res 1996;78:415–23.
26 Ming Z, Parent R, Lavallée M. Nitric oxide-independent dilation of conductance coronary arteries to
acetylcholine in conscious dogs. Circ Res 1997;81:977–87.
27 Matsunaga T, Okumura K, Tsunoda R, Tayama S, Tabuchi T, Yasue H. Role of adenosine in regulation of
coronary flow in dogs with inhibited synthesis of endothelium-derived nitric oxide. Am J Physiol
1996;270:H427–34.
28 Altmann JD, Kinn J, Duncker DJ, Bache RJ. Effect of inhibition of nitric oxide formation on coronary
blood flow during exercise in the dog. Cardiovasc Res 1994;28:119–24.
29 Wight E, Noll G, Lüscher TF. Regulation of vascular tone and endothelial function and its alterations in
cardiovascular disease. Ballière’s Clin Anaesthesiol 1997;11:531–60.
30 Clozel J-P, Clozel M. Effects of endothelin on the coronary vascular bed in open-chest dogs. Circ Res
1989;65:1193–200.
31 Cox DA, Vita J, Treasure CB, et al. Atherosclerosis impairs flow-mediated dilation of coronary arteries in
humans. Circulation 1989;80:458–65.
32 Warren JB, Pons F, Brady AJB. Nitric oxide biology: implications for cardiovascular therapeutics.
Cardiovasc Res 1994;28:25–30.
33 Kourembanas S, Marsden PA, McQuillan LP, Faller DV. Hypoxia induces endothelin gene expression
and secretion in cultured human endothelium. J Clin Invest 1991;88:1054–7.
34 Sharefkin JB, Duamond SL, Eskin SG, McIntire LV, Dieffenbach CW. Fluid flow decreases
preproendothelin mRNA levels and suppresses endothelin-1 peptide release in cultured human endothelial
cells. J Vasc Surg 1991;14:1–9.
35 Pearson PJ, Lin PJ, Schaff HV. Production of endothelium-derived contracting factor is enhanced after
coronary reperfusion. Ann Thorac Surg 1991;51:788–93.
Page 164
36 Lerman A, Sandok EK, Hildebrand FL Jr, Burnett JC Jr. Inhibition of endothelium-derived relaxing factor
enhances endothelin-mediated vasoconstriction. Circulation 1992;85:1894–8.
37 Celermajer DS, Sorensen KE, Spiegelhalter DJ, Georgakopoulos D, Robinson J, Deanfield JE. Aging is
associated with endothelial dysfunction in healthy men years before the age related decline in women. J
Am Coll Cardiol 1994;24:471–6.
38 Treasure CB, Klein JL, Vita JA, et al. Hypertension and left ventricular hypertrophy are associated with
impaired endothelium-mediated relaxation in human coronary resistance vessels. Circulation 1993;87:86–
93.
39 Zeiher AM, Drexler H, Saurbier B, Just H. Endothelium-mediated coronary blood flow modulation in
humans: effects of age, atherosclerosis, hypercholesterolemia, and hypertension. J Clin Invest
1993;92:652–62.
40 Johnstone MT, Creager SJ, Scales KM, Cusco JA, Lee BK, Creager MA. Impaired endothelium-
dependent vasodilation in patients with insulin dependent diabetes mellitus. Circulation 1993;88:2510–16.
41 Quillen JE, Rossen JD, Oskarsson HJ, Minor RL, Lopez AG, Winniford MD. Acute effect of cigarette
smoking on the coronary circulation: constriction of epicardial and resistance vessels. J Am Coll Cardiol
1993;22:642–7.
42 Muller JM, Davis MJ, Chilian WM. Integrated regulation of pressure and flow in the coronary
microcirculation. Cardiovasc Res 1996;32:668–87.
43 Bassenge E, Heusch G. Endothelial and neuro-humoral control of coronary blood flow in health and
disease. Rev Physiol Biochem Pharmacol 1990;116:77–165.
44 Bardenheuer H, Schrader J. Supply-to-demand ratio for oxygen determines formation of adenosine by the
heart. Am J Physiol 1986;250:H173–80.
45 Olsson RA, Bünger R. Metabolic control of coronary blood flow. Prog Cardiovasc Dis 1987;29:369–87.
46 Dole WP. Autoregulation of the coronary circulation. Prog Cardiovasc Dis 1987;29:293–323.
47 Hanley F, Grattan MT, Stevens MB, Hoffman JIE. Role of adenosine in coronary autoregulation. Am J
Physiol 1986;250:H558–66.
48 Olsson RA, Pearson JD. Cardiovascular purinoceptors. Physiol Rev 1990;70:761–845.
49 Glover DK, Ruiz M, Yang JY, et al. Pharmacological stress thallium scintigraphy with 2-
cyclohexylmethylindenehydrazino adenosine (WRC-0470). Circulation 1996;94: 1726–32.
50 Berne RM. The role of adenosine in the regulation of coronary blood flow. Circ Res 1980;47:807–13.
51 Martin SE, Lenhard SD, Schmarkey LS, Offenbacher S, Odle BM. Adenosine regulates coronary blood
flow during increased work and decreased supply. Am J Physiol 1993;264:H1438–46.
52 Rossen JD, Oskarsson H, Minor RL Jr, Talman CL, Winniford MD. Effect of adenosine antagonism on
metabolically mediated coronary vasodilation in humans. J Am Coll Cardiol 1994;23:1421–6.
53 Nichols CG, Lederer WJ. Adenosine triphosphate-sensitive potassium channels in the cardiovascular
system. Am J Physiol 199l;261:H1617–86.
54 Imamura Y, Tomoike H, Narishige T, Takahashi T, Kasuya T, Takeshita T. Glibenclamide decreases
basal coronary blood flow in anesthetized dogs. Am J Physiol 1992;263:H339–404.
55 Duncker DJ, van Zon ND, Pavek TJ, Herrlinger SK, Bache RJ. Endogenous adenosine mediates coronary
vasodilation in response to exercise after K+ATP channel blockade. J Clin Invest 1995;95:285–95.
56 Daut J, Maier-Rudolph W, Von Beckerath N, Mehrke G, Günther K, Goedel-Meiner L. Hypoxic dilation
of coronary arteries is mediated by ATP-sensitive potassium channels. Science 1990;247:1341–44.
57 Young MA, Vatner SF. Regulation of large coronary arteries. Circ Res 1986;59:579–96.
58 Feigl EO. Coronary physiology. Physiol Rev 1983;63:1–206.
59 DiCarli MF, Tobes MC, Mangner T, et al. Effects of cardiac sympathetic innervation on coronary blood
flow. N Engl J Med 1997;336:1208–15.
Page 165
60 Cox DA, Hintze TH, Vatner SF. Effects of acetylcholine on large and small coronary arteries in conscious
dogs. J Pharmacol Exp Ther 1983;225:764–9.
61 Feigl EO. EDRF - a protective factor? Nature 1988;331:490–1.
62 Harrison DG, Sellke FW, Quillen JE. Neurohumoral regulation of coronary collateral vasomotor tone.
Basic Res Cardiol 1990;85 (Suppl I): 121–9.
63 Jones CJH, deFily DV, Kuo L, Davis MJ, Chilian WM. α-Adrenergic responses of isolated canine
coronary microvessels. Basic Res Cardiol 1995;90:61–9.
64 Chilian WM, Harrison DG, Haws CW, Snyder WD, Marcus ML. Adrenergic coronary tone during
submaximal exercise in the dog is produced by circulating catecholamines: evidence for adrenergic
denervation supersensitivity in the myocadium but not in coronary vessels. Circ Res 1986;58:68–82.
65 Tiefenbacher CP, DeFily DV, Chilian WM. Requisite role of cardiac myocytes in coronary α1-adrenergic
constriction. Circulation 1998;98:9–12.
66 Lorenzoni R, Rosen SD, Camici PG. Effect of α1-adrenoceptor blockade on resting and hyperemic
myocardial blood flow in normal humans. Am J Physiol 1996;271:H1302–6.
67 Zeiher AM, Drexler H. Wollschläger H, Just H. Endothelial dysfunction of the coronary microvasculature
is associated with coronary blood flow regulation in patients with early atherosclerosis. Circulation
1991;84:1984–92.
68 Kichuk MR, Seyedi N, Zhang X, et al. Regulation of nitric oxide production in human coronary
microvessels and the contribution of local kinin formation. Circulation 1996;94:44–51.
69 Cocks TM, Angus JA. Endothelium-dependent relaxation of coronary arteries by noradrenaline and
serotonin. Nature 1983;305:627–30.
70 Berkenboom G, Unger P, Fang ZY, Fontaine J. Endothelium-derived relaxing factor and protection
against contraction to norepinephrine in isolated canine and human coronary arteries. J Cardiovasc
Pharmacol 1991;17(suppl 3):S127–32.
71 Baumgart D, Ehring T, Kowallik P, Guth BD, Krajcar M, Heusch G. Impact of α-adrenergic coronary
vasoconstriction on the transmural myocardial blood flow distribution during humoral and neuronal
adrenergic activation. Circ Res 1993;73:869–86.
72 Nathan HJ, Feigl EO. Adrenergic coronary vasoconstriction lessens transmural steal during coronary
hypoperfusion. Am J Physiol 1986;250:H645– 53.
73 Morita K, Mori H, Tsujioka K, et al. α-Adrenergic vasoconstriction reduces systolic retrograde coronary
blood flow. Am J Physiol 1997;273:H2746–55.
74 Amenta F, Coppola L, Gallo P, et al. Autoradiographic localization of β-adrenergic receptors in human
large coronary arteries. Circ Res 1991;68:1591–9.
75 Trivella MG, Broten TP, Feigl EO. ß-Receptor subtypes in the canine coronary circulation. Am J Physiol
1990;259:H1575–85.
76 Miyashiro JK, Feigl EO. Feedforward control of coronary blood flow via coronary ß-receptor stimulation.
Circ Res 1993;73:252–63.
77 Feldman RD, Christy JP, Paul ST, Harrison DG. β-Adrenergic receptors on canine coronary collateral
vessels: Characterization and function. Am J Physiol 1989;257: H1634–9.
78 Flynn AE, Coggins DL, Goto M, Aldea GS, Austin RE, Doucette JW, et al. Does systolic subepicardial
perfusion come from retrograde subendocardial flow? Am J Physiol 1992;262:H1759–69.
79 Parent R, Al-Obaidi M, Lavallée M. Nitric oxide formation contributes to ß-adrenergic dilation of
resistance coronary vessels in conscious dogs. Circ Res 1993;73:241–51.
80 Van Winkle DM, Feigl EO. Acetylcholine causes coronary vasodilation in dogs and baboons. Circ Res
1989;65:1580–93.
81 Hodgson JMB, Marshall JJ. Direct vasoconstriction and endothelium-dependent vasodilation:
Mechanisms of acetylcholine effects on coronary flow and arterial diameter in patients with nonstenotic
coronary arteries. Circulation 1989;79:1043–51.
82 Ludmer PL, Selwyn AP, Shook TL, et al. Paradoxical vasoconstriction induced by acetylcholine in
atherosclerotic coronary arteries. N Engl J Med 1986;315:1046–51.
83 Gulbenkian S, Opgaard OS, Ekman R, et al. Peptidergic innervation of human epicardial coronary
arteries. Circ Res 1993;73:579–88.
Page 166
84 Tseng C-J, Robertson D, Light RT, Atkinson JR, Robertson RM. Neuropeptide Y is a vasoconstrictor of
human coronary arteries. Am J Med Sci 1988;296:11–16.
85 Clarke JG, Davies GJ, Kerwin R, et al. Coronary artery infusion of neuropeptide Y in patients with angina
pectoris. Lancet 1987;i:1057–9.
86 Sheikh SP, Håkanson R, Schwartz TW. Y1 and Y2 receptors for neuropeptide Y. FEBS Lett
1989;245:209–14.
87 Gulbenkian S, Edvinsson L, Opgaard OS, Valença A, Wharton J, Polak JM. Neuropeptide Y modulates
the action of vasodilator agents in guinea pig epicardial coronary arteries. Regul Pept 1992;40:351–62.
88 Franco-Cereceda A. Calcitonin gene-related peptide and human epicardial coronary arteries: presence,
release and vasodilator effects. Br J Pharmacol 1991;102:506–610.
89 Yamamoto H, Yoshimura H, Noma M, Kai H, Kikuchi Y. Preservation of endothelium-dependent
vasodilation in the spastic segment of the human epicardial coronary artery by substance P. Am Heart J
1992;123:298–303.
90 Ludman PF, Maseri A, Clark P, Davies GJ. Effects of calcitonin gene-related peptide on normal and
atheromatous vessels and on resistance vessels in the coronary circulation in humans. Circulation
1991;84:1993–2000.
91 Wharton J, Gulbenkian S. Peptides in the mammalian cardiovascular system. Experientia 1989;56:292–
316.
92 Feliciano L, Henning RJ. Vagal nerve stimulation releases vasoactive intestinal peptide which
significantly increases coronary artery blood flow. Cardiovasc Res 1998;40:45–55.
93 Hill MR, Wallick DW, Mongeon LR, Martin PJ, Levy MN. Vasoactive intestinal polypeptide antagonists
attenuate vagally induced tachycardia in the anesthetized dog. Am J Physiol 1995;269:H1467–72.
94 Lüscher TF, Boulanger CM, Dohi Y, Yang ZH. Endothelium-derived contracting factors. Hypertension
1992;19:117–30.
95 Johnson PC. Review of previous studies and current theories of autoregulation. Circ Res 1964;15(suppl
l):2–9.
96 Chilian WM, Layne SM. Coronary microvascular responses to reductions in perfusion pressure: Evidence
for persistent arteriolar vasomotor tone during coronary hypoperfusion. Circ Res 1990;66:1227–38.
97 Rooke GA, Feigl EO. Work as a correlate of canine left ventricular oxygen consumption, and the problem
of catecholamine oxygen wasting. Circ Res 1982;50:273–86.
98 Downey JM. Myocardial contractile force as a function of coronary blood flow. Am J Physiol
1976;230:1–6.
99 Gregg DE. Effect of coronary perfusion pressure or coronary flow on oxygen usage of the myocardium.
Circ Res 1963;13:497–500.
100 Feigl EO. Coronary autoregulation. J Hypertens 1989;7 (suppl 4);S55–8.
101 Dole WP, Yamada N, Bishop VS, Olsson RA. Role of adenosine in coronary blood flow regulation after
reductions in perfusion pressure. Circ Res 1985;56:517–24.
102 Smith TP Jr, Canty JM Jr. Modulation of coronary autoregulatory responses by nitric oxide. Evidence for
flow-dependent resistance adjustments in conscious dogs. Circulation 1993:73:232–40.
103 Broten TP, Feigl EO. Role of myocardial oxygen and carbon dioxide in coronary autoregulation. Am J
Physiol 1992;262:H1231–7.
104 Dole WP, Nuno DW. Myocardial oxygen tension determines the degree and pressure range of coronary
autoregulation. Circ Res 1986;59:202–15.
105 Baron JF, Vicaut E, Hou X, Duvelleroy M. Independent role of arterial O2 tension in local control of
coronary blood flow. Am J Physiol 1990;258:H1388–94.
106 Narishige T, Egashira K, Akatsuka Y, Katsuda Y, Numaguchi K, Sakata M, Takeshita A. Glibenclamide,
a putative ATP-sensitive K+ channel blocker, inhibits coronary autoregulation in anesthetized dogs. Circ
Res 1993;73:771–6.
107 Clapp LH, Gurney AM. ATP-sensitive K+ channels regulate resting potential of pulmonary arterial
smooth muscle cells. Am J Physiol 1992;262:H916–20.
108 Miyoshi Y, Nakaya Y, Wakatsuki T, et al. Endothelin blocks ATP-sensitive K+ channels and depolarizes
smooth muscle cells of porcine coronary artery. Circ Res 1992;70:612–6.
Page 167
109Aversano T, Ouyang P, Silverman H. Blockade of ATP-sensitive potassium channel modulates reactive
hyperemia in the canine coronary circulation. Circ Res 1991;69:618–23.
110Rajagopalan S, Dube S, Canty JM Jr. Regulation of coronary diameter by myogenic mechanisms in
arterial microvessels greater than 100 microns in diameter. Am J Physiol 1995;268:H788–93.
111Kuo L, Davis MJ, Chilian WM. Endothelium-dependent, flow-induced dilation of isolated coronary
arterioles. Am J Physiol 1990;259:H1063–70.
112Kuo L, Davis MJ, Chilian WM. Myogenic activity in isolated subepicardial and subendocardial coronary
arterioles. Am J Physiol 1988;255:H1558–62.
113McHale PA, Dube GP, Greenfield JC Jr. Evidence for myogenic vasomotor activity in the coronary
circulation. Prog Cardiovasc Dis 1987;30:139–46.
114Canty JM Jr. Coronary pressure-function and steady-state pressure-flow relations during autoregulation in
the unanesthetized dog. Circ Res 1988;63:821–36.
115 Hoffman JIE. Transmural myocardial perfusion. Prog Cardiovasc Dis 1987;29:429–64.
116Canty JM Jr, Giglia J, Kandath D. Effect of tachycardia on regional function and transmural myocardial
perfusion during graded coronary pressure reduction in conscious dogs. Circulation 1990;82:1815–25.
117 Bellamy RF. Diastolic coronary artery pressure–flow relations in the dog. Circ Res 1978;43:92–101.
Hoffman JIE. Transmural myocardial perfusion. In: Kajiya F, Klassen GA, Spaan JAE, Hoffman JIE, eds,
118
Coronary circulation - basic mechanism and clinical relevance. Tokyo: Springer-Verlag, 1990:141–52.
119Smolich JJ, Weissberg PL, Broughton A, Korner PI. Comparison of left and right ventricular blood flow
responses during arterial pressure reduction in the autonomically blocked dog: evidence for right
ventricular autoregulation. Cardiovasc Res 1988;22:17–24.
120Bian X, Williams AG Jr, Gwirtz PA, Downey HF. Right coronary autoregulation in conscious,
chronically instrumented dogs. Am J Physiol 1998;275:H169–75.
121Downey JM, Kirk ES. Inhibition of coronary blood flow by a vascular waterfall mechanism. Circ Res
1975;36:753–60.
122Goto M, Tsujioka K, Ogasawara Y, et al. Effect of blood filling in intramyocardial vessels on coronary
arterial inflow. Am J Physiol 1990;258:H1042–8.
123Austin RE Jr, Aldea GS, Coggins DL, Flynn AE, Hoffman JIE. Profound spatial heterogeneity of
coronary reserve. Discordance between patterns of resting and maximal myocardial blood flow. Circ Res
1990;67:319–31.
124Loncar R, Flesche CW, Deussen A. Coronary reserve of high- and low-flow regions in the dog heart left
ventricle. Circulation 1989;98:262–70.
Downey JM, Kirk ES. Distribution of the coronary blood flow across the canine heart wall during systole.
125
Circ Res 1974;34:251–7.
Westerhof N. Physiological hypotheses - intramyocardial pressure. A new concept, suggestions for
126
measurement. Basic Res Cardiol 1990;85:105–19.
127Recchia FA, Senzaki H, Saeki A, Byrne BJ, Kass DA. Pulse pressure-related changes in coronary flow in
vivo are modulated by nitric oxide and adenosine. Circ Res 1996;79:849–56.
128Doucette JW, Goto M, Flynn AE, Austin RE Jr, Husseini WK, Hoffman JIE. Effects of cardiac
contraction and cavity pressure on myocardial blood flow. Am J Physiol 1993;265:H1342–52.
129Chilian WM, Marcus ML. Effects of coronary and extravascular pressure on intramyocardial and
epicardial blood velocity. Am J Physiol 1985;248:H 170–8.
130Goto M, Flynn AE, Doucette JW, et al. Effect of intracoronary nitroglycerin administration on phasic
pattern and transmural distribution of flow during coronary artery stenosis. Circulation 1992;85:2296–
304.
131Canty JF, Klocke F, Mates RE. Pressure and tone dependence of coronary diastolic input impedance and
capacitance. Am J Physiol 1985;248:H700–11.
132Goto M, Flynn AE, Doucette JW, et al. Cardiac contraction affects deep myocardial vessels
predominantly. Am J Physiol 1991;261:H1417–29.
Page 168
133Heineman FW, Grayson J. Transmural distribution of intramyocardial pressure measured by micropipette
technique. Am J Physiol 1985;249:H1216–H23.
134Yada T, Hiramatsu O, Kimura A, et al. In vivo observation of subendocardial microvessels of the beating
porcine heart using a needle-probe videomicroscope with a CCD camera. Circ Res 1993;72:939–46.
135 Hoffman JIE, Spaan JAE. Pressure-flow relations in coronary circulation. Physiol Rev 1990;70:331–90.
136Komaru T, Lamping KG, Eastham CL, Dellsperger KC. Role of ATP-sensitive potassium channels in
coronary microvascular autoregulatory responses. Circ Res 1991;69:1146–51.
137Chilian WM, Eastham CL, Marcus ML. Microvascular distribution of coronary vascular resistance in
beating left ventricle. Am J Physiol 1986;251:H779–88.
138Jones CJH, Kuo L, Davis MJ, Chilian WM. Regulation of coronary blood flow: coordination of
heterogeneous control mechanisms in vascular microdomains. Cardiovasc Res 1995;29:585–96.
139Yokoyama I, Ohtake T, Momomura S-I, Nishikawa J, Sasaki Y, Omata M. Reduced coronary flow
reserve in hypercholesterolemic patients without overt coronary stenosis. Circulation 1996;94:3232–8.
Maseri A, Crea F, Cianflone D. Myocardial ischemia caused by distal coronary vasoconstriction. Am J
140
Cardiol 1992;70:1602–5.
141Marcus ML, Chilian WM, Kanatsuka H, Dellsperger KC, Eastham CL, Lamping KG. Understanding the
coronary circulation through studies at the microvascular level. Circulation 1990;82:l–7.
142 Maseri A. Coronary vasoconstriction: visible and invisible. N Engl J Med 1991;325:1579–80.
143Maseri A, Crea F. Segmental control of vascular tone in the coronary circulation and pathophysiology of
ischemic heart disease. J Appl Cardiovasc Biol 1991;2:163–73.
144Maseri A, Crea F, Kaski JC, Crake T. Mechanisms of angina pectoris in syndrome X. J Am Coll Cardiol
1991;17:499–506.
Chilian WM, Layne SM, Klausner EC, Eastham CL, Marcus ML. Redistribution of coronary
145
microvascular resistance produced by dipyridamole. Am J Physiol 1989;256:H383–90.
146Watanabe J, Levine MJ, Bellotto F, Johnson RG, Grossman W. Effects of coronary venous pressure on
left ventricular distensibility. Circ Res 1990;67:923–32.
147Kuo L, Arko F, Chilian WM, Davis MJ. Coronary venular responses to flow and pressure. Circ Res
1993;72:607–15.
148Spaan JAE. Coronary diastolic pressure–flow relation and zero-flow pressure explained on the basis of
intramyocardial compliance. Circ Res 1985;56:293–309.
149Klocke FJ, Mates RE, Canty JM Jr, Ellis AK. Coronary pressure–flow relationships. Controversial issues
and probable implications. Circ Res 1985;56:310–23.
150Bellamy RF, Lowensohn HS, Ehrlich W, Baer RW. Effect of coronary sinus occlusion on coronary
pressure–flow relations. Am J Physiol 1980;239:H57–64.
Cantin B, Rouleau JR. Myocardial tissue pressure and blood flow during coronary sinus pressure
151
modulation in anesthetized dogs. J Appl Physiol 1992;73:2184–91.
152Kanatsuka H, Ashikawa K, Suzuki T, Komaru T, Suzuki T, Takishima T. Diameter change and pressure–
red blood cell velocity relations in coronary microvessels during long diastoles in the canine left ventricle.
Circ Res 1990;66:503–10.
153Messina LM, Hanley FL, Uhlig PN, Baer RW, Grattan MT, Hoffman JIE. Effects of pressure gradients
between branches of the left coronary artery on the pressure axis intercept and the shape of steady state
circumflex pressure–flow relations in dogs. Circ Res 1985;56:11–19.
154 Mates RE, Klocke FJ, Canty JM Jr. Coronary capacitance. Prog Cardiovasc Dis 1988;31:1–15.
155Fahri ER, Klocke FJ, Mates RE, et al. Tone-dependent waterfall behavior during venous pressure
elevation in isolated canine hearts. Circ Res 1991;68:392–401.
156Tennant R, Wiggers CJ. The effect of coronary occlusion on myocardial contraction. Am J Physiol
1935;112:36l.
Page 169
157Murakami H, Kim S-J, Downey HF. Persistent right coronary flow reserve at low perfusion pressure. Am
J Physiol 1989;256:H1176–84.
158Iwamoto T, Bai X-J, Downey HF. Coronary perfusion-related changes in myocardial contractile force and
systolic ventricular stiffness. Cardiovasc Res 1994;28:1331–6.
159Dijkman MA, Heslinga JW, Sipkema P, Westerhof N. Perfusion-induced changes in cardiac O2
consumption and contractility are based on different mechanisms. Am J Physiol 1996;271:H984–9.
160Dijkman MA, Heslinga JW, Sipkema P, Westerhof N. Perfusion-induced changes in cardiac contractility
depend on capillary perfusion. Am J Physiol 1998;274:H405–10.
161Bai X-J, Iwamoto T, Williams AG Jr, Fan WL, Downey HF. Coronary pressure–flow autoregulation
protects myocardium from pressure-induced changes in oxygen consumption. Am J Physiol
1994;266:H2359–68.
162 Collins P. Coronary flow reserve. Br Heart J 1993;69:279–81.
163Baumgart D, Haude M, Liu F, Ge J, George G, Erbel R. Current concepts of coronary flow reserve for
clinical decision making during cardiac catheterization. Am Heart J 1998;136:136–49.
Gould KL, Lipscomb K, Hamilton GW. Physiologic basis for assessing critical coronary stenosis. Am J
164
Cardiol 1974;33:87–94.
165Olsson RA, Bugni WJ. Coronary circulation. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan
HE, eds The heart and cardiovascular system. Scientific Foundations. NewYork: Raven Press, 1986:987–
1037.
166Canty JM Jr, Klocke FJ. Reduced regional myocardial perfusion in the presence of pharmacological
vasodilator reserve. Circulation 1985;71:370–7.
167Brown BG, Josephson MA, Peterson RB, et al. Intravenous dipyridamole combined with isometric
handgrip for near maximal acute increase in coronary flow in patients with coronary artery disease. Am J
Cardiol 1981;48:1077–85.
168Ludman PF, Poole-Wilson PA. Myocardial perfusion in humans: What can we measure? Br Heart J
1993;70:307–14.
169Domenech RJ, Hoffman JIE, Nobel MIM, Saunders KB, Henson JR, Subijanto S. Total and regional
coronary blood flow measured by radioactive microspheres in conscious and anesthetized dogs. Circ Res
1969;25:581–96.
170Kety SS, Schmidt CF. The determination of cerebral blood flow in man by the use of nitrous oxide in low
concentrations. Am J Physiol 1945;143:53–66.
White CW, Wilson RF, Marcus ML. Methods of measuring myocardial blood flow in humans. Prog
171
Cardiovasc Dis 1988;31:79–94.
172Cannon PJ, Dell RB, Dwyer EM Jr. Measurement of regional myocardial perfusion in man with 133
xenon and scintillation camera. J Clin Invest 1972;51:964–77.
173Morgan SM, Fisher JD, Horwitz LD. Validation of regional myocardial flow measurements with
scintillation camera detection of xenon-133. Invest Radiol 1978;13:132–7.
174Yoshikawa J, Akasaka T, Yoshida K, Takagi T. Systolic coronary flow reversal and abnormal diastolic
flow patterns in patients with aortic stenosis: assessment with an intracoronary Doppler catheter. J Am Soc
Echocardiogr 1993;6:516–24.
175Heller LI, Cates C, Popma J, Deckelbaum LI, Joye JD, Dahlberg ST, et al. Intracoronary Doppler
assessment of moderate coronary artery disease: comparison with 201T1 imaging and coronary
angiography. Circulation 1997;96:484–90.
176Hozumi T, Yoshida K, Akasaka T, Asami Y, Ogata Y, Takagi T, et al. Noninvasive assessment of
coronary flow velocity and coronary flow velocity reserve in the left descending coronary artery by
Doppler echocardiography. Comparison with invasive technique. J Am Coll Cardiol 1998;32:1251–9.
177Kenny A, Shapiro LM. Identification of coronary artery stenoses and poststenotic blood flow patterns
using a miniature high-frequency epicardial transducer. Circulation 1994;89:731–9.
Stoddard MF, Prince CR, Moriis GT. Coronary flow reserve assessment by dobutamine transesophageal
178
Doppler echocardiography. J Am Coll Cardiol 1995;25:325–32.
Doucette JW, Corl PD, Payne HM, et al. Validation of a Doppler guide wire for intravascular
179
measurement of coronary artery flow velocity. Circulation 1992;85:1899–911.
Page 170
180Ganz W, Tamura K, Marcus HS, Donoso R, Yoshida S, Swan HJ. Measurement of coronary sinus blood
flow by continuous thermodilution in man. Circulation 1971;44:181–95.
181Hutchins GD, Schwaiger M, Rosenspire KC, Krivokapich J, Schelbert H, Kuhl DE. Noninvasive
quantification of regional blood flow in the human heart using N-13 ammonia and dynamic positron
emission tomographic imaging. J Am Coll Cardiol 1990;15:1032–42.
182De Silva R, Camici PG. Role of positron emission tomography in the investigation of human coronary
circulatory function. Cardiovasc Res 1994;28:1595–612.
183Uren NG, Melin JA, Bruyne BD, Wijns W, Baudhuin T, Camici PG. Relation between myocardial blood
flow and the severity of coronary artery stenosis. N Engl J Med 1994;330:1782–8.
184Bol A, Melin JA, Vanoverschelde J, et al. Direct comparison of [13N] ammonia and [15O] water estimates
of perfusion with quantification of regional myocardial blood flow by microspheres. Circulation
1993;87:512–25.
185Isada L, Marwick TH, MacIntyre WJ. Physiologic evaluation of coronary flow: the role of positron
emission tomography. Cleve Clin J Med 1993;60:19–24.
186Ito H, Iwakura K. Assessing the relation between coronary reflow and myocardial reflow. Am J Cardiol
1998;81:8G–12G.
187Villanueva FS, Glasheen WP, Sklener J, Kaul S. Characterization of spatial patterns of flow within the
reperfused myocardium by myocardial contrast echocardiography. Implications in determining extent of
myocardial salvage. Circulation 1993;88: 2569–606.
Page 171
5: The pulmonary circulation

KEITH SYKES
William Harvey reported the experiments which led him to conclude that blood must flow
through the lungs in De Motu Cordis in 16281 but it was not until 1661 that Malpighi
described the microscopic appearance of the pulmonary capillaries which provided the
anatomical link between the right and left heart.2 In 1894 Bradford and Dean3 reported that
the pulmonary artery pressure increased during hypoxia, but it was not until 1946 that von
Euler and Liljestrand4 concluded that this was the result of hypoxic pulmonary
vasoconstriction, and suggested that this mechanism might improve the matching of
perfusion to ventilation at the alveolar level. Nissel’s subsequent demonstration of hypoxic
vasoconstriction in the isolated perfused lung confirmed that this was a local response and
not mediated by the autonomic system.5 Meanwhile, the introduction of the cardiac catheter
into clinical practice6 in the 1940s had led to studies of the haemodynamics of the heart and
lungs, and to the subsequent development of closed and, later, open heart surgery in the
1950s. The subsequent introduction of radioisotope methods for measuring the distribution
of ventilation and blood flow, and the development of practical methods of measuring gas
and blood gas tensions, resulted in a vast increase in our understanding of the mechanisms
governing gas exchange. Later studies have shown that the pulmonary circulation has three
other important roles:
1 It has a regulatory function (for example, as part of the renin–angiotensin system).
2 It takes up or metabolises certain drugs [such as propranolol, lignocaine (lidocaine), and
noradrenaline (norepinephrine)].
3 It filters out particulate matter (such as platelet or fat emboli).
The pulmonary circulation is also concerned with the generation of surfactant, which
maintains alveolar stability, and with the exchange of water. As anaesthesia, mechanical
ventilation, and surgery may produce major changes in the pulmonary circulation, it is
important that the anaesthetist should have a clear understanding of the factors that govern
the
Page 172
distribution of pulmonary blood flow. Readers interested in the pharmacological aspects of
the lung are referred to a review by Bakhle.7
The mechanisms affecting the distribution of blood flow in the normal lung will be
considered first. This will be followed by a discussion on the effects of posture,
haemorrhage, mechanical ventilation, and lung disease. The methods of studying the
pulmonary circulation will then be outlined, followed by a brief review of the way in which
anaesthetic and related drugs may alter the distribution of blood flow, and so affect the
efficiency of gas exchange. Finally, we shall consider the problem of pulmonary
hypertension, the effects produced by pulmonary vasodilator drugs, and the
pathophysiology of pulmonary oedema.
Anatomy and physiology

The pulmonary circulation extends from the right heart outflow to the left atrium. Blood
from the right ventricle passes into the thin-walled pulmonary artery which then branches
repeatedly in parallel with the bronchi to supply the pulmonary capillary network
surrounding the alveoli. The walls of the larger arteries contain more elastic tissue than
smooth muscle, but smooth muscle predominates in arteries less than 1 mm in diameter.
The pulmonary capillaries are 7–10 μm in diameter and form a dense network around the
alveoli. They normally contain about 40% of the pulmonary blood volume, the volume
contained in the pulmonary arteries, capillaries, and veins being about 120, 250, and 150 ml
respectively, though these values are greatly influenced by the conditions of measurement.
The blood from the capillaries is collected into the pulmonary veins which run alongside
the arteries, bronchi, and lymphatics in the interstitial space, and finally drains into the left
atrium. Although there is some autonomic control of the pulmonary circulation, it appears
to be relatively unimportant.
There are three requirements for efficient gas exchange:
z the gas exchanging surface between the alveoli and pulmonary capillary blood must have
a large area
z the alveolar–capillary membrane must be thin to ensure that there is little hindrance to
diffusion
z there must be correct matching of ventilation to perfusion.
To achieve a perfect distribution of blood flow throughout the 30 cm height of the erect
adult lung, it would be necessary to have a high pulmonary artery pressure and thick
muscular arteries to overcome the effects of gravity. The lung is, however, unique in that it
has to receive the whole of the cardiac output, and this may vary from 1–2 l/min in severe
shock to 20–25 l/min in exercise. Furthermore, such changes in flow must
Page 173
be accommodated without imposing an undue load on the right heart. These constraints
have led to the development of a low pressure, low resistance, pulmonary circulation in
which distribution is relatively poorly controlled, but in which increases in flow are
accommodated by the recruitment of extra vessels and the distension of vessels that are
already open. The increase in flow thus increases the cross sectional area of the vascular
bed and so decreases the total pulmonary vascular resistance. So effective is this
mechanism that a doubling of flow can be accommodated with a rise of pulmonary artery
pressure of only 2–5 mm Hg when the patient is in the erect position.
Gravity and the distribution of pulmonary blood flow
As there is a close anatomical relationship between the pulmonary vasculature and the
alveoli, the dimensions of the vessels will be affected by the relationship between the
vascular and alveolar pressures, and by changes in lung volume. The effects will depend on
the location of the vessels. The large extrapulmonary vessels are situated outside the lung
and within the mediastinum, and are affected by regional changes in pleural pressure and by
local mechanical distortions around the hilum. As these vessels are large, it is unlikely that
changes in their geometry will have major effects on flow. The intrapulmonary vessels are,
however, smaller and their resistance is greatly affected by changes in the surrounding lung.
These vessels may be subdivided into three groups according to their anatomical location:
z the alveolar vessels

z the extra-alveolar vessels
z the corner vessels.
The alveolar vessels

The alveolar vessels are the pulmonary capillaries which are compressed when lung
volume is increased by an increase in airway pressure. Blood flow through these capillaries
depends on the relationship of the alveolar pressure, the pulmonary artery pressure, and the
pulmonary venous pressure, and so results in the gravitational distribution of blood flow
described by West.8 In the upright lung (with a height of 30 cm in the adult) three zones
may be distinguished.
Zone 1 is the area in the non-dependent part of the lung in which alveolar pressure
exceeds pulmonary artery pressure. This results in compression of the capillaries so that
there is no blood flow or gas exchange. Continued ventilation of alveoli in this zone
contributes to the alveolar dead space. In the supine position, however, the pulmonary
artery pressure is usually greater than the vertical distance between the right atrium and the
non-dependent part of the lung, so there is usually no zone 1 (Fig. 5.1).
Page 174
In zone 2 pulmonary artery pressure is greater than alveolar pressure, which is, in turn,
greater than pulmonary venous pressure, so capillary blood flow is present and increases
linearly down the zone. The cause of the increase in flow is still debated, some workers
suggesting that the capillaries behave like a Starling resistor, so that more vessels are
recruited in the lower parts of the zone, whereas others claim that the increase in flow down
the zone is the result of increased distension of the vessels. It has also been suggested that
there are critical opening pressures, as there are in the systemic circulation, and that these
must be overcome before blood flow occurs. Others have suggested that there are local
control mechanisms which cause different capillary beds to open in sequence.
In zone 3 both pulmonary arterial and venous pressures exceed alveolar pressure. It is
believed that all the capillaries are open, and that the small increase in flow down the zone
results from further distension. A fourth zone of reduced flow in the most dependent parts
of the lung has also been described. The presence of zone 4 was first noted in experimental
perfusion preparations and was attributed to the formation of pulmonary oedema; this may
well be the explanation for a reduced dependent zone flow in patients with left ventricular
failure. Zone 4 has, however, also been demonstrated in normal patients, particularly at low
lung volume. The
Fig 5.1 The three zone model of the lung. Left: upright lung. Centre: blood flow versus lung height. Right:
supine lung. In zone 1 alveolar pressure (Pa1v) exceeds both pulmonary artery pressure and left atrial
pressure so the capillaries are collapsed and there is no blood flow. In zone 2 pulmonary artery pressure
is greater than alveolar pressure, but the alveolar pressure is greater than left atrial pressure, so that the
capillaries behave as Starling resistors. In zone 3 both and exceed Palv. All the capillaries are open
so that the small increase in flow down the zone is due to further distension. As the height of the supine lung
is less than that of the vertical lung, there is usually no zone 1.
Page 175
reduction in flow could be caused by narrowing of the extra-alveolar vessels resulting from
a gravitationally induced decrease in lung volume in dependent zones, or to hypoxic
pulmonary vasoconstriction secondary to decreased ventilation caused by airway closure. It
could also be caused by the increased vascular resistance resulting from the longer pathway
between the hilum and the periphery of the lung.
The extra-alveolar vessels

The extra-alveolar vessels are the arteries and veins that connect the large
extrapulmonary vessels to the alveolar-capillary network. They run parallel to the bronchi
and are surrounded by an interstitial space which contains fluid, lymphatics, and the fibrous
framework of the lung. The transmural pressure, which determines the diameter of these
vessels, depends on the intravascular pressure and the interstitial pressure. The pressure in
the interstitial space is generated by the elastic recoil of surrounding alveolar units and is
believed to approximate to the subatmospheric pleural pressure. In the erect lung at
functional residual capacity the pleural pressure is about –1 kPa (–10 cm H2O) at the apex
and –0·25 kPa (–2·5 cm H2O) at the base, so that the increase in transmural pressure down
the lung is less than would have been predicted from the vertical height of the lung (30 cm).
When the lung is fully expanded, the extra-alveolar vessels are dilated and so have a
minimal effect on the gravitational distribution of blood flow. At residual volume, however,
pleural pressure becomes higher than atmospheric at the base of the lung so that vessels in
this zone are narrowed. Under these circumstances dependent zone flow may actually equal
that in non-dependent zones so the gravitational gradient of flow is abolished.9
There may also be regional variations in interstitial pressure caused by differences in
regional lung expansion resulting from disease. For example, the reduction in lung volume
resulting from an area of atelectasis causes the pleural pressure over that area to be more
subatmospheric (that is, to have a lower absolute pressure) than the pressure in the
remainder of the pleural space. The resulting increase in transmural pressure may increase
the diameter of the extra-alveolar vessels and so decrease the effectiveness of hypoxic
vasoconstriction.10 In general, expansion of the lung will tend to reduce absolute interstitial
pressure, so dilating the extra-alveolar vessels, whereas interstitial oedema will tend to
increase absolute interstitial pressure and so will narrow them. This may account for the
reduced basal blood flow in patients with mitral stenosis.
The corner vessels

The third type of intrapulmonary vessels are the corner vessels. These are small vessels
situated at the junction of three alveoli. They are surrounded by an interstitial space, but it
is not clear whether this is an extension of the
Page 176
space surrounding the extra-alveolar vessels. The importance of the corner vessels is that
they appear to be shielded from the compressive effects of increased alveolar pressure and
so permit some flow to occur in zone 1 conditions. It seems probable that this flow only
occurs in systole and that it has little effect on gas exchange.
Effects of lung volume on pulmonary vascular resistance
The pulmonary vascular resistance is calculated by dividing the driving pressure

(pulmonary artery minus left atrial pressure) by the flow, so it includes the resistance of all
the vessels between the right and left heart. The importance of the differentiation between
intra- and extra-alveolar vessels is that changes in lung volume exert opposing effects on
the two sets of vessels. Expansion of the lung will occur when transpulmonary pressure is
increased, whether this is produced by a reduction in absolute pleural pressure or an
increase in alveolar pressure, and this will compress the intra-alveolar vessels and so
increase their resistance to blood flow. These changes will also tend to increase the size of
zone 1 and so increase alveolar dead space. On the other hand, expansion of the lung, by
whatever means, will increase the diameter of extra-alveolar vessels and so decrease their
resistance. The diameter of the extrapulmonary vessels will also tend to increase when
absolute pleural pressure is decreased, but as these vessels are large the effects on resistance
will be relatively small, and these effects may be offset by local distortions of the hilum
associated with the expansion of the lung.
As the extra-alveolar vessels are narrow at low lung volumes but expanded at high lung
volumes, whereas the pulmonary capillaries are compressed at high lung volumes and open
at low lung volumes, the pressure–volume curve for the whole lung is U shaped, the
resistance being minimal at the normal end expiratory position or functional residual
capacity (Fig. 5.2).
Regional inhomogeneity of blood flow

The gravitational model of the distribution of pulmonary blood flow was derived from
radioisotope studies, which measured distribution in relatively large areas of lung. In recent
years, methods yielding much higher resolution of the spatial distribution of blood flow
have shown that, in humans, there is marked heterogeneity of distribution within an
isogravitational plane, and that there is a radial distribution of flow with the greatest flow in
the centre of lung lobes.11 12 There have also been a large number of studies in quadripeds
which show: that there is a marked heterogeneity of flow between 1 and 2 cm2 samples of
lung taken from isogravitational planes;13 that there is a distribution of flow which favours
the dorsal regions of lung regardless of posture;14 15 and that flow distribution changes little
when gravitational forces are increased by a factor of three.16 These studies
Page 177
Fig 5.2 The contribution of the resistance of the intra-alveolar (IAV) and extra-alveolar (EAV) vessels to the
total pulmonary vascular resistance (PVR) at different lung volumes. Total resistance is minimal at the
functional residual capacity (FRC).
suggest that distribution is mainly governed by variations in the resistive properties of the
pulmonary vasculature resulting from asymmetrical branching, or other anatomical
differences, and that gravity plays a much smaller role than had previously been thought.
When considering the animal evidence, it must be remembered that the erect human tends
to have a relatively higher lung volume, with distended extra-alveolar vessels, lower
smooth muscle tone, and a greater proportion of the serial resistance in the middle
(microvascular) segment than the animals studied. Although recent studies have shown that
the gravitational gradient of distribution in baboons is decreased when they are held upside
down, this finding could be explained by the effects of an alteration in pleural pressure
gradient on extra-alveolar vessels. In the absence of firm evidence to the contrary, it seems
reasonable to conclude that there is an important gravitational component to distribution in
the human, but that there is much more inhomogeneity of distribution than had previously
been assumed. How these local variations in pulmonary blood flow can be matched by
ventilation distribution to minimise ventilation–perfusion inequalities has yet to be
determined.17 18
Clinical implications
Fortunately, the gravitationally induced increase in blood flow down the lung is normally
matched by an increase in ventilation. This increase is generated by the interaction between
the non-linear pressure–volume curve of the lung and the gravitationally induced gradient
of pleural pressure (Fig. 5.3). When the lung is vertical (height 30 cm) the pressure in the
pleural
Page 178
space is about –1 kPa (–10 cm H2O) in the non-dependent areas and about –0·25 kPa (–2·5
cm H2O) in dependent zones at the normal end expiratory lung volume. The resulting
transpulmonary pressure of 1 kPa (10 cm H2O) at the top of the lung and 0·25 kPa (2·5 cm
H2O) at the base causes the upper alveoli to have a larger resting volume than those at the
base. When the transpulmonary pressure is increased by a reduction in absolute pleural
pressure during inspiration, however, the lower alveoli will expand more than the upper
because they lie on a steeper part of the pressure–volume curve. Thus, under normal
conditions, the increase in ventilation down the lung (which is about half that of the
increase in blood flow) minimises ventilation–perfusion inequalities (Fig. 5.4). If, however,
there is dependent airway closure as a result of a loss of lung elastic recoil or of a reduction
in functional residual capacity, there may be no ventilation to dependent zones during the
early part of inspiration, so that these zones develop low ventilation–perfusion ratios and
arterial PO2 (PaO2) is reduced.
Fig 5.3 Distribution of ventilation. Above: during spontaneous ventilation the gravitationally induced gradient
of pleural pressure causes the non-dependent alveoli to lie on the upper, curved part of the lung pressure–
volume curve, whereas the dependent alveoli lie on the lower, steep portion. As a result the increase in
transpulmonary pressure (ΔP) during inspiration causes more ventilation to enter the dependent zones of the
lung. Below: the absence of diaphragmatic activity during controlled ventilation permits the hydrostatic
pressure generated by the abdominal contents to influence distribution. The position of the alveoli on the
pressure–volume curve of the total respiratory system (that is, lung plus chest wall) now causes ventilation to
be preferentially distributed to the non-dependent zones. Note that changes in end expiratory lung volume
may modify the distribution by moving the alveoli to different portions of the P/V curves.
Page 179
During controlled ventilation the distribution of ventilation is determined by the shape of
the total respiratory (lung plus chest wall) pressure–volume curve because the inspiratory
muscles are no longer active. Furthermore, in the supine position, the hydrostatic pressure
produced by the semiliquid abdominal contents exerts an upward pressure on the dependent
areas of the diaphragm so that ventilation is preferentially directed into non-dependent
zones. As the distribution of blood flow is still gravitationally determined there is gross
mismatching of ventilation and perfusion (Figs. 5.3 and 5.4). The situation is exacerbated if
there is a decrease in pulmonary artery pressure resulting from a reduction in blood volume,
peripheral pooling of blood, or the administration of oxygen or a pulmonary vasodilator
drug, because this will result in an increase in zone 1 with a further increase in alveolar
dead space. Similar changes may occur if mean alveolar pressure is increased by
mechanical ventilation with positive end expiratory pressure (PEEP), or if the emptying of
the lung is delayed in patients with increased airway resistance (auto or intrinsic PEEP).
When there is an increase in dead space/tidal volume ratio in the spontaneously breathing
patient with normal respiratory control mechanisms, minute ventilation will tend to increase
to compensate for the increased dead space
Fig 5.4 Distribution of ventilation and blood flow plotted against lung height during spontaneous
respiration (left) and during anaesthesia with controlled ventilation (right). Note that the small alveolar dead
space (VDalv) associated with zone 1 conditions during spontaneous respiration is increased by the greater
ventilation to non-dependent zones during controlled ventilation. In elderly people there is often an area with
low ventilation-perfusion ratios at the base of the lung associated with airway closure. General
anaesthesia usually results in the development of a shunt in dependent lung zones as a result of compression
collapse.
Page 180
so that PCO2 is maintained at normal levels, but if the minute volume is controlled by a
ventilator, PCO2 may increase. If the rest of the lung is normal there will be no effects on
PaO2 other than those arising from any increase in PCO2.
In most patients undergoing anaesthesia or intensive care there is some alveolar collapse
in dependent lung zones and this creates an intrapulmonary right to left shunt.19 This is
usually quantified by expressing the shunt as a percentage of the cardiac output. The PaO2
resulting from a given shunt depends on the alveolar PO2 (PAO2) (which in turn depends on
the inspired PO2 (PIO2), the alveolar PCO2 (PACO2) and the respiratory exchange ratio) and
on the mixed venous (Fig. 5.5). Normally, it is assumed that an increase or
decrease in the percentage shunt means that the volume of collapsed lung has increased or
decreased. The percentage shunt may, however, change with no alteration of the volume of
collapsed lung if the proportion of blood flowing through the oxygenated and collapsed
zones is changed by an alteration in the pulmonary vascular pressures. For
Fig 5.5 Factors governing arterial oxygen tension, PIO2, PAO2, PACO2, inspired and alveolar gas tensions; R,
respiratory exchange ratio (normally 0.8); PaO2, , arterial and mixed venous oxygen tensions;
and , shunt flow and cardiac output; , oxygen consumption per minute. Note that the PaO2 depends
on both the proportion of blood flowing through the shunt and the . depends on the relationship
between and .
Page 181
example, blood flow through a collapsed area of lung is maximal when it is in the
dependent position but can be reduced by rotating the patient so that the collapsed area is
uppermost, with a resultant decrease in shunt and increase in PaO2.20 (The improvement in
oxygenation is not, however, usually sustained because collapse soon develops in the areas
of lung now made dependent, whereas the collapse in the non-dependent zones disappears.)
The opposite effect can be seen in patients with dependent zone collapse when
pulmonary artery pressure and cardiac output are decreased by vasodilator drugs. Under
such circumstances the continued flow through the dependent zone with reduced flow to the
ventilated area of lung will cause an apparent increase in the proportion of shunt, even
though the actual flow through the shunt is unchanged (Fig. 5.6). The application of a high
peak airway pressure or PEEP will also reduce flow through the ventilated non-dependent
zones and so will have a similar effect. Another example is the redistribution of flow which
may be seen during anaesthesia for thoracic surgery with a double lumen tube. When the
upper lung is collapsed the effects of gravity and hypoxic vasoconstriction in the upper lung
decrease the upper lung blood flow so that the shunt is only 20–30% instead of the 45–55%
predicted from the relative volume of each lung. If the mean airway
Fig 5.6 The effect of a decrease in pulmonary artery pressure (Ppa) resulting from a decrease in cardiac output
on percentage shunt in the presence of dependent zone collapse or consolidation. If flow to the ventilated area
of lung is decreased from 3 l/min to 2 l/min whilst flow through the shunt remains at 1 l/min the percentage
shunt will increase from 25% to 33%. Note that the resulting fall in arterial PO2 (PaO2) will be
accentuated by the decrease in mixed venous resulting from the decrease in output.
Page 182
Fig 5.7 Factors that may increase flow through the non-dependent collapsed lung during one lung anaesthesia
in the lateral position. The distribution of blood flow and ventilation with lung height is shown on the left of
the diagram.
pressure in the dependent lung is increased, however, by the use of high peak or end
expiratory pressures, shunt will increase because the compression of capillaries in the
dependent lung increases pulmonary artery pressure and so diverts blood flow into the non-
dependent collapsed lung (Fig. 5.7). The injection of pulmonary vasoconstrictor drugs will
have a similar effect.
Another factor that affects the PaO2, even when the areas of shunt are scattered
throughout the lung and are not changed by alterations in lung volume, is the direct
relationship between shunt and cardiac output. The increase in shunt with cardiac output
occurs when the output is changed by altering blood volume or the administration of
inotropic drugs but the cause is not properly understood. It is possible that the increase in
flow increases pulmonary artery pressure and so opposes hypoxic vasoconstriction. An
increase in output will usually increase , and this may also reduce the magnitude of
the vasoconstrictor response (see below). Although the increase in may increase the
proportion of shunt it will also increase the oxygen saturation of the blood flowing through
the shunt and so tend to offset the effects of the increased percentage shunt on PaO2.
Obviously, these interactions may lead to very variable effects on PaO2.
Factors controlling pulmonary vasomotor tone
As pulmonary vascular tone in the normal human lung is low, vasoconstrictor responses
are easily demonstrated, whereas vasodilator responses are small unless the vascular bed
has been preconstricted. The major feature that distinguishes the pulmonary circulation
from other vascular beds is that it constricts in response to hypoxia. In recent years, it has
also become apparent that the endothelium plays a major role in controlling pulmonary
vascular tone.
Page 183
Hypoxic pulmonary vasoconstriction
The response of the pulmonary circulation to hypoxia is characterised by four important

features:
z First, the response is present in an isolated perfused lung preparation and in isolated
smooth muscle cells from pulmonary arterioles, and so must be controlled by some local
mechanism.
z Second, the magnitude of pulmonary vasoconstriction is related to the PAO2, the
response curve of blood flow against PAO2 being sigmoid in shape with the maximum
decrease in flow occurring between PAO2 values of 8 and 4 kPa, and a reduction in PAO2
to mixed venous levels resulting in a decrease in flow of about 50%. The apparent
anomaly of a reduction in alveolar oxygen tension causing constriction of precapillary
vessels (the pulmonary arteries < 300 μm in diameter) has now been explained by the
demonstration that these small vessels are completely surrounded by alveoli so that
alveolar gases can diffuse directly into the vessel wall.
z Third, the response can also be activated by a reduction in , although a given

decrease in has a smaller effect than a similar decrease in PAO2. This suggests that
the sensor site is situated within the smooth muscle of the arterial wall, but closer to the
surrounding alveoli than to the lumen of the vessel.
z The fourth feature that characterises the response is that it occurs rapidly. In humans,
blockage of an apical lobe bronchus resulted in an approximately 50% decrease in blood
flow within 5 minutes.21 Recent studies in human volunteers have, however,
demonstrated that there is also a slow component to the response; this becomes fully
established after two hours of isocapnic hypoxia and persists for at least eight hours.22
The exact nature of the constrictor mechanism is still not understood. It is known that
hypoxia causes depolarisation of smooth muscle cells from pulmonary arteries, but that it
hyperpolarises cells from systemic arteries. Depolarisation results in calcium entry into
smooth muscle and vasoconstriction. It appears that depolarisation is caused by closing
voltage gated potassium channels in the cell membrane, although it is still not known how
the hypoxia is sensed.23
Hypoxic vasoconstriction results in a diversion of blood flow away from the hypoxic
area of lung, so improving gas exchange.24 It also produces an increase in pulmonary artery
pressure. The magnitude of these effects depends on the volume of lung made hypoxic. If
the hypoxic area is large there will be less normal lung to accommodate the diverted flow,
so that there will be a greater increase in pressure. As the increase in smooth muscle tone is
opposed by the intravascular pressure, there will be less diversion of flow when the rise in
pulmonary artery pressure is increased.25
Page 184
General factors decreasing hypoxic pulmonary vasoconstriction

Chronic hypoxia
Increased mixed venous
Increased pH
Decreased arterial or mixed venous
Increased vascular pressures
Increased volume of hypoxic lung
Increased transpulmonary pressures
Hypothermia
Handling or trauma to lung
Endotoxin
Pneumonia
Cirrhosis
Thus hypoxic pulmonary vasoconstriction is least effective when the volume of hypoxic
lung is large or when pulmonary vascular pressures are increased by disease, fluid overload,
or left heart failure.26 The magnitude of the response to hypoxia is greatest in the newborn,
varies between species and between individuals in any given species, and is reduced by
hypothermia, sepsis,27 trauma to the lung, liver cirrhosis,28 smoking,29 and the action of
many other drugs. Hypoxia increases the heterogeneity of flow and decreases the central to
peripheral gradient of perfusion in lung lobes.30
An increase in PCO2 has a more variable effect on pulmonary vascular tone, although it
usually augments hypoxic pulmonary vasoconstriction and so increases the response to
hypoventilation. A reduction in pH has a similar effect. As collapsed lung equilibrates with
mixed venous blood gas tensions, the higher PCO2 augments the diversion of blood flow
away from a collapsed area of lung.31 It now appears that the reduction in blood flow in
collapsed lung is almost entirely the result of hypoxic vasoconstriction, and that the
mechanical effect produced by narrowing of the extra-alveolar vessels is of only minor
importance.
PACO2 has a major effect on airway muscle tone, an increase causing bronchodilatation
and a decrease bronchoconstriction. Thus alveolar gas concentrations act on both the
airways and the pulmonary circulation in a manner that tends to minimise ventilation–
perfusion inequalities.
Other factors affecting pulmonary vascular tone
In the early 1970s, it became apparent that the pulmonary endothelium metabolised
circulating vasoactive substances. Some, such as bradykinin, were removed, whereas
others, such as angiotensin II were produced.
Page 185
During the next decade it was shown that kinins, peptides, catecholamines, lipoproteins,
and many other substances were metabolised in the lung. With the discovery and
characterisation of the vasodilator, prostacyclin (PGI2), and the vasoconstrictor,
thromboxane, it became apparent that the endothelium played an important role in
controlling the pulmonary circulation itself.32 In the 1980s it was shown that pulmonary and
systemic vessels release a labile endothelium derived relaxing factor (EDRF) which
modifies the vasopressor response to various pharmacological agents and to acute hypoxia.
It is now clear that EDRF is nitric oxide (NO). The NO is synthesised from L-arginine by
nitric oxide synthase. It then diffuses within the cell, or to another cell, where it stimulates
soluble guanylyl cyclase or other haem containing proteins. This results in an increase in
cyclic guanosine monophosphate (GMP) which produces the physiological effect. For
example, in smooth muscle cells cyclic GMP decreases cell calcium which leads to
relaxation of the muscle cell and vasodilation. As nitric oxide has a high affinity for
haemoglobin with the formation of methaemoglobin, it has a half life measured in seconds.
It is also rapidly oxidised to nitrite and nitrate by superoxide radical in the blood vessel wall
or by oxygen in free solution.
Vascular tone is controlled by opposing factors which cause constriction or dilatation.
Dilatation is induced by acetylcholine, bradykinin, angiotensin converting enzyme
inhibitors, and adenine nucleotides, all of which stimulate NO production. It seems likely
that pulsatile flow and local shear stress may play an important role in the control of NO
release in vivo. There is, however, now evidence that NO may also influence blood
pressure by regulating sympathetic nerve activity. Nitric oxide decreases hypoxic
vasoconstriction in the lung, and there is evidence that there is either decreased production
or increased destruction of NO in systemic and pulmonary hypertension and in ischaemic
heart disease. Excessive production of NO may be the cause of the profound vasodilation in
septic shock. Nitric oxide also inhibits platelet aggregation. It modulates tubuloglomerular
feedback in the kidney, inhibits insulin release, controls the relaxation of sphincters along
the gastrointestinal tract, and may also function as a neurotransmitter.33
The NO synthase which subserves intercellular communication is Ca2+ and cadmodulin
dependent, but there is another inducible, Ca2+ independent synthase which releases larger
quantities of NO over longer periods from activated macrophages, thus causing NO to act
as a cytotoxic agent. More recently other NO synthases have been discovered in tissues
other than the reticuloendothelial system, so raising the possibility that NO may be
implicated in the causation of other types of cell damage.
The rapid inactivation of NO by haemoglobin and the ability to administer the gas by
inhalation has enabled NO to be used as a selective pulmonary vasodilator (see page 203).
Page 186
In 1988 an endothelially derived vasoconstrictor substance termed endothelin was
isolated. Subsequently, three peptides have been identified (ET-1, ET-2, and ET-3). These
do not appear to be stored in the endothelium but may be released by such diverse factors as
shear stress, hypoxia, endotoxin, tumour necrosis factor (TNFα), interferon, adrenaline,
angiotensin, thrombin, activated platelets, and some prostanoids. ET-1 is increased in sepsis
but it is not yet clear what role these substances play in the mediation of normal vascular
tone.
Arachidonic acid metabolites are also released in sepsis. The cyclo-oxygenase system
leads to the formation of prostaglandins, thromboxanes, and PGI2, whereas the lipo-
oxygenase system produces leukotrienes.32
Methods of studying the effects of drugs on the

pulmonary circulation
Drugs may alter the distribution of pulmonary blood flow by altering the total flow or
pulmonary vascular pressures, or by modifying pulmonary vascular tone in the oxygenated
or hypoxic areas of lung. Secondary effects may be produced by drugs which affect airway
tone and so alter total or regional lung volume.
Most studies of drugs use measurements of pulmonary vascular resistance or changes in
the distribution of flow as an index of vascular tone. As a result of the complexity of the
pulmonary circulation all the results obtained from these methods must be interpreted with
caution. The most commonly used methods are summarised in Table 5.1.
Table 5.1 Control of variables during studies on the pulmonary circulation
Preparation Variables not Abnormalities

controlled
Isolated perfused lung Perfusate PO2 No neural control or lymph drainage

(constant flow or pressure) No other organs in circuit
Lobar perfusion Left atrial pressure No neural control or lymph drainage
(constant flow or pressure) of lobe
Ventilated or collapsed lobe Cardiac output, No neural control or lymph drainage
(flowmeters) vascular pressures, of lobe
Unilateral hypoxia Cardiac output, PCO2 decreases with blood flow

vascular pressures,
(radioisotopes, , SF6)
Generalised ventilation Cardiac output, Arterial hypoxaemia

hypoxia vascular pressures,
, mixed venous PO2; , oxygen consumption.

Page 187
Methods using the concept of pulmonary vascular resistance
Conceptual problems
This measurement is made by dividing the pressure difference across the lung by the
flow:
where PVR is pulmonary vascular resistance, and are mean pulmonary artery and
left atrial pressures, and is the cardiac output. The normal value is approximately 1·5
mm Hg/1 per min or 0·1 mm Hg/ml per s. To express the result in CGS units it is necessary
to multiply the second figure by 1332, so that the normal value is approximately 100 dyn/s
per cm5. This measurement is useful in that it has conceptual similarities to Poiseuille’s law
for laminar flow through a parallel sided tube:
This relationship tells us that the resistance increases with increased viscosity (η) of the
perfusing fluid, with increasing length (l) of the tube, and is inversely related to the fourth
power of the radius (r). Blood is, however, a non-newtonian fluid (with a viscosity that
changes with flow), blood flow is pulsatile, and the pulmonary vasculature consists of a
branching network of distensible tubes, the cross sectional area of which is augmented by
recruitment of extra vessels when flow or pulmonary artery pressure increases.
Furthermore, the radius and length of the vessels are affected by changes in transpulmonary
pressure and lung volume. It is, therefore, obvious that Poiseuille’s equation cannot be
applied to the pulmonary circulation and that attempts to do so must frequently yield
conflicting results. Nevertheless, the general concept of resistance is of value and the
measurement can be used as an index of vascular tone if appropriate precautions are taken
to control the variables.
Perfusion preparations
The most reliable measurements are obtained by using one of the various forms of
perfused lung preparation in which changes in resistance can be detected by making
simultaneous measurements of flow and the pressure difference between pulmonary artery
and left atrium. As flow measurement is technically more difficult than pressure
measurement, it is usual to measure the changes in pressure across the lung while the lung
is perfused at constant flow. With such a preparation, however, changes in intravascular
pressure oppose the change in smooth muscle tone, so decreasing the magnitude of the
response. Greater sensitivity is obtained
Page 188
Fig 5.8 Pressure–flow curves obtained when vasomotor tone is normal or increased by hypoxia. The vertical
arrow shows the increase in pressure which would be recorded in response to hypoxia during a constant flow
perfusion, whereas the horizontal arrow shows the decrease in flow in response to hypoxia with a constant
pressure perfusion.
by measuring the changes in flow while the preparation is perfused at constant input and
output pressures. Ideally, a number of measurements are made at each stage of the
experiment so that pressure–flow curves can be plotted before and after the intervention
(Fig. 5.8). A shift of the curve then provides strong evidence of a change in vascular tone
provided the other variables have been kept constant. The optimal control of variables is
obtained in the isolated perfused lung preparation in which the lungs are either retained in
the chest or suspended in a box. The lungs are ventilated at constant tidal volume with 5%
carbon dioxide in an oxygen–nitrogen mixture. End expiratory pressure is kept constant and
airway pressure is monitored to ensure that there are no changes in lung mechanics.
Alveolar hypoxia can then be induced by reducing the inspired oxygen concentration to 3–
5%. In the constant flow type of perfusion, blood from a warmed reservoir is pumped by an
occlusive pump through a cannula tied into the pulmonary artery. It is then drained through
a cannula in the left atrium, which is connected to an overflow system to ensure that left
atrial pressure is maintained constant, and both pulmonary artery and left atrial pressures
are measured. Flow can be derived from a previous calibration of the pump, or measured by
an electromagnetic flowmeter, or by a timed diversion of flow from the atrial cannula into a
parallel calibrated reservoir (Fig. 5.9).
In the alternative technique a constant pressure perfusion is effected by pumping the
blood up to a reservoir in which the surface is maintained at a constant level by means of an
overflow, and flow is again measured by collecting the outflow from the lungs over a
measured period, or by using
Page 189
an electromagnetic flowmeter. Both types of preparation are effectively denervated, thus
eliminating possible reflex responses, and all the other variables are rigidly controlled. Such
preparations are ideally suited to the investigation of agents which are administered by
inhalation, or of drugs that are metabolised by the lung. However, there is no bronchial
circulation and the normal routes of drug elimination are not included in the circuit. In
addition, the ligature round the pulmonary artery occludes the lymphatic drainage so that
the preparation tends to become oedematous after several hours of perfusion.
To overcome these problems, many workers use the in situ perfused lobe technique. The
left lower lobe is usually chosen because it has a long bronchus which can be cannulated
easily so that the lobe can be ventilated separately from the rest of the lung. Perfusion of the
lobe is achieved by using a constant flow device to pump blood from the right side of the
Fig 5.9 Perfusion circuit for isolated perfused lung preparations. The lungs can be ventilated at constant
volume with a normoxic or hypoxic gas mixture and the airway (PAW), pulmonary artery , and left
atrial pressures recorded. Blood flow can be measured by diverting the venous outflow into the
measuring cylinder for a known time.
Page 190
heart into a cannula tied into the left lower lobe artery. The blood drains into the left atrium
and then circulates normally. The disadvantage of this technique is that left atrial pressure
and mixed venous PO2 cannot be controlled (although they can be monitored). The whole
body is, however, perfused normally so that normal detoxification mechanisms are not
interfered with. Innervation and lymphatic drainage are destroyed if the cannula is tied in
place with a ligature around the pulmonary artery, but these disadvantages can be overcome
in larger animals by floating a catheter with a terminal balloon into the appropriate branch
of the artery and by ventilating the lobe with a cuffed endobronchial tube.
In vivo studies: errors in the measurement of pulmonary vascular resistance

Most human studies on the effects of drugs have used the concept of pulmonary vascular
resistance. As already pointed out there are many disadvantages to this approach. The first
is that the measurements are likely to be very inaccurate. It is difficult to measure any
vascular pressure with an error less than 1–2 mm Hg and there are many additional sources
of error when left atrial pressure is derived from a pulmonary wedge pressure measurement.
The accuracy of cardiac output measurement34 is
Factors affecting measurements of pulmonary vascular

resistance (PVR)
‘‘Passive” factors Change in PVR
Pulmonary artery pressure increase Decrease

Left atrial pressure increase Decrease
Transpulmonary pressure, increase or
decrease from functional residual
capacity Increase
Interstitial pressure increase Increase
Blood viscosity increase Increase
“Active” factors
Blood gases
PO2 decrease Increase
PCO2, increase Increase (doubtful in

humans)
pH decrease Increase
Autonomic activity (Probably negligible in
humans)
Endogenous substances
Catecholamines, angiotensin, histamine Increase
Acetylcholine, bradykinin, prostacyclin Decrease
Page 191
rarely better than ±10% so that the total error in the measurement of resistance may well
exceed the change resulting from the experimental intervention. The second disadvantage is
that the measured changes may be caused by “passive” or “active” changes in the
pulmonary circulation which are not directly related to the action of the drug under test.
The major factors that may affect the measurement are shown in the box. From these it may
be concluded that a change in pulmonary vascular resistance can only be considered valid
if:
1 increases or decreases while flow and are unchanged
2 decreases when decreases or is unchanged and flow decreases or is

unchanged
3 increases when flow is increased or unchanged and is increased or

unchanged.
One other approach is to construct a pressure–flow curve for each individual by exercising
the subject in the supine position (when most of the lung vessels are fully distended). Under
these circumstances the curve appears to be linear and so can be defined by two points (Fig.
5.10). If it is assumed that exercise does not in itself alter vasomotor tone, a shift in the
curve may be used as an index of vasomotor activity.
It has now been recognised that, although the pressure–flow curves may be reasonably
linear in the physiological range of flows, and the extrapolated /cardiac output
plots pass through the origin in
Fig 5.10 Comparison of single measurements of pulmonary vascular resistance (PVR) made by relating mean
pulmonary artery pressure minus mean left atrial pressure to cardiac output may be
misleading. Thus exercise increases cardiac output from A to B and is associated with a decrease in calculated
PVR (slopes of dotted lines from origin to A and B), whereas vasoconstriction (A to C) apparently produces
no change in PVR. If, however, the pressure–flow curve (A to B) is defined by exercise, a single measurement
(C) may suffice to indicate vasoconstriction.
Page 192
both healthy and diseased lungs under zone 3 conditions, this may not be so when the
measurements are made under zone 2 conditions, or when the lung is hypoxic or diseased.
Under these circumstances, the convexity towards the pressure axis at low flow rates
indicates recruitment of extra vessels as flow is increased, and the intercept of the
extrapolated linear portions of the /cardiac output plots indicates that the
effective output pressure from the lung is no longer the left atrial pressure, but results from
some external factor narrowing the pulmonary vessels. The critical closing pressure defined
by the intercept may be caused by an increase in vasomotor tone, by increased alveolar
pressure resulting from high levels of positive and expiratory pressure, by intravascular
obstruction, or by an increase in interstitial pressure caused by pulmonary oedema.35
Methods using the distribution of flow as an index of pulmonary vascular

tone
In human studies it is often possible to reduce the number of variables by studying the
regional distribution of blood flow rather than pulmonary vascular resistance. One method
of studying the hypoxic response is to measure the redistribution of blood flow in response
to unilateral hypoxia induced by the administration of 8–10% oxygen through one limb of a
double lumen tube. In the earlier studies the distribution of flow was determined by
measuring the oxygen consumption of each lung, because this is directly proportional to
blood flow. Attempts were also made to use carbon dioxide output in a similar manner, but
this technique proved inaccurate because the carbon dioxide output was also affected by the
ventilation to each lung.36 A much simpler way of measuring the distribution of blood flow
is to infuse a solution of a relatively insoluble gas into the pulmonary artery and to measure
the concentration evolved into the alveoli on each side. One method is based on the
infusion of sulphur hexafluoride and measurement of its concentration in expired gas with
infrared analysis.36 Other methods use a radiolabelled isotope of a relatively insoluble gas
such as xenon or krypton.37 This is dissolved in saline and injected into the right side of the
circulation. Most of the radiolabelled isotope is evolved into the alveoli during the first
passage of the blood through the lungs so that the count rate is directly related to the blood
flow. The radioactivity can either be measured in expired gas or detected by scintillation
detectors or a gamma camera placed over the chest. The use of the gamma camera has not
only enabled blood flow to be measured in areas of collapsed lung, but has also permitted
regional variations in perfusion to be studied. The distribution of flow has also been studied
by measuring regional radioactivity after the injection of microspheres or macroaggregates
labelled with radioactive isotopes. By using radioisotopes with different energies, up to six
sequential injections
Page 193
may be made at different stages of the experiment. As the counting may be delayed for
several hours, the method has proved to be of great value when studying the distribution of
blood flow in stressful environments such as during zero or exaggerated gravity.
Another method of studying the redistribution of blood flow after the administration of a
vasoactive drug is to measure the effects on gas exchange. This can be done by the standard
method of calculating percentage shunt and dead space/tidal volume ratio from the oxygen
and carbon dioxide tensions in arterial and mixed venous blood, and mixed expired gas.
Greater precision is, however, obtained by the Wagner inert gas technique in which a
solution of six inert gases of different solubilities is infused into the pulmonary circulation
at a constant rate, and the retention/excretion ratios are measured from an arterial and
mixed expired gas sample. By using a computer to fit the data to a 50 compartment model
of the lung, it is possible to derive values for shunt, dead space, and compartments with
intermediate ventilation–perfusion ratios.38 As with other in vivo measurements it is
important to remember that any alterations in the distribution of blood flow which occur
may be caused by changes in lung volume or haemodynamics, as well as by changes in
vasomotor tone.
It is obvious that the conditions outlined above severely limit the number of useful
observations that can be made. Furthermore, as the normal vascular bed has very little tone,
the action of pulmonary vasodilator drugs can only be studied after constriction has been
induced by some other agent (such as hypoxia). Obviously, vasodilator drugs must also be
studied in the patients in whom they are likely to be used, but such studies often produce
variable results owing to differences in the aetiology of the pulmonary hypertension in each
patient. For these reasons we will first consider the actions of drugs on the normal
pulmonary circulation and on hypoxic pulmonary vasoconstriction. In the last section we
shall consider the problem of pulmonary hypertension and the effects of vasodilator drugs.
Effect of drugs on the normal pulmonary circulation

It has already been pointed out that the distribution of pulmonary blood flow is primarily
controlled by the interrelationship between the pulmonary vascular pressures, the
transpulmonary pressure difference, and the effects of gravity, and it has been shown how
changes in these variables may affect distribution and gas exchange in the normal and
abnormal lung. Many drugs given during anaesthesia have profound haemodynamic effects
which will produce major effects on flow distribution. The changes in cardiac output may
also affect which will, in
Page 194
turn, affect the partitioning of blood flow between normoxic and hypoxic areas of lung. A
number of these drugs may affect pulmonary vasomotor tone and so further modify the
distribution of blood flow. Their effect depends on the pre-existing level of vascular tone.
In normoxic areas of lung, pulmonary vascular tone is low so that pulmonary vasodilators
have little effect whereas constrictors produce major changes. In areas with hypoxic
pulmonary vasoconstriction the reverse is the case. When analysing the effects of drugs on
the pulmonary circulation, one must, therefore, consider not only the changes in pulmonary
haemodynamics and , but also their effects on vasomotor tone in the normoxic and
hypoxic areas of lung (see box).
Anaesthetic drugs
Experimental studies
Although Buckley and colleagues39 had reported that nitrous oxide increased the pressor
response to alveolar hypoxia in dogs, whereas 0.5% halothane decreased it, reversible
depression of the hypoxic vasoconstrictor response by inhalational agents such as
trichloroethylene, halothane, and ether was not demonstrated in an isolated perfused lung
preparation until 1972.40 Subsequent studies using the isolated perfused lung preparation
have demonstrated that halothane decreases vasomotor tone in the normoxic lung and that
diethyl ether, halothane, methoxyflurane, enflurane, isoflurane, and sevoflurane produce a
dose dependent depression of vasoconstriction in the hypoxic lung. Intravenous anaesthetic
agents such as thiopental, pentobarbital, pentazocine, fentanyl, droperidol, ketamine, and
diazepam appear to have no effect on hypoxic vasoconstriction.41–43
In another series of experiments the left lower lobe was ventilated independently of the
rest of the lung and electromagnetic flowmeters used to measure the partition of flow
between the lobe and the rest of the lung. Ventilation of the lobe with nitrogen reduced the
blood flow by 53% and
Drugs affecting hypoxic pulmonary vasoconstriction

Decrease Increase
Inhalational anaesthetic agents Cyclo-oxygenase inhibitors
β Agonists Propranolol
Pulmonary vasodilators Almitrine
α Blockers Lignocaine (lidocaine)
Sodium nitroprusside
Nitroglycerine
Calcium channel blockers NOS inhibitors
Nitric oxide
Page 195
this reduction in flow was approximately halved when isoflurane or fluroxene were added
to the nitrogen in a concentration of 2 MAC; 0·3 MAC nitrous oxide produced slight but
significant inhibition of the response although halothane and enflurane had little effect
(MAC = minimum alveolar concentration). Similar effects were produced when the agent
was added to the whole lung while the lobe was hypoxic.42
The inhalational agents have also been studied in the intact animal. The hypoxic stimulus
was provided by unilateral ventilation hypoxia and the distribution of blood flow between
the two lungs measured by radio-isotope methods. In these experiments anaesthetic
concentrations of trichloroethylene, ether, and nitrous oxide reduced the magnitude of
hypoxic vasoconstriction, whereas 0 ·5–1·5% halothane had little effect. In intact dogs
submitted to whole lung ventilation hypoxia, isoflurane inhibited the pressor response
whereas halothane and enflurane had no effect.
The difference in results between the isolated perfused lung experiments and those using
in vivo preparations are probably caused by the haemodynamic changes associated with the
administration of the anaesthetic agent. Agents such as halothane produce a marked
decrease in cardiac output, which may be exaggerated when the animal is made hypoxic. If
this causes the intravascular pressure to decrease when vascular tone is decreased by
inhibition of hypoxic vasoconstriction, there may be little change in flow through the
hypoxic segment. There may also be changes in the hypoxic stimulus as a result of changes
in .
Human studies
Studies in the human are subject to many variables and must be interpreted with great
caution. The first report of the action of anaesthetic drugs on human pulmonary
haemodynamics was published by Johnson in 1951.44 The first evidence that clinically used
concentrations of inhalational anaesthetic agents could inhibit hypoxic vasoconstriction
was, however, obtained by Bjertnaes in 1978.45 He used unilateral hypoxia as the stimulus
and administered the agent to the hypoxic lung, the blood flow diversion being measured by
the injection of radiolabelled macro-aggregates. Subsequently, Rodgers and Benumof 46
measured PaO2 before and after the administration of halothane or isoflurane, during one
lung anaesthesia that was induced and maintained with either ketamine or methohexital,
and concluded that approximately 1 MAC concentrations of halothane or isoflurane do not
produce significant depression of hypoxic vasoconstriction in humans. These findings were
confirmed by Carlsson and colleagues who measured the diversion of flow in response to
unilateral ventilation hypoxia with a continuous infusion of sulphur hexafluoride (a
relatively insoluble inert gas). They found that inhaled concentrations of 2% enflurane and
1.0–1 ·5% isoflurane had no effect on
Page 196
the diversion of flow.47 48 Subsequent studies49 suggested that, during a more prolonged
period of administration, halothane may indeed cause some depression of hypoxic
vasoconstriction. It would seem reasonable to conclude that, although there is a large
variation in the effect of inhalational agents on hypoxic vasoconstriction, there is no
contraindication to the use of the inhalational agents in most patients undergoing one lung
anaesthesia.
Other drugs
It was Halmagyi and Cotes50 who first reported the occurrence of arterial hypoxaemia
after the administration of bronchodilator drugs such as adrenaline (epinephrine) and
aminophylline to patients with asthma. During the next decade similar changes were
reported with isoprenaline, salbutamol, and orciprenaline, but it was not clear whether the
hypoxaemia resulted from impaired distribution of gas caused by preferential deposition of
the aerosol in relatively well ventilated areas of lung, from maldistribution of blood flow
secondary to haemodynamic changes, or from impaired hypoxic pulmonary
vasoconstriction. Subsequent studies in animals subjected to unilateral ventilation hypoxia
showed that bronchodilator drugs such as salbutamol and orciprenaline could inhibit
hypoxic vasoconstriction.51,52 Isoprenaline is known to dilate vessels in both normoxic and
hypoxic areas of lung, but the other agents have little effect in normoxia and appear to have
a specific action on hypoxic vasoconstriction. These effects have been confirmed by studies
using the multiple inert gas elimination method in patients with asthma which have shown
that blood flow to low ventilation–perfusion areas was doubled after the administration of
nebulised isoprenaline, and that these changes could have accounted for the observed fall in
PaO2.53
Two other β agonists, dopamine and dobutamine, are of particular interest because,
although both increase blood flow to low ventilation–perfusion areas and decrease PaO2,54
they appear to produce their effects by different mechanisms; dobutamine inhibiting the
hypoxic vasoconstrictor response, whereas dopamine vasoconstricts the vessels in the
oxygenated lung and so decreases the flow diversion from the hypoxic area by increasing
the pulmonary artery pressure.55 56 Protamine also produces pulmonary vasoconstriction
and produces hypoxaemia by a similar mechanism.57
The vasodilator drugs nitroglycerine (predominantly a venodilator) and sodium
nitroprusside (acting mainly on the arterial system) have been shown to produce arterial
hypoxaemia in humans, and to depress the diversion of blood flow in response to unilateral
hypoxia in the dog.58 They probably act by releasing nitric oxide. In patients with acute
respiratory distress syndrome (ARDS) sodium nitroprusside produced a decrease in
pulmonary artery pressure and PaO2 with an increase in the shunt
Page 197
component measured by the multiple inert gas method.59 In another study both
nitroglycerine and prostaglandin E1 were found to have similar effects, although PGE1
administration was accompanied by an increase in cardiac output which increased oxygen
delivery.60 As there was no increase in cardiac output or with both nitroprusside and
nitroglycerine, it seems logical to conclude that the increase in shunt was the result of
inhibition of the hypoxic vasoconstrictor mechanism.
The calcium channel blockers also inhibit hypoxic vasoconstriction, although the effects
of these drugs in normal humans are somewhat variable.61 Nifedipine inhibits hypoxic
vasoconstriction in experimental preparations61 and in normal humans.62 It also reduces
pulmonary vascular resistance in patients with primary pulmonary hypertension, or with
pulmonary hypertension secondary to chronic obstructive lung disease.63 Diltiazem has also
been shown to decrease PaO2 and to increase shunt in patients with ARDS.
Drugs that augment hypoxic vasoconstriction
It is apparent that many drugs interfere with the hypoxic vasoconstrictor mechanism and
so impair gas exchange. This raises the question as to whether any advantage would be
gained by the administration of agents which would augment hypoxic vasoconstriction.64
The cyclo-oxygenase inhibitors such as aspirin and indometacin have been shown to
augment hypoxic vasoconstriction, both in the experimental situation and in patients.65
Other agents that appear to have similar actions are alcohol, lidocaine, propranolol, and
almitrine.41 Almitrine seems to have a biphasic action, low doses augmenting the response
and high doses obtunding it.66 67 NOS inhibitors have also been shown to increase hypoxic
vasoconstriction in sepsis,68 whereas the combination of the intravenous administration of a
NOS inhibitor in the presence of unilateral hypoxia, and NO inhalation to the hyperoxic
lung, reduced hypoxic lung blood flow to almost zero.69 Although augmentation of the
response should improve gas exchange, it also increases pulmonary artery pressure, and this
may impair right ventricular function and reduce the efficiency of the hypoxic
vasoconstrictor mechanism. There appears, therefore, to be little clinical indication for the
use of such agents at the present time.
Pulmonary hypertension
Pulmonary hypertension is defined as a chronic increase in pulmonary artery systolic

pressure above 30 mm Hg or a mean pressure greater than 20–25 mm Hg. From a
consideration of Poiseuille’s equation (page 187) it is apparent that pulmonary hypertension
may be expected to occur in the following circumstances:
1 When there is a reduction in the number of vessels perfused.
Page 198
2 When there is a narrowing of the vessels as a result of intimal thickening, muscle
hypertrophy, vascular spasm, or a decrease in the transmural pressure holding them open.
3 When there is an increase in the pulmonary venous pressure.
4 When there is an increase in blood viscosity (for example, as a result of polycythaemia).
The clinical conditions that may cause pulmonary hypertension are listed in the box. For
convenience these are grouped under five main headings:
z those predominantly associated with a reduction in the size of the vascular bed
z those associated with a narrowing of the vessels
z those associated with an increase in pulmonary venous pressure
z primary or idiopathic (cause unknown)70
z diverse aetiology.
It will, however, become apparent that such a categorisation is somewhat artificial for there
are often a number of factors contributing to the hypertension in each patient.
Reduction in perfused vascular bed
One obvious cause of a reduction in the pulmonary vascular bed is surgical resection.
Even if a pneumonectomy is performed, however, the increase in pulmonary artery pressure
is only 5–8 mm Hg provided that the remaining lung is normal. Much greater increases in
pressure are seen if the remaining lung is affected by disease. A reduction in the area of
perfused vascular bed is also produced by pulmonary embolism. This may result from
thromboemboli, amniotic fluid, tumour, fat, or gas bubbles. As the normal pulmonary
circulation has a low resistance, thromboembolism leads to little increase in pressure until
at least 60% of the pulmonary vessels have been occluded. If, however, pulmonary
hypertension is already present (for example, from previous embolisation), right heart
failure may be induced by a relatively small embolus. It has been suggested that the release
of serotonin (5-hydroxytryptamine) or other endogenous substance may accentuate the
hypertension resulting from the obstruction, but there is little evidence that this occurs in
humans. Pulmonary embolism results in perfusion defects on the lung scan, and ventilation
of these areas of lung increases the alveolar component of dead space. Arterial hypoxaemia
is almost invariably present and is caused mainly by an increase in right to left shunt. This
could be the result of right to left shunting through a patent foramen ovale, the
redistribution of blood flow to collapsed areas of lung, or to the presence of a high pressure
pulmonary oedema.
Amniotic fluid embolism typically occurs during or shortly after labour and is commonly
fatal. It is believed that the lethal effects of amniotic fluid
Page 199
Clinical causes of pulmonary hypertension

1 Reduction of vascular bed:
Extensive surgical resection
Pulmonary embolism
Emphysema
Pulmonary fibrosis
2 Narrowing of vessels:
Chronic increases in blood flow
Increased vascular tone
Endogenous vasoconstrictors
Hypoxia and hypercapnia
Decreased transmural pressure
Pulmonary oedema
Increased alveolar pressure
3 Pulmonary venous hypertension
4 Primary (idiopathic)
5 Other: drugs, toxins, parasites, HIV, portal hypertension
emboli result mainly from the thromboses that they induce and from the subsequent
development of disseminated intravascular coagulation.
Destruction of the pulmonary vascular bed appears to be the major cause of hypertension
in patients with emphysema. There is, however, a poor correlation between the magnitude
of the pathological changes and the degree of pulmonary hypertension. It is possible that
this is caused by differences in the site of the lesions. For example, right ventricular
hypertrophy may occur when as little as 14% of the lung is affected by bronchiolar
emphysema, but it rarely develops in patients with panacinar emphysema until 40–70% of
the lung is affected.71
Large areas of fibrosis occur in the lungs of patients with the pneumoconioses,
sarcoidosis, fibrosing alveolitis, and collagen disease such as scleroderma, systemic lupus
erythematosus, and rheumatoid arthritis, and these lead to the development of pulmonary
hypertension. Pulmonary hypertension has also been reported in patients with advanced
tuberculosis and bronchiectasis.
Narrowing of the pulmonary vessels
The presence of pulmonary hypertension in patients with ARDS was first documented in
1977.72 In the acute phase narrowing of the pulmonary vessels may be caused by
endogenous substances, such as thromboxane A2 or B2 and prostaglandin E2, by an
increase in interstitial pressure secondary to pulmonary oedema, and by alveolar hypoxia
and hypercapnia secondary
Page 200
to respiratory failure. In the later phases of the disease, fibrosis and destruction of the
pulmonary vascular bed become important.
Hypoxia and hypercapnia are important causes of pulmonary hypertension in patients
with chronic obstructive lung disease and there is now firm evidence that prolonged oxygen
therapy has beneficial effects on survival.73 74 Pulmonary hypertension in patients with
chronic bronchitis is increased during acute exacerbations. The administration of oxygen
during an acute exacerbation usually results in some reduction in pulmonary artery pressure
and an increase in physiological dead space. Although it was originally believed that the
increase in PCO2 associated with the administration of oxygen resulted from a decrease in
the hypoxic drive to respiration, it now seems probable that the increase is mainly due to
the inability of the patients to increase their minute volume to compensate for the increase
in dead space induced by the redistribution of blood flow. Alveolar hypoxia is an important
cause of pulmonary hypertension in both the neonatal and adult respiratory distress
syndromes and probably accounts for the prevalence of pulmonary hypertension in patients
with kyphoscoliosis. Patients with the primary hypoventilation syndrome and those with
obstructive sleep apnoea may also develop pulmonary hypertension during sleep. This
responds to nocturnal oxygen therapy.
Alveolar hypoxia also appears to be the main cause of pulmonary hypertension in
individuals living at high altitude.75 There is, however, a wide variation in the response
between individuals, whether they are normally domiciled at high altitude or are normally
domiciled at sea level and then taken to high altitude. In those living constantly at high
altitude there is hypertrophy of the media of the muscular pulmonary arteries. When such
people are moved to sea level there is an immediate decrease in pulmonary artery pressure,
which is probably caused by the release of hypoxic pulmonary vasoconstriction, followed
by a more gradual fall, which is probably related to the involution of the muscle fibres.
Sustained high blood flows, such as those resulting from intracardiac shunts, ultimately
produce narrowing of the pulmonary vessels and pulmonary hypertension. Initially there is
medial hypertrophy in the small pulmonary arterioles, and this is later combined with
intimal proliferation, and plexiform and other dilatational lesions. In the more severe cases
pulmonary haemosiderosis and fibrinoid necrosis are seen. The hypertension decreases
when oxygen is inhaled and the pathological lesions regress if the heart defect is corrected
in the earlier stages of the disease, but, if dilatational lesions and plexogenic arteriopathy
have developed, the changes are generally irreversible. The time of onset of these changes
depends, to a large extent, on the location of the shunt. In patients with pre-tricuspid shunts
(for example, atrial septal defects) the changes tend to develop in young adults or in middle
life, whereas patents with post-tricuspid shunts (for example, ventricular septal defects)
tend to retain the
Page 201
fetal pattern of pulmonary circulation and so have pulmonary hypertension from birth.
Pulmonary blood vessels may also be narrowed by a decrease in transmural pressure
caused by a reduction in lung volume or an increase in interstitial pressure resulting from
pulmonary oedema. A more common cause of narrowing is an increase in alveolar pressure
as a result of an increase in airway resistance. This appears to be an important cause of
pulmonary hypertension in patients with asthma and chronic bronchitis.
Pulmonary venous hypertension
Pulmonary venous hypertension may be caused by mediastinal lesions which compress

the pulmonary veins, a myxoma or ball-valve thrombus in the left atrium, mitral or aortic
valve disease, or left ventricular failure. It may also result from pulmonary veno-occlusive
disease.
An acute increase in pressure in the pulmonary venous system results in pulmonary
congestion and a corresponding increase in pulmonary artery pressure. If the congestion is
severe, pulmonary oedema may result. Chronic increases in venous pressure may lead to
pathological changes in the lung but, in contrast to those produced by increases in
precapillary pressure, they affect the whole of the pulmonary vasculature. The pulmonary
veins and venules show medial hypertrophy, arterialisation, dilatation, and eccentric intimal
fibrosis. The microcirculation is characterised by capillary congestion, oedema, dilatation
of interstitial and pleural lymphatics, and alveolar haemosiderosis. Pulmonary arterioles are
often muscularised and both muscular and elastic arteries may be dilated.76
Other causes
There are a number of other clinical conditions in which the cause of the pulmonary
hypertension is even more obscure. For example, there was a Swiss epidemic of pulmonary
hypertension which appeared to be related to the use of the slimming drug aminorex, and
there was another epidemic in Spain associated with the use of contaminated cooking oil.
There is an association between portal and pulmonary hypertension, and pulmonary
hypertension may occur in patients with schistosomiasis when their ova impact in the lung.
Patients with HIV infection may develop pulmonary hypertension, and there is also a
condition known as primary or idiopathic pulmonary hypertension in which the aetiology is
still far from clear.70
Role of pulmonary vasodilators in pulmonary hypertension
The most obvious effect of pulmonary hypertension is that it increases the right
ventricular pressure, workload, and oxygen consumption, and so
Page 202
may lead to right ventricular failure. As the blood flow to the right ventricle occurs during
both systole and diastole it is important to maintain a high systemic pressure to minimise
myocardial ischaemia. A second problem resulting from a high pulmonary artery pressure
is that it may cause high pressure pulmonary oedema by increasing the pressure in the
precapillary vessels. Pulmonary venous hypertension also causes oedema and in both
situations severe arterial hypoxaemia may result. Pulmonary embolism and other conditions
causing chronic hypertension may cause a maldistribution of blood flow as a result of
changes in arteriolar resistance. Localised reductions in flow result in non-perfused alveoli
and an increase in alveolar dead space, but the concomitant increases in pressure may also
oppose the effects of hypoxic vasoconstriction, which redistributes flow away from
underventilated areas of lung, and so may also increase arterial hypoxaemia.
It will be apparent that there are many causes of pulmonary hypertension, and that a
number may be present in any one patient. For example, in chronic obstructive airway
disease the hypertension may be caused by alveolar hypoxia and hypercapnia, destruction
of the vascular bed, polycythaemia (which increases blood viscosity), gas trapping, and
water retention (which increases the static pressure throughout the circulation). Although
little can be done to alter the size of the vascular bed, all the other causes are amenable to
treatment. The effects of treatment will, however, depend on the magnitude and
reversibility of the factors involved in each patient. Similarly in ARDS it seems probable
that in the early stages pulmonary hypertension is predominantly caused by
vasoconstriction from hypoxia and endogenous mediators, augmented by a reduction in the
vascular bed as a result of alveolar collapse, and obstruction of small vessels by leucocytes,
although in the later stages medial hypertrophy, thrombosis, and interstitial fibrosis
dominate the scene. Again, the relative importance of each factor in the individual patient is
difficult to determine. Many of the putative mediators are very short lived and difficult to
measure, and there is no good animal model of ARDS. As a number of therapeutic
strategies currently employed are designed to reduce pulmonary vasomotor tone, it is
necessary to consider some of the implications of this type of therapy.
Pulmonary vasodilator drugs
Conventional vasodilator drugs have three main disadvantages in patients with

pulmonary hypertension:
1 Most pulmonary vasodilators also dilate the systemic vascular bed and so decrease aortic
pressure. This results in a decrease in coronary perfusion which may lead to myocardial
ischaemia when right ventricular stroke work and oxygen consumption are increased by
an excessive afterload.
Page 203
2 The decrease in pulmonary artery pressure may decrease the perfusion of non-dependent
zones and so increase alveolar dead space.
3 Inhibition of the hypoxic vasoconstrictor mechanism may increase flow to poorly
ventilated areas of lung and so increase arterial hypoxaemia.
These disadvantages, and the failure to document improved survival in adult patients
treated with conventional pulmonary vasodilator drugs, suggest that there is little indication
for their use in adult practice at the present time. (This is not the case in the neonate where
pulmonary vasodilators may play an important role in maintaining pulmonary blood flow in
patients with persistent pulmonary hypertension.)
The situation has been radically changed by the recent discovery of the role of nitric
oxide as a physiological vasodilator.77 Nitric oxide is also a bronchodilator.78 It has now
been shown that nitric oxide can vasodilate the pulmonary vascular bed when inhaled in
concentrations of 10–80 p.p.m. (parts per million). The advantage of the inhalational route
of administration is that the nitric oxide only dilates the pulmonary vessels supplying
ventilated alveoli, and does not reach non-ventilated alveoli where vessels are constricted
by hypoxia. Furthermore, it has no effect on systemic vessels because it is immediately
deactivated by combination with haemoglobin.79 The inhalation of 18 p.p.m. of nitric oxide
decreases pulmonary artery pressure and shunt in patients with ARDS, and the therapy has
been continued for many days without apparent toxicity.80 An even more recent study has
demonstrated a significant improvement in oxygenation without any reduction in
pulmonary artery pressure when inhaled in concentrations of 60–230 parts per billion.81
Interestingly, this concentration is similar to that produced by endogenous NO production
in the nasal mucosa,82 which is denied the ARDS patient by the use of a tracheal tube!
Nitric oxide inhalation has also been used with success in neonates with persistent
pulmonary hypertension,83 but in patients with chronic obstructive lung disease, in which
the hypoxaemia is predominantly caused by ventilation–perfusion inequalities, NO may
decrease arterial PO2 because it dilates vessels in areas of lung that had previously been
subject to hypoxic pulmonary vasoconstriction.84 There may also be interactions with
anaesthetic agents.85
It seems strange that an agent that was once a feared contaminant of nitrous oxide should
now be hailed as a therapeutic agent of great promise. However, NO is rapidly oxidised to
nitrogen dioxide (NO2) and requires specially designed apparatus for its administration,
together with careful monitoring.86 As NO is the mediator in so many biological systems,
there are many potential side effects of long term administration. This is an exciting new
form of treatment which can undoubtedly decrease pulmonary artery pressure and improve
arterial oxygenation, but there is no evidence to indicate that it affects mortality from
ARDS.
Page 204
An alternative method of inducing vasodilation in ventilated areas of lung in ARDS is to
nebulise prostacyclin.87–89Inhaled PGE1 has, however, been found to be ineffective.90
Fluid balance in the lung

No review of the pulmonary circulation would be complete without a brief consideration
of its role in the control of water and solute transfer within the lung, and of the mechanisms
causing pulmonary oedema.
Electron micrographs show that the basement membranes of the alveolar epithelium and
endothelium appear to be fused over the thin portion of the interalveolar septum, where
most gas exchange is presumed to take place, but that they are separated in the thick part of
the septum by the interstitial space (Fig. 5.11). Although no direct connection between the
alveolar interstitial space and the lymphatics surrounding the major conducting airways and
blood vessels has been demonstrated, there are lymphatic channels surrounding the small
blood vessels and terminal airways which probably drain the alveolar interstitial space into
the larger lymphatics. From these, the lymph is pumped into the thoracic duct and so back
into the circulation. At the sites where adjacent cells in both the endothelium
Fig 5.11 Structure of air–blood barrier. The alveolar epithelial and endothelial basement membranes appear to
be fused over the thin, gas exchanging, portion of the septum (left), but are separated by the interstitial space
in the thick portion of the septum (right).
Page 205
and alveolar epithelium abut or overlap there are narrow clefts, but, although the clefts
between endothelial cells are about 4 nm in diameter, those between alveolar type I and II
cells in the epithelium are almost completely fused towards the alveolar spaces. Thus fluids
and solutes that move easily between the inside of the capillary and the interstitial space are
prevented from passing into the alveolar space. Liquid movements across membranes can
occur through cellular junctions or cell membranes, or by endocytosis, and each of these
mechanisms can be modelled by ‘‘pores” of given dimensions, thus accounting for the
differences in permeability of different membranes under different conditions.
The transfer of fluid from the inside of the capillary to the interstitial space depends
on the balance between the hydrostatic and colloid osmotic pressures. This is traditionally
described by the Starling equation:
where K is the capillary filtration coefficient, which describes the permeability

characteristics of the endothelial membrane through which exchange occurs, Pmv and Ppmv
are the hydrostatic pressures in the capillary and interstitial space, πmv and πpmv are the
colloid osmotic pressures in the microvascular and perimicrovascular compartments, and σ
is the reflection coefficient, which is an expression of the permeability of the endothelium
to the solute, the most important of which is albumin. A value of unity would indicate that
the endothelium was impermeable to albumin, whereas a value of zero would indicate that
it was freely permeable. Under the second condition, there would be equal concentrations
of solute on both sides of the membrane so there could be no osmotic gradient.
Under normal conditions, pulmonary lymph flow in humans is about 10 ml/h, but it can
probably increase tenfold before alveolar flooding occurs. The pulmonary microvascular
pressure is between 0 and 20 cm H2O, depending on the vertical height of the capillary in
the lung, and is highest at the proximal end of the capillary. Interstitial pressure around
alveolar vessels is difficult to measure but is generally believed to be about 3–5 cm H2O
below atmospheric, whereas interstitial pressure around extra-alveolar vessels is believed to
be close to pleural pressure at the same vertical height. The value of σ is about 0·5 and the
protein content of lymph is about half that of the plasma, so the colloid osmotic pressure
gradient is about 30 – 15=15 cm H2O. This results in a net pressure gradient favouring
filtration, which is greater in dependent zones of the lung, thus accounting for the fact that
interstitial oedema usually appears first in dependent zones. The interstitial space can
accommodate up to 500 ml of fluid with only a small rise in interstitial pressure, but, when
a critical pressure in the interstitial space is exceeded, the fluid may break through the
alveolar epithelium, so causing alveolar flooding.91 This is an all or none
Page 206
phenomenon, so that flooded alveoli may be seen surrounded by apparently normal alveoli.
Causes of pulmonary oedema
There are four major causes of pulmonary oedema: increased hydrostatic pressure,
increased capillary permeability, decreased plasma osmotic pressure, and lymphatic
obstruction.
Increased capillary hydrostatic pressure may result from overtransfusion, an obstruction
to pulmonary venous outflow (for example, caused by left atrial myxoma) or to left heart
failure resulting from mitral valve disease or severe dysrhythmias. It may occur when blood
flow to a restricted vascular bed is suddenly increased by correction of an anatomical defect
(for example, repair of Fallot’s tetralogy), and may also be induced by extreme exercise,
particularly if the vascular bed is already constricted by hypoxia (as in high altitude
pulmonary oedema). High pressure oedema fluid has a low protein content. Extremely high
pulmonary capillary pressures (in the region of 30–50 mm Hg) may create what has been
termed “stress failure” of the pulmonary vasculature.92 Discontinuities appear in the bodies
of endothelial and type I epithelial cells, whereas the basement membrane often remains
intact, and these lesions permit leakage of fluid and protein into the alveoli. Similar
abnormalities are believed to occur when the lung is overdistended in mechanically
ventilated patients with acute lung injury.93 Pulmonary oedema has also been described in
association with upper airway obstruction, severe laryngospasm, or asthma in children. This
is probably caused by large decreases in absolute pleural pressure which increase left
ventricular afterload and also increase venous return, so promoting pulmonary congestion
and pulmonary oedema.94
Increased capillary permeability oedema is seen in situations in which the alveolar
capillary membrane is damaged by aspiration of gastric contents, inhalation of toxic agents
(smoke, irritant gases), or bacterial or viral agents. It is usually diagnosed when oedema
occurs in the face of a pulmonary capillary wedge pressure of less than 18 mm Hg. It is a
common complication of generalised sepsis and is the basic feature of lung damage seen in
ARDS. The oedema fluid has a protein content that approaches that of plasma and seems to
take longer to clear from the lung than high pressure oedema.
Although oedema would be expected to occur frequently in situations where plasma
proteins are reduced (for example, in extreme haemodilution), it is not common. This is
probably because the decrease in plasma proteins results in an increase in interstitial fluid,
which dilutes the protein content of lymph, so restoring the osmotic balance.95
Thoracic duct obstruction is a rare cause of decreased lymph drainage. The duct drains
into the great veins, however, so that an increase in central venous pressure may increase
the likelihood of pulmonary oedema.
Page 207
Other miscellaneous causes of pulmonary oedema are the neurogenic oedema associated
with head injury, the oedema associated with heroin overdose, and that associated with
sudden expansion of a collapsed lung.
Measurement of extravascular lung water

The only accurate way of measuring extravascular lung water is by comparing
postmortem wet/dry weight ratios after correcting for the residual volume of blood
contained in the wet lung. Most of the methods of measuring extravascular lung water that
can be used clinically are based on double indicator dilution techniques. Most indicator
techniques, however, underestimate lung water because the indicator fails to reach all parts
of the lung. Gaseous indicators do not penetrate into collapsed areas of the lung, and
intravascular indicators similarly fail to reach non-perfused areas of the lung. Considerable
energy has been expended in attempts to improve the double indicator dilution method in
which one indicator (usually a dye) is chosen to remain within the circulation, whereas a
second indicator (usually “coolth” or tritiated water) diffuses into the extravascular water as
well. The extravascular water is then derived from the difference between the two
circulation volumes. Compression of extra-alveolar vessels by interstitial oedema, however,
inevitably results in localised reductions of pulmonary blood flow, so errors of up to 40%
may occur.96 Thoracic electrical impedance measurements are also subject to large errors.
Although measurements using positron emission tomography (PET)97 and magnetic
resonance imaging (MRI)98 have proved reasonably accurate, the techniques are not
suitable for routine use. Computed tomography scans have provided invaluable insights
into the pathophysiological changes in ARDS99 but are not routinely available in the
intensive care unit. Most clinicians therefore continue to rely on radiological estimates of
the severity of pulmonary oedema, together with measurements of gas exchange, as a guide
to therapy.
Conclusions
The low resistance of the pulmonary circulation results in relatively poor control of the
distribution of blood flow. Distribution is primarily dependent on the interrelationship of
transpulmonary and vascular pressures, but is modified by differences in regional
conductance associated with the length of the vascular pathways and the fractal branching
pattern of the pulmonary vasculature. Distribution is further modified at alveolar level by
the influence of alveolar and mixed venous gas tensions. Pulmonary vascular tone is also
influenced by NO and other mediators derived from the endothelium. Drugs may modify
the distribution of flow by changing cardiac output and pulmonary vascular pressures, but a
number of drugs
Page 208
may alter distribution by altering pulmonary vascular tone in normoxic or hypoxic areas of
lung. Drugs that decrease the effectiveness of hypoxic pulmonary vasoconstriction may
increase blood flow to underventilated lung regions and so cause arterial hypoxaemia.
Selective vasodilation of ventilated lung regions by nitric oxide or PGI2 may, however,
reduce shunt and pulmonary artery pressure. Although the complexity of the system makes
it difficult to predict the effects of drugs on gas exchange, a knowledge of the potential
mechanisms will enable the clinician to understand many of the phenomena observed in
routine clinical practice, and should thus improve the standard of patient care.
1 Harvey W. An anatomical disssertation concerning the movement of the heart and blood in living
creatures. Translated by GWhitteridge. Oxford: Blackwell Scientific, 1977.
2 Malpighi M. Duae epistolae de pulmonibus. Florence, 1661.
3 Bradford JR, Dean HP. The pulmonary circulation. J Physiol 1894; 16:34–96.
4 Euler US von, Liljestrand G. Observations on the pulmonary arterial blood pressure in the cat. Acta
Physiol Scand 1946; 12:301–20.
5 Nisell O. Effects of oxygen and carbon dioxide on the circulation of isolated and perfused lungs of the
cat. Acta Physiol Scand 1948;16:121–7.
6 Cournand A, Ranges HA. Catheterization of right auricle in man. Proc Soc Exp Biol Med 1941;46:462–6.
7 Bakhle YS. Pharmacokinetic and metabolic properties of lung. Br J Anaesth 1990; 65:79–93.
8 West JB. Blood flow. In: West JB, ed, Regional differences in the lung. London: Academic Press,
1977:85–165.
9 Hughes JMB, Glazier JB, Maloney JE, West JB. Effect of lung volume on the distribution of pulmonary
blood flow in man. Respir Physiol 1968;4:58–72.
10 Chen L, Williams JJ, Alexander CM, Ray RJ, Marshall C, Marshall BE. The effect of pleural pressure on
the hypoxic pulmonary vasoconstrictor response in closed chest dogs. Anesth Analg 1988; 67:763–9.
11 Amis TC, Jones HA, Hughes JMB. Effect of posture on inter-regional distribution of pulmonary perfusion
and VA/Q ratios in man. Respir Physiol 1984;56:169–82.
12 Hakim TS, Lisbona R, Dean GW. Gravity-independent inequality in pulmonary blood flow in humans. J
Appl Physiol 1987;63:1114–21.
13 Glenny RW, Robertson HT. Regional differences in the Jung. A changing perspective on blood flow
distribution. In: Hlastala MP, Robertson HT, eds. Complexity in structure and function of the lung. New
York: Marcel Dekker Inc, 1998:461–81.
14 Beck KC, Rehder K. Differences in regional conductances in isolated dog lungs. J Appl Physiol
1986;61:530–8.
15 Beck KC, Vettermann J, Rehder K. Gas exchange in dogs in the prone and supine positions. J Appl
Physiol 1992;72:2292–7.
16 Hlastala MP, Chornuk MA, Self DA, et al. Pulmonary blood flow redistribution by increased gravitational
force. J Appl Physiol 1998;84:1278–88.
17 Robertson HT. Measurement of regional ventilation by aerosol deposition. In: Hlastala MP, Robertson
HT, eds, Complexity in structure and function of the lung. New York: Marcel Dekker Inc, 1998:379–99.
18 Swenson ER, Domino KB, Hlastala MP. Physiological effects of oxygen and carbon dioxide on VA/Q
heterogeneity. In: Hlastala MP, Robertson HT, eds, Complexity in structure and function of the lung. New
York: Marcel Dekker Inc, 1998:511–47.
19 Hedenstierna G. Effects of anaesthesia on respiratory function. In: Pearl RG, ed, The lung in anaesthesia
and intensive care. Ballière’s Clinical Anaesthesiology 10.1. London: Ballière Tindall, 1996:1–16.
Page 209
20 Gattinoni L, Pelosi P, Vitale G, Pesenti A, D’Andrea L, Mascheroni D. Body position changes redistribute
lung computed-tomographic density in patients with acute respiratory failure. Anesthesiology 1991;74:15–
23.
21 Morrell NW, Nijran KS, Biggs T, Seed WA. Magnitude and time course of acute hypoxic pulmonary
vasoconstriction in man. Respir Physiol 1995;100:271–81.
22 Dorrington KL, Clar C, Young JD, Jonas M, Tansley JG, Robbins PA. Time course of human pulmonary
vascular response to 8 hours of isocapnic hypoxia. Am J Physiol 1997;273(Heart Circ Physiol 42)H1126–
34.
23 Barman SA. Potassium channels modulate hypoxic pulmonary vasoconstriction. Am J Physiol
1998;275:L64–70.
24 Marshall BE, Hanson CW, Frasch F, Marshall C. Role of HPV in pulmonary gas exchange and blood
flow distribution. Intensive Care Med 1994;20:291–7;379–89.
25 Marshall BE, Marshall C, Benumof J, Saidman LJ. Hypoxic pulmonary vasoconstriction in dogs: effects
of lung segment size and oxygen tension. J Appl Physiol 1981;51:1543–51.
26 Benumof JL, Wahrenbrock EA. Blunted hypoxic pulmonary vasoconstriction by increased lung vascular
pressures. J Appl Physiol 1975;38:846–50.
27 Graham LM, Vasil A, Vasil ML, Voelkel NF, Stenmark KR. Decreased pulmonary vasoreactivity in an
animal model of chronic Pseudomonas pneumonia. Am Rev Respir Dis 1990;142:221–9.
28 Rodriguez-Roisin R, Roca J, Agusti AG, Mastai R, Wagner PD, Bosch J. Gas exchange and pulmonary
vascular reactivity in patients with liver cirrhosis. Am Rev Respir Dis 1987;135:1085–92.
29 Dupuy PM, Lancon JP, Francoisc M, Frostell CG. Inhaled cigarette smoke selectively reverses human
hypoxic vasoconstriction. Intensive Care Med 1995;21:941–4.
30 Mann CM, Domino KB, Walther SM, Glenny RB, Polissar NL, Hlastala MR Redistribution of pulmonary
blood flow during unilateral hypoxia in prone and supine dogs. J Appl Physiol 1998;84:2010–19.
31 McFarlane PA, Gardaz J-P, Sykcs MK. CO2 and mechanical factors reduce blood flow in a collapsed lung
lobe. J Appl Physiol 1984;57:739–43.
32 Curzen NP, Jourdan KB, Mitchell JA. Endothelial modification of pulmonary vascular tone. Intensive
Care Med 1996;22:596–607.
33 Al-Ali MK, Howarth P. Nitric oxide and the respiratory system in health and disease. Respir Med
1998;92:710–15.
34 Gómez CMH, Palazzo MGA. Pulmonary artery catheterization in anaesthesia and intensive care. Br J
Anaesth 1998;81:945–56.
35 Leeman M. The pulmonary circulation in acute lung injury: a review of some recent advances. Intensive
Care Med 1991;17:254–60.
36 Carlsson AJ, Hedenstierna G, Blomqvist H, Strandberg A. Separate lung blood flow in anesthetized dogs:
a comparitive study between electromagnetometry and SF6 and CO2 elimination. Anesthesiology
1987;67:240–6.
37 Sykes MK, Hill AEG, Loh L, Tait AR. Evaluation of a new method for continuous measurement of the
distribution of blood flow between the two lungs. Br J Anaesth 1977;49:285–92.
38 Hedenstierna G. Respiratory measurement. London: BMJ Books, 1998:148–61.
39 Buckley MJ, McLauchlin JS, Fort L, Saigusa M, Morrow DH. Effects of anesthetic agents on pulmonary
vascular resistance during hypoxia. Surg Forum 1964;15: 183–4.
40 Sykes MK, Loh L, Seed RF, Kafer ER, Chakrabarti MK. The effect of inhalational anaesthetics on
hypoxic pulmonary vasoconstriction and pulmonary vascular resistance in the perfused lungs of the dog
and the cat. Br J Anaesth 1972;44:776–87.
41 Sykes MK. Anaesthetics and the pulmonary circulation. In: Altura BM, Halevy S, eds, Cardiovascular
actions of anaesthetics and drugs used during anaesthesia, Vol 2. Regional blood flow and clinical
consderations. Basel: Karger, 1986:92–125.
42 Eisenkraft JB. Effects of anaesthetics on the pulmonary circulation. Br J Anaesth 1990;65:63–78.
43 Ishibe Y, Gui X, Uno H, Shiokawa Y, Umeda T, Suekane K. Effect of sevoflurane on hypoxic pulmonary
vasoconstriction in the perfused rabbit lung. Anesthesiology 1993;79:1348–53.
Page 210
44 Johnson SR. Effect of some anaesthetic agents on circulation in man-with special reference to the
significance of pulmonary blood volume for the circulatory regulation. Acta Chir Scand 1951;suppll58:l–
143.
45 Bjertnaes LJ. Hypoxia-induced pulmonary vasoconstriction in man: inhibition due to diethyl ether and
halothane anesthesia. Acta Anaesthesiol Scand 1978;22:570–88.
46 Rogers SN, Benumof JL. Halothane and isoflurane do not decrease PaO2 during one-lung ventilation in
intravenously anesthetized patients. Anesth Analg 1985;64:946–54.
47 Carlsson AJ, Bindslev L, Hedenstierna G. Hypoxia-induced pulmonary vasoconstriction in the human
lung; the effect of isoflurane anesthesia. Anesthesiology 1987;66:312–16.
48 Carlsson AJ, Hedenstierna G, Bindslev L. Hypoxia-induced pulmonary vasoconstriction in human lung
exposed to enflurane anaesthesia. Acta Anaesthesiol Scand 1987;31:57–62.
49 Benumof JL, Augustine SD, Gibbons JA. Halothane and isoflurane only slightly impair oxygenation
during one-lung ventilation in patients undergoing thoractomy. Anesthesiology 1987;67:910–15.
50 Halmagyi DF, Cotes JE. Reduction in systemic blood oxygen as a result of procedures affecting the
pulmonary circulation in patients with chronic pulmonary disease. Clin Sci 1959;18:475–89.
51 Reyes A, Sykes MK, Chakrabarti MK, Tait A, Petrie A. The effect of salbutamol on hypoxic pulmonary
vasoconstriction in dogs. Bull Eur Physiopathol Respir 1978;14:741–53.
52 Reyes A, Sykes MK, Chakrabarti MK, Carruthers B, Petrie A. Effect of orciprenaline on hypoxic
pulmonary vasoconstriction in dogs. Respiration 1979;38:185–93.
53 Wagner PD, Dantzker DR, Iacovoni VE, Tomlin WC, West JB. Ventilation–perfusion inequality in
asymptomatic asthma. Am Rev Respir Dis 1978;118:511–24.
54 Rennotte MT, Reynaert M, Clerbaux Th, et al. Effects of two inotropic drugs, dopamine and dobutamine,
on pulmonary gas exchange in artifically ventilated patients. Intensive Care Med 1989;15:160–5.
55 Gardaz JP, McFarlane PA, Sykes MK. Mechanisms by which dopamine alters blood flow distribution
during lobar collapse in dogs. J Appl Physiol 1986;60:959–64.
56 NomotoY, Kawamura M. Pulmonary gas exchange effects by nitroglycerin, dopamine and dobutamine
during one lung ventilation in man. Can J Anaesth 1989;36:273–7.
57 KimYD, Michalik R, Lees DE, Jones M, Hanowell S, MacnamaraTE. Protamine induced arterial
hypoxaemia: the relationship to hypoxic pulmonary vasoconstriction. Can Anaesth Soc J 1985;32:5–11.
58 D’Oliveira M, Sykes MK, Chakrabarti MK, Orchard C, Keslin J. Depression of hypoxic pulmonary
vasoconstriction by sodium nitroprusside and nitroglycerine. Br J Anaesth 1981;53:11–17.
59 Radermacher P, Huet Y, Pluskwa F, et al. Comparison of ketanserin and sodium nitroprusside in patients
with severe ARDS. Anesthesiology 1988;68:152–7.
60 Radermacher P, Santak B, Becker H, Falke KJ. Prostaglandin E1 and nitroglycerin reduce pulmonary
capillary pressure but worsen ventilation-perfusion distributions in patients with adult respiratory distress
syndrome. Anesthesiology 1989;70:601–6.
61 Naeije R, Mélot C, Mols P, Hallemans R. Effect of vasodilators on hypoxic pulmonary vasoconstriction in
normal man. Chest 1982;82:404–10.
62 Kennedy T, Summer W. Inhibition of hypoxic pulmonary vasoconstriction by nifedipine. Am J Cardiol
1982;50:864–8.
63 Mélot C, Hallemans R, Naeije R, Mols P, Lejeune P. Deleterious effect of nifedipine on pulmonary gas
exchange in chronic obstructive pulmonary disease. Am Rev Respir Dis 1984;130:612–6.
64 Mélot C, Naeije R, Mols P, Hallemans R, Lejeune P, Jaspar N. Pulmonary vascular tone improves
pulmonary gas exchange in the adult respiratory distress syndrome. Am Rev Respir Dis 1987;136:1232–6.
65 Adnot S, Defouilloy C, Brun-Buisson C, Piquet J, de Cremoux H, Lemaire F. Effects of indomethacin on
pulmonary hemodynamics and gas exchange in patients with pulmonary artery hypertension, interference
with hydralazine. Am Revs Respir Dis 1987;136:1243–9.
66 Reyes A, Roca J, Rodriguez-Roison R, Torres A, Ussetti P, Wagner PD. Effect of almitrine on
ventilation–perfusion distribution in adult respiratory distress syndrome. Am Rev Respir Dis
1988;137:1062–7.
Page 211
67 Mélot C, Deschamps P, Hallemam R, Decroly P, Mols P. Enhancement of hypoxic pulmonary
vasoconstriction by low dose almitrine bismesylate in normal humans. Am Rev Respir Dis 1989;136:111–
19.
68 Fischer SR, Deyo DJ, Bone HG, McGuire R,Traber LD,Traber DL. Nitric oxide synthase inhibition
restores hypoxic pulmonary vasoconstriction in sepsis. Am J Respir Crit Care Med 1997;l56:833–9.
69 Freden F, Wei SZ, Berglund JE, Frostell C, Hedenstierna G. Nitric oxide modulation of pulmonary blood
flow distribution in lobar hypoxia. Anesthesiology 1995;82:1216–25.
70 Rubin LJ. Primary pulmonary hypertension. N Engl J Med 1997;336:111–17.
71 Harris P, Heath D. The human pulmonary circulation, 3rd edn. Edinburgh: Churchill Livingstone,
1986:511.
72 Zapol WM, Snider MX Pulmonary hypertension in severe acute respiratory failure. N Engl J Med
1977;296:476–80.
73 Stuart-Harris C, Bishop JM, Clark TJH, et al. Long term domiciliary oxygen therapy in chronic hypoxic
cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981;i:681–6.
74 Nocturnal oxygen therapy trial group 1980. Continuous or nocturnal oxygen therapy in hypoxic
obstructive airways disease. Ann Intern Med 1980;93:391–8.
75 Naeije R. Pulmonary circulation at high altitude. Respiration 1997;64:429–34.
76 Edwards WD. Pathology of pulmonary hypertension. Cardiovasc Clin 1988;18:321–59.
77 Barnes PJ, Belrisi MG. Nitric oxide and lung disease. Thorax 1993;48:1034–43.
78 Dupuy PM, Shore SA, Drazen JM, Frostell C, Hill WA, Zapol WM. Bronchodilator action of inhaled
nitric oxide in guinea pigs. J Clin Invest 1992;90:421–8.
79 Frostell C, Fratacci M-D, Wain JC, Jones R, Zapol WM. Inhaled nitric oxide. A selective pulmonary
vasodilator reversing hypoxic pulmonary vasoconstriction. Circulation 1991;83:2038–47.
80 Roissant R, Falke KJ, López F, Slama K, Pison U, ZapolWM. Induced nitric oxide for the adult
respiratory distress syndrome. N Engl J Med 1993;328:399–405.
81 Gerlach H, Pappert D, Lewandowski K, Rossaint R, Falke KJ. Long-term inhalation with evaluated low
doses of nitric oxide for selective improvement of oxygenation in patients with adult respiratory distress
syndrome. Intensive Care Med 1993;19:443–9.
82 Gerlach H, Rossaint R, Pappert D, Knorr M, Falke KJ. Autoinhalation of nitric oxide after endogenous
synthesis in nasopharynx. Lancet 1994;343:518–19.
83 Adnot S, Raffestin B, Eddahibi S. NO in the lung. Respir Physiol 1995;101:109–120.
84 Barbera JA, Roger N, Roca J, Rovira I, Higenbottam TW, Rodriguez-Roisin R. Worsening of pulmonary
gas exchange with nitric oxide in chronic obstructive pulmonary disease. Lancet 1996;347:436–40.
85 Nakamura K, Mori K. Nitric oxide and anesthesia. Anesth Analg 1993;77:877–9.
86 Young JD, Dyer OJ. Delivery and monitoring of inhaled nitric oxide. Intensive Care Med 1996;22:77–86.
87 Zwissler B, Kemming G, Habler O, et al. Inhaled prostacyclin (PGI2) versus inhaled nitric oxide in adult
respiratory distress syndrome. Am J Respir Crit Care Med 1996;154:1671–7.
88 Walmroth D, SchneiderT, Shermuly R, Olschewski H, Grimminger F, SeegerW. Direct comparison of
inhaled nitric oxide and aerosolized prostacyclin in acute respiratory distress syndrome. Am J Respir Crit
Care Med 1996;153:991–6.
89 Olschewski H, Walmroth D, Shermuly R, Ghofrani A, Grimminger F, Seeger W. Aerosolized
prostacyclin and iloprost in severe pulmonary hypertension. Ann Intern Med 1996;124:820–4.
90 Sinclair SE, Albert RK. Altering ventilation–perfusion relationships in ventilated patients with acute lung
injury. Intensive Care Med 1997;23:942–50.
91 Staub NA. Pathophysiology of pulmonary edema. In: Staub NA, Taylor AE, eds, Edema. New York:
Raven Press, 1984.
92 Wallin CJ, Rundgren M, Hjelmqvist H, Eriksson S, Leksell LG. Effects of rapid colloid expansion on
pulmonary microvascular pressure and lung water in the conscious sheep. Respir Physiol 1997;108:225–
31.
93 Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies (state of the
art). Am J Respir Crit Care Med 1998;157:1–30.
Page 212
94 Pinsky MR. The hemodynamic consequences of mechanical ventilation: an evolving story. Intensive Care
Med 1997;23:493–503.
95 Sanchez de Léon R, Paterson JL, Sykes MK. Changes in colloid osmotic pressure and plasma albumin
concentration associated with extracorporeal circulation. Br J Anaesth 1982;54:465–73.
96 Wallin CJ, Rosblad PG, Leksell LG. Quantitative estimation of errors in the indicator dilution
measurement of extravascular lung water. Intensive Care Med 1997;23:469–75.
97 Velasquez M, Haller J, Amundsen T, Schuster JP. Regional lung water measurements with PET:
accuracy, reproducibility, and linearity. J Nucl Med 1991;32:719–25.
98 Mayo JR, MacKay AL, Whittall KP, Baile EM, Pare PD. Measurement of lung water content and pleural
pressure gradient with magnetic resonance imaging. J Thorac Imaging 1995;10:73–81.
99 Desai SR, Hansell DM. Lung imaging in the adult respiratory distress syndrome: current practice and new
insights. Intensive Care Med 1997;23:7–15.
Page 213
6: Regulation of the cardiovascular system

NIRAJ NIJHAWAN, DAVID C WARLTIER
Regulation of the cardiovascular system has two major homoeostatic goals: to maintain
arterial pressure relatively constant and to provide sufficient perfusion to tissues to meet
regional metabolic demands. When the requirements of an entire organism for blood flow
are altered, the compensatory cardiovascular response involves changing the arterial
perfusion pressure. On the other hand, when the perfusion to a particular organ system is
altered, cardiovascular compensation occurs by adjusting the calibre of specific blood
vessels. The most important variable that is controlled in the regulation of the circulation is
arterial blood pressure. Arterial pressure has often been related to Ohm’s law in physics.
This implies that blood pressure (analogous to voltage) is directly proportional to the
product of cardiac output (current) and peripheral vascular resistance (resistance). Neural
and humoral control of arterial pressure are mediated through alterations of both cardiac
output and/or peripheral resistance.
The physical characteristics of the vascular system are dynamic and more complex than
previously realised. Age and certain pathophysiological conditions, such as hypertension,
diabetes, and atherosclerosis, result in a progressive stiffening of the conduit (large artery)
vessels related to intimal and smooth muscle changes. Vascular smooth muscle hypertrophy
as a primary or a secondary phenomenon results in a hypersensitivity to control
mechanisms. In any discussion pertaining to cardiovascular regulatory mechanisms,
pathophysiological changes must be considered. Interestingly, ventricular hypertrophy may
proceed with elevated or normal blood pressure. This appears to be related to the very
important pulsatile component of ventricular load, which is not necessarily captured by
measurements of blood pressure.1 An understanding of the differential impact of conduit
and resistance vessels will expand future investigations into cardiovascular control
mechanisms as well as pharmocotherapy.
Acute mechanisms for regulation of arterial pressure are coordinated in the
cardiovascular control centres of the brain stem. These centres regulate both cardiac output
and peripheral resistance and, therefore, are able to
Page 214
exert powerful control of arterial pressure. The brain stem cardiovascular centres are in turn
influenced by impulses from other neural centres, as well as by numerous sensors within
and external to the circulation.
Further regulation of the circulation is mediated by endogenous chemical substances,
which are released into the circulation and have a direct effect on the heart and/or
vasculature in both physiological and pathophysiological states. Many tissues have the
ability to regulate blood flow by local mechanisms, preferentially altering regional
perfusion without changes in cardiac output and systemic vascular resistance. Finally, the
kidneys are the long term regulators of the entire circulation by manipulating the overall
fluid status of an organism. This chapter addresses these different mechanisms individually,
but ultimately all are closely integrated and function in unison to regulate the circulation.
Neural control of the heart and vasculature

Central nervous system cardiovascular centres
The critical central nervous system (CNS) cardiovascular control centres are located in
the medulla oblongata and lower pons.2 The centres for circulatory control are in close
proximity to those regulating respiration, and together both comprise the CNS areas crucial
for survival of the organism.
The centres for circulatory control have two major divisions, the vasomotor and the
cardiac regions, which supply nervous innervation to the peripheral vasculature and heart,
respectively.3 4 The two divisions are neither anatomically nor functionally distinct, as
significant overlap in both of these properties exists. Further subdivisions with more
specialised functions have also been demonstrated in animal experiments utilising precise
microelectrode stimulation, electrical ablation, and surgical trans-section techniques. These
subdivisions function in a highly integrated and coordinated manner to control arterial
pressure.
The vasomotor centre is located bilaterally in the reticular substance of the medulla and
lower third of the pons.3 Although the exact organisation of the vasomotor centre has not
been fully defined, experiments have identified certain areas that have specific functions.
The vasoconstrictor area, also termed C-l, is located in the anterolateral portions of the
upper medulla and contains a high concentration of neurons that secrete noradrenaline
(norepinephrine). The C-l area has been proposed to be one of the sites of action of α2-
adrenoceptor agonists such as clonidine and dexmedetomidine. C-l neurons descend in the
spinal cord and synapse with cells in the intermediolateral cell column. The neurons of the
intermediolateral cell column are preganglionic neurons, which synapse
Page 215
further with adrenergic neurons. Adrenergic neurons send vasoconstrictor fibres to the
periphery via the sympathetic nervous system (SNS).
A vasodilator region referred to as area A-l is positioned bilaterally, but more medially,
in the lower half of the medulla.3 Neurons from this area project rostrally to and inhibit
activity of the vasoconstrictor area (C-l), causing vasodilation. Finally, a sensory area,
designated A-2, is located bilaterally in the tractus solitarius in the posterolateral portions of
the medulla and lower pons. This region receives sensory neural input predominantly from
the glossopharyngeal (cranial nerve IX) and vagus (cranial nerve X) nerves. Neurons
arising from A-2 project to the vasoconstrictor and vasodilator areas modulating outputs
from these regions. Thus, sensory area A-2 integrates reflex control for multiple circulatory
functions. The baroreceptor reflex is a typical example.
The cardiac control centre can also be subdivided.5The cardioinhibitory area is well
characterised and positioned in the nucleus ambiguous and the dorsal nucleus of the vagus
nerve. Parasympathetic vagal efferents arising from this area send impulses to decrease
heart rate and, to a lesser extent, reduce atrial contractility. Few parasympathetic fibres,
however, innervate ventricular myocardium. The precise location of the cardiostimulatory
area is less well defined, but appears to be present in the lateral medulla. Stimulation of this
more diffuse region increases heart rate and myocardial contractility via activation of the
SNS.
Both cardiostimulatory and cardioinhibitory areas, as well as the vasoconstrictor area, are
tonically active, and these regions continuously emit low levels of efferent impulses at a
rate of 0·5–2 impulses per second.6 Activity of the vasoconstrictor area is responsible for a
state of partial arteriolar contraction referred to as vasomotor tone. As the areas that control
vasoconstriction and cardiac stimulation lie together laterally and superiorally in the
cardiovascular control centre, these sites constitute the pressor area (Fig. 6.1).4 Those areas
that cause vasodilation and cardiac inhibition lie more medially, and form the depressor
area (Fig. 6.1). Microelectrode stimulation of the pressor area increases heart rate,
myocardiac contractility, and peripheral vascular resistance in experimental animals.
Conversely, stimulation of the depressor area reduces heart rate and arterial pressure.
The cardiovascular control centres receive neural input from other regions within the
brain.7 The reticular substance of the pons, mesencephalon, and diencephalon send both
excitatory and inhibitory impulses into the cardiovascular centres. The hypothalamus also
has significant excitatory and inhibitory control, especially of the vasoconstrictor region.
Finally, numerous areas of the cerebral cortex (for example, motor cortex, anterior temporal
lobe, frontal cortex, anterior cingulate gyms, amygdala, septum, and hippocampus) also
project to the cardiovascular centres and elicit cardiovascular alterations associated with
emotions.
Page 216
Recently, imidazoline receptors have been found to have a distinct and important role in
regulation of the cardiovascular system.8 This receptor system, localised centrally, may be
more specific for haemodynamic control than closely related α2-adrenergic receptors. These
receptors have traditionally been thought to be the key receptor system for central control
of the sympathetic system. Although a specific endogenous imidazoline compound has not
been isolated, the importance of the imidazoline receptors has been further highlighted by a
correlation between mean arterial pressure and concentrations of ‘‘imidazoline like
substances”.8 In the late
Fig 6.1 Schematic diagram of the brain stem of a cat demonstrating locations of the depressor (horizontal
lines) and pressor (cross hatched) areas. (Reprinted with permission from Alexander.4)
Page 217
1980s, specific imidazoline binding sites, insensitive to catecholamines, were isolated.
Imidazoline analogues cause a dose dependent reduction in arterial pressure when injected
into the rostral ventrolateral medulla, whereas clonidine (an α2 agonist known for reduction
in centrally driven sympathetic tone) has minimal effect. Furthermore, research into
imidazoline receptors has shown that the central sedative properties of α2 agonists can be
separated from their beneficial cardiovascular effects. This is an exciting area of
investigation with the potential for revealing mechanisms of essential hypertension, as well
as the development of novel specific antihypertensive drugs.
Autonomic nervous system
The cardiovascular control centres of the brain stem ultimately activate/deactivate the
autonomic nervous system (ANS) which provides innervation of cardiac and vascular
smooth muscle. The ANS represents the efferent or motor component of cardiovascular
control and consists of two complementary divisions: the sympathetic and parasympathetic
nervous systems (SNS and PNS, respectively). The SNS and PNS are commonly
considered physiologically to be antagonistic, producing opposite effects on innervated
tissues.
The SNS and PNS are bipolar, each consisting of two interconnected neurons.9 The first,
proximal neuron originates within the CNS, but does not make direct contact with the
effector organ. This neuron is referred to as the preganglionic neuron and relays impulses
from the CNS to the autonomic ganglion. Autonomic ganglia contain the cell bodies of the
second, distal neuron (the postganglionic neuron), which innervates the effector organ.
Both sympathetic and parasympathetic divisions have preganglionic neurons that are
myelinated, slow conducting type B fibres with diameters of less than 3 μm (Table 6.1).
Impulses are conducted in these fibres at 3–15 m/s. Postganglionic neurons of both
divisions are unmyelinated type C fibres with diameters of less than 2 μm and conduct
impulses at 0.5–2 m/s (Table 6.1).
Sympathetic nervous system

Preganglionic neurons of the SNS arise from the thoracic and first two lumbar (T1–L2)
segments of the spinal cord.10 The cell bodies of the preganglionic neurons are located in
the intermediolateral grey column of the 14 spinal cord segments and have fibres that leave
the spinal cord in the ventral (anterior) motor nerve roots (Fig. 6.2). These fibres pass via
white (myelinated) communicating rami into one of 22 pairs of ganglia which comprise the
paravertebral sympathetic chain. Once in the ganglia of the paravertebral sympathetic
chain, the preganglionic fibre may follow any one of three courses. First, the fibre may
synapse with the cell bodies of the postganglionic neuron in the ganglion at the level of
exit. Second, the fibre
Page 218
Table 6.1 Nerve fibre classification
Fibre type Diameter (μm) Myelin Conduction

velocity (m/s)
Type A
a 12–20 + 120
β 5–12 + 120
γ 3–6 + 5–40
δ 2–5 + 5–40
ε 2 + 5
Type B <3 + 3–15
Type C 0·3–1·2 – 0·5–2
+, myelinated; –, unmyelinated.
Fig 6.2 Schematic diagram of preganglionic fibres of the sympathetic nervous system. Preganglionic fibres
exiting white rami can make synaptic connections in one of three ways: first, synapse can occur in
paravertebral ganglia at the level of exit; second, preganglionic fibres can travel up or down the paravertebral
sympathetic chain and synapse at other levels; third, fibres can exit the paravertebral chain without synapsing
and travel to peripheral collateral ganglia. (Reprinted with permission from Lawson.10)
Page 219
may course upwards or downwards in the paravertebral sympathetic chain and synapse with
postganglionic neurons at other levels. Finally, preganglionic fibres may travel for variable
distances through the paravertebral chain and exit, without synapsing, to more peripheral,
unpaired, collateral sympathetic ganglia where they synapse with postganglionic neurons
(Fig. 6.2).10 Thus, the cell bodies of postganglionic neurons are located either in the ganglia
of the paired paravertebral sympathetic chains or in the more peripheral unpaired collateral
ganglia. In contrast to the PNS, the SNS ganglia are almost always positioned closer to the
spinal cord than the effector organs innervated. The coeliac and inferior mesenteric ganglia
represent examples of unpaired peripheral collateral ganglia that are formed by the
convergence of preganglionic neurons with numerous postganglionic cell bodies. Many
fibres from postganglionic neurons pass back into spinal nerves via grey (unmyelinated)
communicating rami. These neurons subsequently travel with the spinal nerves to innervate
vascular smooth muscle.
The distribution of the SNS neurons to each organ is determined partly by the embryonic
position from which the organ originates.11 This is significant because the first five thoracic
preganglionic neurons ascend into the neck to form three unique paired ganglia. These
ganglia are the superior cervical, middle cervical, and stellate ganglia. The stellate ganglia
are formed by fusion of the inferior cervical and first thoracic SNS ganglia. These three
pairs of ganglia provide sympathetic innervation to the head, neck, upper extremities, heart,
and lungs.
The distribution of the sympathetic system has considerable importance after intrathecal
injection of local anaesthetics. As the level of the block rises above the T6 level, patients
will have a greater degree of hypotension corresponding to the T6–9 sympathetic control of
the mesenteric vessels. Anaesthetic blocks below T10 are associated with lesser degrees of
hypotension, given the smaller number of veins available to sequester intravascular volume.
Further cephalad spread of a spinal anaesthetic block reaching levels above T4 causes
progressive interference with cardiac sympathetic function, resulting in even greater
hypotension with a diminished compensatory reflex tachycardia contributed byTl–T4
fibres.
Parasympathetic nervous system

The cell bodies of the preganglionic neurons of the PNS are located in the brain stern and
the sacral segments of the spinal cord. Cranial nerves III (oculomotor), VII (facial), IX
(glossopharyngeal), and X (vagus) all contain preganglionic PNS neurons.10 The
preganglionic cell bodies of the second, third, and fourth sacral nerves are located in the
intermediolateral grey column of the spinal cord. The paired vagus nerves contain about
75% of all fibres of the PNS, and have the most extensive distribution of any of the
parasympathetic nerves, including innvervation of the heart.
Page 220
Preganglionic neurons of the PNS are quite different from the analogous preganglionic
nerves in the SNS. PNS preganglionic fibres are longer and pass uninterrupted to ganglia
near or in the innervated organ, which are generally not visible. As a consequence, the
postganglionic neurons of the PNS are short because of the location of the corresponding
ganglion. The proximity of the PNS ganglia to the effector organs limits the distribution of
the postganglionic neurons, in contrast to postganglionic SNS fibres which are more
widespread.
All preganglionic neurons of either division of the ANS release the neurotransmitter
acetylcholine and are classified as cholinergic. The neurotransmitter released from
postganglionic neurons in the PNS is acetylcholine. The neurotransmitter between
postganglionic adrenergic neurons and effector organs in the SNS is noradrenaline
(norepinephrine). Several types of non-adrenergic, non-cholinergic nerves have also been
described in the past 20 years, revealing cardiovascular control mechanisms to have
substantially greater levels of complexity.12
ANS effects on the circulation

The heart and peripheral vasculature are innervated by both the SNS and PNS.
Alterations in heart rate (chronotropism), the strength of contraction (inotropism), and
coronary blood flow are produced by both divisions of the ANS (Table 6.2). Cardiac vagal
fibres are primarily distributed to the sinoatrial and atrioventricular nodes and to the atria,
but are only minimally distributed to the ventricular myocardium. PNS stimulation leads to
decreases in the rate of sinoatrial node discharge and reduces atrioventricular node
excitability, causing slowing of impulse conduction to the ventricles via muscarinic
receptors. Strong vagal discharge may cause complete sinoatrial node arrest or interrupt
impulse conduction from the
Table 6.2 Effects of adrenergic and cholinergic stimulation on the cardiovascular
system
Adrenergic response Cholinergic response
Heart
Sinoatrial node Tachycardia Bradycardia
Atrioventricular node Increased conduction Decreased conduction
His–Purkinje system Increased automaticity Minimal
and conduction velocity
Myocardium Increased contractility Minimal
Vasculature
Skin and mucosa Constriction Dilatation
Skeletal muscle Constriction (α)> Dilatation
dilatation (β2)
Coronary Constriction (α1) and Dilatation (EDRF) and
dilatation (β2) constriction
Pulmonary Constriction Dilatation (?)
EDRF, endothelium derived relaxing factor (NO).

Page 221
atria to the ventricles. The PNS has little effect on myocardial contractility because of the
lack of vagal efferent distribution to the ventricles. Any decrease in contractile force after
PNS stimulation is most probably secondary to declines in heart rate. The human heart is
tonically stimulated by both the PNS and SNS but vagal tone predominates. The degree of
vagal tone is greatest in young individuals. Total pharmacological blockade of the ANS or
cardiac denervation (for example, heart transplantation) results in higher resting heart rates
by inhibition of this dominant vagal tone.
The SNS has the same supraventricular distribution as the PNS, but it also provides
greater innervation of the ventricular myocardium.13 Postganglionic fibres of the SNS arise
in the paired stellate ganglia. The right stellate ganglion sends fibres primarily to the
anterior epicardial surface and interventricular septum of the heart. Stimulation of these
fibres leads to increases in heart rate via activation of β1-adrenergic receptors. The left
stellate ganglion distributes fibres to the lateral and posterior surfaces of both ventricles.
Left stellate ganglion stimulation increases ventricular contractility while causing only
minimal increases in heart rate. Experimental investigations have revealed that normal
basal sympathetic tone maintains cardiac contractility at approximately 20% greater levels
than the denervated heart.
SNS activation causes coronary arteriolar vasodilation indirectly through increases in
myocardial oxygen consumption and coupling of myocardial
metabolic requirements to blood flow.14 Simultaneous α-adrenergic activation,
however, causes large epicardial coronary artery vasoconstriction, limiting metabolically
driven coronary vasodilation by about 30%. Parasympathetic activation causes coronary
vasodilation in experimental animal models because of an acetylcholine mediated release of
nitric oxide (endothelium derived relaxing factor). The direct stimulation of human
coronary vascular smooth muscle by PNS, in the absence of a normally functioning
vascular endothelium, may produce constriction leading to coronary spasm.
The peripheral vasculature is innervated by both divisions of the ANS, but the SNS has
far greater importance in the regulation of vascular tone.15 The distribution of
parasympathetic nerves is relatively limited and PNS stimulation dilates vessels partially
via endothelial mechanisms. The SNS causes vasoconstriction by stimulation of α-
adrenergic receptors. The vasculature of the skin, kidneys, spleen, and mesentery has
extensive sympathetic innervation. Vascular beds of the brain, heart, and muscle have
significantly less innervation.
Vascular tone is the sum of the muscular forces intrinsic to the blood vessel wall,
opposing an increase in vessel diameter.6 The degree of tone is influenced by ANS activity,
humoral and local metabolic substances, and autacoids with dilator or constrictor actions.
Basal vasomotor tone results partially from low level, continuous impulses (0·5–2 impulses
per second)
Page 222
from the lateral portion of the vasomotor centre in the medulla oblongata that pass via the
SNS to maintain partial arteriolar and venular constriction. Circulating adrenaline from the
adrenal medulla may add to regional constriction dependent on the relative concentration of
α-constrictor versus β2-vasodilator receptors. Increasingly, experimental evidence points to
a larger role of chemical mediators in the control of basal vasomotor tone.
Basal vasomotor tone is normally maintained at approximately half-maximal
constriction, enabling the arteriole to further constrict or dilate. Without basal vasomotor
tone, the arteriole could only constrict during SNS stimulation. Thus, vasodilation can be
produced by a reduction in the tonic sympathetic nerve activity without elicitation of
“opposing” PNS. Basal vasomotor tone produces little resistance to flow in the venule
compared with the arteriole. The importance of the degree of SNS stimulation on the
venous circulation is, however, to increase or reduce venous capacitance. A small change in
venous capacitance can produce large alterations in venous return and cardiac preload
because up to 80% of the total blood volume can be stored in the veins.
Reflex control of the circulation
Intrinsic reflexes
Cardiovascular reflexes represent rapidly acting mechanisms to control the circulation
using the central and autonomic nervous systems.16 Reflex control of the circulation can be
initiated either from within the cardiovascular system (intrinsic reflexes) or from other
organs or systems (extrinsic reflexes). Intrinsic reflexes are the most important short term
regulators of arterial pressure. These reflexes are produced by changes in arterial pressure
or special chemical stimuli. Alterations in blood pressure are sensed by stretch receptors,
pressoreceptors, baroreceptors, or mechanoreceptors. Chemoreceptors are sensitive to
chemical stimuli, regulate respiration, and, secondarily, also influence the circulation.
Arterial baroreceptor reflexes – Arterial baroreceptors are specialised, pressure sensitive,
nerve endings in walls of the aortic arch and internal carotid arteries just above the carotid
bifurcation (carotid sinus) (Fig. 6.3). Afferent fibres from the baroreceptors travel in the
aortic and carotid sinus nerves, which join the vagus and glossopharyngeal nerves,
respectively, and connect with the cardiovascular centres in the medulla, commonly in the
nucleus tractus solitarius. Cells from the nucleus tractus solitarius project to the C-l area
and inhibit this region by secretion of γ-aminobutyric acid (GABA). There is a small tonic
discharge from baroreceptor afferents at a normal arterial pressure. When the baroreceptor
endings are stretched, action potentials are generated and propagated at a frequency that is
approximately proportional to the pressure change in the artery. Hence, increased arterial
pressure sensed by the baroreceptors will increase the
Page 223
frequency of impulses travelling to the CNS. Afferent input produces greater activity in the
medullary depressor area and inhibits the pressor and cardiac areas, causing decreases in
myocardial contractility and heart rate, and reducing vasoconstrictor tone of both arterioles
and veins. Therefore, increased blood pressure leads to reflex activity aimed at reducing the
pressure back to a normal set point (the depressor reflex). The opposite effect for declines
in blood pressure is also true (the pressor reflex). The arterial baroreceptor reflex provides a
negative feedback mechanism for homoeostasis of arterial pressure.16
The baroreceptor reflex plays an important role in rapid control of arterial pressure, for
example when rising from a recumbent position. The baroreceptor response can be
experimentally elicited during sudden decreases or increases in arterial pressure produced
by intravenous administration of sodium nitroprusside or phenylephrine, respectively
(Smyth’s procedure) (Fig. 6.4). The slope of the plot of heart rate (or R–R interval) versus
systolic pressure during rapid changes in pressure obtained
Fig 6.3 Diagram of the central connections of the aortic and carotid sinus baroreceptors. (Reprinted with
permission from Smith and Kampine.7)
Page 224
Fig 6.4 Baroreceptor reflex in a human subject. (a) A recording of the baroreceptor response after intravenous
infusion of sodium nitroprusside. Note the increase in heart rate and sympathetic nerve activity in response to
the decline in blood pressure. (b) The baroreceptor mediated decrease in heart rate and sympathetic nerve
activity after an intravenous infusion of phenylephrine. (Reprinted with permission from Ebert.17)
by this technique is a measure of the “sensitivity” of the baroreflex.17 It has been proposed
that such sensitivity may be reduced in patients prone to sudden death. Clamping of the
common carotid artery or damage to the carotid sinus nerve during carotid endarterectomy
surgery may elicit dramatic haemodynamic changes through alterations in the arterial
baroreflex.
Arterial baroreceptors respond most effectively to the rate of change of arterial pressure.
The response is greatest to changes of arterial pressure in the physiological range (80–150
mm Hg). Several subcategories of baroreceptors exist that respond to different pressure
ranges. Arterial baroreceptors respond more actively to declines in arterial pressure than
increases. Evidence suggests that the carotid baroreceptors are more sensitive to pressure
changes than the aortic baroreceptors and operate at lower ranges of arterial pressure.
Experimental investigations have determined that arterial baroreceptors primarily influence
reflex control of cardiac rate and contractility. Control of systemic vascular resistance is of
secondary importance. The vessels most influenced by arterial baroreceptors are splanchnic
arterioles and venules, with the venules being important in increasing venous return to the
heart during the pressor reflex. Evidence also suggests that the arterial baroreceptors
participate in stimulating the renin–angiotensin system via an increase in sympathetic
tone.18 Decreased baroreceptor responsiveness occurs with advancing age, hypertension,
and coronary artery disease.
An important property of arterial baroreceptors is the ability to adapt to prolonged
changes in arterial pressure. Arterial baroreceptors continue to function in hypertensive
individuals or even during acute hypertensive episodes or exercise, but reset at a higher
blood pressure range.19 The higher
Page 225
pressure ultimately forms a new baseline range, but can be reversed if the increase in
pressure is relieved. This resetting of the baroreceptor range demonstrates that these
reflexes probably have no role in long term blood pressure regulation but, instead, respond
only to acute changes in arterial pressure. In fact, resetting can be demonstrated within
minutes to hours. The efferent portion of the baroreceptor reflex arc is blocked by many
drugs used for the treatment of hypertension. As a result, orthostatic hypotension
commonly occurs and, if severe, may lead to syncope. This condition may also be present
secondary to pathological processes such as diabetes. Such “autonomic insufficiency” can
lead to haemodynamic instability especially during anaesthesia.
Atrial and vena caval low pressure baroreceptors – The right and left atria and inferior
and superior vena cava near the junction with the right atrium also contain specialised low
pressure mechanoreceptors that respond to increases in central venous pressure.7 These
baroreceptors respond to pressure change but in a much lower range that arterial
baroreceptors. The low pressure baroreceptors send impulses via large myelinated fibres in
the vagus nerves to the CNS when the atria or vena cava are distended. The efferent portion
of the reflex consists of SNS fibres to the sinoatrial node and subsequently causes
tachycardia. This increase in heart rate caused by atrial stretch is known as the Bainbridge
reflex and is abolished by vagotomy. Although the Bainbridge reflex has been observed in
numerous species, the heart rate response to atrial filling in humans is complicated by
numerous other factors, including the dominant arterial baroreflex, so this reflex probably
plays only a secondary role.
Other baroreceptors have been isolated that are also stimulated by filling and distension
of the atria but send impulses via unmyelinated vagal fibres to the CNS.11 The reflex heart
rate response is the opposite of that which occurs in the Bainbridge reflex, and the overall
response is analogous to that of the arterial baroreceptors. An increase in venous return
increases and positive pressure ventilation reduces discharge from the receptors. Arterial
distension results in decreases in SNS activity, causing a decline in vasoconstriction of
skeletal muscle, renal, and mesenteric arterioles, an increase in splanchnic venous
capacitance, and a reduction in heart rate. The decrease in SNS activity is accompanied by a
decline in renin secretion. Reduced circulating angiotension and aldosterone also decrease
arteriolar vasoconstriction and lead to a diminution in plasma volume.
As with arterial baroreceptors, atrial baroreceptors adapt to a continuous increase in
pressure by resetting.7 This has been demonstrated in experimental animals where
congestive heart failure leads to prolonged atrial distension, increased atrial pressure and
attenuation of this reflex. This adaptation is reversed if the heart failure is relieved. It is also
important to note that these atrial baroreceptors are stimulated not only by stretch but also
by atrial muscle contraction. Hence, it is believed that the
Page 226
diuresis observed in certain clinical and experimental pathological conditions, such as
paroxysmal atrial tachycardia and atrial fibrillation, may be the result of the unusual
contractile activity in the atrial wall. Atrial natriuretic peptide may also play an important
role. The low pressure baroreceptors are significant in the control of extracellular fluid
volume. When volume is reduced, these receptors cause a reflex release of vasopressin and
enhanced sympathetic tone, which activates the renin–angiotensin–aldosterone axis. In
addition, SNS reflex constriction of the afferent arterioles occurs to conserve intravascular
volume.15These actions increase volume and homoeostasis is maintained.
Ventricular reflexes – The ventricles contain receptors that are also stimulated by stretch
or by strong ventricular contraction.7 These receptors provide afferent input to the medulla
via unmyelinated vagal fibres. The medulla responds by decreasing sympathetic tone and
causing bradycardia and vasodilation. A very similar reflex response (that is, bradycardia
and vasodilation accompanied by apnoea) can be elicited by injecting the drug, veratridine,
into the heart or coronary circulation (especially the left circumflex perfusion territory in
canine experiments). The unusual coronary chemoreceptor response is referred to as the
Bezold–Jarisch reflex. Investigation suggests that this reflex may also be triggered by
intracoronary injections of other pharmacological agents, including serotonin, capsaicin,
nicotine, bradykinin, histamine and digitalis. A similar reflex has been noted after injection
of contrast media during coronary angiography and in certain pathological conditions when
specific metabolites accumulate in the coronary circulation, for example, in myocardial
necrosis. It has been proposed that the coronary chemoreceptor reflex may be elicited
during inferior wall myocardial infarction.
Arterial chemoreceptors – There are also arterial chemoreceptors located in the carotid
and aortic bodies, small masses of tissue lying in close proximity to the carotid sinus and
the aortic arch receptors (see Fig. 6.3), which have prominent effects on respiration and the
circulation. The carotid body is the major chemoreceptor. The special nerve endings
respond to decreases in the arterial partial pressure of oxygen (PaO2), increases in the
arterial partial pressure carbon dioxide (PaCO2), and increases in arterial hydrogen ion
concentration. The afferent pathway is located in the same nerves as the adjacent
baroreceptors. Arterial chemoreceptors serve primarily to cause an increase in respiratory
minute volume, but secondarily these receptors produce sympathetic vasoconstriction
during hypotension. This response is additive to that produced by arterial baroreceptors.
The ‘‘secondary” circulatory reflex actions improve oxygen delivery to heart and brain
through generalised peripheral vasoconstriction and increased arterial pressure. The
increased arterial pressure occurs during hypotension secondary to severe depletion of
intravascular volume and subsequent reduction in blood flow and ischaemia of the carotid
and aortic bodies. The
Page 227
chemoreceptor response also contributes to formation of Mayer waves during recording of
arterial pressure. Decreases in perfusion of the chemoreceptors during hypotension causes
activation of the reflex and results in increases in arterial pressure. Increases in flow then
deactivate the chemoreceptors and declines in arterial pressure occur. The repetitive
cyclisation leads to large swings in pressure (Mayer waves) at a frequency of 2–3
cycles/min.
Extrinsic reflexes
Receptors of the afferent limbs of extrinsic reflex arcs are external to the circulatory
system. These reflexes are less consistent than intrinsic reflexes and, in normal
circumstances, play only a minor role in circulatory control. On the other hand, extrinsic
reflexes are important and protective during certain types of environmental stresses and
pathophysiological circulatory states. Afferent impulses enter the CNS via somatic nerves
but the central processing of these reflexes is still uncertain. Examples of extrinsic reflexes
include pain and cold, oculocardiac, CNS ischaemic, and Cushing reflexes.
Pain reflex – Pain, depending on severity, produces variable haemodynamic responses.
Mild to moderate pain results in tachycardia and increases in arterial pressure mediated by
the somatosympathetic reflex. This is a common finding in the postoperative period if
analgesia is inadequate. Severe pain, as experienced by deep bone trauma or stretching of
abdominal or perineal viscera, may elicit bradycardia, hypotension, and, at times,
circulatory collapse and syncope.
Cold reflex – Cutaneous thermosensitive nerve endings respond to cold temperature and
send impulses through somatic afferent fibres to the hypothalamus. This results in
cutaneous vasoconstriction and piloerection. An example of this reflex is the cold pressor
test in which application of intense local cold, such as immersion of a hand in ice water,
leads to stimulation of both pain and cold receptors with a subsequent increase in arterial
pressure. In certain patients with coronary artery disease, the cold pressor test can produce
angina by either reflex coronary vasoconstriction or abruptly increased left ventricular
afterload.
Oculocardiac reflex – Receptors stimulated by pressure or stretch in the extraocular
muscles, conjunctiva, and globe send impulses through the ophthalmic division of the
trigeminal nerve (cranial nerve V) to the CNS.20 This leads to bradycardia and hypertension
and possibly to more severe cardiac arrhythmias including asystole. This reflex does have a
tendency to fatigue with repeated stimulation, most probably at the level of the
cardioinhibitory centre.
CNS ischaemic reflex – The CNS ischaemic response occurs when severe hypotension
(as in circulatory shock) reduces perfusion and causes hypoxia of the medullary vasomotor
centre.21 Chemoreceptors in the vasomotor
Page 228
centre sense local increases in PCO2 and decreases in pH. As a result, there is an intense
increase in SNS activity leading to a profound and generalised vasoconstriction.
Simultaneously, an increase in PNS activity reduces heart rate. This reflex does not become
active until mean arterial pressure decreases below 50 mm Hg (6·7 kPa) and is maximal at
mean pressures of 15–20 mm Hg (2·0–2·7 kPa). The CNS ischaemic response does not
participate in regulation of normal arterial pressure but is an emergency control system to
restore cerebral blood flow when it is dangerously reduced.
The Cushing reflex – The Cushing reflex is another reflex that has an origin directly in
the CNS. When intracranial pressure is acutely elevated, cerebral vessels are compressed.
The decrease in cerebral perfusion pressure causes a reduction in arterial blood flow, and
the CNS ischaemic response occurs. The decrease in blood flow to the vasomotor area
results in an increase in SNS activity. This leads to progressive elevations in arterial
pressure in an effort to exceed intracranial pressure and maintain adequate cerebral
perfusion. Simultaneously, decreases in heart rate are observed which are mediated by the
baroreceptor reflex.
Spinal shock and autonomic hyperreflexia

When the spinal cord is acutely trans-sected, all cord functions, including reflexes
mediated through the CNS, are interrupted below the level of injury.20 Vasoconstrictor tone
regulated by the ANS is disrupted below the cord lesion and, if the spinal cord injury occurs
at a high level (that is, C6–C7), spinal shock may result. Spinal shock is characterised by a
fall in blood pressure as a result of lack of arteriolar and venular constriction as well as a
loss of skeletal muscle pump action. The efferent limb of cardiovascular reflexes cannot
compensate for the reduction in arterial pressure. This condition is analogous to the loss of
control of blood pressure during spinal anaesthesia. Spinal shock typically lasts from one to
three weeks.
After this period, local reflexes in the spinal cord below the level of disruption gradually
return and a chronic stage, characterised by SNS overactivity or autonomic hyperreflexia,
begins. Hyperreflexia occurs in response to cutaneous or visceral stimulation below the
level of cord lesion in 85% of patients with spinal cord injury at or above the T6 level. The
stimulus sends afferent impulses into the spinal cord, causing local activation of
preganglionic sympathetic nerves and subsequent vasoconstriction. This reflex is normally
modulated by inhibitory impulses from higher centres, but as a result of the trans-section,
such impulses are blocked. The vasoconstriction causes hypertension and elicits the
baroreceptor depressor reflex. The reflex mediated decline in SNS activity and increase in
PNS activity occur only above the level of the cord trans-section, whereas vasoconstriction
continues below this level. With high levels of
Page 229
trans-section, circulatory reflexes are insufficient to offset the effects of vasoconstriction,
leading to persistent and sometimes severe hypertension.
Humoral control of the circulation

Catecholamines
The adrenal medulla is unique in that this gland is innervated by preganglionic SNS
fibres which pass directly to it from the spinal cord.10 22 Embryologically, cells of the
adrenal medulla are derived from neural tissue and are analogous to postganglionic
neurons. The adrenal medulla secretes primarily adrenaline (epinephrine) (80%) and also
noradrenaline (nor-epinephrine) in response to SNS stimulation. As adrenaline and
noradrenaline are released into the blood and exert functions at distal sites, these
catecholamines function as hormones.
The release of catecholamines by the adrenal medulla occurs after acetylcholine is
secreted by preganglionic SNS fibres. The cardiovascular response to the secreted
adrenaline and noradrenaline is similar to direct stimulation by the SNS. The effects of
those hormones are, however, significantly prolonged (10–30 seconds) compared with the
duration of action of noradrenaline as a neurotransmitter. Adrenal medullary secretion may
be considered to be additive to that of SNS stimulation in that some vessels that are poorly
innervated are also constricted by circulating catecholamines.
Renin–angiotensin system
The renin–angiotensin system is another important humoral regulator of the

cardiovascular system, particularly under conditions of stress.18 23 The enzyme renin is
synthesised in and released from juxtaglomerular cells of the renal cortex. Juxtaglomerular
cells are modified vascular smooth muscle cells located in the tunica media of the afferent
arteriole immediately proximal to the glomerulus. Renin is secreted into the blood in
response to a decrease in renal artery pressure, reduced sodium delivery to the distal tubule
(sensed by osmoreceptors of the macula densa region) and SNS stimulation via activation
of β1-adrenergic receptors. Renin cleaves the hepatically synthesised α2-globulin,
angiotensinogen, to form the decapeptide, angiotensin I. Angiotensin I is physiologically
inactive but is rapidly hydrolysed to form the octapeptide angiotensin II by angiotensin
converting enzyme which is found in high concentrations in the vascular endothelium of the
lungs.
Angiotensin II produces vasoconstriction of arterioles in most vascular beds. It also
stimulates SNS ganglion cells and facilitates impulse transmission in the SNS. Activation
of presynaptic angiotensin receptors
Page 230
increases noradrenaline (norepinephrine) release. Angiotensin II directly stimulates the
adrenal cortex to synthesise and secrete aldosterone. Aldosterone causes salt and water
retention leading to expansion of plasma volume which further increases arterial pressure.
Decreases in blood pressure or declines in sodium delivery to the macula densa (for
example, haemorrhage, dehydration) lead to formation of angiotensin II. Angiotensin II
increases arterial pressure by elevating the peripheral vascular resistance and by causing
aldosterone secretion with subsequent salt and water retention. The renin–angiotensin–
aldosterone axis system is of major importance in maintaining blood pressure, especially
during periods of hypovolaemia, sodium deprivation, or inadequate cardiac output. Also,
this system has been found to play a role in basal vascular tone as well as having an
interplay with other regulatory systems such as the kinins.24, 25
Vasopressin
Vasopresson or antidiuretic hormone is a peptide synthesised in the supraoptic and

paraventricular nuclei of the brain stem, which is transported to and released from the
posterior pituitary gland.26 Vasopressin secretion from the pituitary occurs in response to
multiple physiological stimuli, including: an increase in plasma osmolality sensed by
osmoreceptors located in the hypothalamus; a decrease in plasma volume detected by
cardiopulmonary receptors, particularly in the left atrium; and increased plasma
concentrations of angiotensin II. Vasopressin binds to specific receptors which cause the
collecting ducts of the kidney to increase free water reabsorption (inhibit diuresis). An
increase in plasma volume occurs, which has a negative feedback on further Vasopressin
release.
The vasoconstriction produced by vasopressin is generalised, affecting most regional
circulations. Skin and gastrointestinal tract arteries are markedly constricted, whereas large
epicardial coronary arteries and pulmonary arteries are constricted to a lesser extent. Severe
and preferential constriction of small coronary collaterals can result in areas of regional
myocardial ischaemia. Large quantities of vasopressin are released by the posterior
pituitary gland during haemorrhage, and the constrictor action exerts an adjunctive pressor
effect to that produced by baroreceptor reflexes.
Endothelium derived nitric oxide
Nitric oxide (NO) produced by vascular endothelium (formerly known as endothelium

derived relaxing factor or EDRF) is a cell messenger that has important physiological and
pathophysiological actions.27 28 The primary effect of NO on the circulation is to relax
vascular smooth muscle. A variety of endogenous mediators and pharmacological agents
exerts
Page 231
haemodynamic effects by causing a release of NO from vascular endothelium. For
example, acethylcholine will produce vasodilation in the presence of an intact endothelium
in isolated vascular ring preparations. In contrast, acetylcholine causes vasoconstriction
after endothelial denudation.
Nitric oxide is synthesised by at least two major NO synthase (NOS) isoforms. One is
expressed constitutively in neurons and vasculature, and requires calcium and calmodulin
binding for activation. This constitutive enzyme is involved in cell communication and is
activated by an increase in intracellular calcium. The enzyme is in a soluble form in neural
tissue, but it is membrane-bound in vascular endothelium. The other isoenzyme is
expressed after induction by cytokines or endotoxin and participates in host defence. This
inducible isoform has calmodulin bound as a subunit and produces NO continuously
without requiring calcium. This isoenzyme may contribute to the pathophysiology
associated with syndromes characterised by an overproduction of cytokines such as septic
shock. It is present in macrophages, but is not normally found in endothelial cells or
vascular smooth muscle unless induced by cytokines. The NOSs are mixed function mono-
oxygenases which use NADPH. L-Arginine is oxidised in a stepwise manner to form NO
and citrulline as primary products. NO synthases can be specifically and competitively
inhibited by L-arginine analogues (for example, Ng-nitro-L-argenine methylester [L-
NAME] and Ng-mono-methyl-L-arginine [L-NMMA]). The inhibitors have been used to
elucidate the physiological and pathophysiological roles of NO. Administration of
inhibitors of the synthesis of NO to experimental animals leads to considerable increases in
arterial pressure, substantiating the premise that a basal release of NO normally occurs and
that defects in NO synthesis may play a role in the aetiology of hypertension. The same
inhibitors may also be useful in increasing arterial pressure in septic shock. The primary
biological function of NO appears to be the activation of soluble guanylyl cyclase, which
subsequently increases the cyclic guanosine 3’:5’-monophosphate (cGMP) content of
several tissues including vascular smooth muscle. The cGMP may cause relaxation of
vascular smooth muscle by several mechanisms, including the activation of cGMP
dependent protein kinase, which in turn activates a calcium ATPase that causes extrusion of
calcium from the cell or improves calcium uptake into the sarcoplasmic reticulum.
Nitric oxide plays an important and diverse role in cardiovascular regulation.
Physiologically, it has significant roles in the maintenance of basal vascular tone and the
modulation of platelet–vessel wall interactions. Conversely, an overproduction or
underproduction of NO also has pathophysiological implications in hypotension associated
within certain shock states, essential hypertension, and atherosclerosis. The release of NO
may stem from stimuli such as intracellular calcium, changes in oxygen tension, vessel wall
sheer stress, and other humoral mediators of vascular tone. NOS inhibitors may provide
important avenues for the control of
Page 232
certain hypotensive states, including septic shock where traditional therapies including
administration of catecholamines have limited effectiveness. Furthermore, return of
catecholamine sensitivity after the use of NOS inhibitors in patients with septic shock have
been described. NO may also serve as the active moiety of certain antihypertensive agents
that act to donate it.29 L-Arginine may be another important agent that serves to control
hypertension by being converted to NO via NOS.30
Atrial natriuretic peptide
Atrial natriuretic factor (ANP) is a peptide synthesised and stored in human atrial
myocytes and secreted in response to distension of the atria (increased vascular volume or
increased atrial pressure), adrenaline (epinephrine), vasopressin, or morphine.31 Studies
have determined that there is a larger amount of ANP synthesised and stored in the right
than in the left atrium. ANP acts directly on the arterial and venous vasculature and kidneys
to reduce arterial pressure and intravascular volume. It decreases blood pressure by relaxing
vascular smooth muscle and sympathetic tone. ANP inhibits renin release and aldosterone
secretion, thereby interfering with sodium retention and leading to natriuresis. Further renal
effects include dilatation of afferent arterioles and possible constriction of efferent
arterioles of the glomerulus, leading to increased filtration fraction and diuresis.
Controversy exists about whether ANP inhibits sodium uptake in the inner medullary
collecting ducts causing further natriuresis. ANP suppresses antidiuretic hormone secretion.
No direct inotropic or chronotropic effects caused by this peptide have been observed, but
some evidence indicates that at least a portion of the vasodilator action of ANP is mediated
by NO.
Other humoral substances
There are several other endogenous substances that can be released into the circulation
and affect the heart and vasculature. These include adenosine, histamine, the plasma kinins
(kallidin and bradykinin), serotonin, and endothelins.
Adenosine is a ubiquitous endogenous nucleotide that stimulates A1-, A2-, and A3-
receptors to produce haemodynamic actions.11 It has inhibitory effects on cardiac impulse
conduction through the atrioventricular node (negative dromotropic effect). Adenosine is
also a potent vasodilator. The local regulation of flow in the coronary circulation and other
regions during an increase in oxygen demand is at least partially related to the release of
adenosine.22
Histamine is one of several naturally occurring endogenous substances collectively
referred to as autacoids.7 Other autacoids include prosta-
Page 233
glandins, angiotensin II, serotonin, and plasma kinins. Histamine is located in mass cells in
the lungs, skin, and gastrointestinal tract and in basophils throughout the blood. The
predominant circulatory effects of histamine are caused by dilatation of arterioles and
capillaries. Histamine causes vasodilation by binding to H1- and H2-receptors directly on
blood vessels. Histamine induced vasodilation leads to flushing, decreased systemic
vascular resistance, reduced arterial pressure, and increased capillary permeability. This
autacoid also has positive inotropic properties by directly stimulating cardiac H2-receptors
and indirectly by causing adrenaline (epinephrine) release from the adrenal medulla.
Stimulation of cardiac H2-receptors also causes positive chronotropic effects. Finally,
coronary arteries have been shown to be vasoconstricted as a result of H1-receptor
stimulation and vasodilated after H2-receptor stimulation.32
Plasma kinins are among the most potent endogenous vasodilators known.33 Two
examples of plasma kinins are kallidin and bradykinin, which are polypeptides formed by
cleavage of α2-globulin kininogens by kallikrein enzymes. Plasma kinins are approximately
10 times more potent as vasodilators than histamine. Kinins also cause increased capillary
permeability and tissue oedema. vasodilation leads to a marked reduction in arterial
pressure. In contrast, plasma kinins have been shown to constrict large veins, which
subsequently elevates venous return. This increase in venous return increases stroke volume
and cardiac output. Many of the actions of bradykinin are mediated by release of NO from
vascular endothelium.
Kinins have been found to be increasingly important and may provide new pathways for
understanding cardiovascular control mechanisms and therapies. These entities are clearly
connected to the renin–angiotensin system. There is an inverse relationship between plasma
kinin levels and angiotensin converting enzyme (ACE) and renin, respectively.25 The renin–
angiotensin system is linked to the kinins via ACE. This enzyme metabolises kinins to
inactive products in addition to activating angiotensin II.
Serotonin (5-hydroxytryptamine) is also an endogenous vasoactive autacoid synthesised
from tryptophan.34 About 90% of endogenous serotonin exists in enterochromaffin cells of
the gastrointestinal tract and the remainder in the CNS and platelets. The circulatory actions
of serotonin are dependent on the specific vascular bed. It produces vasodilation in blood
vessels in skeletal muscles and skin, while causing vasoconstriction particularly in
splanchnic and renal vessels and, to a lesser extent, in cerebral and pulmonary vessels. It is
also a potent venoconstrictor, but the resulting effects of increased venous return on cardiac
output may be obscured by baroreceptor responses mediated by reflexes.
Within the past decade, a very important peptide system (endothelins) has been
uncovered not only for its properties in the control of vascular tone but also for growth and
development of the vascular system. Endothelins
Page 234
are possibly the most potent of all endogenous vasoconstrictors. These polypeptides may be
involved in basal vascular tone and blood pressure in normal physiological states.
Endothelins have a short half life and are therefore felt to have primarily a local action. The
role of this system in human hypertension is unclear and difficult to ascertain given that
plasma levels may not correlate accurately with their local effects. Finally, a number of
endothelin antagonists are presently being developed as alternative antihypertensive agents
or new drugs for the treatment of congestive heart failure.35
Local regulation of the circulation

Pressure autoregulation
Flow is locally controlled in certain vascular beds by a process termed “autoregulation”36

(Fig. 6.5). By definition, autoregulation is the ability of an organ to maintain a relatively
constant blood flow in the presence of changes in arterial perfusion pressure. The kidneys,
brain, and heart are organs that exhibit autoregulation, whereas the skin and lungs are
organs with minimal ability for autoregulation.
Fig 6.5 Pressure autoregulation in the normal coronary circulation. Note that during maximum vasodilation,
autoregulation is abolished (dashed line) and pressure–flow relationships resemble those of a rigid/elastic
tube. (Reprinted with permission from Goldberg and Warltier.37)
Page 235
A schematic diagram of the changes in blood flow to an organ over a wide range of
perfusion pressures in the presence or absence of autoregulation is shown in Fig. 6.5. In the
absence of autoregulation, pressure–flow relationships are linear, and increases in driving
pressure lead to direct increases in perfusion. An autoregulatory curve is characterised by a
large range of pressures during which flow remains relatively constant. vasodilation is
achieved at lower perfusion pressures by relaxation of smooth muscle, and vascular smooth
muscle constriction occurs at higher perfusion pressures to maintain constant flow. The
ability to autoregulate assumes that a set point of basal vasomotor tone allows for this
dilatation or constriction to occur. Flow varies directly with pressure when the limits of
autoregulation are exceeded. In regional beds that are maximally vasodilated, for example
by a drug such as dipyridamole, the process of autoregulation is eliminated and flow is
directly dependent on driving pressure (Fig. 6.5).37 Autoregulation of blood flow is affected
to only a small extent by neural and humoral influences. Experimental investigations have
demonstrated that the ability to autoregulate flow is largely an intrinsic property and even
occurs in denervated tissues. As a result of this, autoregulation is considered to be a local
phenomenon affected primarily by the active tone of arterioles.
Two major theories have been advanced to explain the mechanism of autoregulation.36
The first of these is the myogenic theory which suggests that elevations in perfusion
pressure lead to stretch and increases in tension of vascular smooth muscle cells. The
distension of the smooth muscle directly causes vasoconstriction to maintain flow constant
despite an increased driving pressure. Conversely, at low perfusion pressures, there is less
muscle tension, vascular smooth muscle cells relax, and blood flow is maintained despite
the decrease in pressure. Therefore, the degree of tension smooth muscle is exposed to is
the stimulus for regulation, and subsequent constriction/dilatation serves as the mediator of
this mechanism. Autoregulation is advantageous to vital organs such as the brain, heart, and
kidneys because blood flow is optimised even during periods of hypo- and hypertension. It
has also been proposed that the myogenic mechanism protects capillaries from excessively
high blood pressures which could cause these fragile vessels to rupture.
The metabolic theory of autoregulation is based on the state of oxygenation in the
surrounding tissues. A reduction in perfusion pressure leads to a decrease in blood flow and
tissue PO2 with concomitant increases in tissue PCO2 and other metabolites related to
normal cellular activity (for example, lactic acid, adenosine, K+, and H+). The reduced PO2,
increased PCO2, and vasodilator substances directly cause arteriolar relaxation, and blood
flow increases. High perfusion pressures supply ample O2, remove CO2, and wash out
vasodilators leading to vasoconstriction and decreases in tissue blood flow. Regardless of
the validity of either the myogenic or the
Page 236
metabolic theory, both are based on the premise that autoregulation of tissue flow is a
negative feedback mechanism, maintaining constancy of arterial flow during large changes
in perfusion pressure.
Matching of flow and metabolism
When tissues have greater metabolic activity, such as skeletal muscle during exercise or
myocardium during increases in heart rate, arterioles in the tissue dilate and flow increases
independently of changes in perfusion pressure. This process is referred to as active or
functional hyperaemia.19 The mechanism of functional hyperaemia is very similar to the
metabolic theory of autoregulation. As the requirement of tissues for O2 increases, there is a
concomitant accumulation of byproducts of metabolism. For example, as more ATP is
consumed during periods of increased metabolic activity, adenosine is released into the
interstitium. The accumulation of this and other vasodilator substances leads to smooth
muscle relaxation in arterioles with increases in tissue perfusion. This represents a powerful
regulator of flow in which perfusion is closely coupled with metabolic demands.
Investigations have demonstrated that, during intense exercise, active hyperaemia can
increase skeletal muscle blood flow as much as 20-fold.
Reactive hyperaemia
When blood flow is interrupted by an arterial occlusion for a brief period and then
suddenly restored, the resulting flow greatly exceeds previous levels for a short time before
returning to the usual resting levels (Fig. 6.6). The increase in flow during early reperfusion
is termed ‘‘reactive hyperaemia”. Reactive hyperaemia is most probably related to the
metabolic theory of tissue perfusion. Lack of perfusion causes a deprivation of oxygen and
accumulation of vasodilating metabolites. The resulting degree and duration of the excess
blood flow during the reactive hyperaemic response are proportional to the length of the
blood flow interruption and severity of oxygen debt (Fig. 6.6).38 This response emphasises a
close connection between delivery of nutrients to tissues and the regulation of tissue
perfusion. Reactive hyperaemia will occur after ischaemia (partial or total occlusion of
arterial supply) or hypoxia (decreased PaO2). The heart and brain have large, skeletal
muscle intermediate, and liver, lung, and skin relatively small reactive hyperaemia
responses.
Page 237
Fig 6.6 Average values (lower histogram) and chart recordings (top panel) of reactive hyperaemic responses
of canine coronary blood flow (CBF) after 10, 30, 60, 75, 90, and 120 seconds of total coronary occlusion.
Note that with increasing time of occlusion there is an increasing hyperaemic response. (Reprinted with
permission from Warltier et al.38)
Long term regulation of the circulation

Whereas circulatory reflexes provide acute and rapid control of the circulation, the
concept of long term regulation is based upon the balance of blood volume and urinary
output.22 The kidneys provide the major long term control of the circulation. Reflex
changes in haemodynamics, as mediated by baroreceptors, play no important role in long
term adjustments because the receptors rapidly adapt to continued increases or decreases in
pressure. Long term regulation follows a simple renal–body fluid mechanism: increases in
arterial pressure produce increases in salt and water output through the kidneys, hence
reducing extracellular fluid volume, blood volume, and venous return. Reduced venous
return decreases cardiac output and arterial pressure. After several weeks, cardiac output
returns to previous levels while reductions in systemic vascular resistance develop to
maintain the lower arterial pressure. Conversely, a decline in blood pressure stimulates the
kidneys to retain fluid which ultimately results in elevation of arterial pressure. Renal
perfusion pressure and urinary output are directly related. This process occurs
independently of any hormonal actions, but it is significantly amplified by the renin–
angiotensin system.
The renal–body fluid mechanism is continuous and overriding, and undergoes no
adaptation. It requires that small alterations in fluid volume
Page 238
lead to substantial changes in arterial pressure. It is also unique because it has the ability to
return the blood pressure completely back to normal values. This is in contrast to short term
regulating mechanisms, which cannot entirely re-establish normal arterial pressure. Overall,
the renal–body fluid mechanism develops over hours to days and regulates the circulation
over an infinite time span.
Conclusion
Several mechanisms involving the CNS, ANS, humoral and chemical substances, and
circulatory reflexes act in combination to provide homoeostasis of the circulation. By
functioning together to regulate cardiac output, peripheral vascular resistance, and venous
capacitance, relatively constant arterial pressure and adequate tissue perfusion are
maintained. The cardiovascular control mechanisms act in unison to provide important
adaptive responses to stresses such as haemorrhage or exercise. Without such reflexes, even
the simple act of assuming an upright position would be met with abrupt and large declines
in cardiac output and arterial pressure threatening the status of the organism.
Acknowledgements
This work was supported by the Anesthesiology Research Training Grants GM 08377.
1 Mitchell GF, Pfeffer JM, Pfeffer MA. The heart and conduit vessels in hypertension. Med clin North Am
1997;81:1247–71.
2 Dampney RA, Goodchild AK, Tan E. Identification of cardiovascular cell groups in the brain stem. Clin
Exp Hypertens [A] 1984;6:205–20.
3 Hilton SM, Spyer KM. Central nervous regulation of vascular resistance. Annu Rev Physiol 1980;42:399–
441.
4 Alexander RS. Tonic and reflex functions of medullary sympathetic cardiovascular centers. J
Neurophysiol 1946;9:205–17.
5 Levy MN, Martin PJ, Stuesse SL. Neural regulation of the heart beat. Annu Rev Physiol 1981:43:443–53.
6 Calaresu FR, Yardley CP. Medullary basal sympathetic tone. Annu Rev Physiol 1988;50:511–24.
7 Smith JJ, Kampine JP. Circulatory physiology: the essentials. Baltimore, MD: Williams & Wilkins, 1990.
8 Dontenwill M, Tibirica E, Greney H, et al. Role of imidazoline receptors in cardiovascular regulation. Am
J Cardiol 1994;73:3–6A.
9 Janig W. Pre- and postganglionic vasoconstrictor neurons: differentiation, types, and discharge properties.
Annu Rev Physiol 1988;50:525–39.
10 Lawson NW. Autonomic nervous system physiology and pharmacology. In: Barash PG, Cullen BF,
Stoelting RK, eds. Clinical anesthesia. Philadelphia: JB Lippincott, 1992:319–84.
11 Katz AM. Physiology of the heart. New York: Raven Press, 1992.
Page 239
12 Ahluwalia A, Cellek S. Regulation of the cardiovascular system by non-adrenergic non-cholinergic
nerves. Curr Opin Nephrol Hypertens 1997;6:74–9.
13 Hoffman BB, Lefkowitz RJ. Adrenergic receptors in the heart. Annu Rev Physiol 1982;44:475–84.
14 Feigl EO. Coronary physiology. Physiol Rev 1983;63:1–205.
15 Donald DE, Shepherd JT. Autonomic regulation of the peripheral circulation. Annu Rev Physiol
1980;42:429–39.
16 Shepherd JT, Mancia G. Reflex control of the human cardiovascular system. Rev Physiol Biochem
Pharmacol 1986;105:l–99.
17 Ebert TJ. Autonomic balance and cardiac function. Curr Opin Anaesthesiol 1992;5:3–10.
18 Reid IA, Morris BJ, Ganong WF. The renin–angiotensin system. Annu Rev Physiol 1978;40:377–410.
19 Ludbrook J. Reflex control of blood pressure during exercise. Annu Rev Physiol 1983;45:155–68.
20 Stoelting RK, Dierdorf SF. Anesthesia and co-existing disease. New York: Churchill Livingstone, 1993.
21 Guyton AC. Acute hypertension in dogs with cerebral ischemia. Am J Physiol 1948;154:45–54.
22 Vatner SF, Cox DA. Circulatory function and control. Textbook of internal medicine. Philadelphia: JB
Lippincott, 1992.
23 Mirenda JV, Grissom TE. Anesthetic implications of the renin–angiotensin system and angiotensin-
converting enzyme inhibitors. Anesth Analg 1991;72:667–83.
24 Johnston CI. Franz Volhard Lecture. Renin–angiotensin system: a dual tissue and hormonal system for
cardiovascular control. J Hypertens 1992;10:S13–26.
25 Scholkens BA. Kinins in the cardiovascular system. Immunopharmacology 1996; 33:209–16.
26 Hays RM. Agents affecting the renal conservation of water. Pharmacological basis of therapeutics.
NewYork: Macmillan, 1985.
27 Rich GF, Johns RA. Nitric oxide and the pulmonary circulation. Adυ Anesth 1994;11:1–25.
28 Johns RA. Endothelium, anesthetics, and vascular control. Anesthesiology 1993;79:1381–91.
29 Welch G, Loscalzo J. Nitric oxide and the cardiovascular system. J Cardiovasc Surg 1994;9:361–71.
30 Maxwell AJ, Cooke JP. Cardiovascular effects of L-arginine. Curr Opin Nephrol Hypertens 1998;7:63–
70.
31 Cogan MG. Renal effects of atrial natriuretic factor. Annu Rev Physiol 1990;52:699–708.
32 Ginsburg R, Bristow MR, Stinson EB, Harrison DC. Histamine receptors in the human heart. Life Sci
1980;26:2245–9.
33 Regoli D.Neurohumoral regulation of precapillary vessels: the kallikrein–kinin system. J Cardioυasc
Pharmacol 1984;6(suppl 2):S401–12.
34 Marwood JF, Stokes GS. Serotonin (5HT) and its antagonists: involvement in the cardiovascular system.
Clin Exp Pharmacol Physiol 1984;11:439–55.
35 Parris RJ, Webb DJ. The endothelin system in cardiovascular physiology and pathophysiology. Vasc Med
1997;2:31–3.
36 Olsson RA. Local factors regulating cardiac and skeletal muscle blood flow. Annu Rev Physiol
1981;43:385–95.
37 Goldberg AH, Warltier DC. The coronary circulation: Importance for anesthesiologists. Semin Anesth
1990;9:232–44.
38 Warltier DC, Gross GJ, Brooks HL. Pharmacologic- vs. ischemia-induced coronary artery
vasodilation.Am J Physiol 1981;240:H767–74.
Page 240
7: Cerebral circulation
DAVID K MENON
The brain receives 15% of the resting cardiac output (700 ml/min in the adult) and accounts
for 20% of basal oxygen consumption. Mean resting cerebral blood flow (CBF) in young
adults is about 50 ml/100 g brain per min. This mean value represents two very different
categories of flow: 70 and 20 ml/100 g per min for grey and white matter, respectively.
Regional CBF (rCBF) and glucose consumption decline with age, along with marked
reductions in brain neurotransmitter content, and less consistent decreases in
neurotransmitter binding.1
Functional anatomy of the cerebral circulation

Arterial supply
Blood supply to the brain is provided by the two internal carotid arteries and the basilar
artery, which divides into the two posterior cerebral arteries. The anastomoses between
these two sets of vessels gives rise to the circle of Willis (Fig. 7.la). Functionally significant
hypoplasia of the anterior and posterior communicating arteries is, however, common, and
a classic “normal” polygonal anastomotic ring is found in less than 50% of brains.2 Despite
the anatomical variations described, certain patterns of regional blood supply from
individual arteries are generally recognised (Fig. 7.1b). Cerebral ischaemia associated with
systemic hypotension classically produces maximal lesions in areas where the zones of
blood supply from two vessels meet, resulting in “watershed” infarctions. The presence of
anatomical variants may, however, substantially modify patterns of infarction following
large vessel occlusion. For example, in some individuals, the proximal part of one anterior
cerebral artery is hypoplastic, and flow to the ipsilateral frontal lobe is largely provided by
the contralateral anterior cerebral artery, via the anterior communicating artery. Occlusion
of the single dominant anterior cerebral artery in such a patient may result in massive
infarction of both frontal lobes: the unpaired anterior cerebral artery syndrome.
Page 241
Fig 7.1 (a) Classic anatomy of the circle of Willis. ICA, internal carotid artery; AComA, anterior
communicating artery; ACA, anterior cerebral artery (A1; precommunicating segment of ACA; A2,
postcommunicating segment of ACA); MCA, middle cerebral artery; PComA, posterior communicating
artery; PCA, posterior cerebral artery (P1, precommunicating segment of PCA; P2, postcommunicating
segment of PCA). (b) Classic patterns of blood flow distribution in the brain, showing contributions for the
main components of the circle of Willis. Areas of the brain that are at the junction of two arterial territories
are most at risk of hypoperfusion during global ischaemic insults, and may suffer watershed infarction.
Page 242
Microcirculation
The cerebral circulation is protected from systemic blood pressure surges by a specially
designed branching system and two resistance elements: the first of these lies in the large
cerebral arteries and the second in vessels with a diameter of less than 100 μm. The
architecture of the cerebral microvasculature is highly organised and follows the columnar
arrangement seen with neuronal groups and physiological functional units.3 Pial vessels on
the surface of the brain give rise to arterioles that penetrate the brain at right angles to the
surface and also to capillaries at all laminar levels. Each of these arterioles supplies a
hexagonal column of cortical tissue, with intervening boundary zones, an arrangement that
is responsible for columnar patterns of local blood flow, redox state,4 and glucose
metabolism seen in the cortex during hypoxia or ischaemia.5 Capillary density in the cortex
is one third of adult levels at birth, doubles in the first year, and reaches adult levels at four
years. In the adult animal capillary density is related to the number of synapses, rather than
the number of neurons or mass of cell bodies in a given region,6 and can be closely
correlated with the regional level of oxidative metabolism.7, 8 Conventionally, functional
activation of the brain is thought to result in “capillary recruitment”, implying that some
parts of the capillary network are non-functional during rest. Recent evidence suggests,
however, that all capillaries may be persistently open,8 and ‘‘recruitment” involves changes
in capillary flow rates with homogenisation of the perfusion rate in a network.9
Venous drainage
The brain is drained by a system of infra- and extracerebral venous sinuses, which are
endothelialised channels in folds of dura mater (Fig. 7.2). These sinuses drain into the
internal jugular veins, which, at their origin receive minimal contributions from extra
cerebral tissues. Measurement of oxygen saturation in the jugular bulb (Sjvo2) thus provides
a useful measure of cerebral oxygenation. It has been suggested that the supratentorial
compartment is preferentially drained by the right internal jugular vein, whereas the
infratentorial compartment is preferentially drained by the left internal jugular vein. More
recent data suggest, however, considerable interindividual variation in cerebral venous
drainage.10
Cerebral blood volume: physiology and potential for therapeutic

intervention
Most of the intracranial blood volume of about 200 ml is contained in these venous
sinuses and pial veins, which constitute the capacitance vessels of the cerebral circulation;
reduction in this volume can buffer rises in the volume of other intracranial contents (the
brain and CSF). Conversely,
Page 243
Fig 7.2 Venous sinuses of the brain, showing drainage of the superior sagittal sinus into the right jugular vein
via the transverse and sigmoid sinus. Note that the first extracranial tributary of the internal jugular vein (the
common facial vein or CFV) enters it below the lower level of the body of the second cervical vertebra.
Consequently, a retrograde jugular catheter with its tip above this level samples blood that drains exclusively
from intracranial contents.
when compensatory mechanisms to control intracranial pressure (ICP) have been

exhausted, even small increases in cerebral blood volume (CBV) can result in steep rises in
ICP (Fig. 7.3).
The position of the system on this curve can be expressed in terms of the pressure
volume index (PVI), which is defined as the change in intracranial volume that produces a
tenfold increase in ICP. This is normally about 26 ml,11 but may be markedly lower in
patients with intracranial hypertension, who are on the steep part of the intracranial
pressure–volume curve (see effects of arterial carbon dioxide tension, PaCO2 on CBV).
With the exception of oedema reduction by mannitol, the only intracranial constituent
whose volume can be readily modified by the anaesthetist via physiological or
pharmacological interventions is the CBV. Although the CBV forms only a small part of
the intracranial volume, and such interventions only produce small absolute changes
(typically about 10 ml or less), they may result in marked reductions in ICP in the presence
of intracranial hypertension. Conversely, inappropriate anaesthetic management may cause
the CBV to increase. Again, although the absolute
Page 244
Fig 7.3 Intracranial pressure–volume curve. The pressure volume index (PVI), defined as the change in
intracranial volume required to cause a tenfold increase in ICP, increases non-linearly, from approximately 25
ml in normals (A on the curve), to as little as 5 ml in patients with raised ICP (B on the curve).
magnitude of such an increase may be small, it may result in steep rises in ICP in the
presence of intracranial hypertension.
The appreciation that pharmacological and physiological modulators may have
independent effects on CBV and CBF is an important one for two reasons. First,
interventions aimed at reducing CBV in patients with intracranial hypertension may have
prominent effects on CBF and result in cerebral ischaemia12 (Fig. 7.4). Conversely, drugs
that produce divergent effects on CBF may have similar effects on CBV, and using CBF
measurement to infer effects on CBV and hence ICP may result in erroneous conclusions.13
Determinants of cerebral perfusion
The inflow pressure to the brain is equal to the mean arterial pressure (MAP) measured at
the level of the brain. The outflow pressure from the intracranial cavity depends on the ICP,
because collapse of intracerebral veins is prevented by the maintenance of an intraluminal
pressure 2–5 mm Hg above ICP. The difference between the MAP and the ICP thus
provides an estimate of the effective cerebral perfusion pressure (CPP):
Microcirculatory transport and the blood–brain barrier

Endothelial cells in cerebral capillaries contain few pinocytic vesicles, and are sealed
with tight junctions, with no anatomical gap. Consequently,
Page 245
Fig 7.4 Relative effects of PaCO2 on cerebral blood flow (CBF) and voume (CBV). Hyperventilation is aimed
at reducing CBV in patients with intracranial hypertension, but may be detrimental because of its effects on
CBF. Note that the slope of CBF reactivity to PaCO2 is steeper than that for CBV (about 25% per kPa PaCO2
vs 20% per kPa PaCO2, respectively).
unlike other capillary beds, the endothelial barrier of cerebral capillaries presents a high
electrical resistance and is remarkably non-leaky, even to small molecules such as mannitol
(molecular weight, Mw, 180 daltons). This property of the cerebral vasculature is termed
the “blood-brain barrier” (BBB); it resides in three cellular components (the endothelial
cell, astrocyte, and pericyte), and one non-cellular structure (the endothelial basement
membrane) (Fig. 7.5). A fundamental difference between brain endothelial cells and the
systemic circulation is the presence of inter-endothelial tight junctions termed the “zona
occludens”. The BBB is a function of the cerebral microenvironment rather than an
intrinsic property of the vessels themselves, and leaky capillaries from other vascular beds
develop a BBB if they are transplanted to the brain or exposed to astrocytes in culture.14
Passage through the BBB is not simply a function of molecular weight; lipophilic
substances traverse the barrier relatively easily, and several hydrophilic molecules
(including glucose) cross the BBB via active transport systems to enter the brain interstitial
space.15 (Fig. 7.6). In addition, the BBB maintains a tight control of relative ionic
distribution in the brain extracellular fluid. These activities are energy requiring and
account for the fact that the mitochondrial density is exceptionally high in these endothelial
cells, accounting for 10% of cytoplasmic volume.16 Although the BBB is disrupted by
ischaemia, this process takes hours or days rather than minutes, and much of the cerebral
oedema seen in the initial period after ischaemic insults is cytotoxic rather than vasogenic.
Page 246
Fig 7.5 The components of the blood–brain barrier (see text).
Consequently, mannitol retains its ability to reduce cerebral oedema in the early phases of
acute brain injury.
Measurement of rCBF
All clinical and many laboratory methods of measuring CBF or rCBF are indirect and
may not produce directly comparable measurements. It is also important to treat results
from any one method with caution, and attribute any observed phenomena to physiological
effects only when demonstrated by two or more independent techniques. Methods of
measuring CBF may be regional or global, and applicable either to humans or primarily to
experimental animals. All of these methods have advantages and disadvantages (Table 7.1).
All methods that provide absolute estimates of rCBF use one of two principles: either they
measure the distribution of a tracer or they estimate rCBF from the wash-in or wash-out
curve of an indicator. Other techniques do not directly estimate rCBF, but can be used
either to measure a related flow variable (such as arterial flow velocity) or to infer changes
in flow from changes in metabolic parameters. Some techniques have been used for the
measurement of CBF are described briefly.
Page 247
Fig 7.6 Correlation between brain uptake index and oil/water partition coefficients for different substrates.
Although blood–brain barrier (BBB) permeability, in general, increases with lipid solubility, note that several
substances, including glucose and amino acids, show high penetration as a result of active transport or
facilitated diffusion. (After Oldendorf.15)
The Kety–Schmidt technique17
The Kety–Schmidt technique involves the insertion of catheters into a peripheral artery
and the jugular bulb. A diffusible tracer such as 10–15% inhaled nitrous oxide (N2O) is
administered, and paired arterial and jugular venous samples of blood are obtained at rapid
intervals for measurement of N2O levels. The resultant plot of concentration versus time
produces an arterial and a venous curve (Fig. 7.7). The jugular venous level of N2O rises
more slowly than the arterial levels, because N2O is being taken up by the brain as it is
delivered. The rate of equilibration of the two curves measures the rate at which N2O is
being delivered to the brain, and thus provides a means of measuring global CBF.
Xenon-133 wash-out
An array of collimated scintillation counters is positioned over the head to plot the
regional decay in radioactivity after the intracarotid18 or intra-
Page 248
Table 7.1 Methods of measuring cerebral blood flow
Technique Global/ Comments

regional
Human and laboratory methods

Kety–Schmidt17 Global Uses rate of uptake of N2O to measure global
CBF. Requires jugular bulb and arterial catheters.
Repeated measures possible.
133 Xe wash-out18 20 Regional Classic method uses wash-out curve of radioactive
xenon after intracarotid injection to estimate CBF.
Summated curves show fast and slow wash-out
components (? grey and white matter). Primarily
looks at cortex, poor resolution. Modifications
include intravenous/ inhalational/intra-aortic
administration of xenon. Repeated measures
possible.
Dynamic x ray CT Regional Looks at wash-out of stable xenon138/i.v.
contrast139 after inhalation. Repeated measures
possible, but not at rapid intervals.
PET (H215O)140 Regional Uses distribution of radiolabelled markers to
estimate rCBF Repeated measures possible.
Expensive equipment. Good resolution.
SPECT141 Regional Uses distribution of radiolabelled tracer to
estimate rCBF. Cheaper than PET, but resolution
poorer. Repeated measurements difficult. Only
measures relative CBF.
NIROS (+O2 wash-in) ? Detects rate of change of cerebral oxygenation
142 Regional state after step change in FIO2. Poorly defined
volume and resolution. ?Accurate.
Continued/
Page 249
Table 7.1 Continued
Technique Global/ Comments

regional
Laboratory methods
H2 clearance143 Regional Measures wash-out of H2 after inhalational
administration. Very localised measurement (1–2
mm3) with hydrogen electrode. Requires
craniotomy. Repeated measures possible.
Autoradiography144 Regional Uses distribution of radiolabelled tracer ([14C]-
iodoantipyrine) to estimate rCBF. Excellent
resolution. Single measurement only.
Radiolabelled Regional Uses distribution of radiolabelled microspheres
microspheres145 (15μm diameter) to estimate rCBF; Resolution
not as good as autoradiography. Repeated
measurements with different radiolabels possible.
Radiolabel injection in left atrium or aorta.
Indirect or non-quantitative measures
MRI146 Regional Utilises change in regional oxygenation during
functional activation to show changes in rCBF.
Superb resolution, absolute measures not
possible. Excellent resolution, repeated measures
easy. Modification with external label may
permit quantitation, but repeated measures more
restricted.
Doppler (Regional) Measures flow velocity in middle cerebral artery
ultrasonography147 using Doppler ultrasonography. Indirect measure
of CBF.
NIROS26 (Regional) Measures regional haemoglobin oxygenation and
cytochrome redox state in restricted and poorly
defined volume.
MRS148 Regional Provides information regarding intracellular pH
and tissue levels of ATP and lactate. Poor
resolution. Repeatable.
Abbreviations: PET, positron emission tomography; SPECT, single photon emission

tomography; NIROS, near infrared optical spectroscopy; MRI, magnetic resonance
imaging; MRS, magnetic resonance spectroscopy; FIO2, fraction of inspired O2.
Page 250
Fig 7.7 The Kety–Schmidt method of measuring global cerebral blood flow (CBF). The rate of increase in
arterial and jugular venous concentrations of a diffusible tracer gas (N2O in this instance) are
compared. A rapid equilibration implies high CBF, whereas a slow equilibration is evidence of low CBF.
aortic19 injection of 133Xe. The wash-out curve for radioactivity is biexponential, and may
be resolved into two monoexponential components, which represent a fast wash-out and a
slow wash-out component. Although these are often referred to as grey matter and white
matter components, it must be emphasised that there is no basis to support such an
anatomical distinction, because the two curves represent pharmacokinetic compartments
rather than specific neuroanatomical structures. The technique does provide two
dimensional information regarding rCBF, but is invasive and primarily looks at superficial
cortical blood flow. Further, intracarotid injection permits the assessment of only a single
cerebral hemisphere at a time. One modification involves the inhalational20 or intravenous
administration of 133Xe; although this makes the technique less invasive, problems arise
because of recirculation and contamination by extracranial tissues. The simultaneous
presence of activity in both cerebral hemispheres also leads to the “look-through’’
phenomenon, where rCBF reductions on one side may be missed because of activity sensed
in deeper or contralateral tissues.
Tomographic rCBF measurement: dynamic CT, SPECT, PET, and fMRI
Tomographic information regarding rCBF may be obtained by quantifying the wash-out

of a radiodense contrast agent, using rapid sequential x ray
Page 251
computed tomographic imaging (dynamic CT scanning). In the past, inhaled stable xenon
has most often been used as the contrast agent. More recent studies have been performed
using standard radioiodinated intravenous contrast agents.
Single photon emission tomography (SPECT) and positron emission tomography (PET)
use γ emitting and positron emitting isotopes, respectively, to produce tomographic images
of rCBF. PET, in addition to imaging cerebral blood flow (Fig. 7.8), can provide
quantitative tomographic information on cerebral blood volume, oxygen metabolism, and
oxygen extraction fraction (OEF). Functional magnetic resonance imaging (fMRI) produces
tomographic images of rCBF in one of two ways. One technique uses an intravenous MR
contrast agent (for example, gadopentate dimeglumine, Magnevist) in much the same way
as the previous techniques in this section use other agents. MRI can also produce, without
the use of external contrast agents, tomographic images of changes in rCBF after functional
activation by imaging the increases in MR signal intensity produced by the decreases in
regional deoxyhaemoglobin levels, which occur during flow metabolism coupling.
Detailed discussion of these techniques is beyond the scope of this chapter, and the
interested reader is referred to the references in Table 7.1 for details.
Fig 7.8 PET images of cerebral blood flow (CBF) in a patient with a left temporoparietal contusion after a
head injury. Note the marked heterogeneity in CBF values in the region of the contusion.
Page 252
Continuous clinical monitoring of the adequacy of cerebral perfusion, in general, tends to
use techniques other than those outlined in Table 7.121 (Fig. 7.9). The parameter most
commonly monitored in head injured
Fig 7.9 Continuous record of multimodality monitoring in a patient with acute head injury during a plateau
wave in the intracranial pressure (ICP). Note the fall in the cerebral perfusion pressure (CPP) and reduction in
the middle cerebral artery flow velocity (FV) measured using transcranial Doppler and reduction in laser
Doppler signal intensity (FLUX), suggesting reduced capillary flow. The ischaemia associated with the
reduction in CPP produces jugular venous desaturation (SJO2 catheter) and a fall in oxyhaemoglobin signal
(NIRS). (Recording courtsey of Dr Marek Czosnyka.)
Page 253
patients is the cerebral perfusion pressure, although many centres are increasingly using
fibreoptic jugular venous oximetry22 and transcranial Doppler measurement of middle
cerebral artery flow velocity.23 Monitoring of the processed EEG24 or evoked potentials25
provides information regarding the consequences of reduced CBF, and this technique has
been used in the context of cardiopulmonary bypass and carotid endarterectomy. Near
infrared optical spectroscopy26 and laser Doppler flowmetry27 are investigational techniques
whose roles have not been clearly defined.
Transcranial Doppler ultrasonography
Transcranial Doppler ultrasonography (TCD) measures the velocity of red blood cells
(RBCs) flowing through the large vessels at the base of the brain using the Doppler shift
principle. As the diameter of these basal vessels is not affected by common physiological
variables, such as MAP and PaCO2, flow velocity (FV) in these vessels provides an index of
flow. Although many of the intracranial arteries may be studied, the middle cerebral artery
is most commonly insonated (Fig. 7.10) because it is easy to detect, receives a substantial
proportion of the blood flow from the internal
Fig 7.10 Transcranial Doppler ultrasonography (TCD) waveform from the left middle cerebral artery. Note
the three axes on the TCD waveform, which include time on the x axis and red blood cells (RBCs) flow
velocity on the y axis. The grey scale of the waveform is the third axis, and represents the population density
of RBCs at any given velocity.
Page 254
carotid artery, and allows easy probe fixation. Provided that the angle of insonation and the
diameter of the vessel insonated remain constant, relative changes in CBF velocity correlate
closely with changes in CBF.28–30 Changes in TCD velocities and waveform patterns can
be used to detect cerebral ischaemia, hyperaemia, and vasospasm.31 In addition, the
characteristics of the TCD waveform may be used to provide a non-invasive estimate of
cerebral perfusion pressure (CPP).32
Jugular venous oximetry
Cerebral oxygenation has conventionally been assessed by jugular bulb oximetry.

Conventionally, the superior saggital sinus is thought to drain primarily into the right
internal jugular vein, and it is common practice to place jugular bulb catheters on this side
in order to monitor the oxygenation in the supratentorial compartment. More recent data
suggest that supratentorial venous drainage is less lateralised, and a case has been made for
bilateral jugular bulb catheterisation.33 Normal jugular bulb oxygen saturations (SJVO2)
tend to run at 65–70%. Reductions in SJVO2 or increases in arteriojugular differences in
oxygen content (DaJO2) to greater than 9 ml/dl provide useful markers of inadequate CBF33
(Fig. 7.11) and can guide therapy,34 and SJVO2 values below 50% have been shown to be
associated with a worse outcome in head injury.35 Conversely, marked elevations in SJVO2
may provide evidence of cerebral hyperaemia.
Fig 7.11 Use of jugular bulb oximetry to assess the adequacy of cerebral blood flow (CBF). Reductions in
CBF force an increase in oxygen extraction by the brain if cerebral metabolic requirements for O2 (CMRO2)
requirements are to be met. This is reflected by a fall in jugular venous oxygen saturation (SJVO2); (a) normal
and (b) reduced CBF.
Page 255
Physiological determinants of regional cerebral blood

flow and volume
Flow–metabolism coupling
Increases in local neuronal activity are accompanied by increases in regional cerebral

metabolic rate (rCMR). Until recently, the increases in rCBF and oxygen consumption
produced during such functional activation were thought to be closely coupled to the
cerebral metabolic rate of utilisation of O2 (CMRo2) and glucose (CMRglu). However, it
has now been clearly shown that increases in rCBF during functional activation tend to
track glucose utilisation, but may be far in excess of the increase in oxygen consumption.36
This results in regional anaerobic glucose utilisation, and a consequent local decrease in
oxygen extraction ratio and an increase in local haemoglobin saturation. The resulting local
decrease in deoxyhaemoglobin levels is used by functional MRI techniques to image the
changes in rCBF produced by functional activation. Despite this revision of the
proportionality between increased rCBF and CMRo2 during functional activation in the
brain, the relationship between rCBF and CMRglu is still accepted as linear.
The cellular mechanisms underlying these observations are elucidated by recent
publications, which have highlighted the role played by astrocytes in the regulation of
cerebral metabolism.37 These data suggest that astrocytes utilise glucose glycolytically and
produce lactate, which is transferred to neurons where it serves as a fuel in the citric acid
cycle.38 Astrocytic glucose utilisation and lactate production appear to be, in large part,
coupled by the astrocytic reuptake of glutamate released at excitatory synapses (Fig. 7.12).
The regulatory changes involved in flow–metabolism coupling have a short latency
(about 1 s) and may be mediated by either metabolic or neurogenic pathways. The former
category includes the increases in perivascular K+ or adenosine concentrations that follow
neuronal depolarisation. The cerebral vessels are richly supplied by nerve fibres, and the
mediators thought to play an important part in neurogenic flow metabolism coupling are
acetylcholine39 and nitric oxide,40 although roles have also been proposed for 5-
hydroxytryptamine, substance P, and neuropeptide Y.
Autoregulation
Autoregulation refers to the ability of the cerebral circulation to maintain CBF at a

relatively constant level in the face of changes in CPP by altering cerebrovascular
resistance (CVR) (Fig. 7.13). Although autoregulation is maintained irrespective of whether
changes in CPP arise from alterations in MAP or ICP, autoregulation tends to be preserved
at lower levels when falls in CPP are the result of increases in ICP rather than decreases in
MAP
Page 256
Fig 7.12 Relationship of astrocytes to oxygen and energy metabolism in the brain. Glucose taken up by
astrocytes undergoes glycolysis for generation of ATP to meet astrocytic energy requirements (for glutamate
reuptake, predominantly). The lactate that this process generates is shuttled to neurons, which utilise it
aerobically in the citric acid cycle.
caused by hypovolaemia.41 42 One possible reason for this may be the cerebral
vasoconstrictive effects of the massive levels of catecholamines secreted in haemorrhagic
hypotension, because lower MAP levels are tolerated in hypotension if the fall in blood
pressure is induced by sympatholytic agents,43 44 or occurs in the setting of autonomic
failure.45 Autoregulatory changes in CVR probably arise from myogenic reflexes in the
resistance vessels, but these may be modulated by activity of the sympathetic system or the
presence of chronic systemic hypertension.46 Thus, sympathetic blockade or cervical
sympathectomy shifts the autoregulatory curve to the left, whereas chronic hypertension or
sympathetic activation shifts it to the right. These modulatory effects may arise from
angiotensin-mediated mechanisms. Primate studies suggest that nitric oxide is unlikely to
be important in pressure autoregulation.47
In reality, the clear cut autoregulatory thresholds seen with varying CPP in Fig. 7.13a are
not observed; the autoregulatory “knees” tend to be more gradual, and there may be wide
variations in rCBF at a given value of CPP in experimental animals and even in
neurologically normal individuals.48 It
Page 257
Fig 7.13 Effect of changes in cerebral perfusion pressure (CPP), PaCO2, and PaO2 on cerebral blood flow
(CBF). (a) Note the increase in slope of the CBF/PaCO2 curve as basal CBF increases from 20 ml/100 g per
min (white matter) to 50–70 ml/ 100 g per min (grey matter), (b) Note that maintenance of CBF with
reductions in CPP is achieved by cerebral vasodilation, which results in reductions in cerebrovascular
resistance. This results in an increase in cerebral blood volume (CBV), which has no detrimental effects in
healthy subjects. These CBV increases may, however, result in critical increases in intracranial pressure (ICP)
in patients with intracranial hypertension, who operate on the steep part of the intracranial pressure–volume
curve.
Page 258
has been demonstrated that symptoms of cerebral ischaemia appear when the MAP falls
below 60% of an individual’s lower autoregulatory threshold.49 Generalised extrapolation
from such individualised research data to the production of “safe” lower limits of MAP for
general clinical practice is, however, hazardous for several reasons:
1 There may be wide individual scatter in rCBF autoregulatory efficiency, even in normal
subjects.
2 The coexistence of fixed vascular obstruction (for example, carotid atheroma or vascular
spasm) may vary the MAP level at which rCBF reaches critical levels in relevant
territories.
3 The autoregulatory curve may be substantially modulated by the mechanisms used to
produce hypotension. Earlier discussion made the distinction between reductions in CPP
produced by haemorrhagic hypotension, intracranial hypertension, and pharmacological
hypotension. The effects on autoregulation may also vary with the pharmacological agent
used to produce hypotension. Thus, neuronal function is better preserved at similar levels
of hypotension produced by halothane, nitroprusside, or isoflurane in comparison to
trimethephan.50
4 Autoregulatory responses are not immediate: estimates of the latency for compensatory
changes in rCVR range from 10 to 60 s.51
Some recent studies suggest that, especially in patients with impaired autoregulation, the
cardiac output and pulsatility of large vessel flow may be more important determinants of
rCBF than CPP itself.52
Arterial carbon dioxide tension
Cerebral blood flow is proportional to PaCO2, subject to a lower limit below which
vasoconstriction results in tissue hypoxia and reflex vasodilatation, and an upper limit of
maximal vasodilatation (Fig. 7.13a). On average, in the middle of the physiological range,
each KiloPascal change in PaCO2produces a change of about 15 ml/100 g per min in CBF.
The slope of the PaCO2/CBF relationship depends, however, on the baseline normocapnic
rCBF value, being maximal in areas where it is high (for example, grey matter; cerebrum)
and least in areas where it is low, (for example white matter; cerebellum, and spinal cord).
Moderate hypocapnia (PaCO2 of about 3.5 kPa) has long been used to reduce CBV in
intracranial hypertension, but this practice is under review for two reasons:
1 The CO2 response is directly related to the change in perivascular pH; consequently, the
effect of a change in PaCO2 tends to be attenuated over time (hours) as brain extracellular
fluid (ECF) bicarbonate levels fall to normalise interstitial pH.53
2 It has now been shown that “acceptable” levels of hypocapnia in head injured patients can
result in dangerously low rCBF levels.12 54
Page 259
Prostaglandins may mediate the vasodilation produced by CO2;55more recent work
suggests that nitric oxide may also be involved,56 perhaps in a permissive capacity.57
Effects on CBV and ICP

Grubb et al58 studied the CBF/PaCO2 response curve in primates and demonstrated that
the CBF changed by approximately 1.8 ml/100 g per min for each change in PaCO2 of 1
mm Hg (133 Pa). In the same experiment, however, the CBV/PaCO2 curve was much flatter
(about 0·4 ml/100 g per mm Hg [0·3 ml/100 g per kPa] change in PaCO2). It follows from
these figures that, although a reduction in PaCO2 from 40 to 30 mm Hg (5·3 to 4 kPa)
would result in about a 40% reduction in CBF (from a baseline of about 50 ml/100 g per
min), it would only result in a 0·4% reduction in intracranial volume. This may seem trivial,
but, in the presence of intracranial hypertension, the resultant 5 ml decrease in intracranial
volume in an adult brain could result in a halving of ICP because the system operates on the
steep part of the intracranial compliance curve.59
Arterial oxygen pressure and content
Classic teaching is that CBF is unchanged until PaCO2 levels fall below approximately 7
kPa, but rises sharply with further reductions60 (see Fig. 7.13a). Recent TCD data from
humans, however, suggest cerebral thresholds for cerebral vasodilatation as high as 8·5 kPa
(about 89–90% arterial O2 saturation or SaO2).61 This non-linear behaviour is because tissue
oxygen delivery governs CBF; the sigmoid shape of the haemoglobin–O2 dissociation
curve means that the relationship between CaO2 (arterial O2 content) and CBF is inversely
linear. These vasodilator responses to hypoxaemia appear to show little adaptation with
time,62 but may be substantially modulated by PaCO2 levels.63 64 Nitric oxide does not
appear to play a role in the vasodilatory response to hypoxia.56
Some studies suggest that hyperoxia may produce cerebral vasoconstriction, with a 10–
14% reduction on CBF with inhalation of 85–100% O2, and a 20% reduction in CBF with
100% O2 at 3·5 atmospheres.65 There are no human data to suggest that this effect is
clinically significant.
Haematocrit
As in other organs, optimal O2 delivery in the brain depends on a compromise between

the oxygen carrying capacity and the flow characteristics of blood; previous experimental
work suggests that this may be best achieved at a haemotocrit of about 40%. Some recent
studies in the setting of vasospasm after subarachnoid haemorrhage have suggested that
modest haemodilution to a haematocrit of 30–35% may improve neurological outcome by
improving rheological characteristics66 and increasing rCBF.
Page 260
This may, however, result in a reduction in O2 delivery if maximal vasodilatation is already
present and, as clinical results in the setting of acute ischaemia have not been uniformly
successful, this approach must be viewed with caution.
Autonomic nervous system
The autonomic nervous system mainly affects the larger cerebral vessels, up to and
including the proximal parts of the anterior, middle, and posterior cerebral arteries. β1-
Adrenergic stimulation results in vasodilatation whereas α2-adrenergic stimulation
vasoconstricts these vessels. The effect of systemically administered α or β agonists is less
significant. Significant vasoconstriction can, however, be produced by extremely high
concentrations of catecholamines (for example, in haemorrhage) or centrally acting α2
agonists (for example, dexmedetomidine).
Pharmacological modulation of CBF

Inhaled anaesthetics
All the potent fluorinated agents have significant effects on CBF and CMR. The initial
popularity of halothane as a neurosurgical anaesthetic agent was reversed by the discovery
that it was a potent cerebral vasodilator, producing decreases of 20–40% in cerebrovascular
resistance in normocapnic individuals at 1·2–1·5 MAC (MAC = minimum alveolar
concentration).67 68 In another study, 1% halothane was shown to result in clinically
significant elevations in ICP in patients with intracranial space occupying lesions.69
Preliminary studies with enflurane and isoflurane suggested that these agents might produce
smaller increases in cerebrovascular resistance (CVR) at equivalent doses.70 As enflurane
may produce epileptogenic activity, its use in the context of neuroanaesthesia decreased.
Several studies, however, compared the effects of isoflurane and halothane on CBF, with
conflicting results. Although some studies showed that halothane produced larger decreases
in CVR, others found no difference. Examination of the patterns of rCBF produced by these
two agents provides some clues to the origin of this discrepancy. Halothane selectively
increases cortical rCBF, while markedly decreasing subcortical rCBF, whereas isoflurane
produces a more generalised reduction in rCBF.71 72 A review of published comparisons of
the CBF effects of the two agents suggests that studies that estimated CBF using techniques
that preferentially looked at the cortex (for example, 133Xe wash-out) tended to show that
halothane was a more potent vasodilator, whereas most studies that have used more global
measures of hemispheric CBF (for example, the Kety–Schmidt technique) have found little
difference between the two agents at levels of around 1 MAC (Fig. 7.14).
Page 261
Both agents tend to reduce global CMR, but the regional pattern of such an effect may
vary, with isoflurane producing greater cortical metabolic suppression73 (reflected by its
ability to produce EEG burst suppression at higher doses). Both the rCBF and rCMR effects
of the two anaesthetics are markedly modified by baseline physiology and other
pharmacological agents. Thus, CBF increases produced by both agents are attenuated by
hypocapnia (more so with isoflurane74 75), and thiopental attenuates the relative preservation
of cortical rCBF seen with halothane. It is difficult to predict accurately what the effect of
either agent would be on CBF in a given clinical situation, but this would be a balance of its
suppressant effects on rCMR (with autoregulatory vasoconstriction) and its direct
vasodilator effect (which is partially mediated via both endothelial and neuronal nitric
oxide).56 76
Although initial reports suggests that halothane could ‘‘uncouple” flow and
metabolism,76 more recent studies clearly show that at concentrations commonly in use for
neuroanaesthesia (0·5–1 MAC) neither halothane77 nor isoflurane77 78 completely disrupts
flow–metabolism coupling, although
Fig 7.14 Comparison of mean ±SD CBF in animals anaesthetised with 0·5–1·5 MAC isoflurane or halothane,
either in the same study or in comparable studies from a single research group with identical methodology
within a single publication.149–160 In studies shown on the left, CBF was estimated using techniques likely to
be biased towards cortical flow (for example, 133Xe wash-out) and shows that halothane produces greater
increases in CBF. In studies on the right, CBF was estimated using techniques that measured global CBF (for
example, the Kety–Schmidt method); the difference in effects on CBF between the two agents is much less
prominent.
Page 262
their vasodilator effects may alter the slope of this relationship. These vasodilator effects
may become more prominent at higher concentrations.
In equi-MAC doses, nitrous oxide is probably a more powerful vasodilator than either
halothane or isoflurane;79 80 this fact, coupled with its lack of CMR depression,81 produces a
particularly unfavourable pharmacodynamic profile in patients with raised ICP.82 83 Further,
the vasodilatation produced by nitrous oxide is not decreased by hypocapnia,84 although the
resulting increases in ICP can be attenuated by the administration of other CMR
depressants such as the barbiturates.85
Although initial studies suggested that desflurane86 and sevoflurane87 had effects on the
cerebral vasculature that appear very similar to isoflurane, more recent studies have shown
distinct differences between these agents.
Although high dose desflurane, like isoflurane, can produce EEG burst suppression, this
effect may be attenuated over time.88 It is not known whether this adaptation represents a
pharmacokinetic or a pharmacodynamic effect. Initial clinical reports suggest that
desflurane may cause a clinically significant rise in ICP in patients with supratentorial
lesions,89 despite its proven ability to reduce CMRo2 as documented by EEG burst
suppression.88 90 These increases in ICP, which are presumably related to cerebral
vasodilatation, appear to be independent of changes in systemic haemodynamics.91
In humans, sevoflurane produces some increase in TCD flow velocities at high doses (≥
1.5 MAC), but these appear to be less marked than desflurane, and were reported to be
unassociated with increases in ICP in patients with supratentorial space occupying lesions.92
In other studies, 1.5 MAC sevoflurane caused no increase in middle cerebral artery flow
velocities,93 and did not affect CO2 reactivity or pressure autoregulation.94
Intravenous anaesthetics
Thiopentone,95 etomidate,96 and propofol97 98 all reduce global CMR to a minimum of

approximately 50% of baseline, with a coupled reduction in CBF, although animal studies
suggest small differences in the distribution of rCBF changes with individual agents.
Decreases in CBV have been demonstrated with barbiturates99 and probably occur with
propofol and etomidate as well. Maximal reductions in CMR are reflected in an isoelectric
EEG, although burst suppression is associated with only slightly less CMR depression.95
Initial doubts that CBF reductions produced by propofol were secondary to falls in MAP
have proved to be unfounded.97 98
Even high doses of thiopental or propofol100 do not appear to affect autoregulation, CO2
responsiveness, or flow–metabolism coupling.
Opioids
Although high doses (3 mg/kg) of morphine and moderate doses of fentanyl (15 μg /kg)
have little effect on CBF and CMR, high doses of
Page 263
fentanyl (50–100 μg/kg)101 and sufentanil102 depress CMR and CBF. Results with alfentanil,
in doses of 0.32 mg/kg, show no reduction in rCBF.103 These effects are variable and may
be prominent only in the presence of N2O, where CMR may be reduced by 40% from
baseline.104 Bolus administration of large doses of fentanyl or alfentanil may be associated
with increases in ICP in patients with intracranial hypertension,105 probably as a result of
reflex increases in CBF that follow an initial decrease in CBF (caused by reductions in
MAP and cardiac output produced by large bolus doses of these agents). These effects are
unlikely to be clinically significant if detrimental haemodynamic and blood gas changes
can be avoided.
Other drugs
Ketamine can produce increases in global CBF and ICP,106 with specific increases in
rCMR and rCBF in limbic structures.107 These changes may be partially attenuated by
hypocapnia, benzodiazepines, or halothane.108 Sedative doses of benzodiazepines tend to
produce small decreases in CMR and CBF;109 however, there is a ceiling effect and
increasing doses do not produce greater reductions in these variables.110 α2 Agonists such as
dexmedetomidine reduce CBF in humans.111 There are good data, in animal models at
least112 to show that CBF reductions produced by intraventricular dexmedetomidine are
probably direct vascular effects of the agent, and not exclusively the consequence of either
systemic hypotension or coupled falls in rCBF arising from reductions in neuronal
metabolism.
Most non-depolarising neuromuscular blockers have little effect on CBF or CMR,
although large doses of d-tubocurarine may increase CBV and ICP secondary to histamine
release and vasodilatation. In contrast, suxamethonium (succinylcholine) can produce
increases in ICP, probably secondary to increases in CBF mediated via muscle spindle
activation. These effects, however, are transient and mild113 114 and can be blocked by prior
precurarisation115 if necessary; they provide no basis for avoiding suxamethonium in
patients with raised ICP when its rapid onset of action is desirable for clinical reasons.
CBF in disease
Ischaemia
Graded reductions in CBF are associated with specific electrophysiological and

metabolic consequences (Table 7.2). Some of these thresholds for metabolic events are well
recognised, but others, such as the
Page 264
Table 7.2 Electrophysiological and metabolic consequences of graded reductions in cerebral blood fl
(CBF)
CBF Electrophysiological/metabolic consequence

(ml/100 g per min)
>50 Normal neuronal function

? Immediate early gene activation
? Cessation of protein synthesis
? Cellular acidosis
20–23 Reduction in electrical activity
12–18 Cessation of electrical activity
8–10 ATP rundown, loss of ionic homoeostasis
<8 Cell death (also depends on other modifiers: duration, CMR, etc)
CMR, cerebral metabolic rate.

development of acidosis, cessation of protein synthesis, and the failure of osmotic
regulation, have only recently received attention.116 Ischaemia is thus a continuum between
normal cellular function and cell death; cell death, however, is not merely a function of the
severity of ischaemia, but is also dependent on its duration and several other circumstances
that modify its effects. Thus, the effects of ischaemia may be ameliorated by the CMR
depression produced by hypothermia or drugs, exacerbated by increased metabolic demand
associated with excitatory neurotransmitter release, or compounded by other mechanisms of
secondary neural injury (such as cellular calcium overload or reperfusion injury) (Fig.
7.15).
Fig 7.15 Relationship of cerebral blood flow (CBF) to the presence of ischaemia under conditions of varying
metabolism. Changes in CBF levels compared with physiological levels may be misleading, because a
diagnosis of ischaemia or hyperaemia demands that CBF levels be assessed in the context of metabolic
requirements.
Page 265
Head injury
Severe head injury is accompanied by both direct and indirect effects on CBF and
metabolism, which show both temporal and spatial variations. CBF may be high, normal, or
low soon after the ictus, but is typically reduced.117 Of patients undergoing CBF studies
within 6–8 hours of a head injury, 30% have significant cerebral ischaemia.118 Global
hypoperfusion in these studies was associated with a 100% mortality rate at 48 hours, and
regional ischaemia with significant deficits. CBF patterns also vary with relation to the time
after injury116 (Fig. 7.16). Initial reductions are replaced, especially in patients who achieve
good outcomes, by a period of relative increase in CBF, which towards the end of the first
week post ictus may be replaced by reductions in CBF that are the consequence of
vasospasm associated with subarachnoid haemorrhage.120 CBF changes are non-uniform in
the injured brain. Blood flow tends to be reduced in the immediate vicinity of intracranial
contusions,121 122 and cerebral ischaemia associated with hyperventilation may be extremely
regional and not reflected in global monitors of cerebrovascular adequacy123 (Fig. 7.17).
Elevations in ICP result in reductions in CPP and cerebral ischaemia, which lead to
secondary neuronal injury. There is strong evidence that maintenance of a CPP above 60
mm Hg improves outcome in patients with head injury and raised ICP.124 Traditionally,
patients with intracranial hypertension have been nursed head-up in an effort to reduce ICP.
It is important to realise, however, that such manoeuvres will also reduce the effective
MAP at the level of the head and run the risk of reducing CPP. Feldmann et al125 suggest
that a 30° head-up elevation may provide the optimal balance by reducing ICP without
decreasing CPP.
Fig 7.16 Spectrum of cerebral blood flow (CBF) patterns after severe head injury. Following an initial period
of ischaemia lasting less than 24 hours, CBF begins to rise and may exceed normal values on days 2 to 4. CBF
may fall to subnormal levels at later time points, chiefly as a result of the presence of vasospasm secondary to
traumatic subarachnoid haemorrhage. CBF levels may never rise in some patients, especially those who have
a poor outcome.
Page 266
Fig 7.17 PET image of cerebral blood flow (CBF) showing the effect of hyperventilation within the first 24
hours after head injury. Despite the maintenance of SJO2 values at acceptable levels, hyperventilation results
in increases in the volume of brain tissue where CBF falls below recognised thresholds of ischaemia.
Hypertensive encephalopathy
Current concepts of the causation of hypertensive encephalopathy are based on the forced
vasodilatation hypothesis.126 Severe acute or sustained elevations in MAP overcome
autoregulatory vasoconstriction in the resistance vessels and result in forced vasodilatation.
These vasodilated vessels, exposed to high intraluminal pressures, leak fluid and protein
and result in cerebral oedema, which is multifocal and later diffuse.
Subarachnoid haemorrhage
Cerebral autoregulation and CO2 responsiveness are grossly distorted after subarachnoid
haemorrhage (SAH), more so in patients in worse clinical grades127 (Fig. 7.18). Such
patients may be unable to compensate for reductions in MAP produced by anaesthetic
agents and develop clinically significant deficits.128 Clinically significant vasospasm after
SAH occurs in up to 30–40%129 of patients, typically several days after the initial bleed and
may be caused by one or more of several mechanisms. NO may be taken up by
haemoglobin in the extravasated blood or be inactivated to peroxynitrite (ONOO–) by
superoxide radicals (O–) produced during ischaemia and reperfusion. Alternatively, spasm
may be secondary to lipid peroxidation of the vessel wall by various oxidant species,
including superoxide and peroxynitrite. Other authors have proposed a role for
endothelin.130 131 Vasospasm tends to be worst in patients with the largest
Page 267
Fig 7.18 Effect of Hunt and Hess grade of aneurysmal subarachnoid haemorrhage on (a) pressure
autoregulation and (b) cerebral blood flow (CBF) reactivity to changes in PaCO2.
Page 268
amounts of subarachnoid blood,132 suggesting that the blood itself contributes to the
phenomenon. Vasospasm is associated with parallel reductions in rCBF and CMRo2 in the
regions affected.
The clinical impact of late vasospasm has been substantially modified by the routine use
of Ca2+ channel blockers such as nimodipine133 and by the routine use of hypertensive
hypervolaemic haemodilution134 (triple H therapy). Triple H (3H) therapy involves the use
of colloid administration (with venesection if needed) to increase filling pressures and
reduce haematocrit to 30–35%. If moderate hypertension is not achieved with volume
loading, vasopressors and inotropes are used to maintain mean blood pressures as high as
120–140 mm Hg. The hypertensive element of this therapy protects non-autoregulating
portions of the cerebral vasculature from hypoperfusion, whereas the haemodilution
improves rheological characteristics of blood and facilitates flow through vessels whose
calibre is reduced by spasm. Such interventions have been shown to produce clinically
useful improvements in rCBF in regions of ischaemia.135
Mechanisms in rCBF control

Some of the mechanisms involved in cerebrovascular control are shown in Fig. 7.19.
Several of these have been referred to earlier. In addition, the level of free Ca2+ is
important in determining vascular tone, and arachidonate metabolism can produce
prostanoids that are either vasodilators (for example, prostacyclin, PGI2) or
vasoconstrictive (for example, thromboxane, TxA2). Endothelin (ET), produced by
endothelin converting enzyme (ECE) in endothelial cells, balances the vasodilator effects of
nitric oxide in a tonic matter by exerting its influences at ETA receptors in the vascular
smooth muscle.
Nitric oxide in the regulation of cerebral haemodynamics56
Recent interest has focused on the role of NO in the control of cerebral haemodynamics.
NO is synthesised in the brain from the amino acid L-arginine by the constitutive form of
the enzyme nitric oxide synthase (NOS). This form of the enzyme is calmodulin-dependent
and requires Ca2+ and tetrahydrobiopterin for its activity; it differs from the inducible form
of the enzyme, which is present in mononuclear blood cells and is activated by cytokines.
Under basal conditions, endothelial cells synthesise NO which diffuses into the muscular
layer and, via a cGMP-mediated mechanism, produces relaxation of vessels. There is strong
evidence to suggest that NO exerts a tonic dilatory influence on cerebral vessels. It is
important to emphasise that data on NO obtained from peripheral vessels cannot always be
translated to the cerebral vasculature; for example, some
Page 269
of the endothelium derived relaxant factor (EDRF) activity in cerebral vessels may be
caused by compounds other than NO. There is growing evidence that carbon monoxide
(CO), produced by haem oxygenase, may be responsible for significant cerebral
vasodilatation, especially when NO production is reduced.136
Nitric oxide plays an important role in cerebrovascular responses to functional activation,
excitatory amino acids, hypercapnia, ischaemia, and SAH. Further, NO may play an
important part in mediating the vasodilatation produced by volatile anaesthetic agents,56
although other mechanisms, including a direct effect on the vessel wall, cannot be
excluded.
Fig 7.19 Mechanisms involved in the regulation of regional cerebral blood flow (rCBF) in health and disease.
The diagram shows a resistance vessel in the brain in the vicinity of a neuron (N) and an astrocyte (A). Other
abbreviations: E, endothelium; M, muscular layer; PGs=prostaglandins; TxA2, thromboxane A2; ET,
endothelin; ECE, endothelin converting enzyme; ETA, ETA receptor; NO, nitric oxide; CO, carbon monoxide;
DA, dopamine. The inset box shows the detail of the vessel wall and adjacent glial cell process. See text for
details.
Page 270
Neurogenic flow–metabolism coupling
Although the last 10 years have focused on flow–metabolism coupling being effected by
a diffusable extracellular mediator, there is now accumulating evidence to suggest that
dopaminergic neurons may play a major part in such events, and in addition may control
blood–brain barrier permeability.137
1 Edvinsson L, Mackenzie ET, McCulloch J. The aged brain. In: Cerebral blood flow and metabolism.
New York: Raven Press, 1993:647–60.
2 Alpers BJ, Berry RG, Paddison RM. Anatomical studies in the circle of Willis in normal brains. Arch
Neurol Psychiatry 1959;81:409–18.
3 Collins RC. Intracortical localization of 2-deoxyglucose metabolism on-off metabolic columns. In:
Passonneau JV, Hawkins RA, Lust WD, Welsh FA, eds, Cerebral metabolism and neural function.
Baltimore: Williams &Wilkins, 1980:338–51.
4 Welsh FA. Regional evaluation of ischaemic metabolic alterations. J Cereb Blood Flow Metab
1984;4:309–16.
5 Pulsinelli WA, Duffy TE. Local cerebral glucose metabolism during controlled hypoxaemia in rats.
Science 1979;204:626–9.
6 Duning HS, Wolff HG. The relative vascularity of various parts of the central and peripheral nervous
system in the cat and its relation to function. J Comp Neurol 1937;67:433–50.
7 Sokoloff L, Reivich M, Kennedy C, et al. The [14C]-deoxyglucose method for measurement of local
cerebral glucose utilization: Theory, procedure, and normal values in the conscious and anesthetized
albino rat. J Neurochem 1977;28:897–916.
8 Göbel U, Theilen H, Kuschinsky W. Congruence of total and perfused capillary network in rat brains.
Circ Res 1990;66:271–81.
9 Kuschinsky W, Paulson OB. Capillary circulation in the brain. [Review]. Cerebrovasc Brain Metabol Rev
1992;4:261–86.
10 Beards SC, Yule S, Kassner A, Jackson A. Anatomical variation of cerebral venous drainage: the
theoretical effect on jugular bulb blood samples. Anaesthesia 1998;53:627–33.
11 Shapiro K, Marmarou A, Shulman K. Characterization of clinical CSF dynamics and neural axis
compliance using the pressure-volume index. I. The normal pressure–volume index. Ann Neurol
1980;7:508–13.
12 Chesnut RM. Hyperventilation in traumatic brain injury: Friend or foe? Crit Care Med 1997;25:1275–8.
13 Todd MM, Weeks J. Comparative effect of propofol, pentobarbital and isoflurane on cerebral blood flow
and volume. J Neurosurg Anesthesiol 1996;8:296–303.
14 Janzer RC, Raff MC. Astrocytes induce blood–brain barrier properties in endothelial cells. Nature
1987;325:253–7.
15 Oldendorf WH. Lipid solubility and drug penetration of the blood brain barrier. Proc Soc Exp Biol Med
1974;147:813–16.
16 Oldendorf WH, Cornford ME, Brown WJ. The large apparent work capability of the blood brain barrier.
A study of the mitochondrial content of capillary endothelial cells in brain and other tissues of the rat. Ann
Neurol 1977;1:409–17.
17 Kety SS and Schmidt CF. The determination of cerebral blood flow in man by use of nitrous oxide in low
concentrations. Am J Physiol 1945;143:53–66.
18 Ingvar DH, Lassen NA. Quantitative determination of regional cerebral blood flow in man. Lancet
1961;ii:806–7.
19 Prough DS, Stump DA, Roy RC, et al. Response of cerebral blood flow to changes in carbon dioxide
tension during hypothermic cardiopulmonary bypass. Anesthesiology 1986;64:576–81.
Page 271
20 Mallett BL, Veall N. Investigation of cerebral blood flow in hypertension, using radioactive xenon
inhalation and extracranial recording. Lancet 1963;i:1081–2.
21 Cruz J, Raps EC, Hoffstad OJ, Jaggi JL, Gennarelli TA. Cerebral oxygenation monitoring [Review]. Crit
Care Med 1993;21:1242–6.
22 Bodenham A, Webster NR. New practical beside procedures on the intensive care unit. Balliere’s Clin
Anesthesiol 1992;6:425–41.
23 Mayberg TS, Lam AM. Management of central nervous system trauma. Curr Opin Anaesthesiol
1993;6:764–71.
24 Faught E. Current role of electroencephalography in cerebral ischaemia [Review]. Stroke 1993;24:609–
13.
25 Zentner J, Schramm J. Monitoring the central nervous system. Curr Opin Anaesthesiol 1993; 6:784–90.
26 Wyatt JS. Noninvasive assessment of cerebral oxidative metabolism in the human newborn [Review]. J R
Coll Phys Lond 1994;28:126–32.
27 Bolognese P, Miller JI, Heger IM, Milhorat TH. Laser-Doppler flowmetry in neurosurgery [Review]. J
Neurosurg Anesthesiol 1993;5:151–8.
28 Bishop CCR, Powell S, Rutt D, et al. Transcranial Doppler measurement of the middle cerebral flow
velocity: A validation study. Stroke 1986; 17:913–15.
29 Dahl A, Russell D, Nyberg-Hanson R, et al. A comparison of regional cerebral blood flow and middle
cerebral artery flow velocities:simultaneous measurements in healthy subjects. J Cereb Blood Flow Metab
1992;12:1049–54.
30 Matta BBF, Lam AM. Isoflurane and desflurane do not dilate the middle cerebral artery appreciably
(Abstract). Br J Anaesth 1995;74:486–7P.
31 Menon DK. Monitoring the central nervous system. Curr Anaesth Crit Care 1997;6:254–63.
32 Czosnyka M, Matta BF, Smielewski P, et al. Cerebral perfusion pressure in head-injured patients: a
noninvasive assessment using transcranial Doppler ultrasonography. J Neurosurg 1998;88:802–8.
33 Metz C, Holzschuh M, Bein T, et al. Monitoring of cerebral oxygen metabolism in the jugular bulb:
Reliability of unilateral measurements in severe head injury. J Cereb Blood Flow Metab 1998;18:332–43.
34 Cruz J. The first decade of continuous monitoring of jugular bulb oxyhemoglobin saturation: management
strategies and clinical outcome. Crit Care Med 1998;26344–51.
35 Shienberg M, Kanter MJ, Robertson CS, et al. Continuous monitoring of jugular venous oxygen
saturation in head-injured patients. J Neurosurg 1992;76:212–17.
36 Fox PT, Raichle ME, Mintun MA, Dence C. Nonoxidative glucose consumption during focal physiologic
neural activity. Science 1988;241:462–1.
37 Tsacopoulos M, Magistretti PJ. Metabolic coupling between glia and neurons. J Neurosci 1996;16:877–
85.
38 Pellerin L, Magistretti PJ. Glutamate uptake into astrocytes stimulates anaerobic glycolysis: a mechanism
coupling neuronal activity to glucose utilization. Proc Natl Acad Sci USA 1994;91:10625–9.
39 Edvinsson L, Mackenzie ET, McCulloch J. Perivascular nerve fibres in brain vessels. In: Cerebral blood
flow and metabolism. New York: Raven Press, 1993:57–91.
40 Kontos HA. Nitric oxide and nitrosothiols in cerebrovascular and neuronal regulation [Review]. Stroke
1993;24:1155–8.
41 Miller JD, Stanek A, Langfitt TW. Concepts of cerebral perfusion pressure and vascular compression
during intracranial hypertension. Prog Brain Res 1972;35:411–32.
42 Miller JD, Stanek AE, Langfitt TW. Cerebral blood flow regulation during experimental brain
compression. J Neurosurg 1973;39:186–96.
43 Fitch W, MacKenzie ET, Harper AM. Effects of decreasing arterial blood pressure on cerebral blood flow
in the baboon: Influence of the sympathetic nervous system. Circ Res 1975;37:550–7.
44 Heistad DD, Marcus ML, Sandberg S, Abboud FM. Effect of sympathetic nerve stimulation on cerebral
blood flow and on large cerebral arteries of dogs. Circ Res 1977;41:342–50.
45 Thomas D, Bannister RG. Preservation of autoregulation of cerebral blood flow in autonomic failure. J
Neurol Sci 1980;44:205–12.
Page 272
46 Strandgaard S, Olesen J, Skinhoj E, Jassen NA. Autoregulation of brain circulation in severe arterial
hypertension. BM J 1973;1:507–10.
47 Thompson BG, Pluta RM, Girton ME, Oldfield EH. Nitric oxide mediation of chemoregulation but not
autoregulation of cerebral blood flow in primates. J Neurosurg 1996;84:71–8.
48 Bentsen N, Larsen B, Lassen NA. Chronically impaired autoregulation of cerebral blood flow in long-
term diabetes. Stroke 1975;6:497–502.
49 Strandgaard S. Autoregulation of CBF in hypertensives. The modifying influence of prolonged
antihypertensive treatment on the tolerance to acute, drug induced hypotension. Circulation 1976;53:720–
7.
50 Michenfelder JD, Theye RA. Canine systemic and cerebral effects of hypotension induced by
hemorrhage, trimethaphan, halothane, or nitroprusside. Anesthesiology 1977; 46:188–95.
51 Aaslid R, Lindegaard K-F, SortebergW, Nornes H. Cerebral autoregulation dynamics in humans. Stroke
1989;20:45–52.
52 Davis DH, Sundt TM Jr. Relationship of cerebral blood flow to cardiac output, mean arterial pressure,
blood volume, and alpha and beta blockade in cats. J Neurosurg 1980;52:745–54.
53 Koehler RC, Traystman RJ. Bicarbonate ion modulation of cerebral blood flow during hypoxia and
hypercapnia. Am J Physiol 1982;243:H33–40.
54 Stringer WA, Hasso AN, Thompson JR, Hinshaw DB, Jordan KG. Hyperventilation-induced cerebral
ischaemia in patients with acute brain lesions: demonstration by Xenon-enhanced CT. Am J Neuroradiol
1993;14:465–84.
55 Pickard JD, MacKenzie ET. Inhibition of prostaglandin synthesis and the response of baboon cerebral
circulation to carbon dioxide. Nature (New Biol) 1973;245:187–8.
56 Maktabi MA. Role of nitric oxide in regulation of cerebral circulation in health and disease. Curr Opinion
Anaesthesiol 1993;6:799–83.
57 Okamoto H, Hudetz AG, Roman RJ, et al. JP Neuronal NOS-derived NO plays permissive role in cerebral
blood flow response to hypercapnia. Am J Physiol 1997;272;H559–66.
58 Grubb RL Jr, Raichle ME, Eichling JO, Ter-Pogossian MM. The effects of changes in PaCO2 on cerebral
blood volume, blood flow and vascular mean transit time. Stroke 1974;5:630–9.
59 Kosteljanetz M. Acute head injury: pressure–volume relations and cerebrospinal fluid dynamics.
Neurosurgery 1986;18:17–24.
60 McDowall DG. Interrelationships between blood oxygen tension and cerebral blood flow. In: Payne JP,
Hill DW, eds, Oxygen measurements in blood and tissues. London: Churchill, 1966:205–14.
61 Gupta AK, Menon DK, Czosnyka M, et al. Thresholds for hypoxic cerebral vasodilatation in volunteers.
Anesth Analg 1997;85:817–20.
62 Krasney JA, Jensen JB, Lassen NA. Cerebral blood flow does not adapt to sustained hypoxia. J Cereb
Blood Flow Metab 1990;10:759–64.
63 Krasney JA, McDonald B, Matalon S. Regional circulatory responses to 96 hours of hypoxia in conscious
sheep. Respir Physiol 1984;57:73–88.
64 Krasney JA, Hajduczok G, Miki K, Matalon S. Peripheral circulatory responses to 96 hours of eucapnic
hypoxia in conscious sheep. Respir Physiol 1985;59:197–211.
65 Purves MJ. The physiology of the cerebral circulation. Cambridge: Cambridge University Press, 1972.
66 Adams HP. Prevention of brain ischaemia after aneurysmal subarachnoid haemorrhage. Neurol Clin
1992;10:251–68.
67 Christensen MS, Hoedt-Rasmussen K, Lassen NA. Cerebral vasodilatarion by halothane anesthesia in
man and its potentiation by hypotension and hypercapnia. Br J Anaesth 1967;39:927–31.
68 Wollman H, Alexander SC, Cohen PJ, Chase PE, Melman E, Behar MG. Cerebral circulation of man
during halothane anesthesia. Effects of hypocarbia and of d-tubocurarine. Anesthesiology 1964;25:180–4.
69 Jennett WB, Barker J, Fitch W, McDowall DG. Effects of anaesthesia on intracranial pressure in patients
with space-occupying lesions. Lancet 1969;i:61–4.
Page 273
70 Murphy FL, Kennell EM, Johnstone RE. The effects of enflurane, isoflurane, and halothane on cerebral
blood flow and metabolism in man. Abstracts of the Meeting of The American Society of
Anesthesiologists 1974:61–2.
71 Hansen TD, Warner DS, Todd MM, Vust LJ, Trawick DC. Distribution of cerebral blood flow during
halothane versus isoflurane anesthesia in rats. Anesthesiology 1988;69:332–7.
72 Hansen TD, Warner DS, Todd MM, Vust LJ. Effects of nitrous oxide and volatile anaesthetics on cerebral
blood flow. Br J Anaesth 1989;63:290–5.
73 Scheller MS, Todd MM, Drummond JC. Isoflurane, halothane and regional cerebral blood flow at various
levels of PaCO2 in rabbits. Anesthesiology 1986;64:598–604.
74 Drummond JC, Todd MM. The response of the feline cerebral circulation to PaCO2 during anesthesia
with isoflurane and halothane and during sedation with nitrous oxide. Anesthesiology 1985;62:268–73.
75 Okamoto H, Meng W, Ma J et al. Isoflurane-induced cerebral hyperaemia in neuronal nitric oxide
synthase gene deficient mice. Anesthesiology 1997;86:875–84.
76 Kuramoto T, Oshita S, Takeshita H, Ishikawa T. Modification of the relationship between cerebral
metabolism, blood flow, and the EEG by stimulation during anesthesia in the dog. Anesthesiology
1979;51:211–17.
77 Hansen TD, Warner DS, Todd MM, Vust LJ. The role of cerebral metabolism in determining the local
cerebral blood flow effects of volatile anaesthetics: evidence for persistent flow–metabolism coupling. J
Cereb Blood Flow Metab 1989;9:323–8.
78 Maekawa T, Tommasino C, Shapiro HM, Kiefer-Goodman J, Kohlenberger RW. Local cerebral blood
flow and glucose utilisation during isoflurane anesthesia in the rat. Anesthesiology 1986;65:144–51.
79 Sakabe T, Kuramoto T, Kumagae S, Takeshita H. Cerebral responses to the addition of nitrous oxide to
halothane in man. Br J Anaesth 1976;48:957–62.
80 Lam AM, Mayberg TS, Eng CC, Cooper JO, Bachenberg LK, Mathisen TL. Nitrous oxide-isoflurane
anesthesia causes more cerebral vasodilation than an equipotent dose of isoflurane in humans. Anesth
Analg 1994;78:462–8.
81 Reasoner D, Warner DS, Todd MM, McAllister A. Effects of nitrous oxide on cerebral metabolic rate in
rats anaesthetized with isoflurane. Br J Anaesth 1990;65:210–15.
82 Henriksen HT, Jorgensen PB. The effect of nitrous oxide on intracranial pressure in patients with
intracranial disorders. Br J Anaesth 1973;45:486–92.
83 Moss E, McDowall DG. ICP increases with 50% nitrous oxide in oxygen in severe head injuries with
controlled ventilation. Br J Anaesth 1979;51:757–61.
84 Kaieda R, Todd MM, Warner DS. The effects of anaesthetics and PaCO2 on the cerebrovascular,
metabolic, and electroencephalographic responses to nitrous oxide in the rabbit. Anesth Analg
1989;68:135–43.
85 Phirman JR, Shapiro HM. Modification of nitrous oxide induced intracranial hypertension by prior
induction of anaesthesia. Anesthesiology 1977;46:150–1.
86 Young WL. Effects of desflurane on the central nervous system. Anesth Analg 1992;75(suppl 4):32–7.
87 Scheller MS, Tateishi A, Drummond JC, Zornow MH. The effects of sevoflurane on cerebral blood flow,
cerebral metabolic rate for oxygen, intracranial pressure and the electroencephalogram are similar to those
of isoflurane in the rabbit. Anesthesiology 1988;68:548–51.
88 Lutz LJ, Milde JH, Milde LN. The cerebral functional, metabolic, and hemodynamic effects of desflurane
in dogs. Anesthesiology 1990;73:125–31.
89 Muzzi DA, Losasso TJ, Dietz NM, Faust RJ, Cucchiara RF, Milde LN. The effect of desflurane and
isoflurane on cerebrospinal fluid pressure in humans with supratentorial mass lesions. Anesthesiology
1992;76:720–4.
90 Rampil IJ, Lockhart SH, Eger El, Weiskopf RB. Human EEG dose response to desflurane. Anesthesiology
1990;73:A1218.
91 Tonner PH, Scholz J, Krause T, Paris A, von Knobelsdroff G, Schulte an Esch J. Administration of
sufentanil and nitrous oxide blunts cardiovascular responses to desflurane but does not prevent an increase
in middle cerebral artery flow velocity. Eur J Anaesthesiol 1997;14:389–96.
Page 274
92 Bundgaard H, von Oettingen G, Larsen KM, et al. Effects of sevoflurane on intracranial pressure, cerebral
blood flow and cerebral metabolism. A dose–response study in patients subjected to craniotomy for
cerebral tumours. Acta Anaesthesiol Scand 1998;42:621–7.
93 Heath KJ, Gupta S, Matta BF. The effects of sevoflurane on cerebral heamodynamics during propofol
anesthesia. Anesth Analg 1997;85:1284–7.
94 Gupta S, Heath K, Matta BF. Effect of incremental doses of sevoflurane on cerebral pressure
autoregulation in humans. Br J Anaesth 1997;79:469–72.
95 Kassel NF, Hitchon PW, Gerk MK, Sokoll MD, Hill TR. Alterations in cerebral blood flow, oxygen
metabolism, and electrical activity produced by high dose sodium thiopental. Neurosurgery 1980;7:598–
603.
96 Milde LN, Milde JH, Michenfelder JD. Cerebral functional, metabolic, and haemodynamic effects of
etomidate in dogs. Anesthesiology 1985;63:371–7.
97 Ramani R, Todd MM, Warner DS. The cerebrovascular, metabolic and electroencephalographic effects of
propofol in the rabbit – a dose response study. J Neurosurg Anesth 1992;4:110–19.
98 Van Hemelrijck J, Van Aken H, Plets C, Goffin J, Vermaut G. The effects of propofol on intracranial
pressure and cerebral perfusion pressure in patients with brain tumours. Acta Anaesthesiol Belg
1989;40:95–100.
99 Weeks J, Todd MM, Warner DS, Katz J. The influence of halothane, isoflurane, and pentobarbital on
cerebral plasma volume in hypocapnic and normocapnic rats. Anesthesiology 1990;73:461–6.
100 Fox J, Gelb AW, Enns J, Murkin JM, Farrar JK, Manninen PH. The responsiveness of cerebral blood
flow to changes in arterial carbon dioxide is maintained during propofol–nitrous oxide anesthesia in
humans. Anesthesiology 1992;77:453–6.
101 Carlsson C, Smith DS, Keykhah MM, Englebach I, Harp JR. The effects of high dose fentanyl on cerebral
circulation and metabolism in rats. Anesthesiology 1982;57:375–80.
102 Keykhah MM, Smith DS, Carlsson C, Safo Y, Englebach I, Harp JR. Influence on sufentanil on cerebral
metabolism and circulation in the rat. Anesthesiology 1985;63:274–80.
103 McPherson RW, Krempasanka E, Eimerl D, Traystman RJ. Effects of alfentanil on cerebral vascular
reactivity in dogs. Br J Anaesth 1985;57:1232–8.
104 Murkin JM, Ferrar JK, Tweed WA, McKenzie FN, Guiraudon G. Cerebral autoregulation and
flow/metabolism coupling during cardiopulmonary bypass: the influence of PaCO2. Anesth Analg
1987;66:825–32.
105 Sperry RJ, Bailey PL, Reichman MV, Peterson PB, Pace NL. Fentanyl and sufentanil increase intracranial
pressure in head trauma patients. Anesthesiology 1992;77:416–20.
106 Åkeson J, Böjrkman S, Messeter K, Rosén I, Heifer M. Cerebral pharmacodynamics of anaesthetic and
subanaesthetic doses of ketamine in the normoventilated pig. Acta Anaesthesiol Scand 1993;37:211–18.
107 Davis DW, Mans AM, Biebuyck JF, Hawkins RA. The influence of ketamine on regional brain glucose
use. Anesthesiology 1988;69:199–205.
108 Menon DK, Burdett NG, CarpenterTA, Hall LD. Functional MRI of ketamine-induced changes in rCBF:
An effect at the NMDA receptor? [Abstract], Br J Anaesth 1993;7:767.
109 Forster A, Juge O, Morel D. Effects of midazolam on cerebral blood flow in human volunteers.
Anesthesiology 1982;56:453–5.
110 Fleischer JE, Milde JH, Moyer TP, Michenfelder JD. Cerebral effects of high-dose midazolam and
subsequent reversal with RO 15–1788 in dogs. Anesthesiology 1988;68:234–42.
111 Zornow MH, Fleischer JE, Scheller MS et al. Dexmedetomidine and alpha 2-adrenergic agonist,
decreases cerebral blood flow in the isoflurane anesthetised dog. Anesth Analg 1990; 70:624–30.
112 McPherson RW, Koehler RC, Kirsch JR, Traystman RJ. Intraventricular dexmedetomidine decreases
cerebral blood flow during normoxia and hypoxia in dogs. Anesth Analg 1997;84:139–47.
113 Ducey JP, Deppe AS, Foley FT. A comparison of the effects of suxamethonium, atracurium and
vecuronium on intracranial haemodynamics in swine. Anaesth Intensive Care 1989;17:448–55.
Page 275
114Kovarik WD, Lam AM, Slee TA, Mathisen TL. The effect of succinylcholine on intracranial pressure,
cerebral blood flow velocity and electroencephalogram in patients with neurologic disorders [Abstract].
Anesthesiology 1991;75:207.
115Stirt AJ, Grosslight KR, Bedford RF, Vollmer D. ‘‘Defasciculation” with metocurine prevents
succinylcholine-induced increases in intracranial pressure. Anesthesiology 1987;67:50–3.
Siesjo BK. Pathophysiology and treatment of focal cerebral ischaemia. Part I: Pathophysiology. J
116
Neurosurg 1992;77:169–84.
117Robertson CS, Contant CF, Gokaslan ZL, Narayan RK, Grossman RG. Cerebral blood flow,
arteriovenous oxygen difference, and outcome in head injured patients. J Neural Neurosurg Psychiatry
1992; 55:594–603.
118Bouma GJ, Muizelaar JP, Stringer WA, Choi SC, Fatouros P, Young HF. Ultra early evaluation of
regional cerebral blood flow in severely head-injured patients using Xenon-enhanced computerized
tomography. J Neurosurg 1992;77:360–9.
119Martin NA, Patwardhan RV, Alexander MJ, et al. Characterization of cerebral hemodynamic phases
following severe head trauma: hypoperfusion, hyperaemia and vasospasm. J Neurosurg 1997;87:9–19.
120Martin NA, Doberstein C, Zane C, Caron MJ, Thomas K, Becker DP. Posttraumatic cerebral arterial
spasm: transcranial Doppler ultrasound, cerebral blood flow and angiographic findings. J Neurosurg
1992;77:575–83.
121Marion DW, Darby J, Yonas H. Acute regional cerebral blood flow changes caused by severe head
injuries. J Neurosurg 1991;74:407–14.
122McLaughlin MR, Marion DW. Cerebral blood flow within and around cerebral contusions. J Neurosurg
1996;85:871–6.
123Menon DK, Minhas P, Herrod NJ, et al. Cerebral ischaemia associated with hyperventilation: a PET
study. Anesthesiology 1997;87:A176.
124Chan KH, Dearden NM, Miller JD, Andrews PJ, Midgley S. Multimodality monitoring as a guide to
treatment of intracranial hypertension after severe brain injury. Neurosurgery 1993;32:547–52.
125Feldman Z, Kanter MJ, Robertson CS, et al. Effect of head elevation on intracranial pressure, cerebral
perfusion pressure and cerebral blood flow in head injured patients. J Neurosurg 1992;76:207–11.
126Lassen NA, Agnoli A. The upper limit of autoregulation of cerebral blood flow on the pathogenesis of
hypertensive encephalopathy. Scand J Clin Lab Invest 1973;30:113–16.
127Voldby B, Enevoldsen EM, Jensen FT. Cerebrovascular reactivity in patients with ruptured intracranial
aneurysm. J Neurosurg 1985;62:59–67.
128Pickard JD, Matheson M, Patterson J, Wyper D. Prediction of late ischemic complications after cerebral
aneurysm surgery by the intraoperative measurement of cerebral blood flow. J Neurosurg 1980;53:305–8.
129Kassell NF, Peerless SJ, Durward QJ, Beck DW, Drake CG, Adams HP. Treatment of ischaemic deficits
from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery
1982;11:337–43.
Yamaura I, Tani E, MaedaY, Minami N, Shindo H. Endothelin-1 of canine basilar artery in vasospasm. J
130
Neurosurg 1992;76:99–105.
131Clozel M, Watanabe H. BQ-123, a peptidic endothelin ETA receptor antagonist, prevents the early
cerebral vasospasm following subarachnoid hemorrhage after intracisternal but not intravenous injection.
Life Sci 1993;52:825–34.
132Jakobsen M, Skjødt T, Enevoldsen E. Cerebral blood flow and metabolism following subarachnoid
hemorrhage: effect of subarachnoid blood. Acta Neurol Scand 1991;8:226–33.
133Pickard JD, Murray GD, Illingworth R, et al. Effect of oral nimodipine on cerebral infarction and
outcome after subarachnoid hemorrhage: British aneurysm nimodipine trial. BM J 1989;298:636–42.
134OrigitanoTC, WascherTM, Reichman OH, Anderson DE. Sustained increased cerebral blood flow with
prophylactic hypervolemic haemodilution (“Triple-H” Therapy) after subarachnoid haemorrhage.
Neurosurgery 1990;27:729–38.
Page 276
135Darby JM, Yonas H, Marks EC, Durham S, Snyder RW, Nemoto EM. Acute cerebral blood flow
response to dopamine-induced hypertension after subarachnoid hemorrhage. J Neurosurg 1994;80:857–
64.
136Zakhary R, Gaine SP, Dinennan JL, et al. Heme oxygenase 2: Endothelial and neuronal localisation and
role in endothelial-dependent relaxation. Proc Natl Acad Sci USA 1996;93:795–8.
Iadecola C. Neurogenic control of the cerebral microcirculation: is dopamine minding the store? Nature
137
Neurosci 1998;1:363–4.
138Yonas H, Gur D, Latchaw RE, Wolfson SK. Xenon computed tomographic blood flow mapping. In:
Wood JH, ed, Cerebral blood flow. Physiologic and clinical aspects. New York: McGraw-Hill, 1987:220–
42.
Miles KA, Hayball M, Dixon AK. Colour perfusion imaging: a new application of computed tomography.
139
Lancet 1991;337:643–5.
140Herscovitch MD, Powers WJ. measurement of regional cerebral blood flow by positron emission
tomography. In: Wood JH, ed, Cerebral blood flow. Physiologic and clinical aspects. New York:
McGrawHill, 1987:257–71.
141Holman BL, Hill TC. Perfusion imaging with single-photon emission computed tomography. In: Wood
JH, ed, Cerebral blood flow. Physiologic and clinical aspects. New York: McGraw-Hill, 1987:243–56.
142Elwell CE, Owen-Reece H, Cope M, et al. Measurement of adult cerebral haemodynamics using near
infrared spectroscopy. Acta Neurochir 1993;suppl 59:74–80.
143Farrar JK. Hydrogen clearance technique. In: Wood JH, ed, Cerebral blood flow. Physiologic and clinical
aspects. New York: McGraw-Hill, 1987:275–87.
144Ginsburg MD. Autoradiographic measurement of local cerebral blood flow. In: Wood JH, ed, Cerebral
blood flow. Physiologic and clinical aspects. New York: McGraw-Hill, 1987:299–308.
Warner DS, Kassell NF, Boarini DJ. Microsphere cerebral blood flow determination. In: Wood JH, ed,
145
Cerebral blood flow. Physiologic and clinical aspects. New York: McGraw-Hill, 1987:288–98.
146Pritchard JW, Rosen BR. Functional study of the brain by NMR. J Cereb Blood Flow Metab
1994;14:365–72.
147 Newell DW, Aaslid R, eds, Transcranial Doppler. New York: Raven Press, 1992.
Menon DK, Peden CJ, Hall AS, Sargentoni J, Whitwam JG. Magnetic resonsance for the anaesthetist:
148
Part I: Principles, applications, safety aspects. Anaesthesia 1992;47:240–55.
149Madsen JB, Cold GE, Eriksen HO, Eskesen V, Blatt-Lyon B. CBF and CMRo2 during craniotomy for
small supratentorial cerebral tumours in enflurane anaesthesia. A dose–response study. Acta Anaesthesiol
Scand 1986;30:633–6.
150Madsen JB, Cold GE, Hansen ES, Eardrum B. Cerebral blood flow, cerebral metabolic rate of oxygen and
relative CO2-reactivity during craniotomy for supratentorial cerebral tumours in halothane anaesthesia. A
dose–response study. Acta Anaesthesiol Scand 1987;31:454–7.
151Madsen JB, Cold GE, Hansen ES, Eardrum B. The effect of isoflurane on cerebral blood flow and
metabolism in humans during craniotomy for small supratentorial cerebral tumours. Anesthesiology
1987;66:332–6.
152Kaieda R, Todd MM, Warner DS. The effects of anaesthetics and PaCO2 on the cerobrovascular,
metabolic, and electroencephalographic responses to nitrous oxide in the rabbit. Anesth Analg
1989;68:135–43.
153Todd MM, Drummond JC. A comparison of the cerebrovascular and metabolic effects of halothane and
isoflurane in the cat. Anesthesiology 1984;60:276–82.
154Eintrei C, LeszniewskiW, Carlsson C. Local application of 133Xenon for measurement of regional
cerebral blood flow (rCBF) during halothane, enflurane and isoflurane anesthesia in humans.
155Scheller MS, Todd MM, Drummond JC. Isoflurane, halothane and regional cerbral blood flow at various
levels of PaCO2 in rabbits. Anesthesiology 1986;64:598–604.
156Michenfelder JD, SundtTM, Fode N, Sharborough FW. Isoflurane when compared to enflurane and
halothane decreases the frequency of cerebral ischemia during carotid endarterectomy. Anesthesiology
1987;67:336–40.
Page 277
Gelman S, Fowler KC, Smith LR. Regional blood flow during isoflurane and halothane anaesthesia.
157
AnesthAnalg 1984;63:557–65.
158Stullken EH Jr, Milde JH, Michelfelder JD, Tinker JH. The nonlinear responses of cerebral metabolism to
low concentrations of halothane, enflurane, isoflurane and thiopental. Anesthesiology 1977;46:28–34.
159HansenTD, Warner DS, Todd MM, Vust LJ, Trawick DC. Distribution of cerebral blood flow during
halothane versus isoflurane anesthesia in rats. Anesthesiology 1988;69:332–7.
160Boarini DJ, Kassell NJ, Coesler HC, Butler M, Sokoll MD. Comparison of systemic and cerbrovascular
effects of isoflurane and halothane. Neurosurgery 1984;15:400–9.
Page 278
8: Renal, splanchnic, skin, and muscle

circulations
NGUYEN D KIEN, JOHN A REITAN
Renal circulation
The major functions of the kidneys include:
1 Excretion of the end products of systemic metabolism while retaining essential nutrients.
2 Regulation of the volume and composition of body fluids.
3 Production of endocrine substances, including renin, prostaglandins, and kinins, which
are important for the control of the pressure, volume, and flow of blood.
Anatomy
The kidneys are bilateral, bean shaped organs, which lie in a retro-peritoneal position on
either side of the vertebral column beneath the diaphragm. An adult human kidney weighs
between 115 and 170 g with the upper and lower borders situated between the twelfth
thoracic and third lumbar vertebrae. The right kidney is slightly more caudal in position
because of the liver. Located on the concave border facing the vertebral column is an
indentation called the hilus through which the ureter, blood and lymph vessels, and a nerve
plexus pass into the renal sinus. The cut surface of a bisected kidney reveals an outer region
called the cortex and an inner region called the medulla. The cortex is divided into the outer
cortical layer and the inner juxtamedullary layer. The medulla is made up of 4–18 conical
pyramids. The base of each pyramid faces the cortex and the apex (called the papilla)
extends into the renal sinus and is covered by a funnel shaped calyx. It is through these
calyces that the urine is drained from the kidney into the pelvis and the ureter.
The kidney is innervated by the renal plexus of the sympathetic division of the
autonomic nervous system. The nerve fibres enter the kidney
Page 279
alongside the arterial vessels. Their branches supply the renal vasculature throughout the
cortex and the outer region of the medulla.
The functional unit of the kidney is the nephron. There are approximately 106 nephrons
in each kidney. Each nephron consists of a glomerulus, a capillary network through which
plasma is filtered, and the uriniferous tubule, a long and cylindrical tube where filtered
plasma is modified into urine.
Glomerulus
The glomerulus is composed of a capillary network that invaginates into the dilated blind
end of the nephron called Bowman’s capsule.1 The glomerulus is the filtration barrier
between blood and urine that is responsible for the production of an ultrafiltrate of plasma.
The glomerular capillaries are covered by a thin fenestrated endothelium. The endothelial
cells form an initial barrier to the passage of blood constituents from the capillary to
Bowman’s capsule. Beneath the endothelium is the basement membrane of the glomerulus.
This extracellular membrane, consisting of fibrils in a glycoprotein matrix, serves as a
retaining wall of large sized proteins. The largest cells of the glomerulus are the visceral
epithelial cells, which are partially responsible for the synthesis and maintenance of the
glomerular basement membrane. The mesangial cells are similar to the visceral epithelial
cells. These cells contain filaments and are capable of phagocytosis. They produce
prostaglandins and appear to have a role in the counter regulation of the effect of
vasoconstrictors.
Juxtaglomerular apparatus
The Juxtaglomerular apparatus is an area of the nephron where the distal tubal comes
into contact with the arterioles.2 It is composed of:
z Juxtaglomerular granular cells

z extraglomerular mesangium
z the macula densa.
The Juxtaglomerular granular cells are modified smooth muscle cells which produce,
store, and release renin. The extraglomerular mesangium is connected with the
intraglomerular mesangium, and is composed of cells that are similar to the mesangial cells.
The extraglomerular mesangium has been suggested as a functional link of the mesangium,
glomerular arterioles, and macula densa.
The Juxtaglomerular apparatus is a major structural component of the renin–angiotensin
system. It is involved in the autoregulation of renal blood flow and glomerular filtration
rate. Changes in sodium or chloride concentration at the macula densa, or alterations in the
volume and stretch of the afferent arteriole, are thought to affect the control of renin
release.
Page 280
Renal tubule
The renal tubule consists of the proximal tubule, the loop of Henle, and the distal tubule.
Bowman’s capsule opens into the first section of the renal tubule, called the proximal
tubule, which lies within the cortex. The proximal tubule contains a large number of
lysosomes that are responsible for the normal turnover of intracellular constituents by
autophagocytosis. Proteins are absorbed from the tubule lumen by endocytosis or
pinocytosis. The proximal tubule is lined with cuboidal epithelium with microvilli that
increase the surface area for reabsorption and secretion. The proximal tubule plays a major
role in the reabsorption of various ions, water, and organic solutes such as glucose and
amino acids. Approximately half of the filtrate is reabsorbed in the proximal tubule.
The transition from the proximal tubule to the loop of Henle occurs abruptly at the outer
layer of the medulla. The loop of Henle consists of a straight segment of the proximal
tubule, the descending thin limb, and the thick ascending limb. The descending thin limb
has permeability properties that are important for maintaining medullary hypertoxicity and
the delivery of a dilute fluid to the distal tubule.
The distal tubule begins close to the macula densa and extends into the cortex where two
or more nephrons combine to form a cortical collecting duct. The distal tubule is involved
in the active transport of sodium chloride. Its function is regulated by various hormones
including vasopressin, parathyroid hormone, and calcitonin, which exert their effects by
activating the adenylate cyclase system. The cells of the distal tubule are similar to those of
the proximal tubule, except they have fewer microvilli. The distal tubule connects to the
collecting duct via the connecting tubule which plays an important role in potassium
secretion.
The collecting duct extends from the cortical region through the medulla to the tip of the
papilla. It collects fluid from several nephrons, travels along the conical pyramid, and
terminates at the minor calyx. The collecting duct is also involved in potassium secretion
and urine acidification.
There are two types of nephrons separated by their locations in the kidney. About 85% of
nephrons are located in the cortex and are called cortical nephrons. The remaining 15% are
found close to the medulla and are called the juxtamedullary nephrons. The nephrons filter
approximately 180 litres of plasma each day through the glomerular component. Only 1%
is excreted as urine, the remaining plasma fluid being reabsorbed into the circulation
through the tubules.
Vascular supply
The renal vasculature is characterised by two capillary networks surrounding the
glomerulus and the tubule.3 The renal artery usually divides into the anterior and posterior
main branches before entering the renal parenchyma. These main branches divide further
into segmental
Page 281
arteries which give rise to the interlobar arteries in the renal sinus. These vessels advance to
the junction of the cortex and medulla where they divide and form arcuate arteries. Their
divisions tend to lie in a plane parallel to the kidney surface at the corticomedullary
junction. As the arcuate arteries advance towards the kidney surface and branch into the
interlobular arteries (also known as cortical radial arteries), further divisions occur to form
the afferent arterioles. The arterioles eventually divide into several branches which form the
capillary network of the glomerulus. The wall structure of the intrarenal arteries and the
afferent arterioles resembles that of vessels found in other organs, indicating that these
vessels can regulate the glomerular capillary flow through their well developed smooth
muscles.4 Afferent arterioles are composed of one to three layers of smooth muscles. As the
arterioles approach the glomerulus, the smooth muscle cells are replaced by granular cells
of the juxtaglomerular apparatus.
The glomerular capillaries exit Bowman’s capsule to form the efferent arterioles. The
efferent arterioles are smaller in size than the afferent arterioles. This feature is a
contributing factor in raising the glomerular pressure. The efferent arterioles supply the
renal tubules in the form of a capillary network called the peritubular capillaries. Near the
corticomedullary junction, the efferent arterioles become larger, longer, and more muscular
than those in the outer cortex. The arteriolar smooth muscle is replaced by pericytes as the
efferent arterioles extend into the medulla to form long loops of thin walled vessels called
vasa recta. The peritubular capillaries eventually reunite to form interlobular veins which
converge into the arcuate and interlobar veins. They run between the pyramids and give rise
to several trunks that leave the kidney through a single renal vein at the hilus. Unlike the
arterial system, which lacks collateral vessels, the venous network anastomoses at several
levels. This allows normal drainage of blood even when a large venous branch is occluded.
Control of the renal circulation
The kidneys receive approximately 20% of cardiac output. They are capable of
increasing flow even further, although they constitute less than 0.5% of the total body
weight. This marked renal blood flow is well in excess of that required to provide renal
tissue with sufficient oxygen and nutrients. Therefore, renal blood flow is regulated to
maintain an optimum delivery of filtrate to the nephrons and adequate reabsorption of fluid
back into the vascular system. The factors that control renal circulation are divided into
intrinsic factors (autoregulation and renal nerves) and extrinsic factors (hormonal and other
endogenous vasoactive agents).
Autoregulation
Regulation of the renal blood flow depends closely on changes of vascular resistance
secondary to constriction or relaxation of vascular smooth
Page 282
muscles. Autoregulation is the intrinsic ability of the kidney to maintain a relatively
constant blood flow over a range of renal perfusion pressure from 75 to 180 mm Hg.
Outside this pressure range, afferent arteriolar resistance is less responsive and flow
becomes pressure dependent.5 As this vascular reaction is demonstrable in isolated kidneys,
autoregulation of renal blood flow is generally assumed to be mediated by factors intrinsic
to the kidney. It is well accepted that blood flow to the renal cortex is autoregulated.
Whether there is also blood flow autoregulation in the medulla remains controversial.
Currently, two theories are proposed to explain autoregulation. One is the myogenic
theory first discussed by Bayliss in 1902.6 It has been shown that a vasoconstrictor is
released from renal vessels when the transmural pressure difference increases, supporting a
relationship between the endothelium and the vascular smooth muscle, which mediates the
autoregulation. The second theory postulates a tubuloglomerular feedback mechanism
which enables autoregulation of both renal blood flow and glomerular filtration rate. Within
the autoregulatory range, there is a coupling between renal blood flow and glomerular
filtration rate, which supports the postulate that the principal autoregulatory actions on renal
vascular resistance occur at preglomerular arterioles.7 The tubuloglomerular feedback
theory is based on a relationship between the distal tubular Na+ delivery and the intrarenal
release of renin.8 Changes in distal tubular flow and/or solute can affect renal blood flow
and thus modulate glomerular filtration rate. Recent data support a close interaction
between the tubuloglomerular feedback and the myogenic mechanisms (Fig 8.1). Thus, the
renal vascular bed is believed to consist of a rapid (myogenic
Fig 8.1 An example of autoregulation in the kidney that demonstrates the flow and filtration stability between
75 and 180 mm Hg blood pressure.——— renal blood flow; – – – glomerular filtration rate.
Page 283
response) and a slow (tubuloglomerular feedback mechanism) component of renal
autoregulation.
Renal nerves
The renal nerves contain both afferent and efferent nerve fibres which release
noradrenaline (norepinephrine).9 Both noradrenaline and adrenaline (epinephrine) released
from the adrenal medulla activate α1-adrenoceptors and cause vasoconstriction, which
decreases both renal blood flow and glomerular filtration rate. Renal nerves also release
dopamine (DA) which activates specific dopamine receptors existing in abundance in the
renal tissues. Activation of the DA1-receptors causes significant increases in both cortical
and medullary blood flows.10 In addition, the release of neuropeptideY and noradrenaline
from sympathetic activation can stimulate Y-receptors and contribute to the modulation of
renal vascular resistance.11
Renin-angiotensin system
Renin is a proteolytic enzyme that is synthesised in the epithelial cells of the
juxtaglomerular apparatus and secreted into the surrounding interstitium. It cleaves
angiotensin I from angiotensinogen. Angiotensin I is converted into angiotensin II which
exerts both direct and indirect adrenergic actions on renal arterioles. Angiotensin II can also
act as a circulating hormone. It is a potent vasoconstrictor; however, its most important
action is to stimulate aldosterone production and secretion by the adrenal cortex. Thus, an
increase in renin release from the kidney will lead to an increase in Na+ reabsorption
secondary to a higher blood level of aldosterone. The increased reabsorption of Na+ will
facilitate water movement from the interstitial space and increase plasma volume. The
renin–angiotensin system shuts off once the volume deficit is corrected.12 Although
angiotensin II has potent vasoconstrictor effects on the afferent arterioles, its primary action
appears to be on the efferent arterioles. Vasoconstriction of the efferent arterioles can
contribute to the maintenance of the glomerular filtration, particularly when renal plasma
flow is reduced. There are two main angiotensin II receptors: AT1 and AT2. Activation of
these receptors increases the sensitivity of the tubuloglomerular feedback mechanisms and
augments autoregulation. Blockade of the renin–angiotensin system, however, has no effect
on autoregulation.13
Antidiuretic hormone
This hormone, also called vasopressin, is synthesised in the hypothalamus and released
from the posterior pituitary gland. The release of antidiuretic hormone (ADH) can be
stimulated by changes in the volume and osmolality of body fluids or by activation of the
sympathetic nervous system. This hormone acts on the collecting tubules where it inhibits
Page 284
diuresis and increases water reabsorption leading to increased plasma volume. Thus, ADH
modulates urinary concentration via an increase in osmotic water permeability and a
decrease in medullary blood flow by constricting juxtamedullary arterioles.13 Although
ADH is involved in the maintenance of the volume and osmolality of plasma, the
magnitude and direction of such involvement remain controversial. The conflicting results
may result from different dosages of ADH, varying states of fluid balance, and the
influence of other vasoactive agents.
Prostaglandins
Prostaglandin E2 (PGE2) and prostacyclin (PGI2) are produced within the kidney during
haemorrhagic hypovolaemia. The production of these substances is stimulated by
sympathetic nerve activity and angiotensin II. During haemorrhage, prostaglandin synthesis
occurs within the kidneys, which causes vasodilation in the afferent and efferent arterioles
to prevent severe renal vasoconstriction and ischaemia. Therefore, prostaglandins appear to
play a role in modulating the renal vascular effects of other vasoactive agents, including
vasoconstrictor hormones and bradykinins. As prostaglandin inhibitors do not significantly
alter renal blood flow autoregulation, the contribution of prostaglandins to the control of
basal or resting renal blood flow is considered to be minimal.14
Atrial natriuretic peptide

Atrial natriuretic peptide (ANP) is a recently discovered peptide that appears to be
involved in the regulation of renal function and Na+ balance. ANP is released primarily
from cardiac atria in response to atrial distension secondary to volume expansion. This
peptide is a rapidly acting, potent natriuretic and diuretic substance, which can also exert
direct vasodilator effects on the systemic vasculature. There are membrane receptors for
ANP throughout the renal cortical and medullary vasculature, particularly at the glomerular
capillaries. Infusion of ANP increases renal blood flow, glomerular filtration rate, papillary
plasma flow, and sodium excretion caused by a direct vasodilator effect.15 As ANP can
inhibit renin secretion and aldosterone release, and increase urinary kallikrein excretion,
part of the renal effects of ANP may be indirect. Nevertheless, ANP appears to have an
important role in the control of Na+ balance during conditions of altered plasma volume.
The Na+ excretion may be partially mediated by the vasodilator effect of ANP which
increases the glomerular filtration rate.
Adenosine
It has been proposed that adenosine may play a role in the regulation of renal blood flow
and glomerular filtration.16 Administration of adenosine causes significant changes in renal
vascular resistance, glomerular filtration rate, and renin release. Responses of renal vascular
resistance to intrarenal
Page 285
infusion of adenosine include an initial transient vasoconstriction of the afferent arterioles,
possibly by interaction with the renin–angiotensin system, and is followed by a dilatory
phase that may be associated with a marked vasodilation caused by both direct and indirect
mechanisms. The vasoconstriction of the afferent arterioles combined with the vasodilation
of the efferent arterioles can result in sustained decreases in glomerular filtration pressure
and rate. In addition, adenosine can exert a direct and powerful inhibition of renin release
by the juxtaglomerular cells. In general, adenosine appears to be an important mediator in
the control of renal blood flow. Further studies are, however, necessary to delineate the
effects of adenosine on both the renal and the systemic vasculature.
Kinins
Administration of bradykinin released from plasma kallikrein causes marked renal
vasodilation. Infusion of kinin antagonist decreases papillary blood flow by 20% without
changing outer cortical blood flow, indicating that kinins exert a vasodilatory influence on
the papillary vessels. The role of kinin in regulating medullary haemodynamics is not,
however, clear because more than 90% of the renal kallikrein is found in the cortex and a
very small fraction in the medulla and papilla.
Nitric oxide
The role of endothelium derived relaxing factor (EDRF) or nitric oxide (NO) is being
extensively investigated. Recent data suggest that about 30% of renal vascular resistance
may be controlled by NO. The collecting duct and vasa recta capillaries seem to be the
major sites of NO synthesis.17 NO inhibitor infused into the renal medulla decreases
papillary blood flow without changes in cortical blood flow, renal blood flow, or mean
arterial pressure.18 Systemic inhibition of NO, however, significantly decreases renal blood
flow without affecting the glomerular filtration rate.19 During NO blockade, renal
autoregulation remains intact. Apparently, NO blockade shifts the myogenic response to a
lower renal arterial pressure and thus modulates autoregulation. It has been speculated that
NO blockade inhibits the myogenic response, but autoregulation is sustained by a strong
activation of the tubuloglomerular feedback mechanism.7 In the presence of NO deficiency
in the cells of the macula densa, the tubuloglomerular feedback mechanism becomes
increasingly sensitive, which leads to elevated sodium and water retention and thereby
hypertension. The release of NO in response to changes in vessel tension is postulated to be
associated with the activation of membrane receptors or to involve a complex enzymatic
pathway. Inflammation mediators such as tumour necrosis factor modify the release of NO
from either endothelial or non-endothelial cells.20
Page 286
Other vasoactive substances such as serotonin and histamine have been proposed as
mediators involved in the control of renal haemodynamics. Current data, however, suggest
that their role in regulating the renal circulation appears to be limited. Future studies should
examine the possible role of endothelin in the local control of renal blood flow.
Fig 8.2 demonstrates the main contributors to renal circulatory control.
Effects of anaesthetic drugs on renal blood flow
Anaesthetic drugs are associated with significant sympathetic and endocrine changes,
particularly those with sympathomimetic properties such as pentobarbital (pentobarbitone),
and ketamine can stimulate the release of catecholamines. An increased blood level of
adrenaline causes significant renal vasoconstriction and triggers renin release from the
juxtamedullary nephrons. The formation of angiotensin II from renin leads to profound
vasoconstriction of the renal vessels, and associated decreases in both the renal blood flow
and glomerular filtration rate. Surgical stress may also induce further release of
catecholamines, ADH, and aldosterone into the circulation. Opioids generally decrease
regional vascular resistance by both local and systemically mediated mechanisms, but
changes in perfusion pressure are moderated by reflex increases in sympathetic activity.
They also appear to have no significant effect on renal blood flow. A low concentration of
propofol has no effect on blood pressure, heart rate, or renal nerve activity. At moderate to
high doses, renal nerve activity decreased by 22–50%, but the effect of propofol on renal
blood flow still has to be elucidated.
Fig 8.2 Factors controlling renal blood flow.

Page 287
In healthy volunteers in the absence of surgical stimuli, as well as in patients undergoing
surgery, inhalational anaesthetic drugs are associated with dose dependent decreases in
renal blood flow, glomerular filtration rate, and urine production. These anaesthetic drugs
have both direct and indirect effects on renal blood flow. The direct effects on the renal bed
can be intensified in certain clinical conditions, including dehydration, pain, and loss of
blood volume. They are accompanied by indirect factors such as depressed myocardial
performance, altered vascular resistance, or decreased intravascular volume associated with
most inhalational anaesthetic agents. Data from laboratory animals clearly demonstrate
changes in active ion transport during anaesthesia. It has been suggested that an interaction
with adrenaline is involved in the augmented Na+ transport, whereas the inhibition of Na+
transport is related to a direct effect of inhalational anaesthetic drugs. At concentrations of
3% or higher, halothane causes significant decreases in renal blood flow and oxygen
consumption. The effects of enflurane and isoflurane on renal blood flow are considerably
less than those of halothane. All inhalational anaesthetic drugs at high concentrations,
however, possess potent vasodilator properties which may lead to loss of autoregulation.
Spinal or epidural anaesthesia may cause significant reductions in glomerular filtration
rate and renal plasma flow when arterial blood pressure is markedly reduced. When the
reduction of perfusion pressure is corrected, renal haemodynamics are normalised,
suggesting that perfusion pressure is an important factor in sustaining normal circulation in
the kidney. No change in plasma renin levels is observed with either spinal or epidural
anaesthesia. Therefore, the decrease in renal blood flow during spinal or epidural
anaesthesia is possibly caused by a reduction in venous return and decreased cardiac output.
In summary, anaesthetic drugs are associated with an increase in renal vascular
resistance. It is still, however, controversial whether anaesthetic drugs alter renal
autoregulation.
Splanchnic circulation
Anatomy
The splanchnic circulation is composed of gastric, small intestinal, colonic, pancreatic,

hepatic, and splenic circulations.21 22They are arranged in parallel with one another and fed
by three arteries:
1 The coeliac artery, which perfuses the hepatic artery, the stomach, the spleen, and part of
the pancreas.
2 The superior mesenteric artery which supplies branches to the pancreas, small intestines,
and part of the colon.
Page 288
3 The inferior mesenteric artery which supplies the rest of the colon.
The stomach is composed of three histologically distinct layers: the mucosa, submucosa,
and muscularis externa. The mucosa receives approximately three quarters of resting blood
flow. Gastric arteries pierce the muscularis layer and form a network of branches into the
submucosal area. This submucosal plexus gives rise to smaller branches that form a
secondary arcade within the smooth muscles at the base of the mucosa. The venous system
is parallel to the arterial vessels.
The mesenteric vasculature to the small and large intestine is composed of several
circuits coupled both in series and in parallel. The three parallel circuits serve the
muscularis, the submucosa, and the mucosa. Each of these circuits possesses a series
coupled component consisting of resistance arterioles, precapillary sphincters, the
capillaries themselves, postcapillary sphincters, and the venous capacitance vessels. The
resistance arterioles are the primary determinants of vascular resistance and they regulate
blood flow both to the splanchnic bed as a whole and through each of the parallel circuits.
The mucosa has an enormous surface area for absorption created by villi and microvilli.
The mucosa is the metabolically active area in the gut and receives well over half of the
total resting organ blood flow. The small intestinal villi are oxygenated via a countercurrent
circulation whereby a central, small artery diffuses oxygen across to the parallel
submucosal veins. As a result of this anatomical configuration, the tips of the villi have the
lowest tissue oxygen tension (PO2), and ischaemia to this area may be induced by relatively
small changes in blood flow.23 The submucosa and mucosa both have a parallel capillary
plexus. Venous drainage from the small intestinal veins joins with veins from the colon,
which subsequently join with splenic veins to form the portal system.24
The pancreas receives its blood supply from several branches of the coeliac and superior
mesenteric arteries. It also drains into the portal system.
The splenic circulation is unusual in that the internal red pulp has a complex mesh like
structure which filters the blood passing through it. Blood flowing through the spleen may
bypass the red pulp and constitutes a fast compartment, whereas blood perfusing the red
pulp has a considerably increased transit time and may be considered to be the slow
compartment. Splenic blood flow is approximately 250 ml/min. The oxygen content of
splenic venous blood is quite high because of the fast compartment and adds significantly
to the portal oxygen saturation.
The liver is unique in that it has both an arterial and a venous afferent blood supply. The
hepatic artery provides approximately 30–40% of total hepatic blood flow, with the rest
coming from the portal veins. In the normal resting adult, about 500 ml/min perfuse the
liver via the hepatic artery whereas an additional 1300 ml/min flow via the portal system.
The
Page 289
portal blood flows through hepatic sinusoids, which puts it in close contact with the
metabolic cells of the liver parenchyma. Hepatic arterial blood flow also supplies nutrients
to the connective tissue within the liver itself, especially the walls of the collecting ducts
which form the intrahepatic biliary system. The hepatic sinuses are lined with endothelium
similar to that found in capillaries, but the permeability of this endothelium is extreme and
allows for rapid, easy diffusion of components within the blood into the hepatic
parenchyma and back. The sinusoids drain into venules which form the hepatic veins; these,
in turn, empty into the inferior vena cava. There is a well organised intrahepatic lymph
system as well, which accounts for more than half of the total body lymph flow.
Overall, splanchnic blood flow requires about a quarter of the cardiac output in normal
resting adults and, by coincidence, the splanchnic capacitance venous system serves as a
large reservoir for almost a quarter of the total blood volume of the body.25
Control of the splanchnic circulation
The resistance arterioles are the primary determinant of vascular resistance in the
splanchnic system, and therefore regulate blood flow through this bed as a whole and
through each parallel circuit. In this sort of control system, the relationship between flow
and resistance can be described by the haemodynamic version of Ohm’s law:
In the case of the mesenteric circulation, this would be the hydrostatic difference between
the arterial and venous pressures divided by flow across the gut.
The control of splanchnic blood flow is by a combination of neuroreflex and hormonal
factors.23 26 Neural control of the mesenteric circulation is almost exclusively sympathetic
in origin. The parasympathetic fibres originating from the vagi have little effect on
splanchnic blood flow, although they are most important in the regulation of secretion and
motility of the gut. The sympathetic postganglionic fibres act directly on the vascular
smooth muscle of the arterioles. In this way increased sympathetic activity decreases blood
flow to splanchnic organs. In addition, sympathetic outflow to the splanchnic bed contracts
the venous smooth muscles of the capacitance veins in the splanchnic circulation, and may
expel a large volume of pooled blood from the splanchnic reservoir into the systemic
circulation. Most of these sympathetic ganglia arise from the coeliac plexus with lesser
contributions from the superior and inferior mesenteric plexus. Through the sympathetic
system, the mechanoreceptor reflexes – particularly the low pressure cardiopulmonary
receptor systems – are closely
Page 290
involved in splanchnic arterial and venous vascular tone.27 Interestingly, β2-adrenoceptors
are also present in the mesenteric circulation and activation of these receptors causes
vasodilation.
Circulating substances that may alter vascular resistance in the splanchnic bed include
adrenaline, noradrenaline, angiotensin II, vasopressin, and gastrointestinal peptides such as
glucagon, vasoactive intestinal peptide (VIP), and cholecystokinin.28 The role of adenosine
in control of the splanchnic circulation is controversial, but research has shown: (1) its
presence in normal gut, and (2) that blockade of its receptors alters mesenteric blood flow
significantly.29 Thus, adenosine probably plays an important local role in increasing
regional blood flow. In fact, most of the gastrointestinal peptides are vasodilators; they
rarely reach a concentration high enough to be vasoactive in the systemic circulation as a
whole. The catecholamines, including angiotensin and vasopressin, probably achieve a
concentration that is centrally vasoactive only under circumstances of significant shock.
In addition to the classic gastrointestinal hormones, various vasoactive substances,
including histamine, serotonin, bradykinin, and prostaglandins, which are produced and
stored in the splanchnic organs, have been shown to affect organ blood flow in this
region.30 Many of the changes observed with these substances are in patients or study
models with compromised splanchnic function.24 The action of these hormones may be
independent or in combination with other regulatory mechanisms.
Regional production and circulatory effects from vasoactive substances have been shown
in normal subjects.32 The potent vasoconstrictor, endothelin-1 (ET-1) induces paracrine
actions in the mesenteric vessels through its local production and breakdown. NO inhibits
the formation of ET-1 in humans and thereby counteracts its constricting action on the gut’s
vasculature.33 Apparently, this interactive relationship between ET-1 and NO is another
step in the complex regulatory pathways for the splanchnic circulation.
Autoregulation in the splanchnic circulation is demonstrated by compensatory dilatation
of the resistance arterioles in response to an acute reduction of perfusion pressure which
serves to restore the decreased tissue perfusion partially. Splanchnic autoregulation is less
pronounced than in the cerebral, cardiac, or renal circulation. The response is evident,
however, as a mechanism whereby initial levels of hypoperfusion are rapidly ameliorated
by marked arteriolar vasodilation which partially restores blood flow. This phenomenon is a
variation of the postischaemic hyperperfusion mechanism seen in many areas of the body,
and is the result of a direct myogenic response to the reduction in perfusion pressure and
the accumulation of local ischaemic metabolites in the region, including adenosine, which
may be the principal metabolic mediator of autoregulation. Interestingly, the oxygen
consumption in the small intestine is even
Page 291
more rigorously autoregulated than blood flow, with the result that oxygen uptake in this
organ remains constant when arterial perfusion pressures are varied fourfold.34 The portal
venous system does not autoregulate so that, as the portal venous pressure and flow are
raised, resistance either remains constant or may decrease.
In addition to autoregulation of blood flow within the individual organ as a whole, the
splanchnic circulation responds to reductions in perfusion pressures by redistributing blood
flow to various levels within the individual organ. This redistribution is achieved by
changes in the relative resistance of the arterioles and precapillary sphincters, gating the
parallel vascular circuits in the various layers of the organ. In shock, for example, this
response usually favours the mucosa at the expense of the muscularis layers. The gut is
protected from ischaemic injury by its unique ability to increase oxygen extraction as much
as sixfold, thereby maintaining oxygen consumption at near normal levels over a broad
range of flows and avoiding the usual hypoxic sequelae, particularly in areas at risk such as
the tips of the villi. This protective mechanism is the result not only of the ‘‘mass effect” of
rapid diffusion of oxygen along a steeper concentration gradient, but also of the opening of
hypoperfused capillary beds. The net result is an increase in perfused capillary density as
flow is reduced within the mucosa of the gut. This provides an important defence against
splanchnic ischaemia.
An illustrated overview of the central mechanisms for splanchnic blood flow is shown in
Fig 8.3.
Fig 8.3 Factors influencing blood flow through the splanchnic circulation.
Page 292
The effects of anaesthetic drugs on the splanchnic blood flow
Animal studies have shown that some intravenous anaesthetic drugs, particularly
propofol, result in an increase in total liver blood flow with contributions from both the
hepatic artery and portal systems. These data mark propofol as a vasodilator in this region.
Human studies have, however, indicated an initial fall in hepatic flow followed by a return
towards normal levels. Barbiturates also may cause a decrease in overall splanchnic blood
flow, most pronounced in the portal system because of the relative venoplegia caused by
higher doses of these drugs. Midazolam may also reduce splanchnic flow initially, with a
rebound increase from the splanchnic reservoir shortly thereafter.
Opioids, in small doses, produce a central sympathetic withdrawal and relative increase
in vascular capacitance in this circulation.35
The change in splanchnic blood flow after regional anaesthesia of the central neural axis
is dictated by the level of block obtained. Spinal and epidural anaesthetic drugs at T10 and
below have little effect upon splanchnic flow. Raising the block level to the T4 dermatome
reduces total splanchnic blood flow by at least 25% in humans. Inclusion of the sympathetic
outflow from the coeliac and mesenteric plexus produces the withdrawal of resting venous
tone and the subsequent reduction in transhepatic flow.
With inhalational anaesthetic agents, there is most often a reduction in portal blood flow,
particularly by halothane. Isoflurane and enflurane apparently increase hepatic arterial
blood flow which facilitates an increase in oxygen delivery to the liver with these
anaesthetic agents.36 Moreover, noradrenaline spillover (a result of increased sympathetic
nerve activity) was reduced in the mesenteric circulation of pigs during isoflurane
anaesthesia, thus suggesting a relative vasodilation by the anaesthetic in that region.37
Of equal importance in the anaesthetised patient are the physiological consequences of
ventilation. During inspiration with controlled ventilation, the diaphragm compresses the
liver, which causes an increase in hepatic venous pressure and decreased transhepatic
conductance. With expiration, there is a reversal of this and splanchnic flow increases
dramatically. Positive end expiratory pressure will decrease mesenteric arterial flow, portal
blood flow, and total splanchnic blood flow.38 Interestingly, hypocapnia, which often
accompanies controlled ventilation, reduces portal blood flow because of an increase in the
resistance in mesenteric and splenic arteries and portal veins. On the other hand, hepatic
arterial resistance is decreased and hepatic arterial blood flow thereby maintained. Increases
in circulating carbon dioxide levels most commonly increase portal and total hepatic blood
flows, probably mediated through carbon dioxide as a direct vasodilator and the increase in
cardiac output resulting from stimulation of
Page 293
the central nervous system. Concurrently, there is an initial transient decrease in hepatic
arterial flow followed by an increase towards the normocapnic control values. This
phenomenon in the hepatic artery is compatible with a typical escape phenomenon of
hepatic arterial vasculature from sudden sympathetic stimulation.
Cutaneous circulation
Anatomy
The largest organ in the body is the skin with a surface area of approximately 1·9 m2 and
a weight of about 2 kg in a 70 kg adult. The blood flow to the skin, including the
microcirculation, is approximately 10 times greater than the metabolic needs of the
cutaneous system. Under normothermic conditions, the venular plexus in the cutaneous
circulation has the potential of being one of the larger reservoirs of blood volume in the
body. The gross anatomy of the cutaneous circulation reflects the ability of the skin
circulation to dissipate or preserve heat within the body as a whole and the anatomy reflects
this functional need.39 40
The cutaneous vascular system is divided into three interconnected levels:
1 The deep subdermal or subcutaneous plexus
2 The middle or cutaneous plexus
3 The superficial or subpapillary plexus.
The subdermal plexus is the major vascular network of the overlying skin. Vessels of this
plexus generally run in the subcutaneous fatty or areolar tissue. The arterial blood supply to
the skin comes primarily from muscular cutaneous arteries which perforate the
subcutaneous tissue from underlying muscle. These arteries and their concomitant veins
ascend through the reticular dermis to the papillary dermis. Here the artery forms a
superficial arteriolar plexus with terminal arterioles which project into capillary loops
through the epidermis. The collecting venous system forms a double layered horizontal
network at the subcutaneous dermal junction, and returns into the subcutaneous and
muscular area via collecting venules. Total skin flow is made up of the blood perfusing
through these vessels as well as bypasses (shunts) at deeper levels – primarily in the hands
and feet – where the arteriovenous anastomoses are prominent in changing skin flow for
thermal regulation.41 42
The skin is the primary site of exchange of body heat with the external environment.43
Hence, changes in cutaneous blood flow in response to various metabolic states and
environmental conditions provide the main mechanism by which temperature homoeostasis
occurs. Skin blood flow is, under normal circumstances, about 5–6% of the resting cardiac
output.
Page 294
This can decrease markedly in a cold environment when heat retention is necessary.
Alternatively, the skin vessels may dilate to increase flow up to seven times the normal
state when heat loss is required by hypermetabolic states. Specialised areas of the skin,
including the palm, fingers, sole, toes, and the face, possess the capability for remarkable
vasodilation and constriction.44 There may be as much as a 75-fold change in flow from
cold to hot environments.
Control of the cutaneous circulation
The potential for skin vessels to generate great increases in vascular conductance makes
this circulation an important regional flow area during changes in the environment and
during anaesthesia.
Cutaneous resistance vessels and the venous plexus in the subcutaneous dermal junction
are richly innervated with sympathetic vasoconstrictor nerves which maintain a relatively
high degree of neurogenic activity and, hence, vascular tone. This predominant
vasoconstrictor tone is mediated by hormonal action of circulating catecholamines at
postjunctional α-adrenergic receptors. In addition, there is an active dilator system that is
activated under thermal, physical, or emotional stress. The mechanism for this active
dilatation is not clear and may involve the release of a yet-to-be-determined
neurotransmitter on cutaneous blood vessels or, more probably, involves a release of a
vasodilator substance from activated sweat glands.45 The substance most commonly
mentioned or identified is bradykinin. Recent studies have shown, however, that calcitonin
gene related peptide (CGRP), a vasoactive polypeptide, can increase regional blood flow in
the skin under resting conditions at the expense of other organ flow, primarily the
splanchnic circulation. From this work it has been proposed that normal regional blood
flow changes in the skin may be mediated to some extent by CGRP acting as a local
vasodilator and produced by neuronal activity.46
In the context of total body circulatory control, investigators have shown that the
sympathetic vasomotor fibres in the skin vasculature exert significant influence on overall
homoeostasis. Such control is exerted from both the low pressure cardiopulmonary and the
high pressure arterial baroreceptor areas. These conclusions deviate from the previous
literature in which baroreflex sympathetic vasoconstriction in the human skin was proposed
to be more or less selectively mediated from cardiopulmonary low pressure receptors.47
The afferent control of resistance vessels in the skin is interesting in that arterioles in
non-peripheral areas are innervated by two distinct sympathetic nerve types:
1 Adrenergic vasoconstrictor nerves similar to those in all arterioles.
2 A specialized sympathetic active vasodilator nerve.
Page 295
Arterioles in the peripheral skin seem to be innervated solely by sympathetic
vasoconstrictor nerves. The remarkable dilatation that occurs in these areas under thermal
stress is solely the result of withdrawal of sympathetic support.48
Temperature sensors lie predominantly in the preoptic area of the anterior hypothalamus,
although they may be situated in the abdominal viscera and spinal cord as well. Increased
heat to these areas causes vasodilation and cold elicits vasoconstriction. Such reflex activity
is mediated through cardiovascular centres in the brain. Increased thermal content in the
blood causes inhibition of normal sympathetic vasoconstriction at the preganglionic neuron
level, which results in opening of arteriovenous anastomoses (shunts) in the extremities –
the nose, ears, and mouth. Under normal sympathetic activity these shunts are closed, and
withdrawal of sympathetic activity opens them for thermal regulation. Aortic and carotid
chemoreceptor stimulation secondary to acidaemia, hypercapnia, or hypoxaemia may cause
a decrease in sympathetic tone to the cutaneous arterioles with subsequent vasodilation.
This effect is the opposite of that seen in the muscle, splanchnic, and renal circulations.49
Regional cutaneous blood flow control to non-peripheral areas includes a cholinergic
pathway to eccrine sweat glands in the skin. Recent work has shown that some afferent
sympathetic preganglionic and postganglionic fibres innervate and activate the glands as
well. When the cholinergic and sympathetic pathways to the glands are activated, they
trigger the release of bradykinin alone or other local vasodilator substances, possibly NO,
serotonin (5-HT) or other peptides.50 51
The only parasympathetic innervation to affect skin blood vessels reaches the sweat
glands via the sudomotor nerve. Activation of this system increases the output of the
enzyme kallikrein. This substance, in turn, splits the polypeptide bradykinin from globulin
in the perivascular interstitial tissues and induces a most potent vasodilatory effect.
Figure 8.4 illlustrates the control mechanisms for regulation of cutaneous blood flow.
Effects of anaesthetic drugs on the cutaneous blood flow
In humans, propofol administration causes an increase in flow, recorded by laser

Doppler, which is probably the result of decreased sympathetic tone caused by the
anaesthetic drug plus the direct vasodilating effect of propofol itself. A similar, though
smaller, increase in skin blood flow may be observed with thiopental and other short acting
barbiturates as well.
Opiate anaesthesia, in particular morphine anaesthesia, has been shown to increase
forearm skin flow, and the neurocomponent is thought to be mediated centrally rather than
by a local inhibition of adrenergic receptors. In addition, there is a centrally mediated effect
upon the venous system, in that a significant venodilatation occurs approximately five
minutes after
Page 296
delivery of centrally acting morphine. Most probably, in both cases, sympathetic mediated
vasoconstriction is reduced and resultant vasodilation occurs.
With local anaesthetic drugs there is vasodilation of the arterioles and venules at high
concentrations; however, small amounts of local anaesthetic drugs have a mild constrictive
effect on the precapillary arteriolar vessels. Mepivacaine may also have a postcapillary
constrictive effect in small concentrations. Overall, at concentrations used in clinical
practice, the direct acting vasodilatory effects of the local anaesthetic drugs predominate.52
Most of the potent inhalational anaesthetic drugs cause some degree of sympathoplegia
which results in opening of the arteriovenous anastomoses in the extremities – nose, ears,
and mouth. These shunts, normally closed, are opened during anaesthesia and may lead to
an accelerated body hypothermia. Indeed, the relative vasodilation in the cutaneous vessels
may be one of the causes of dilated superficial veins whenever these agents are used.
Under regional anaesthesia, a relative sympathoplegia again occurs and conductance
within the skin of the affected limb will increase, whereas blood flow through the cutaneous
tissues in non-blocked limbs will decrease by reflex compensatory mechanisms. At least a
T8 epidural sensory level is, however, necessary for a complete blockade of the distal
intraneurally recorded skin sympathetic activity in the lower extremity.53
In summary, the cutaneous circulation is greatly influenced by anaesthesia and, although
normally possessing thermoregulatory capacities, it may be significantly inhibited by
anaesthetic drugs to mimic poikilothermic adaptation.
Muscle circulation
Anatomy
Skeletal muscle comprises approximately 40% of the total body mass. The density of the
capillary networks varies greatly between the two basic muscle fibre types that are present
in humans.54 The slow twitch, or red, fibres are numerous in more slowly contracting
muscles which help maintain posture and fulfil isometric functions. The fast twitch, or
white, fibres are most numerous in the fast contracting muscles such as those used for
running and quick movement. Numerous animal studies have demonstrated that resting
blood flow and capillary density, as well as oxygen consumption, vary with the type of
muscle that is being perfused. In general, the capillary density for slow twitch fibres is
approximately two to three times that seen in fast twitch fibres in the same muscles. Other
animal studies have shown mat oxidative enzymes, as well as blood flow, are several
Page 297
times higher in the slow contracting muscles compared with the fast twitch muscles, which
have a much higher activity of glycolytic enzymes and rely more on anaerobic metabolism.
Apparently, slow twitch muscles respond better to prolonged activation and long term
aerobic function.
In resting muscle, the precapillary arterioles exhibit intermittent asynchronous
contractions and relaxations which, in effect, limit the number of capillary beds that are
being perfused. This very action allows sudden recruitment and an increase in the number
of nutrient capillary beds that perfuse the muscle when muscle activity begins. As a result
of this, total blood flow through resting skeletal muscle varies between 2 and 5 ml/min per
100 g. With exercise, this can increase in trained athletes to more than 125 ml/min per 100
g.55 Dynamic rhythmic exercise has a large, rapid effect on vascular contracture which is
not explained by known neural, metabolic, myogenic, or hydrostatic influences.56 The
muscle relaxation between contractions draws blood from the arteries into the veins, and
effectively reduces arterial driving pressure. During the subsequent contraction venous
blood is pumped centrally raising the central venous pressure. The consequent increase in
muscle blood flow, coupled with the decrease in arterial pressure and raised central venous
pressure, markedly increases the calculated conductance. This rhythmic contraction and
relaxation, in effect, pumps blood across the muscle bed.
The anatomical architecture of the muscle and the increase in intramuscular pressure
causing hindrance to blood flow itself determine the increase in perfusion pressure
observed during exercise. In the forearm, the circulation is initially arrested during
voluntary isometric contraction of more than 70% maximal. In the calf, tensions of only
20–30% of maximal voluntary contraction are necessary to interrupt blood flow.54 The
increase in vascular resistance during strong muscle contraction occurs chiefly in the larger
supply – the branched arteries down into the muscle bed. The manner in which this vascular
bed is occluded keeps red cells in the capillary and so provides a continuous, albeit
dwindling, supply of nutrient oxygen during contraction.
Overall, the sequence of vascular segments in the muscle beds is similar to those seen in
other parts of the bodies – that is, large arteries, small branched arteries, arterioles,
precapillary arterioles, capillaries, then postcapillary venules, venules, and collecting veins
which return blood to the central circulation. At the capillary level, only between 20% and
25% of the circuits are open in the resting muscle and these have decreased conductance
from sympathetic tone and intramuscular pressure.
Control of the muscle circulation
Blood flow to skeletal muscle depends on the pressure gradient through the vessel
complex and the calibre in the resistance vessels. Local
Page 298
modulation of vessel size is dependent on chemical and physical events near or within the
environment of the resistance vessels, as a result of changes in nervous innervation of the
vessels and circulating vasoactive agents.
The resting tone in the resistance arterioles is relatively high in the non-active muscle.57
There is an intrinsic basal vascular resistance, or contractile property, in these small vessels
and those of 50–100 μm in diameter tend to have more myogenic tone than the larger
vessels up to 400 μm in diameter. In isolated muscle, the change from continuous to
pulsatile perfusion causes a gradual increase in vascular resistance, presumably by the
periodic stretch of the arterioles, which provides a continuing stimulus to the smooth
muscle cells in the vessel walls.58 59This phenomenon was a concept initiated by Bayliss
early in this century, who suggested that distension of the resistance blood vessels by
intravascular pressure contributes to the basal vascular resistance through a direct action on
the vascular smooth muscle. The phenomenon of autoregulation in the muscle vascular bed
relies on this myogenic theory and resides primarily at the precapillary arterioles. Recent
observations strongly suggest that myogenic regulatory mechanisms contribute directly or
indirectly to circulatory homoeostasis in the following ways: intravenous pressure induces a
tonic excitatory activation which initiates an intrinsic myogenic basal tone in the arterial
microvessels. This, in turn, induces variable vascular resistance (total peripheral resistance)
mainly from the myogenic tone, and basically serves to maintain normal arterial pressure at
rest. Lastly, it produces an improved nutritional flow and exchange characterised by blood
flow recruitment, capillary recruitment, and adjustments to the capillary perfusion/defusion
ratio for optimum exchange of nutrients in a heterogeneous capillary network.60 This
myogenic control seems to initiate and maintain basal vascular tone. In synergism with
metabolic vasodilators, it modifies the autoregulation of blood flow through the muscle
itself. Both the humoral and neurogenic β-adrenergic effects depress myogenic reactivity
and counter the α-adrenergic constriction.
Nervous control
The direct vasoconstricting activity of the sympathetic nervous system on the muscle
blood flow is through the α-adrenergic receptor system. Vasoconstrictor fibres emerge from
the ventral roots of the lower thoracic through upper lumbar segments with maximum
outflow through L3. In addition, there is cholinergic innervation of the resistance vessels in
the muscle bed. This is limited to cholinergic vasodilator fibres from the lumbar region for
lower limb innervation. The sympathetic nerves follow the somatic nerves to the vessels
innervated and continue along the adventitial surface of the vessels. Only the outer layers of
smooth muscle cells within the vessels are in contact with noradrenergic vessels. These
axons form varicosities which are demyelinated Schwann cells. Within these varicosi-
Page 299
ties, the vesicles for storage of neurotransmitter substance are situated. At the neuroeffector
junction, arterial baroreceptors and chemoreceptors, the cardiopulmonary low pressure
mechanoreceptors, the receptors in the skeletal muscle, and those originating from centres
in the brain signal alterations in the amount of noradrenaline delivered by the terminal
nerves into the junctional clefts near the vascular smooth muscle cells. The actual release of
noradrenaline is initiated by action potentials generated within the ganglionic cell body.
Both sympathetic nerve stimulation and exogenous noradrenaline cause vasoconstriction
for skeletal muscle resistance vessels. Small amounts of noradrenaline metabolites appear
in the venous drainage after nerve stimulation. This indicates that reuptake into the
sympathetic nerve terminals must be the main route for terminating action of the released
noradrenaline on the muscle vessels. In animal studies, complete ablation of sympathetic
activity to a resting limb muscle results in less than a threefold increase in blood flow.
Conversely, maximal stimulation of the noradrenergic nerves to the same limb produces a
decrease in blood flow of about 75%. Interestingly, simple somatic motor denervation, in
the presence of an intact sympathetic innervation of striated muscle, will increase blood
flow by 25%. This is thought to be the result of overcoming resting somatic muscle tone
which causes relative decompression of the blood supply to the muscle, thus opening up
flow.54
Compared with the local mechanisms regulating muscle blood flow, the noradrenergic
nerves control relatively small portions of the maximal flow available to the muscle bed.
Nevertheless, because of the great muscle mass within the body, small amounts of variation
in flow resistance may permit major changes in total body vascular resistance.61
Constriction of the muscle resistance vessels occurs solely by activation of the
sympathetic noradrenergic nerves or muscle compression itself. On the other hand,
neurogenic vasodilation can occur either by withdrawal of noradrenergic activity or by
release of substances that lead to relaxation of vascular smooth muscle. The most important
effectors of this type are sympathetic, cholinergic, and histaminergic systems.
The arteriolar resistance vessels within the muscle vascular bed contain receptors of the
β2 subtype. Activation of these receptors causes relaxation of the smooth muscle which can
be prevented by β-adrenergic antagonists. Via this system, adrenaline dilates skeletal
resistance muscles. The β2-adrenoceptors in blood vessels within the muscle respond
primarily to adrenaline released from the adrenal medulla rather than to any noradrenaline
released from adrenergic nerves. The β2 activation is an additional mechanism to induce
large scale vasodilation in the working muscle.
Histamine is found in relatively high concentration in the walls of arteries and veins of
the muscle bed. Activation of these histamine receptors causes marked vasodilation of the
resistance arterioles similar to that seen with
Page 300
potassium and adenosine. Although resting tone is modulated by endothelial derived NO
(EDNO), its role in hyperaemia and exercise is controversial.62 Factors such as shear stress
in the vessel wall, acetylcholine, free Ca2+, and circulating insulin all play a role in
inducing NO synthase (NOS) and local NO production. PGI2 has also been shown to affect
muscle arteriolar conductance during low flow states.63 Other metabolic vasodilators
include potassium, histamine, hypoxaemia, and increased osmolality.
There are several factors that modify the amount of noradrenaline from sympathetic
nerve endings. Metabolic acidosis, in addition to relaxing active smooth muscles, depresses
the contractile response of blood vessels to sympathomimetic amines and nerve stimulation.
Potassium ions in large concentrations cause small vessel vasodilation via inhibitory effects
on neurotransmission and a direct depressant effect on the vascular smooth muscle cell.
Hyperosmolarity causes vasodilation of many vascular beds because of a reduced release of
noradrenaline in the sympathetic nerve endings. Adenosine has a relaxing action on the
muscle arterioles by a direct effect on vascular smooth muscle and, indirectly, by an
inhibitory effect on noradrenergic neurotransmission.
A number of prejunctional receptors have been demonstrated, the activation of which
depresses noradrenaline release from sympathetic nerve endings as well. These include
specific α2-adrenoceptors, muscarinic receptors activated by acetylcholine, histamine, and
5-HT receptors. In most animals, and most probably in humans, there is a cholinergic
vasodilator pathway that originates in the motor cortex. The descending pathway has
discrete relays in the hypothalamus and continues through the mesencephalon descending
via the lateral spinothalamic tract. These cholinergic nerves run in the sympathetic nerves
to the muscle vessels and they innervate almost exclusively the small arteries and arterioles.
The acetylcholine released on activation of these fibres causes an instant dilatation by
activating prejunctional muscarinic receptors, thereby reducing the noradrenaline release
from the sympathetic nerve endings. They act on postjunctional muscarinic receptors in the
vascular smooth muscle cells, which results in dilatation. Cholinergic nerves are generally
quiescent. Stimulation of specific hypothalamic areas that produce defence reactions (in
times of rage or fear) will, however, increase muscle blood flow mediated by this indirect
action without actual muscle activity. In isolated blood vessels from many species, removal
of the endothelium abolishes this relaxation induced cholinergic transmitter. Consequently,
it is felt that most of this cholinergic reaction may come by way of NO release.64
Teleologically the resting vasodilation caused by this cholinergic mechanism “primes the
pump” for subsequent physical activity. Two excellent reviews of muscle blood flow and
its regulation during exercise have been published recently.63 65
Page 301
Reflex regulation
Changes in the activity of both the carotid and aortic high pressure mechanoreceptors
result in inverse changes in blood flow to the limb muscles via the sympathetic nervous
system. Activation of carotid and aortic chemoreflexes leads to constriction of arteriolar
resistance vessels which is most pronounced in the skeletal muscle bed. Recent work has
shown that the low pressure mechanoreceptors (atriopulmonary receptors) are much less
important than previously thought in causing reflex sympathetic activation and
vasoconstriction in the human skeletal muscle circulation during stress.61 Indeed, the high
pressure system predominantly controls the resting tone for these resistance arterioles.
Reflexes originate also from receptors in the skeletal muscle. This reflex, via activation
of the adrenergic system, may be a prime cause for the rise in blood pressure in an isolated
exercising limb. This suggests that intramuscular mechanoreceptors generate part of the
reflex drive during induced contractions of the skeletal muscle.66 A summary of factors that
influence blood flow in the muscle circulation is illustrated in Figure 8.5.
Effects of anaesthetic drugs on the muscle blood flow
As mentioned before, somatic motor denervation of the muscle leads to a 25% decrease
in resting vascular tone in the muscle circulation. Similarly, the use of neuromuscular
relaxants in supine, anaesthetised patients shows a fall in calculated vascular resistance
compared with the unrelaxed state.
A dual effect by propofol occurs in the muscle bed of humans. An initial increase in
blood flow occurs after infusion of propofol which may result, in part, from the direct
vasodilating effect of the drug. A sustained fall in calculated muscle bed vascular resistance
may, however, continue thereafter because of depressed cardiac and muscle sympathetic
baroreflex sensitivities. When propofol is used to control the stress response during surgery,
the vasodilating effects of the drug override the neural vasoconstriction induced by
baroreflex from the surgical stimulation.67
Opioids also have a dual effect on vascular tone in the muscular bed. The direct action of
morphine is that of venoconstriction, whereas the centrally mediated effects include
withdrawal of venous tone in the muscle bed. The short acting synthetic opioids have
produced graded, dose related decreases in calculated vascular resistance across the muscle
bed in animal studies. Certainly large doses of opioids given acutely may cause marked
hypotension through a decrease in calculated vascular resistance.68 Much of this vascular
resistance change occurs in the muscle circulation.
The potent inhalational anaesthetic drugs have a direct vasodilating effect on most
skeletal muscle beds. Reflex vasoconstriction from humoral factors, such as vasopressin
released during the relative hypotension induced by the potent inhalational agents, however,
causes an overriding
Page 302
Fig 8.4 Control mechanisms in the cutaneous circulation.
Fig 8.5 Control mechanisms for circulation through skeletal muscle.

Page 303
vasoconstriction in many of the beds as well. On the whole there is little change in the
skeletal muscle blood flow during most inhaled anaesthesia.
With regional blockade, particularly in segments that involve the sympathetic outflow
tract to the lower extremities, blood flow to the muscle beds of the lower limbs increases.
This is brought about by a reduction in the resting sympathetic tone to the blood vessels.
Active vasodilation does not take place, but rather myogenic autoregulation limits the
amount of dilatation that evolves.
In summary, control of the skeletal muscle vascular resistance involves complex
interactions among neurogenic, autoregulatory, and metabolic systems. From the
anaesthetic standpoint, resting control by sympathetic innervation, vasopressin levels, and
autoregulatory mechanisms make a significant contribution towards total vascular
resistance. The redistribution of cardiac output seen in haemorrhage and mild hypothermia,
and with anaesthetic drugs is influenced markedly by the striated muscle circulation.
1 Kanwar YS, Venkatachalam MA. Ultrastructure of glomerulus and juxtaglomerular apparatus. In: Renal
physiology, Vol 1, Windhager EE, ed, Handbook of physiology, Section 8. New York: Oxford University
Press, 1992:3–40.
2 Tisher CC, Madsen KM. Anatomy of the kidney. In: Brenner, BM, Rector FC Jr, eds, The kidney, vol I.
3 Dworkin LD, Brenner BM. The renal circulations. In: Brenner BM, Rector FC Jr, eds, The kidney, vol I.
4 Venkatachalam MA, Kriz W. Anatomy. In: Heptinstall RH, ed, Pathology of the kidney, vol I. Boston:
Little, Brown &38; Co., 1992:1–92.
5 Jones RD, Berne RM. Intrinsic regulation of skeletal muscle blood flow. Circ Res 1964;14:126–38.
6 Bayliss WM. On the local reactions of the arterial wall to changes in internal pressure. J Physiol (Lond)
1902;28:220–6.
7 Navar LG. Integrating multiple paracrine regulators of renal microvascular dynamics. Am J Physiol
1998;274(3 Part 2):F433–44.
8 Hall JE, Brands MW. The renin–angiotensin–aldosterone systems. Renal mechanisms and circulatory
homeostasis. In: Seldin DW, Giebisch B, eds, The kidney. Physiology and pathophysiology, vol 2. New
York: Raven Press, 1992:1455–1504.
9 KonV. Neural control of renal circulation. Mineral Electrolyte Metab 1989;15:33–43.
10 Kien ND, Moore PG, Jaffe RS. Cardiovascular function during induced hypotension by fenoldopam or
sodium nitroprusside in anesthetized dogs. Anesth Analg 1992;74:72–8.
11 Bischoff A, Michel MC. Renal effects of neuropeptide Y. Pflügers Arch – Eur J Physiol 1998;435:443–
53.
12 Keeton TK, Campbell WB. The pharmacologic alteration of renin release. Pharmacol Rev 1980;32:81–
227.
13 Navar LG, Inscho EW, Majid DSA, Imig JD, Harrison-Bernard LM, Mitchell KD. Paracrine regulation of
the renal microcirculation. Physiol Rev 1996;76:425–536.
14 Venuto R, O’Dorisio CT, Ferris TF, Stein JH. Prostaglandins and renal function II. The effects of
prostaglandin inhibition on autoregulation of blood flow in the intact kidney of the dog. Prostaglandins
1975;9:817–28.
15 Pallone TL, Silldorff EP, Turner MR. Intrarenal blood flow: Microvascular anatomy and the regulation of
medullary perfusion. Clin Exp Pharmacol Physiol 1998;25:383–92.
Page 304
16 Spielman WS, Thompson CI. A proposed role for adenosine in the regulation of renal hemodynamics and
renin release. Am J Physiol 1982;242:F423–35.
17 TeradaY, Tomita K, Nonogushi H, Marumo F. Polymerase chain reaction localization of constitutive
nitric oxide synthase and soluble guanylate cyclase messenger RNAs in microdissected rat nephron
segments. J Clin Invest 1992;90:659–65
18 Mattson DL, Roman RJ, Cowley AW Jr. Role of nitric oxide in renal papillary blood flow and sodium
excretion. Hypertension 1992;19:766–70.
19 Tollins JP, Palmer RMJ, Mondala S, Raij L. Role of endothelium-derived relaxing factor in the
hemodynamic response to acetylcholine in vivo. Am J Physiol 1990;258:665–72.
20 Lüscher TF, Bock HA. The endothelial L-arginine/nitric oxide pathway and the renal circulation. Klin
Wochenschr 1991;69:603–9.
21 Reilly PM, Bulkley GB. Vasoactive mediators and splanchnic perfusion. Crit Care Med 1993;21:S55–68
22 Rosenblum JD, CM Boyle, LB Schwartz. The mesentric circulation. Anatomy and physiology. Surg Clin
North Am 1997;77:289–306.
23 Takala J. Determinants of splanchnic blood flow. Br J Anaesth 1996;77:50–8.
24 Nishida O, Moriyasu F, Nakamura R, et al. Relationship between splenic and superior mesenteric venous
circulation. Gastroenterology 1990;98:721–5.
25 Guyton AC. Textbook of medical physiology, 9th edn. Philadelphia: WB Saunders, 1996:179–80.
26 Parks DA, Jacobson ED. Physiology of the splanchnic circulation. Arch Intern Med 1985;145:1278–81.
27 Escourrou P, Raffestin B, PapelierY, Pussard E, Rowell LB. Cardiopulmonary and carotid baroreflex
control of splanchnic and forearm circulations. Am J Physiol 1993;264:H777– 82.
28 Stadeager C, Hesse B, Henriksen O, et al. Effects of angiotensin blockade on the splanchnic circulation in
normotensive humans. J Appl Physiol 1989;67:786–91.
29 Jacobson ED, Pawlik WW. Adenosine mediation of mesenteric blood flow. J Physiol Pharmacol
1992;43:3–19.
30 Holzer P. Peptidergic sensory neurons in the control of vascular functions: mechanisms and significance
in the cutaneous and splanchnic vascular beds. Rev Physiol Biochem Pharmacol 1992;121:49–146.
31 Gatta A, Merkel C. Clinical pharmacology of splanchnic circulation in cirrhosis. Pharmacol Res
1990;22:235–52.
32 Ahlborg G, Weitzberg E, Lundberg JM. Circulating endothelin-1 reduces splanchnic and renal blood flow
and splanchnic glucose production in humans. J Appl Physiol 1995:79:141–5.
33 Ahlborg G, Lundberg JM. Nitric oxide-endothelin-1 interaction in humans. J Appl Physiol 1997:82:1593–
600.
34 Tokics L, Brismar B, Hedenstierna G. Splanchnic blood flow during halothane-relaxant anaesthesia in
elderly patients. Acta Anaesthesiol Scand 1986;30:556–61.
35 Tverskoy M, Gelman S, Fowler KC, Bradley EL. Influence of fentanyl and morphine on intestinal
circulation.Anesth Analg 1985;64:577–84.
36 Gelman S. Effects of anesthetics on splanchnic circulation. In: Altura BM, Halevy S, eds, Cardiovascular
actions of anesthetics and drugs used in anesthesia, vol 2. Basel: Karger, 1986:126–61.
37 Aneman A, Pontèn J, Fändriks L, Eisenhofer G, Friberg P, Biber B. Hemodynamic, sympathetic and
angiotensin II responses to PEEP ventilation before and during administration of isoflurane. Acta
Anaesthesiol Scand 1997;41:41–8.
38 Love R, Choe E, Lippton H, Flint L, Steinberg S. Positive end-expiratory pressure decreases mesenteric
blood flow despite normalization of cardiac output. J Trauma 1995;39:195–9.
39 Tan OT, Stafford TJ. Cutaneous circulation. In: Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austin
KF, eds. Dermatology and general medicine, 3rd edn. New York: McGraw-Hill, 1987:357–67.
40 Swerlick RA. The structure and function of the cutaneous vasculature. J Dermatol 1997:24:734–8.
41 Pavletic MM. Anatomy and circulation of the canine skin. Microsurgery 1991;12:103–12.
Page 305
42 Midttun M, Sejrsen P. Cutaneous blood flow rate in areas with and without arteriovenous anastomoses
during exercise. Scand J Med Sci Sports 1998;8:84–90.
43 Tripathi A, Mack GW, Nadel ER. Cutaneous vascular reflexes during exercise in the heat. Med Sci Sports
Exerc 1990;22:796–803.
44 Sessler DI, Rubinstein EH. Letter to the editor. Anesthesiology 1989;70:371–2.
45 Johnson JM. Exercise and the cutaneous circulation. Exerc Sport Sci Rev 1992;20:59–97.
46 Jager K, Muench R, Seifert H, Beglinger C, Bollinger A, Fischer JA. Calcitonin gene-related peptide
(CGRP) causes redistribution of blood flow in humans. Eur J Clin Pharmacol 1990;39:491–4.
47 Edfeldt H, Lundvall J. Sympathetic baroreflex control of vascular resistance in comfortably warm man.
Analyses of neurogenic constrictor responses in the resting forearm and in its separate skeletal muscle and
skin tissue compartments. Acta Physiol Scand 1993;147:437–47.
48 Johnson JM, Brengelmann GL, Hales JRS, Vanhoutte PM, Wenger CB. Regulation of the cutaneous
circulation. Fed Proc 1986;45:2841–50.
49 Kellogg DL Jr, Johnson JM, KenneyWL, Pergola PE, KoshibaWA. Mechanisms of control of skin blood
flow during prolonged exercise in humans. Am J Physiol 1993;265:H562–8.
50 Holzer P. Neurogenic vasodilation and plasma leakage in the skin. Gen Pharmacol 1998;30:5–11.
51 Joyner MJ, Dietz NM. Nitric oxide and vasodilation in human limbs. J Appl Physiol 1997;83:1785–96.
52 Fruhstorfer H, Wagener G. Effects of intradermal lignocaine and mepivacaine on human cutaneous
circulation in areas with histamine-induced neurogenic inflammation. Br J Anaesth 1993;70:167–72.
53 Lundin S, Kirnö K, Wallin BG, Elam M. Effects of epidural anaesthesia on sympathetic nerve discharge
to the skin. Acta Anaesthesiol Scand 1990;34:492–7.
54 Shepherd JT. Circulation to skeletal muscle. In: Shepherd JT, Abboud FM, eds. Handbook of physiology,
Section 2, The cardiovascular system, Vol 3, Peripheral circulation and blood flow, Part 1. Baltimore:
Williams &Wilkins, 1983:319–70.
55 Carù B, Colombo E, Santoro F, Laporta A, Maslowsky F. Regional flow responses to exercise. Chest
1992;101:S223–5.
56 Sheriff DD, Rowell LB, Scher AM. Is rapid rise of vascular conductance at onset of dynamic exercise due
to muscle pump? Am J Physiol 1993;265:H1227–34.
57 Smith JJ, Porth CJM. Posture and the circulation: the age effect. Exp Gerontol 1991;26:141–62.
58 Bevan JA, Laher I. Pressure and flow-dependent vascular tone. FASEB J 1991;5:2267–73.
59 Grände PO. Myogenic mechanisms in the skeletal muscle circulation. J Hypertens 1989;7:S47–53.
60 Mellander S. Functional aspects of myogenic vascular control. J Hypertens 1989;7:S21–30.
61 Jacobsen TN, Morgan BJ, Scherrer U, et al. Relative contributions of cardiopulmonary and sinoaortic
baroreflexes in causing sympathetic activation in the human skeletal muscle circulation during orthostatic
stress. Circ Res 1993;73:367–78.
62 Reid MB. Role of nitric oxide in skeletal muscle: synthesis, distribution and functional importance. Acta
Physiol Scand 1998:162:401–9.
63 Delp MD, Lauglin MH. Regulation of skeletal muscle perfusion during exercise. Acta Physiol Scand
1998:162:411–19.
64 Vane JR, Boning RM. Endothelium-derived vasoactive factors and the control of the circulation. Semin
Perinatol 1991;15(1):4–10.
65 Saltin B, Radegran MD, Koskolou, Roach RC. Skeletal muscle blood flow in humans and its regulation
during exercise. Acta Physiol Scand 1998;162:421–36.
66 Longhurst JC, Mitchell JH. Reflex control of the circulation by afferents from skeletal muscle. In: Guyton
AC, Young DB, eds, Cardiovascular physiology III, vol 18. Baltimore: University Park Press, 1979:125–
48.
67 Sellgren J, Ejnell H, Elam M, Ponten J, Wallin BG. Sympathetic muscle nerve activity, peripheral blood
flows, and baroreceptor reflexes in humans during propofol anesthesia and surgery. Anesthesiology
1994;80:534–44.
Page 306
68 White DA, Reitan JA, Kien ND, Thorup SJ. Decrease in vascular resistance in the isolated canine
hindlimb after graded doses of alfentanil, fentanyl, and sufentanil. Anesth Analg 1990;71:29–34.
Page 307
9: Microcirculation
JAMES E BAUMGARDNER, ALEX L LOEB, DAVID E LONGNECKER
A major concern of modern anaesthesia and critical care practice involves optimal
preservation of organs and tissues that are at risk for ischaemia. The effects of anaesthetic
agents on the preservation of vital organ function influences the selection of anaesthetic
drugs and techniques. Anaesthetic drugs have been shown, in animal models, to influence
outcome in several abnormal circulatory conditions that compromise perfusion and tissue
delivery of nutrients such as oxygen and glucose. These circulatory abnormalities include
local organ ischaemia, haemorrhagic shock, and septic shock.
Partial or complete tissue ischaemia presents an obvious compromise of nutrient delivery
to tissue. Several anaesthetic drugs have been shown to provide drug specific protective
effects for partial ischaemia, or for complete ischaemia followed by reperfusion.
Anaesthetic drugs have been shown to differ in their effects on biochemical consequences
and histological outcome for incomplete liver ischaemia,1 for complete liver ischaemia and
reperfusion,2 3 after repeated complete cerebral ischaemia episodes,4 and after incomplete
cerebral ischaemia.5–7
Haemorrhagic shock results in selectively reduced blood flow and nutrient delivery to
several organs, including the splanchnic, renal, skin, and muscle circulations. Anaesthetic
drugs have a drug specific effect on outcome in animal models of haemorrhagic shock
(induced by controlled haemorrhage for a fixed period of time, followed by reperfusion).
Ketamine provided a markedly better survival than halothane or pentobarbital
(pentobarbitone) anaesthesia in haemorrhaged rats.8 This study also reported a reduction in
intestinal mucosal lesions with ketamine anaesthesia. More recently, high epidural
anaesthesia combined with general anaesthesia in haemorrhaged dogs improved survival
compared with general anaesthesia alone,9 again suggesting that anaesthetic techniques
alter outcome from compromised organ perfusion. Anaesthetic specific differences in
bacterial translocation and intestinal histological damage have also been demonstrated in
haeSmorrhaged rats.10
Page 308
The circulatory defects in sepsis and septic shock include a hyper-dynamic circulation
with high cardiac output, low systemic resistance, and increased oxygen uptake, with
proportionally greater increases in oxygen delivery. The resulting decreased oxygen
extraction may represent either inadequate nutrient delivery from blood to tissue or
impaired biochemical use of oxygen. Anaesthetics have specific effects on oxygen delivery
, oxygen uptake , and serum lactate in dogs infused with endotoxin. In a dog
model of septic shock, ketamine caused less lactate accumulation than enflurane,11 and
ketamine preserved cardiovascular function better than halothane, isoflurane, or alfentanil.12
In contrast, enflurane anaesthesia in endotoxaemic rats resulted in significantly less
intestinal pathology than ketamine anaesthesia,13 and halothane improved survival
compared with ketamine after reperfusion of ischaemic bowel.14
Several mechanisms have been suggested to explain these protective effects of
anaesthetic drugs:
z they can suppress tissue metabolic requirements and thereby increase tolerance to
reduced nutrient delivery
z they can modulate biochemical and cellular mediators of ischaemic tissue damage, and
thereby reduce permanent damage after an ischaemic insult
z they are known to have drug specific and dose dependent effects on tissue perfusion,
which can increase tissue oxygenation during an ischaemic insult.
Anaesthetic drugs decrease tissue metabolism in several organs; the extent of this
metabolic suppression depends on the specific anaesthetic and the specific organ. For
example, halothane produces more suppression of metabolism in the myocardium than
isoflurane, but isoflurane produces greater inhibition of metabolism in cerebral tissue.
Decreased metabolic requirements may ameliorate the effects of reductions in tissue
nutrients, but these are specific effects of anaesthetic drugs at the local tissue level and not a
uniform consequence of the general anaesthetic state.
Anaesthetic drugs can also modify the biochemical consequences of tissue hypoxia. The
past 15–20 years have witnessed a tremendous growth in our knowledge of the role of
mediators involved with tissue hypoxia, particularly in the brain. Release of various
excitatory neurotransmitters, such as glutamate, increases in intracellular calcium, and
generation of free radicals and lipid peroxidation are all thought to play a role in the
permanent tissue damage that follows tissue hypoxia. Anaesthetic drugs can have important
effects in modulating these changes after ischaemia. For example, ketamine is an antagonist
of the N-methyl-D-aspartate (NMDA) subtype of the glutamate receptor, and excessive
activation of the glutamate receptor pathway is associated with excitotoxicity and cell
death.15 In addition, anaesthetic drugs have drug specific and organ specific effects on
Page 309
tissue microcirculation, and these actions can modify organ perfusion and cell viability in
response to circulatory insults such as shock, ischaemia, or hypoxaemia.
The primary function of the circulatory system is the delivery of nutrients such as oxygen
and glucose and the removal of waste such as carbon dioxide. The efficient
accomplishment of this task depends on the distribution of blood flow within organs, which
is regulated primarily by the resistance vessels in the microcirculation. The
microcirculatory effects of anaesthetic drugs can influence nutrient delivery and the
distribution of blood flow in the following:
z in cases of globally impaired perfusion and/or oxygenation, as occurs in haemorrhagic

and septic shock
z in cases of incomplete tissue ischaemia as might occur in peripheral vascular disease
z during reperfusion after complete ischaemia, such as that after arterial occlusion during
certain surgical procedures.
This chapter provides an introduction and overview of the microcirculation; it covers

anatomy and terminology, the normal function of the microcirculation, physiological
control mechanisms, and the effects of anaesthetic and other drugs on the microcirculation
and microcirculatory control mechanisms. The microcirculation has many biological
functions, including immune functions, endocrine functions, and solute and water
exchange, but the major focus in this chapter is on the primary tasks of distribution of blood
flow, exchange of nutrients, and regulation of vascular capacitance. Wherever possible, the
effects of anaesthetic drugs will be related to probable roles in improving outcome in
abnormal circulatory conditions.
Anatomy
The microvasculature is described as a series of successive branchings of arterioles, with
a decrease in diameter with each branching generation. This arteriolar branching structure
culminates in the capillaries, which then coalesce into successively larger branches of
venules, finally merging into larger veins. Terminology in the microvasculature follows this
branching scheme, with the largest arterioles and venules in a given vital microscopy
preparation (a living tissue examined directly under the microscope) designated as first
order vessels (for example, 1A arterioles and 1V venules) and the next largest integer
assigned to each successively smaller generation of vessels (2A, 3A, etc) down to the
capillary level. The assignment of typical vessels in a given preparation is somewhat
arbitrary, but it is a useful scheme for comparison between preparations, so that a given
vessel of a
Page 310
numbered generation may share some features with the same generation of a different sized
animal, even though the absolute vessel diameters may not be the same.
The distinction between large arterioles and small arteries is not definitive, but generally
arterioles cover a range of sizes from 10 to 150 μm in diameter. Typically, a microvascular
network includes three to five branching orders spanning this size range. The walls of
arterioles contain relatively large amounts of vascular smooth muscle, which participate in
active circulatory regulation; the smallest arterioles – meta-arterioles – contain intermittent
bands of vascular smooth muscle.
Capillaries are long thin tubes of endothelium, typically 5–10 μm in diameter and 50–
1000 μm in length. They are devoid of vascular smooth muscle except for special bands of
muscle at the arteriolar end of the capillary, the precapillary sphincters, which participate in
regulation of capillary density. Precapillary sphincters are present only in some tissues, but
most small arterioles appear to exhibit sphincter-like activity, and thus act as functional
sphincters. In addition to active regulation of capillary density in a given tissue, different
organs and tissues vary widely in their maximal perfused capillary density, with highly
metabolically active tissues generally having a denser capillary network.
Venules of a given branching order are larger than their equivalent arterioles and have
thinner and more distensible walls. As a result, the cross sectional area of a venular
generation is larger, and blood flow is slower, than for the equivalent arteriolar generation.
Larger venules tend to pair with adjacent arterioles of the same branch order (Fig 9.1). The
parallel arrangement of arterioles and venules may provide an opportunity for
countercurrent exchange of gases and solutes, most notably in intestinal mucosa.16
Quantitative description of even a simple microcirculatory network would require an
enormous number of branch lengths, branch diameters, numbers of branches at
bifurcations, and branching angles at each bifurcation. Progress has been made recently in
fractal descriptions of these networks, reducing the entire description to a single fractal
equation with a small number of parameters.17
Function
Resistance
Distribution of cardiac output to individual organs is regulated locally by changes in

vascular resistance, which is controlled by regulation of vascular diameters. Resistance
(defined as blood flow divided by pressure decrease across a vascular bed) in a cylindrical
vessel varies approximately with the
Page 311
Fig 9.1 Drawing of a typical microvascular network in rat cremaster muscle, with numbering system for
vessel generations based on order in branching hierarchy. (Adapted with permission from Hutchins PM,
Goldstone J, Wells R. Effects of haemorrhagic shock on the microvasculature of skeletal muscle. Microvasc
Res 1973;5:131.)
fourth power of the diameter, with the result that small changes in vascular diameter effect
large changes in resistance. In skeletal muscle and intestinal circulations, small arterioles (<
100 μm in diameter in the cat) are responsible for nearly all of the vascular resistance to
blood flow. Small arterioles make a major contribution to resistance in other circulations as
well, but recently it has been recognised that larger arterioles and small arteries make a
significant contribution to resistance in the coronary circulation, and even larger arteries
make a significant contribution in the cerebral circulation (Fig 9.2).18 19
Many organs and tissues can regulate local blood flow to meet tissue metabolic needs.
Autoregulation refers specifically to maintenance of constant tissue blood flow over a range
of perfusing pressures, when all other variables (for example, tissue metabolic rate, arterial
PCO2 or PaCO2, arterial PO2 or PaO2, neurohumoral inputs) are constant. This regulation
may be mediated by response of vascular smooth muscle to changes in distending pressure
(the myogenic hypothesis) or by feedback regulation of
Page 312
Fig 9.2 Pressure, as percentage of aortic pressure, measured in several sizes of arteries and arterioles of the
cat; for brain, heart, mesentery, and skeletal muscle. (Reproduced with permission from Faraci FM, Heistad
DD. Regulation of large cerebral arteries and cerebral microvascular pressure. Circ Res 1990;66:9.)
local metabolic mediators (the metabolic hypothesis). Although the exact mechanism is
unknown, it is evident from microvascular studies that the vessels responsible for
autoregulation are the small arterioles, with increasingly less contribution as vessel
diameter increases and branching order decreases.19 Many tissues also vary local blood
flow in proportion to tissue metabolic rate (again with other variables such as perfusing
pressure held constant), a phenomenon known as metabolic regulation. Small arterioles are
also primarily responsible for this local regulatory response. The mechanism of metabolic
regulation is unknown, but recent studies of the cerebral circulation in particular suggest a
prominent role for regulation by increased production of nitric oxide (NO) from specific
neurons in response to surrounding neuronal activity.20
Microvascular studies have led to a growing appreciation that alterations in vascular
diameters are usually not uniform and tend to be very site specific. For example,
autoregulatory responses do not change all vessel diameters or vessel resistances
proportionately, but tend to affect the most distal arterioles. In contrast, changes in vascular
diameter from systemic neural inputs alter the larger arterioles and small arteries.
Vasoactive and anaesthetic drugs also affect specific branching orders or sizes of arterioles.
For example, nitroglycerine produces dose dependent dilatation of coronary arterioles and
arteries more than 200 μm in diameter, but causes little change in smaller arteriolar
diameters; in contrast, nifedipine causes homogeneous dilatation of all coronary
microvessels.19 The functional implications of the specificity of microvascular responses
are not yet clear, but may in part explain why a group of vasodilators that produce
seemingly
Page 313
equivalent decreases in total vascular resistance can have drug specific differences in
preventing tissue ischaemia.
Capacitance
Vascular capacitance refers to the ability of the circulatory system to store variable
amounts of blood with minimal changes in central filling pressures. Regulation of vascular
capacitance by neurohumoral mechanisms plays an important role in cardiovascular
responses to haemorrhage and other acute changes in blood volume. About 70% of the
blood volume resides in the venous circulation, with most of it contained in venules in the
microcirculation. Neurohumoral regulation controls the diameter of these microcirculatory
venules, especially in the splanchnic circulation, whereas large veins function primarily as
conduits.21
Exchange
The primary function of the microcirculation is the delivery of oxygen and the removal
of carbon dioxide from tissue. Adequate gas exchange depends on tissue blood flow and
diffusion characteristics between the blood and tissues. It is therefore possible to have
adequate overall perfusion but deficient gas exchange. In septic shock, for example, many
tissues receive increased blood flow, but the decreased extraction of oxygen may produce
tissue hypoxia from inefficient use of this blood flow.22
Capillaries have been classically regarded as the only vessels that exchange significant
amounts of oxygen between blood and tissue, with regulation of capillary density playing
the major role in the regulation of the efficiency of gas exchange. As capillary density
increases, the area available for diffusion of gases to tissue increases proportionately. As
the cross sectional area available for blood flow also increases, the flow velocity in each
capillary decreases, and consequently the time available for diffusion (the capillary transit
time) also increases with increasing capillary density. The regulation of capillary density by
means of the vascular smooth muscle at the precapillary sphincters therefore represents a
potential site for anaesthetic effects on vascular smooth muscle to alter the efficiency of gas
exchange. Precapillary sphincters may also play a role in regulating the efficiency of gas
exchange by matching individual capillary perfusion to local metabolism, thereby reducing
the flow/metabolism heterogeneity that can reduce the efficiency of gas exchange.23
In addition to the classic role of capillaries in tissue gas exchange, it is now recognised
that other vessels in a microcirculatory vessel network can exchange gases with tissues and
with each other.24 Diffusional interactions between blood vessels function essentially as
arteriovenous shunting and thereby impair the efficiency of gas exchange. These diffusional
vessel interactions are sensitive to the time available for diffusion (vessel transit
Page 314
time), which is increased by arteriolar or venular dilatation.25 The vascular smooth muscle
of the arterioles and venules therefore represents another potential site for anaesthetic
effects to influence the efficiency of tissue gas exchange.
Efficient tissue gas exchange depends on the total number of open vessels in a
microvascular network, and on the distribution of blood flow within those vessels.
Anaesthetic and other vasoactive drugs that act on vascular smooth muscle probably
influence the efficiency of gas exchange by these mechanisms, although experimental
studies at this level have been limited. In normal tissue, gas exchange is very efficient and
blood flow can be reduced dramatically (by anaesthetic drugs or other mechanisms) in most
organs before the onset of cellular hypoxia. When tissue oxygenation is impaired as a result
of trauma or disease, however, the microvascular effects of anaesthetic drugs may have an
important influence on tissue oxygenation.
Control of the microcirculation

The contractile activity of smooth muscle in the walls of arteries, arterioles, venules, and
veins is controlled by many mechanisms. Vascular smooth muscle tone is influenced by
remote (for example, centrally mediated) control mechanisms, and by local control
mechanisms. Remote control mechanisms can be further subdivided into neural control and
humoral control. Alterations in arteriolar smooth muscle contraction that result from these
influences are responsible for the changes in resistance that determine the distribution of
cardiac output. There is a remarkable variation in basal vascular tone and resistance from
organ to organ. Renal arterioles, for example, have a very low resting vascular tone and can
constrict markedly in response to stimuli such as hypovolaemia, but they have little
capacity to dilate. In contrast, arterioles in the skin have a high resting tone and can both
dilate and constrict in response to thermo-regulatory stimuli.
Neural control
Arteries and arterioles are innervated by sympathetic fibres terminating in the blood
vessel walls. The extent of sympathetic innervation varies from organ to organ. For
example, renal vessels and mesenteric vessels receive dense innervation by adrenergic
neurons, but the cerebral and coronary vessels receive fairly sparse adrenergic innervation.
Accordingly, organs with a rich supply of adrenergic neurons exhibit a much greater
constrictor response to sympathetic neural stimulation than occurs in the cerebral and
coronary circulations. Similarly, neuronal supply along the generations of the vascular tree
is not uniform, but generally tends to be more dense in the
Page 315
larger vessels with a decreasing neuronal supply in the more distal generations.
Consequently, vasoconstrictor fibre stimulation results not only in a change in organ
vascular resistance, but also in a change in the distribution of organ vascular resistance,
with an increased contribution by larger arterioles and arteries.18 19 Autonomic innervation
plays an important regulatory role in coordination of the vascular response to any stimuli
that activate the sympathoadrenal axis, such as hypovolaemia, hypoxia, hypercapnia, tissue
trauma, and pain. Sympathetic innervation probably plays a minimal role in more local
control, such as local regulation of blood flow in response to increased organ metabolism.
Systemic humoral control
Numerous endogenously produced mediators have been shown to constrict or dilate

vascular smooth muscle. Intravascularly administered adrenaline (epinephrine),
noradrenaline (norepinephrine), angiotensin II, vasopressin, endothelin, prostaglandin
PGF2α, and thromboxane cause dose dependent, organ specific, and site specific
vasoconstriction. Intravascularly administered acetylcholine, atrial natriuretic peptide
(ANP), bradykinin, serotonin, adenosine, prostacyclin (PGI2), and PGE2 cause dose
dependent, organ specific, and site specific vasodilation.19 26 27 There is no doubt that many
of these substances play a role in local humoral regulation of vascular smooth muscle tone.
Although our knowledge of the complexities of endogenous mediators in vascular control is
growing rapidly, the role of endogenous mediators in remote humoral control remains
uncertain.
Local control: vascular endothelium
Over the past two decades there has been a tremendous growth in our understanding of
the role of vascular endothelium in mediating effects of vasoactive drugs. For example,
endogenous compounds such as acetylcholine, bradykinin, ATP, histamine, and many
others dilate arterioles by stimulating the release of multiple endothelium derived relaxing
factors (EDRFs), which include nitric oxide (NO), endothelium derived hyper-polarising
factor (EDHF), and eicosanoids. These different EDRFs can act independently or in concert
to influence microvascular diameters and blood flows.28–31
This new understanding of the role of the endothelium in vascular control was initiated
by the discovery32 that NO (then called EDRF) functioned as an important mediator of
vasodilation. EDRF, subsequently identified as NO, is produced by the conversion of
arginine and oxygen into NO and citrulline, a reaction that is catalysed by the enzyme NO
synthase
Page 316
(NOS).33 34 Although many forms of NOS have been identified, they may be classified into
constitutive forms (that is, always expressed), those found in endothelial cells and neurons,
and inducible forms that can be expressed in response to stimuli such as inflammation.
The multiple locations of these isoforms of NOS imply multiple functions for the
gaseous transmitter, and it is now evident that NO is involved in a variety of biological
functions, including local cardiovascular control, neuronal transmission, immune function,
and the control of pathogens (for example, bacteria). This discussion focuses only on
microvascular control functions of NO, but concise reviews are available that describe its
other biological functions.35
The actions of NO on the peripheral circulation can be evaluated by the infusion of an
arginine analogue inhibitor of NOS into intact animals, and recording the resultant changes
in cardiovascular function. We have used this approach to determine the actions of NO on
blood pressure, cardiac output, and regional blood flows in conscious rats, by injecting
radio-labelled microspheres into the circulation before and during the infusion of the NOS
inhibitor NG-monomethyl-L-arginine (L-NMMA), and again after the infusion of L-
arginine, to demonstrate that the changes observed during response to L-NMMA were
reversible by providing more substrate for NO production.36 Inhibition of NO produced
major changes in blood pressure regulation and peripheral circulatory control. The general
circulatory effects are summarised in Table 9.1.
In brief, inhibition of NO caused a doubling of systemic vascular resistance and
increased arterial pressure, accompanied by decreases in heart rate and cardiac output.
Inhibition of NO by L-NMMA decreased local blood flow in the cerebrum, heart,
kidneys, spleen, gastrointestinal tract, portal vein, liver (total flow), skin, ear, and white fat,
whereas flow increased in the hepatic artery (presumably in response to the decrease in
portal and total hepatic
Table 9.1 Effects of EDRF/NO inhibition by L-NMMA and reversal of inhibition by L-
arginine on systemic haemodynamics
Treatment
Parameter Control L-NMMA L-Arginine
CO (ml/min) 90±5 52±5 68±6

HR (beats/min) 441±13 404±17 430±16
MAP (mm Hg) 129±5 145±5 134±4
SVR (mm Hg × min 1.5±0.1 3.0±0.3 2.l±0.2
× g/ml)
CO, cardiac output; HR, heart rate; MAP, mean arterial pressure; SVR, systemic vascular
resistance.
(Reproduced with permission from Greenblatt et al.28 36)
Page 317
Fig 9.3 The percentage increase in mean regional vascular resistances induced by L-NMMA in conscious rats,
reflecting the heterogeneity of the contribution of EDRF/NO to regional vascular control. (Reproduced with
permission from Greenblatt EP, Loeb AL, Longnecker DE. Marked regional heterogeneity in the magnitude
of EDRF/NO-mediated vascular tone in awake rats. J Cardiovasc Pharmacol 1993;21:237.)
blood flow). Vascular resistances were increased in virtually every tissue studied, although
the magnitude of the response was variable among organs and tissues, indicating that the
relative importance of this system depends on local factors (Fig 9.3).
Similar results have been observed in numerous studies in isolated blood vessels or local
microvascular networks, suggesting that the results obtained above can be attributed to
inhibition of NO in the vascular endothelium, although the possibility of remote effects
cannot be ruled out whenever inhibitors of NO are administered systemically. These studies
provide clear evidence that NO is a profound and ubiquitous controller of the peripheral
circulation. It is clear, however, that NO is not the only controller of the peripheral
circulation, and it is impossible to attribute either physiological or pharmacological effects
in the microcirculation to a single mechanism, because multiple mechanisms are involved
in the control of vascular tone.
Effects of anaesthetic drugs on the microcirculation

This section does not present a comprehensive review of the effects of each anaesthetic
on each individual tissue, but focuses instead on a few illustrative studies of the differences
in peripheral circulatory effects of anaesthetic drugs. Wherever possible, comparison is
made between the
Page 318
effects of two anaesthetic drugs in the same study, using the same protocol.
Isoflurane versus halothane
Comparison of the effects of isoflurane against those of halothane emphatically

illustrates that two anaesthetic drugs that appear to be similar on a superficial level can be
quite distinctive on more detailed investigation. When delivered in equipotent
concentrations (minimum alveolar concentration or MAC), these two agents produce
approximately equal decreases in mean arterial blood pressure. However, when cardiac
output and systemic vascular resistance are also measured, it becomes apparent that
halothane lowers blood pressure primarily by decreases in cardiac output, whereas
isoflurane primarily decreases systemic vascular resistance.37–39 Further differences
between these two agents are evident on more detailed investigation of individual organ
blood flow and microcirculatory vascular diameters. In general, isoflurane is the more
potent vasodilator, but the relative decreases in vascular resistance for these agents are
organ specific, and the extent of dilatation also tends to be specific for vessel branching
order within a tissue.
Organ resistances and blood flows

Anaesthetic drugs mediate changes in organ blood flow by changes in vascular diameter
and resistance. Resistance is a quotient of pressure divided by flow, and therefore
measurements of systemic pressure and local (organ or tissue) blood flow can be used to
calculate changes in local resistance. Several techniques are available to measure organ
blood flow in laboratory and clinical studies. Radioactive microspheres, which distribute to
organs in proportion to blood flow, provide a quantitative and absolute measure of tissue
blood flow, but require destruction of the tissue sample for counting and weighing to
achieve the most accuracy, and therefore are limited to laboratory studies. Indicator dilution
techniques (for example, thermal or indocyanine green dilution) also provide absolute and
quantitative blood flow measurements. These techniques require venous outflow sampling,
but are applicable in some clinical studies when an appropriate venous catheter can be
placed. Large vessel flow probes (for example, electromagnetic or Doppler ultrasonic
probes) require invasive placement, but can be implanted for chronic use. Some thermal
and inert tracer clearance methods are invasive (for example, hydrogen clearance
techniques that require implantation of hydrogen microelectrodes), whereas others are non-
invasive and are used clinically, for example, radioactive xenon (133Xe) wash-out. More
recently, laser Doppler velocimetry (LDV) has been used in both laboratory and clinical
studies. This technique measures relative changes in blood flow quantitatively and
continuously.
Page 319
In dogs receiving an opiate general anaesthestic, either halothane or isoflurane was added
and titrated to decrease mean arterial pressure to 60 mm Hg.40 Inspired isoflurane of 1·5%
maintained myocardial blood flow (measured by radioactive microspheres) and decreased
coronary vascular resistance by 40%. In contrast, 1·1 % inspired halothane decreased
myocardial blood flow by 35% and decreased coronary resistance by 13%.
Debaene et al41 induced cirrhosis in rats by bile duct ligation and then compared the
influence of anaesthetic drugs on liver blood flow during mild haemorrhage (removal of
20% of estimated blood volume). One MAC (1·3% inspired in rats) isoflurane anaesthesia
maintained hepatic arterial flow (measured by radioactive microspheres) at prehaemorrhage
values. One MAC (1·0% inspired in rats) halothane anaesthesia caused a significant
decrease in hepatic arterial flow during haemorrhage.
Frink et al1 studied portal venous and hepatic arterial blood flows by means of
chronically implanted probes in healthy dogs. They measured the effects of four different
anaesthetic drugs on portal venous and hepatic arterial blood flow, compared with
conscious controls, for three different doses of anaesthetic drugs (1·0, 1·5, and 2·0 MAC).
Halothane reduced hepatic arterial blood flow in a dose dependent fashion and increased
hepatic arterial vascular resistance. In contrast, isoflurane maintained hepatic arterial flow
at all doses. Isoflurane reduced portal venous flow at the higher doses. Halothane caused
greater decreases in portal venous flow at all doses.
Vollmar et al42 studied liver and pancreatic blood flows (by microspheres) and tissue
oxygen tensions (by multiwire surface electrodes) in rats. Isoflurane and halothane were
titrated to reduce mean arterial pressure to 50 mm Hg (2·3% inspired isoflurane, and 1·0%
inspired halothane). Isoflurane and halothane both caused similar reductions in portal
venous and hepatic arterial blood flows (by radioactive microspheres) compared with
chloralose anaesthetised controls, but isoflurane reduced mean tissue PO2 from 4·08 kPa to
2·33 kPa, whereas halothane caused greater reductions to a mean tissue PO2 of 1·53 kPa.
Halothane maintained pancreatic blood flow and tissue oxygen tension compared with
controls. In contrast, isoflurane increased pancreatic blood flow and tissue oxygen tension.
Jacob et al43 studied arterial blood flow (by means of implanted Doppler probes)
supplying an ileocolic graft after oesophageal reconstruction in humans.43 Fentanyl
infusions of 300 μg/h that were established intra-operatively were maintained immediately
postoperatively and then 0·65 MAC isoflurane or halothane was administered in a crossover
design. Halothane did not change mesenteric blood flow and isoflurane increased
mesenteric blood flow by 38%.
Hansen et al44 studied the distribution of cerebral blood flow in rats at 1 MAC of either
isoflurane or halothane.44 Blood flows to the subcortex were
Page 320
not significantly different between halothane and isoflurane, but halothane caused
significantly greater blood flows in the neocortex.
Vessel diameters
Vessel diameters can be studied in intact tissue under direct microscopic visualisation
(vital microscopy) by means of either transillumination (for a suitably thin tissue) or epi-
illumination of a tissue surface. This invasive technique has been applied primarily in
anaesthetised animals. Some vascular networks have been examined microscopically in
humans (for example, the nail fold),45 but these are fairly specialised circulations and it is
difficult to extrapolate results to other organs of interest. Direct microscopic visualisation
has the unique advantage of providing information about the specific vessels and branching
orders that are influenced by anaesthetic drugs. The studies are, however, technically
demanding and there have been fewer reports of the comparative effects of anaesthetic
drugs using these techniques.
Conzen et al40 studied the microcirculation of the epicardial surface of the left ventricle
in intact, beating dog hearts by means of epi-illumination.40 The dogs were anaesthetised
with opiate infusions and then received isoflurane or halothane titrated to reduce mean
arterial pressure to 60 mmHg (1·1% inspired halothane, 1·5% inspired isoflurane).
Isoflurane caused larger increases in arteriolar diameters than halothane. Both agents
dilated 20–200 μm vessels specifically, and had no effects on larger arterioles or on
precapillary sphincters.
Leon et al46 studied rat diaphragm arteriolar diameters and functional capillary density by
vital microscopy. All rats received pentobarbital anaesthesia, followed by three
concentrations (0·50, 0·75, and 1·0 MAC) of either halothane or isoflurane. Halothane
administration caused dose dependent constriction specifically in the A4 arterioles of
diaphragm muscle, leaving A2 and A3 arterioles unchanged, and also decreased capillary
density in a dose dependent manner. Administration of isoflurane caused no significant
changes in arteriolar diameter or capillary density.
Summary of isoflurane versus halothane

This brief and selective review of recent studies of organ resistances and blood flows
demonstrates the different effects of isoflurane and halothane for heart, liver, pancreas,
mesentery, and brain, and also different effects on specific regions within an organ. In the
case of the liver and pancreas, these agent specific microcirculatory effects have been
associated with changes in tissue oxygen tension. The vital microscopy studies in cardiac
and skeletal muscle demonstrate the specificity of the circulatory effects of anaesthetic
drugs for vessels of a particular size and branching order. Further studies will be required to
connect the specificity of these changes to the specific influence of anaesthetic drugs on the
delivery of nutrients to tissue. In
Page 321
general, isoflurane tends to be a more potent vasodilator than halothane, and as a result
tends to maintain tissue blood flow, although there are exceptions for selected tissues and
for specific pathologies. The clinical importance is that, in patients with circulatory
pathology and compromised tissue oxygenation, the microcirculatory effects of anaesthetic
drugs have the potential to influence tissue preservation in the perioperative period.
Parenteral anaesthetic drugs
Compared with the inhalational anaesthetic drugs, considerably fewer studies of the
effects of parenteral anaesthetic drugs on organ blood flows and microvascular diameters
have been reported. Data are particularly scarce for direct comparisons of two or more
parenteral anaesthetic techniques, for example high dose opiate anaesthesia compared with
propofol infusion. There have, however, been several microvascular studies comparing
ketamine with other anaesthetic techniques, and a selective review of these reports further
supports the concept of drug specific effects on organ blood flows and branching
generations. Interest in ketamine arises in part because it is used in laboratory animals as
well as in humans, and also because of its unique beneficial properties in haemorrhagic
shock.8 47 Ketamine may also be uniquely beneficial in septic shock,12 48 although the data
in this area are more conflicting.13 14
Organ resistances and blood flows

Debaene et al41 compared the effects of several anaesthetic drugs, including ketamine and
enflurane, on portal venous blood flow and hepatic arterial blood flow during mild
haemorrhage (removal of 20% of estimated blood volume) in cirrhotic rats. During
haemorrhage, enflurane (2·2% inspired) decreased portal venous flow more than ketamine
(1·5 mg/kg/per min i.v.), but after reinfusion of the shed blood the portal venous blood
flows were similar. During haemorrhage, enflurane decreased hepatic arterial blood flow
and ketamine resulted in an unchanged hepatic arterial blood flow.
In isolated intestinal loops in dogs, Tverskoy et al49 found that ketamine (both 8 and 16
mg/kg i.v.) caused increased vascular resistance and decreased blood flow compared with
pentobarbital.
Seyde and Longnecker50 studied the effects of four anaesthetic drugs (ketamine 1
mg/kg/per min i.v., halothane 1·2% inspired, isoflurane 1·4% inspired, and enflurane 2·2%
inspired) on organ blood flows (measured by microspheres) and organ vascular resistances
in rats. Results were compared with the effects in conscious controls, during both
normovolaemia and moderate hypovolaemia (removal of 30% of estimated blood
volume).50 Before haemorrhage, ketamine maintained cerebral blood flow at conscious
values, whereas the inhaled agents increased cerebral blood flow. After haemorrhage,
ketamine, enflurane, and halothane reduced
Page 322
cerebral blood flow significantly compared with conscious animals, but isoflurane did not.
At baseline, isoflurane maintained myocardial blood flow at conscious values, and
ketamine, enflurane, and halothane progressively reduced myocardial blood flow. After
haemorrhage, myocardial blood flow increased in conscious animals, did not change under
isoflurane or halothane anaesthesia, and decreased under ketamine or enflurane anaesthesia.
Before haemorrhage, portal venous flow was increased, compared with conscious animals,
by ketamine anaesthesia and was reduced compared with controls by halothane anaesthesia;
portal flow was decreased in all groups after haemorrhage. Hepatic arterial blood flow after
haemorrhage was greatest in animals receiving isoflurane anaesthesia, although ketamine
resulted in an increase in hepatic arterial blood flow, compared with normovolaemia.
Several other organ blood flows were studied, and the study demonstrates the drug specific
effects of the anaesthetic drugs on organ vascular resistances, as well as the unique
properties of ketamine.
Miller et al51 compared the distribution of cardiac output in conscious rats with that
during halothane, enflurane, or ketamine anaesthesia.51 Halothane (1·3% inspired)
decreased cardiac output and increased the percentage of cardiac output going to the brain,
kidney, liver, and large intestine. Enflurane (2·2% inspired) also increased the percentage of
cardiac output going to the liver, spleen, and large intestine, but the cardiac output did not
decrease compared with the conscious state. Ketamine 125mg/kg i.m., tended to decrease
cardiac output (although not significantly), and the percentage of the cardiac output going
to the brain increased, whereas the percentage of the cardiac output going to muscle
decreased, and that going to the heart, intestine, liver, and kidneys did not change.
In rats haemorrhaged to a mean arterial pressure of 60 mm Hg, ketamine anaesthesia (1·5
mg/kg per min i.v.) resulted in increased blood flow to the brain, heart, kidneys, intestine,
and liver, compared with pentobarbital anaesthesia (6 mg/kg per h i.v.)52
Gaab et al53 used multiwire microelectrodes on the surface of rat cerebral cortex to
measure tissue PO2 profiles, local blood flow profiles (by means of hydrogen clearance),
and local metabolism (for example, the rate of oxygen depletion during carotid occlusion).
Ketamine 320 mg/kg i.p., or pentobarbital, 65 mg/kg i.p. resulted in almost identical tissue
PO2 profiles, but very different local blood flow profiles. Ketamine anaesthesia resulted in
greater mean blood flows compared with pentobarbital anaesthesia. Ketamine also resulted
in greater tissue metabolism compared with pentobarbitone anaesthesia.
Dempsey et al54 measured cerebral blood flow in cats after the release of middle cerebral
artery occlusion. In cats anaesthetised by pentobarbital, pre-treatment with indometacin
(indomethacin) resulted in increased
Page 323
postischaemic cerebral blood flow compared with cats that did not receive indometacin pre-
treatment. In cats anaesthetised by ketamine, pre-treatment with indometacin did not affect
postischaemic cerebral blood flow.
Vessel diameters
Longnecker et al55 measured arteriolar diameters microscopically in rat cremaster muscle
during haemorrhage, and compared the effects of enflurane anaesthesia and ketamine
anaesthesia. They also measured tissue PO2 with oxygen microelectrodes. Severe
haemorrhage (removal of blood to decrease mean arterial pressure to 35 mm Hg for 30 min)
during enflurane anaesthesia (2·2 vol% inspired) resulted in constriction in 1A, 3A, and 4A
arterioles and decreased tissue PO2 in skeletal muscle. Haemorrhage during ketamine
anaesthesia (125 mg/kg i.m. with supplements of 30 mg/kg i.m. as needed) resulted in less
constriction of the larger arterioles and increases in diameter in the 4A arterioles, and no
significant decreases in tissue PO2.
Longnecker and Harris56 57 reported changes in arteriolar and venular diameters in bat
wing, a mammalian skin microvasculature, and compared values with those during
ketamine or halothane anaesthesia. Ketamine anaesthesia (120 mg/kg i.m.) produced
dilatation of the small arterioles (30–65 μm) and no change in venular diameters, compared
with conscious controls. A smaller dose of ketamine, 40 mg/kg i.m., did not change
arteriolar or venular diameters. In contrast, 0·71 MAC of halothane (0·81% inspired) caused
arteriolar dilatation and no change in venular diameters, whereas 1·25 MAC of halothane
(1·42% inspired) caused more dilatation of arterioles and dilatation of the venules as well.
Summary of parenteral anaesthetic drugs

There have been relatively few studies of the microcirculatory effects of parenteral
anaesthetic drugs, and generalisations about their effects are therefore difficult. It is clear,
however, from the studies on organ blood flows and vessel diameters reviewed here, that
the effects of parenteral anaesthetic drugs are agent specific, organ specific, pathology
specific (that is, haemorrhage vs normovolaemia) and vessel generation specific (that is,
fourth order arterioles versus first and second order arterioles). Further studies will be
required to connect the specificity of these changes to the specific influence of anaesthetic
drugs on delivery of nutrients to tissue.
Mechanisms of anaesthetic effects on the

microcirculation
The microcirculation is controlled by a number of remote (neural, humoral, and
hormonal) and local (metabolic, endothelial, and myogenic)
Page 324
regulatory mechanisms, and it is the summation of these actions that result in overall
microcirculatory control. Similarly, the anaesthetic drugs alter numerous biological
functions, including cellular metabolism, neuronal activity and transmission, humoral and
hormonal control, and direct actions on vascular smooth muscle. It is possible, however, to
demonstrate the effects of anaesthetic drugs on specific microvascular control systems in
some situations, and these at least enhance the logic of the argument that the peripheral
vascular actions of the anaesthetic drugs can be explained by actions on physiological
systems that control the microcirculation. The actions of nitrous oxide (N2O), halothane,
and isoflurane are used as examples here.
Nitrous oxide has been shown to activate the sympathetic nervous system, as evidenced
by increased splanchnic nerve activity when N2O was administered to cats receiving
halothane anaesthesia.58 We studied the peripheral circulatory actions of N2O, when added
to a halothane anaesthesia in rats, in order to determine whether the reported changes in
sympathetic activity in the splanchnic nerve were associated with changes in blood flow
and vascular resistance in the splanchnic viscera.59 Blood flows were measured using
radiolabelled microspheres. N2O, when substituted for nitrogen in the breathing mixture,
caused increased vascular resistance in the kidneys, small bowel, and spleen, and decreased
blood flow in the kidneys, small bowel, spleen, and liver. Cardiac output decreased in
response to N2O and the cerebral blood flow increased subsequent to cerebral vasodilation.
The increase in vascular resistance and the decrease in blood flow to the splanchnic viscera
are consistent with the increase in sympathetic nerve activity seen in the splanchnic nerve
of cats, and suggests that the effects of N2O in the splanchnic viscera can be attributed to an
action of sympathetic nerves. (This mechanism does not, of course, explain the actions of
N2O on the cerebral circulation, and serves to emphasise the diverse actions of anaesthetic
drugs on various peripheral vascular circulations.)
Although both halothane and isoflurane produce arterial hypotension in humans and
animals, the mechanism(s) of this effect differ considerably. Halothane decreases blood
pressure primarily by decreasing cardiac output,38 50 whereas isoflurane acts primarily as a
peripheral vasodilator.39 51 We reasoned that the peripheral vascular actions of isoflurane
might result from its actions on the NO system, and tested this hypothesis by administering
L-NMMA, a blocker of NO synthesis, in the presence of either halothane or isoflurane
anaesthesia.60 Systemic and regional haemo-dynamics were measured with the
radiolabelled microsphere technique.
Blockade of NO by L-NMMA produced greater increases in arterial blood pressure and
systemic vascular resistance in animals anaesthetised with isoflurane, indicating that the
NO pathway was more prominent under isoflurane anaesthesia (Fig 9.4).
Page 325
Fig 9.4 Comparison of changes in systemic haemodynamics induced by L-NMMA during 1 MAC of
halothane anaesthesia (open columns) versus 1 MAC of isoflurane anaesthesia (filled columns) in rats. CO,
cardiac output; HR, heart rate; MAP, mean arterial pressure; SVR, systemic vascular resistance. (Reproduced
with permission from Greenblatt and Loeb.60)
Similarly, blockade of the NO system produced greater increases in vascular resistance in

the heart, kidneys, gastrointestinal tract, hepatic artery, and skin in those receiving
isoflurane anaesthesia (Fig 9.5).
These results provide evidence that the NO pathway acts during both halothane and
isoflurane anaesthesia to produce vasodilation, but the system is much more active during
isoflurane anaesthesia and this mechanism contributes at least in part to the vasodilation
that is observed during isoflurane anaesthesia.
More recent studies demonstrated that the contribution of NO to agonist-stimulated
microvascular responses was greater during isoflurane than during halothane anaesthesia.61
Acetylcholine and bradykinin induced vasodilation in fourth order rat cremaster muscle
arterioles in a NO dependent manner during isoflurane anaesthesia, because vasodilation
was attenuated by pre-treatment of the NOS inhibitor L-NMMA. In contrast, responses to
these agonists were unaltered by L-NMMA during halothane or ketamine anaesthesia. The
NO independent vasodilation seen in the presence of halothane or ketamine was the result
of EDHF release from vascular endothelium, because combining NOS inhibition with an
increase in K+ in the muscle superfusate to inhibit EDHF action prevented the dilatation.61
62 These data indicated that both EDHF and NO contribute to microvascular responsiveness
during anaesthesia, and that the relative contributions of the different vasodilators to
vascular control can be altered by specific anaesthetic drugs.
The responses to agonists during isoflurane anaesthesia were similar to those observed in
the presence of inhibitors of EDHF action, and suggested that anaesthesia could modify
vascular cell membrane potential, one of the
Page 326
primary controls over the vascular smooth muscle contractile state. Evidence to support the
concept that isoflurane and other anaesthetic drugs may influence vascular smooth muscle
membrane potential and EDHF action has been provided by a number of investigators.
Lischke and co-workers63 64 have shown that isoflurane, halothane, enflurane, sevoflurane,
desflurane, etomidate, and thiopental can inhibit EDHF activity, with isoflurane being the
most potent of the volatile anaesthetic drugs. Yamazaki et al65 have shown that halothane,
isoflurane, and sevoflurane can directly hyperpolarise vascular smooth muscle cells in the
rat mesenteric microcirculation. Hyperpolarisation of the vascular smooth muscle would be
expected to induce relaxation and vasodilation.
Hyperpolarisation of vascular smooth muscle by the anaesthetic drug may reduce the
sensitivity of the muscle to the additional influences of EDHF, while at the same time
inducing vasodilation directly. This mechanism may be responsible for both the direct
vasodilator activities of anaesthetic drugs such as isoflurane and modification of the relative
contribution of different EDRFs to microvascular control.
Together, these results indicate that the anaesthetic drugs act on at least some of the
known peripheral circulatory control mechanisms to produce
Fig 9.5 Comparison of changes in regional vascular resistances induced by L-NMMA during 1 MAC of
halothane (open columns) versus 1 MAC of isoflurane (filled columns) in rats. (Reproduced with permission
from Greenblatt and Loeb.60)
Page 327
their effects on organ blood flows and vascular resistances. They do not, however, imply
that the mechanisms described here are the only mechanisms that may be active during
anaesthesia. For example, peripheral vasoconstriction may occur during hypovolaemia
under anaesthesia50 and presumably other remote or local factors may predominate under
other circumstances. The results with the inhalational anaesthetic drugs illustrate, however,
that an understanding of microcirculatory control mechanisms, combined with an
understanding of the actions of the anaesthetic drugs, allows one to make reasonable
assumptions about the mechanisms that may explain the peripheral circulatory actions of
the anaesthetic drugs.
1 Frink EJ, Morgan SE, Coetzee A, Conzen PF, Brown BR. The effects of sevoflurane, halothane,
enflurane, and isoflurane on hepatic blood flow and oxygenation in chronically instrumented greyhound
dogs. Anesthesiology 1992;76:85–90.
2 Nordstrom G, Winso O, Biber B, Hasselgren PO. Influence of pentobarbital and chloralose on metabolic
and hemodynamic changes in liver ischaemia. Ann Surg 1990;212:23–9.
3 Nagano K, Gelman S, Parks D, Bradley EL. Hepatic circulation and oxygen supply–uptake relationships
after hepatic ischemic insult during anesthesia with volatile anesthetics and fentanyl in miniature pigs.
Anesth Analg 1990;70:53–62.
4 Holder DS. Effects of urethane, alphaxolone/alphadolone, or halothane with or without neuromuscular
blockade on survival during repeated episodes of global cerebral ischemia in the rat. Lab Anim
1992;26:107–13.
5 Rasool N, Faroqui M, Rubenstein EH. Lidocaine accelerates neuroelectrical recovery after incomplete
global ischemia in rabbits. Stroke 1990;21:929–35.
6 Helfaer MA, Kirsch JR, Traystman RJ. Anesthetic modulation of cerebral hemodynamic and evoked
responses to transient middle cerebral artery occlusion in cats. Stroke 1990;21:795–800.
7 Sano T, Patel PM, Drummond JC, Cole DJ. A comparison of the cerebral protective effects of etomidate,
thiopental, and isoflurane in a model of forebrain ischemia in the rat. Anesth Analg l993;76:990–7.
8 Longnecker DE, Sturgill BC. Influence of anesthetic agent on survival following hemorrhage.
9 Shibata K, Yamamoto Y, Murakami S. Effects of epidural anesthesia on cardiovascular response and
survival in experimental haemorrhagic shock in dogs. Anesthesiology 1989;71:953–9.
10 LaRocco MT, Rodriguez LF, Chen CY, et al. Reevaluation of the linkage between acute hemorrhagic
shock and bacterial translocation in the rat. Circ Shock 1993;40:212–20.
11 VanderLinden P, Gilbart E, Engleman E, deRood M, Vincent JL. Adrenergic support during anaesthesia
in experimental endotoxin shock: norepinephrine versus dobutamine. Acta Anaesthesiol Scand
1991;35:134–40.
12 VanderLinden P, Gilbart E, Engelman E, Schmartz D, deRood M, Vincent JL. Comparison of halothane,
isoflurane, alfentanil, and ketamine in experimental septic shock. Anesth Analg 1990:70:608–17.
13 Schaefer CF, Bracken DJ, Tompkins P, Wilson MF. Choice of anaesthetic alters the circulatory shock
pattern as gauged by conscious rat endotoxemia. Acta Anaesthesiol Scand 1987;31:550–6.
14 Bavister PH, Longnecker DE. Influence of anaesthetic agents on the survival of rats following acute
ischaemia of the bowel. Br J Anaesth 1979;51:921–5.
15 Lipton SA, Rosenberg PA. Excitatory amino acids as a final common pathway for neurologic disorders. N
Engl J Med 1994;330:613–22.
16 Lundgren O, Haglund U. The pathophysiology of the intestinal countercurrent exchanger. Life Sci
1978;23:1411–22.
Page 328
17 Glenny RW, Robertson HT, Yamashiro S, Bassingthwaighte JB. Applications of fractal analysis to
physiology. J Appl Physiol 1991;70:2351–67.
18 Faraci FM, Heistad DD. Regulation of large cerebral arteries and cerebral microvascular pressure. Circ
Res 1990;66:8–17.
19 Marcus ML, Chilian WM, Kanatsuka H, Dellsperger KC, Eastham CL, Lamping KG. Understanding the
coronary circulation through studies at the microvascular level. Circulation 1990;82:l–7.
20 Iadecola C. Regulation of the cerebral microcirculation during neural activity: is nitric oxide the missing
link? Trends Neurol Sci 1993;16:206–14.
21 Hainsworth R. The importance of vascular capacitance in cardiovascular control. News Physiol Sci
1990;5:250–4.
22 Nelson DP, King CE, Dodd SL, Schumacker PT, Cain SM. Systemic and intestinal limits of O2 extraction
in the dog. J Appl Physiol 1987;63:387–94.
23 Piiper J, Pendergast DR, Marconi C, Meyer M, Heisler N, Cerretelli P. Blood flow distribution in dog
gastrocnemius muscle at rest and during stimulation. J Appl Physiol 1985;586:2068–74.
24 Ellsworth M, Ellis C, Popel A, Pittman R. Role of microvessels in oxygen supply to tissue. News Physiol
Sci 1994;9:119–23.
25 Secomb TW, Hsu R. Simulation of O2 transport in skeletal muscle: diffusive exchange between arterioles
and capillaries. Am J Physiol 1994;267:H1214–21.
26 Carmines PK, Fleming JT. Control of the renal microvascularure by vasoactive peptides. FASEB J
1990;4:3300–9.
27 Vanlersberghe C, Lauwers MH, Camu F. Prostaglandin synthetase inhibitor treatment and the regulatory
role of prostaglandins on organ perfusion. Acta Anaesth Belg 1992;43:211–25.
28 Saito Y, Eraslan A, Lockard V, Hester RL. Role of venular endothelium in control of arteriolar diameter
during functional hyperemia. Am J Physiol 1994;267:H1227–31.
29 Nagao T, Vanhoutte PM. Endothelium-derived hyperpolarizing factor and endothelium-dependent
relaxations. Am J Respir Cell Mol Biol 1993;8:1–6.
30 Koller A, Sun D, Huang A, Kaley G. Corelease of nitric oxide and prostaglandins mediates flow-
dependent dilatation of rat gracilis muscle arterioles. Am J Physiol 1994;267: H326–32.
31 Garland CJ, Plane F, Kemp BK, Cocks TM. Endothelium-dependent hyperpolarization: a role in the
control of vascular tone. Trends Pharmacol Sci 1995;16:23–30.
32 Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth
muscle by acetylcholine. Nature 1980;288:373–6.
33 Ignarro LJ, Buga GM, Wood KS, Byrns RE, and Chaudhuri G. Endothelium-derived relaxing factor
produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci USA 1987;84:9265–9.
34 Palmer RMJ, Ferridge AG, and Moncada S. Nitric oxide release accounts for the biological activity of
endothelium-derived relaxing factor. Nature 1987;327:524–6
35 Lowenstein CJ, Dinerman JL, Snyder SH. Nitric oxide: a physiologic messenger. Ann Intern Med
1994;120:227–37.
36 Greenblatt EP, Loeb AL, Longnecker DE. Marked regional heterogeneity in the magnitude of EDRF/NO-
mediated vascular tone in awake rats. J Cardiovasc Pharmacol 1993;21:2350.
37 Longnecker DE. Effects of general anaesthetic drugs on the microcirculation. Microcirc Endothel Lymph
1984;1:129–50.
38 Eger EI II, Smith NT, Cullen DJ, Cullen BF, Gregory GA. A comparison of the cardiovascular effects of
halothane, fluroxene, ether, and cyclopropane in man: a resumé. Anesthesiology 1971;34:25–41.
39 Stevens WC, Cromwell TH, Halsey MJ, Eger EI II, Shakespeare TF, Bahlman SH. The cardiovascular
effects of a new inhalation anaesthetic, Forane, in human volunteers at arterial carbon dioxide tension.
40 Conzen PF, Habazettl H, Vollmar B, Christ M, Baier H, Peter K. Coronary microcirculation during
halothane, enflurane, isoflurane, and adenosine in dogs. Anesthesiology 1992;76:261015–70.
Page 329
41 Debaene B, Goldfarb G, Braillon A, Jolis P, Lebrec D. Effects of ketamine, halothane, enflurane, and
isoflurane on systemic and splanchnic hemodynamics in normovolemic and hypovolemic cirrhotic rats.
42 Vollmar B, Conzen PF, Kerner T, et al. Blood flow and tissue oxygen pressures of liver and pancreas in
rats: Effects of volatile anesthetics and of haemorrhage. Anesth Analg 1992;75:421–30.
43 Jacob L, Boudaoud S, Payen D, et al. Isoflurane, and not halothane, increases mesenteric blood flow
supplying esophageal ileocoloplasty. Anesthesiology 1991;74:699–704.
44 Hansen TD, Warner DS, Todd MM, Vust LJ, Trawick DC. Distribution of cerebral blood flow during
halothane versus isoflurane anaesthesia in rats. Anesthesiology 1988;69:332–7.
45 Fagrell B, Fronek A, Intaglietta M. A microscope–television system for studying flow velocity in human
skin capillaries. Am J Physiol 1977;233:H318–21.
46 Leon A, Boczkowski J, Dureuil B, Vicaut E, Aubier M, Desmonts J. Diaphragmatic microcirculation
during halothane and isoflurane exposure in pentobarbital-anesthetized rats. J Appl Physiol 1992;73:1614–
18.
47 Longnecker DE, McCoy S, Drucker WR. Anesthetic influence on response to haemorrhage in rats. Circ
Shock 1979;6:55–60.
48 Worek FS, Blumel G, Zeravik J, Zimmermann GJ, Pfeiffer UJ. Comparison of ketamine and pentobarbital
anaesthesia with the conscious state in a porcine model of Pseudomonas aeruginosa septicaemia. Acta
Anaesthesiol Scand 1988;32:509–15.
49 Tverskoy M, Gelman S, Fowler KC, Bradley EL. Effects of anaesthesia induction drugs on circulation in
denervated intestinal loop preparation. Can Anaesth Soc J 1985;32:516–24.
50 Seyde WC, Longnecker DE. Anesthetic influences on regional hemodynamics in normal and
haemorrhaged rats. Anesthesiology 1984;61:686–98.
51 Miller ED, Kistner JR, Epstein RM. Whole-body distribution of radioactively labelled microspheres in the
rat during anesthesia with halothane, enflurane, or ketamine. Anesthesiology 1980;52:296–302.
52 Idvall J. Influence of ketamine anesthesia on cardiac output and tissue perfusion in rats subjected to
haemorrhage. Anesthesiology 1981;55:297–304.
53 Gaab MR, Poch B, Heller V. Oxygen tension, oxygen metabolism, and microcirculation in vasogenic
brain edema. Adv Neurol 1990;52:247–56.
54 Dempsey RJ, Roy MW, Meyer KL, Donaldson DL. Indomethacin-mediated improvement following
middle cerebral artery occlusion in cats. J Neurosurg 1985;62:874–81.
55 Longnecker DE, Ross DC, Silver IA. Anesthetic influence on arteriolar diameters and tissue oxygen
tension in hemorrhaged rats. Anesthesiology 1982;57:177–82.
56 Longnecker DE, Miller FN, Harris PD. Small artery and vein response to ketamine HC1 in the bat wing.
Anesth Analg 1974;53:64–8.
57 Longnecker DE, Harris PD. Dilatation of small arteries and veins in the bat during halothane anesthesia.
58 Fukunaga AF, Epstein RM. Sympathetic excitation during nitrous oxide–halothane anesthesia in the cat.
59 Seyde WC, Ellis JE, Longnecker DE. The addition of nitrous oxide to halothane decreases renal and
splanchnic flow and increases cerebral blood flow in rats. Br J Anaesth 1986;58:63–8.
60 Greenblatt EP, Loeb AL, Longnecker DE. Endothelium-dependent circulatory control – a mechanism for
the differing peripheral vascular effects of isoflurane versus halothane. Anesthesiology 1992;77:1178–85.
61 Loeb AL, Godeny I. Longnecker DE. Anesthetics alter the relative contributions of NO and EDHF in the
rat cremaster muscle microcirculation. Am J Physiol 1997;273:H6l8– 27.
62 Adeagbo AS, Triggle CR. Varying extracellular [K+]: A functional approach to separating EDHF- and
EDNO-related mechanisms in perfused rat mesenteric arterial bed. J Cardiovasc Pharmacol 1993;21:423–
9.
63 Lischke V, Busse R, Hecker M. Inhalation anesthetics inhibit the release of endothelium-derived
hyperpolarizing factor in the rabbit carotid artery. Anesthesiology 1995;83:574–82.
64 Lischke V, Busse R, Hecker M. Volatile and intravenous anesthetics selectively attenuate the release of
endothelium-derived hyperpolarizing factor elicited by bradykinin in the
Page 330
coronary microcirculation. Naunyn-Schmiedeberg’s Arch Pharmacol 1995; 352:346–9.
65 Yamazaki M, Stekiel TA, Bosnjak ZJ, Kampine JP, Stekiel WJ. Effects of volatile anesthetic agents on in
situ vascular smooth muscle transmembrane potential in resistance- and capacitance-regulating blood
vessels. Anesthesiology 1998;88:1085–95.
Page 331
10: Anaesthesia and the cardiovascular

system
WOLFGANG BUHRE, ANDREAS HOEFT
The objective of anaesthesia is to provide analgesia, unconsciousness, suppression of reflex

responses to surgical stimuli, and muscle relaxation, if required. In general, this goal is
achieved by administering a combination of various agents, such as volatile and intravenous
anaesthetic drugs, opioids, benzodiazepines, and muscle relaxants. In principle, all drugs
used in anaesthetic practice affect the performance of the cardiovascular system, either by
direct effects on the heart and the vascular system or indirectly by altering neurohumoral
control of the circulation. It is very difficult to obtain a clear pharmacodynamic profile of
an anaesthetic drug under in vivo conditions because both direct and indirect effects
interact. Moreover, most anaesthetic drugs alter vascular tone and myocardial performance
simultaneously. Thus, it is difficult to distinguish whether a decrease in blood pressure
and/or stroke volume is the result of changes in myocardial loading conditions or of a direct
negative inotropic effect of the anaesthetic agent itself. In vitro studies using isolated heart
and vessel preparations, in which loading conditions can be carefully controlled, are
therefore a rational approach to investigate the cardiovascular effects of anaesthetic drugs.
Such studies have shown that most of the volatile and many of the intravenous anaesthetic
drugs exert direct negative inotropic effects on the myocardium, and some of them cause
systemic vasodilation. The isolated models are, however, relatively artificial, and the
experimental results are not always in accordance with clinical observations. For instance,
opioids as well as benzodiazepines are thought to be more or less haemodynamically inert,
that is, to elicit no negative inotropic effects on the myocardium or cause significant
vasodilation.1–5 However, in some patients induction of anaesthesia with these drugs leads
to dramatic hypotension, even if small doses are given.6 Conversely, patients with
compromised myocardial function and documented low ejection fraction often tolerate
anaesthetic drugs with known negative inotropic properties surprisingly well. Thus, clinical
experience suggests that, at least in some patients, centrally mediated indirect effects of
anaesthesia (in particular the depression of the
Page 332
sympathetic drive) may be as important as the direct effects of the anaesthetic drug on the
myocardium or vascular smooth muscle.
In fact, the interaction of anaesthesia with central control mechanisms of the circulation
seems to be an intrinsic dilemma of anaesthesia. Even if the ideal anaesthetic drug with no
adverse effects on the myocardium or vascular smooth muscle could be found, it is
questionable whether the centrally mediated effects of anaesthesia on autonomic control of
the cardiovascular system could be separated from the desired goals of anaesthesia.
Basically (with the possible exception of ketamine), centrally mediated depression of
cardiovascular performance is more or less common to all anaesthetic techniques,
regardless of whether volatile anaesthetic or intravenous anaesthetic agents are employed.
In the clinical setting, it is unlikely that loading conditions (that is, preload, afterload)
remain unchanged during the study of the effects of anaesthetic drugs on the cardiovascular
system. At the same time, most parameters of myocardial function are clearly load
dependent. The clinical application of echocardiography and specific indicator dilution
techniques may overcome these problems, because calculation of load independent
parameters may become possible in clinical practice.
Effects of anaesthesia on the sympathetic system

Cardiovascular homoeostasis is largely regulated by the autonomic nervous system,
which controls heart rate, myocardial contractility, vascular resistance, and the tone of the
venous capacitance vessels. Anaesthesia alters basic sympathetic tone as well as the
sympathetic response to painful (surgical) stimuli. It is essential for the anaesthetist to be
familiar with the interaction of anaesthesia with the autonomous nervous system in order to
evaluate and treat haemodynamic disorders that may occur during anaesthesia. Current
knowledge about the interaction of anaesthetic agents with the sympathetic nervous system
is based mainly on studies of plasma catecholamine levels, assessment of baroreceptor
reflex response, and direct recordings of sympathetic nerve activity.7–12
Sympathetic nerve activity
Muscle sympathetic nerve activity can be recorded from the peroneal nerve. A very thin
epoxy coated needle (0·2 mm) with a small tip (5μm) is placed within the peroneal nerve
below the bony prominence at the head of the fibula. Using a reference electrode and a
special set-up with a differential preamplifier and a bandpass filter, identification of
characteristic muscle sympathetic nerve activity is possible (Fig 10.1). Sympathetic nerve
activity is quantified by the number of bursts per minute or number of bursts per 100
cardiac cycles, or as total activity calculated from the
Page 333
Fig 10.1 Respiratory movements, muscle sympathetic nerve activity, and blood pressure in a patient
undergoing induction of anaesthesia with propofol. The upper panel shows condensed recordings, the lower
panel depicts selected periods in an enlarged time scale. Induction of anaesthesia with propofol decreases
muscle sympathetic nerve activity and blood pressure (upper panel). Tracheal intubation causes a dramatic
increase in muscle sympathetic nerve activity. (Modified from Sellgren et al.11)
product of bursts per minute and mean burst amplitude. Efferent bursts frequently occur in
pulse synchronous groupings and are often phase locked to late expiration and early
inspiration efforts.7 This pattern is thought to be caused by baroreceptor
modulation.10Effects of induction of anaesthesia on sympathetic nerve activity
Effects of induction of anaesthesia on sympathetic nerve activity

There are a number of investigations on the effects of induction of anaesthesia, tracheal
intubation, and surgical stimulation on muscle sympathetic nerve activity. Most of the work
has been performed by the groups of Ebert et al. and Sellgren et al.7 9–15 Induction of
anaesthesia decreased sympathetic outflow (Fig 10.2). Propofol and thiopentone
(thiopentone) had the most pronounced effects,7 10–11 whereas etomidate preserved muscle
sympathetic nerve activity7(Fig 10.2). Moreover, the decrease of muscle sympathetic nerve
activity after propofol is more pronounced in unpremedicated patients than in those who
received benzodiazepine premedication7 11. Obviously, baseline sympathetic drive seems to
be an important determinant of the impact of anaesthesia on
Page 334
sympathetic outflow and the resulting haemodynamic effects. These findings are in
accordance with the clinical observation that propofol occasionally causes severe
hypotension, particularly in patients who presumably are under an increased sympathetic
drive in the conscious state, for example, patients with borderline hypovolaemia or
compensated myocardial failure. Induction of anaesthesia with methohexital
(methohexitone) has also been reported to decrease muscle sympathetic nervous activity.10
These experimental data would support the clinical impression that etomidate is the
induction anaesthetic agent of choice for patients with compromised cardiovascular
performance and compensatory increased sympathetic drive. Laryngoscopy as well as
surgical stimulation are associated with an immediate and often dramatic increase of
sympathetic activity (Fig 10.1). Under such conditions, the pulse synchronous
Fig 10.2 Effect of various anaesthetics on muscle sympathetic nerve activity. Muscle sympathetic nerve
activity is given as bursts per minute and percent change of baseline values. Induction of anaesthesia with
thiopentone (TH, 4 mg/kg), propofol (P, 2·5 mg/kg), and propofol after premedication with diazepam (P &
pm, 2–2·5 mg/kg and 0·25 mg/kg, respectively), is associated with a significant decrease in muscle
sympathetic nerve activity, whereas etomidate (ETO, 0·3 mg/kg) exerts only minor effects. Isoflurane alone
(ISO 1·2 vol% and ISO 0·6 vol%) also inhibits muscle sympathetic nerve activity. In contrast, nitrous oxide
(N2O, 70%) causes sympathetic hyperactivity, which is counterbalanced by combination with isoflurane (ISO
0·6 vol% & N2O 70%, ISO 0·3 vol% and N2O 70%). (Data from Ebert et al.7 12 and Sellgren et al.10)
Page 335
rhythmicity is lost and a more continuous activity pattern is observed in some patients.11
Effects of maintenance of anaesthesia on sympathetic nerve activity

As maintenance of anaesthesia is inevitably associated with suppression of cerebral
activity, it is to be expected that all maintenance anaesthetic agents, that non-specifically
inhibit neuronal activity, will also decrease sympathetic outflow. This has been
demonstrated for propofol and halothane.10 11 In contrast to other volatile anaesthetic
agents, larger concentrations of desflurane are associated with increasing sympathetic
outflow, and sympathetic activation is amplified during rapid increase in the inspired
desflurane concentration.14 There is evidence that this effect is caused by effector sites
within the lungs.15 Desflurane caused marked increases in heart rate and blood pressure.
These haemodynamic responses may pose a risk for myocardial ischaemia to patients with
ischaemic heart disease (Fig 10.3).15 Isoflurane and sevoflurane were not associated with
increases in muscle sympathetic nerve activity, even when the concentration was rapidly
increased.16 17 Nitrous oxide (N2O) is also known to enhance sympathetic drive under
certain circumstances. In healthy volunteers, a progressive increase in muscle sympathetic
nerve activity was observed when low doses of N2O (25% and 40%) are inhaled via a
facemask.9 These changes are associated with a significant increase in blood pressure.
Sellgren et al11 also observed that sympathetic nerve activity is enhanced by N2O in
comparison to the conscious state and that the combination of isoflurane and N2O increases
sympathetic nervous activity compared with isoflurane alone11 (see Fig 10.2). Fentanyl 3
μg/kg given before induction of anaesthesia to a spontaneously breathing patient induces a
temporary slight increase in muscle sympathetic nerve activity,10 possibly because of
associated hypercapnia. Data for sufentanil or alfentanil are currently not available.
Plasma catecholamines
Several authors have measured plasma catecholamine concentrations during induction

and maintenance of anaesthesia.18 Noradrenaline (nor-epinephrine) is thought to correlate
more or less with overall sympathetic nervous activity,19 whereas adrenaline (epinephrine)
concentrations vary with adrenal release.20 Interpretation of circulating plasma
catecholamine concentrations is, however, hampered by the fact that measurable circulating
noradrenaline is the result of an ‘‘overspill” phenomenon at the nerve ending, that is, the
net effect of release and reuptake mechanisms.20 21 Furthermore, data on plasma
catecholamines from earlier studies have to be interpreted cautiously because baseline
levels of adrenaline and noradrenaline were barely in the range of detection.20 Despite these
Page 336
Fig 10.3 Effects of rapid advancement in the anaesthetic concentration of volatile anaesthetic agents in human
volunteers. The rapid increase in the inspired concentration was performed after a 30 min stabilisation period
at 0·8 (sevoflurane) or 1·0 MAC (isoflurane, desflurane) concentration. In contrast to all other anaesthetic
agents, an increase in the inspired concentration of desflurane leads to a significant increase in sympathetic
nerve activity (SNA), resulting in an increase in mean arterial pressure (MAP) and heart rate (not shown).
SNA is measured as burst frequency per 100 heart beats. , desflurane; , isoflurane; , sevoflurane. (Data
from Ebert et al.14 16 Reproduced with permission from Lippincott, Williams and Wilkins. Anesthesiology
1993;79:444–53 and Anesthesia Analgesia 1995;81:l l–22.)
Page 337
limitations, changes in plasma noradrenaline levels seem to correlate well with muscle
sympathetic nerve activity in healthy volunteers.22
Induction of anaesthesia is usually associated with a decrease in plasma catecholamine
concentrations.23 A complete suppression of noxious stimuli is, however, difficult to
achieve, even with very high doses of opioids. Neither very high doses of fentanyl (50 μg or
100 μg/kg) nor high doses of sufentanil (10, 20, or 30 μg/kg) prevented increases in plasma
catecholamines during sternotomy.24This study confirmed earlier results of Sonntag et al.25
who showed that neither high doses of sufentanil (10 μg/kg bolus followed by continuous
infusion of 0·15 μg/kg min) nor moderate doses of sufentanil (1 μg/kg bolus followed by
continuous infusion of 0·015 μg/kg per min) in combination with N2O (30% O2/70% N2O)
prevented a response to surgical stimuli as judged by increases in blood pressure and
noradrenaline plasma concentrations (Fig 10.4). Interestingly, despite significant increases
in circulating noradrenaline concentrations no concomitant change in the adrenaline
concentration levels was observed.25
Fig 10.4 Plasma levels of adrenaline (epinephrine) and noradrenaline (nor-epinephrine) in patients undergoing
cardiac surgery with sufentanil anaesthesia. Neither high dose sufentanil (10 μg/kg + 0.10 μg/kg per min) nor
sufentanil–nitrous oxide anaesthesia (1 μg/kg + 0.015 μg/kg per min in 30% O2/70% N2O) are able to prevent
the noradrenaline response to surgical stimulation (sternotomy). No additional hypnotics were used in this
study. , Noradrenaline: sufentanil–nitrous oxide; , Noradrenaline: high dose sufentanil; , adrenaline:
sufentanil–nitrous oxide; , adrenaline: high dose sufentanil. (Data from Sonntag et al.25)
Page 338
Similarly, in patients anaesthetised with sufentanil and midazolam, a significant increase in
noradrenaline concentrations occurred during sternotomy but adrenaline concentrations
remained unchanged.26 In the same study, thoracic epidural anaesthesia in combination with
light general anaesthesia (N2O and midazolam) blocked the haemodynamic and humoral
(noradrenaline) response to sternotomy. Moreover, compared with general anaesthesia,
lower concentrations of noradrenaline and adrenaline were observed during bypass.
Possibly, thoracic epidural anaesthesia is an appropriate method to block the sympathetic
response to surgical stimuli.26 In contrast, desflurane increases plasma noradrenaline
concentrations dose dependently in healthy volunteers.17 27
In summary, induction and maintenance of anaesthesia with volatile and intravenous
anaesthetic drugs are generally associated with depression of the sympathetic drive and
decreased catecholamine concentrations. Pure opioid anaesthesia is not able to block the
sympathetic response to intense surgical stimulation. In contrast to other volatile
anaesthetic drugs, desflurane can cause sympathetic hyperactivity, most probably mediated
by airway receptor sites, resulting in tachycardia, hypertension and increased muscle
sympathetic nerve activity.16 Another exception is ketamine, which produces increases in
heart rate and arterial pressure. This effect can be attributed to the central sympathomimetic
effects of ketamine which include the block of reuptake of monoamines into adrenergic
nerves.
Baroreceptor reflex
Baroreflex control of heart rate can be studied in conscious and anaesthetised subjects by
intravenous administration of vasoactive drugs, such as phenylephrine (pressor test) and
sodium nitroprusside (depressor test). A bolus of phenylephrine (approximately 150 μg) is
administered, preferably via a central venous catheter. As a result of the baroreceptor
reflex, the increase in blood pressure results in a reflex slowing of heart rate. Baroreceptor
sensitivity is defined as the slope of arterial pressure change divided by the R–R interval
change in the ECG. In a similar way, baroreceptor reflex, after an abrupt decline in blood
pressure, can be determined. Usually boli of sodium nitroprusside (100 μg) are used.
Pressor and depressor tests are affected by anaesthetic drugs to various degrees.28
Of the three induction anaesthetic drugs, propofol, etomidate, and thiopentone, etomidate
has the least effect on baroreceptor reflex function, followed by propofol and thiopental
(Fig 10.5).7 The latter almost abolished the baroreceptor response when given in equipotent
doses. Methohexital also inhibits the baroreceptor reflex in experimental animals.28 A
significant decrease in pressor baroreceptor response after administration has also been
observed for midazolam and diazepam.29
Page 339
Fig 10.5 Effects of intravenous induction anaesthetic agents on baroreceptor response. Baroreceptor response
to pressor (phenylephrine 150 μg) and depressor stimulus (sodium nitroprusside 100 μg) is assessed by
cardiac baroslopes, that is, the ratios of R–R interval change to systolic blood pressure changes. Induction of
anaesthesia with thiopentone (TH, 4 mg kg–1) and propofol (P, 2·5 mg/kg) significantly inhibits baroreceptor
response whereas etomidate (ETO, 0·3 mg/kg) preserves the activity of the baroreceptor reflex. (Data from
Ebert et al.7 12)
Inhaled anaesthetic agents depress the arterial baroreceptor reflex in both humans and
animals.30 31 Early studies had demonstrated that both isoflurane and halothane attenuate the
cardiac baroreflex.31 It was, however, hypothesised that isoflurane and enflurane preserve
cardiac baroreflex function because a dose dependent increase in heart rate was observed
with both anaesthetic agents.32 The depression of the baroreflex response caused by
isoflurane is less pronounced than with equipotent doses of halothane or enflurane.26 Recent
data clearly demonstrate that halothane, isoflurane, and enflurane do not differ with respect
to reflex heart rate response to both increasing and decreasing blood pressure32 (Fig 10.6).
As the effects of desflurane and sevoflurane seem to be very similar to those obtained with
isoflurane, it can be concluded that there are no clinically relevant differences between
inhaled anaesthetic agents with respect to baroreflex control.32
Page 340
Fig 10.6 Changes in cardiac baroreflex response associated with isoflurane (ISO), halothane (HAL), and
enflurane (ENF). Percentage change (from conscious baseline) in cardiac baroreflex slope during sodium
nitroprusside (falling pressure response, top) and phenylephrine (rising pressure response, bottom). Slope was
calculated as the linear relationship between mean arterial presure and R–R interval. No significant
differences between isoflurane, halothane, and enflurane were observed. Thus, the results suggest that reflex
heart rate response to both decreasing and increasing blood pressure does not differ between isoflurane,
halothane, and enflurane in this experimental setting. (Modified from Muzi and Ebert.32 Reproduced by
permission of Academic Press, Inc. Advances in Pharmacology 1994;31:379–87.)
Page 341
Heart rate variability
Analysis of small oscillations in heart rate, termed “heart rate variability” (HRV),
provides a non-invasive estimate of autonomic reflex function.33 34 Typically, two major
components are seen in a heart rate spectrum. The high frequency component (0·15–0·5 Hz)
is said to be mediated by the parasympathetic nervous system. The low frequency
component (0·04–0·15 Hz) is influenced by both the sympathetic and the parasympathetic
nervous system, and is related to waves in arterial pressure mediated by the baroreceptor
reflex. The ratio between low and high frequency (LF/HF) is considered to be a useful
indicator of cardiac sympathetic nerve activity.35 The LF/HF ratio has been used as an
estimate of autonomic reflex activity in several clinical studies comparing different
anaesthetic or premedication regimens.36–43 In practice, analogue ECG recordings are
digitised and the spectral power distribution is calculated using fast Fourier
transformation.33–35 The total power spectrum is divided into the LF and HF component,
and the ratio between LF and HF (LF/ HF) as well as the total power (LF + HF) of HRV is
then calculated.
The LF/HF ratio increased in elderly unpremedicated patients, but not in young patients
after arrival in the operating room.36 Both midazolam 0·06 μg/kg i.m. and diazepam 0·2
μg/kg orally decreased the LF/HF ratio, suggesting that both drugs are equally effective in
attenuating sympathetic nerve activity. The route of administration (that is, intramuscularly
or orally) is apparently of no clinical relevance.36 A significant reduction in total power, the
LF and the HF component of HRV in elderly compared with young patients was observed,
consistent with a decline in parasympathetic nerve activity with age.36 Recently,
Michaloudis et al.37 studied the effects of midazolam 0·08 μg/kg, morphine 0·15 μg/kg and
clonidine 2 μg/kg as premedication, on total HRV and the LF and HF components of
HRV.37 In accordance with previous results,36 they observed that midazolam reduces both
the LF and the HF component of HRV. Although the ratio between both remained
unchanged, however, morphine and clonidine decreased the LF more than the HF
component, suggesting parasympathetic dominance.37 Latson and O’Flaherty42 studied the
effects of surgical stimulation and different anaesthetic regimens on changes in total heart
rate variability. They compared two groups of patients receiving either isoflurane–N2O or
continuous propofol infusion for laparoscopic tubal ligation. In both groups, total HRV
decreased significantly after induction of anaesthesia. After skin incision, HRV recovered
in patients receiving propofol, but it remained depressed in patients receiving isoflurane–
N2O.42 These results suggest that a shift in autonomic balance towards sympathetic
dominance took place only in patients receiving propofol.42 These results are somewhat
surprising because it is well documented that propofol blunts the sympathetic response
when compared
Page 342
with other anaesthetic regimens. In the study by Latson and O’Flaherty, however, only
small doses of fentanyl were given to patients receiving propofol.42 Thus, the observed
increase in HRV after surgical stimulation may be the result of inappropriate analgesia or
insufficient anaesthetic depth.42 The effects of fentanyl 7·5 μg/kg and different dosages of
midazolam (0·075, 0&3183;1, 0·2 mg/kg) on HRV were studied by Zickman et al43 in
patients scheduled for coronary bypass surgery. A significant decrease in the LF component
was observed, whereas the HF component was only slightly decreased. The authors
concluded that this anaesthetic induction technique decreases sympathetic, but not
parasympathetic autonomous nervous activity.43 As expected, the use of ketamine results in
an increase in sympathetic activity reflected by an increase in the LF component.40
A number of studies investigated the effects of volatile anaesthetic agents on changes in
HRV during surgery and recovery from anaesthesia.38 41 Comparing isoflurane and
desflurane, total power of HRV was back to control values by 60 min after the end of
surgery in desflurane treated patients, but remained suppressed in patients who underwent
isoflurane anaesthesia.38 These findings suggest that neural reflex control is restored earlier
after desflurane anaesthesia. In contrast, no significant differences were observed between
halothane or isoflurane.41 When comparing the effects of induction of anaesthesia with
thiopentone 4 mg/kg and etomidate 0·3 mg/kg, Latson et al44 found that both induction
agents decrease total HRV by 89% and 58%, respectively. The decrease was, however,
more pronounced with thiopentone than with etomidate.44
In summary, it seems that determination of HRV allows characterisation of the effects of
anaesthesia on autonomic control.
Effects of anaesthesia on the heart

The effect of anaesthetic drugs on the myocardium has been the subject of numerous
experimental and clinical investigations. As outlined above, it is very difficult to
differentiate, under in vivo conditions, between direct myocardial effects of anaesthetic
agents, direct vascular effects, and indirect effects mediated via the autonomic nervous
system. A reasonable approach is to investigate the influence of anaesthetic agents on
myocardial contractility in isolated in vitro models, such as cultured myocytes, papillary
muscles, or isolated working heart preparations. Preload and afterload can be well
controlled in these preparations, and any interference with sympathetic or parasympathetic
innervation is avoided. For the same reasons, it is, however, very difficult to apply results
obtained in vitro to the clinical situation. Basically, the more isolated a preparation is, the
more
Page 343
artificial the physiological environment. For instance, isolated working heart preparations
are deprived of any basic sympathetic drive, and coronary oxygen supply is often limited
because of asanguineous crystalloid perfusion. Papillary muscle preparations are devoid of
a blood supply, and the experiments are often performed at room temperature in order to
decrease metabolic rate and to facilitate sufficient oxygen supply by diffusion. Cultured
myocytes or skinned muscle fibres are obviously very artificial models and far from clinical
reality.
Myocardial contractility
Nevertheless, isolated models can be useful to identify the pure effect of substances on
the myocardium itself. Several induction agents have been shown to elicit negative
inotropic properties in papillary muscle preparations or isolated working heart models.45–56
Most recently, for example, Stowe and co-workers50 investigated the effects of etomidate,
ketamine, midazolam, propofol, and thiopental on cardiac function and metabolism in an
isolated guinea pig heart model at equimolar doses. In principle, all induction agents
(including midazolam and etomidate) produce a dose dependent decrease in myocardial
contractility (Fig 10.7) On a molar basis, propofol (less so midazolam and etomidate)
depresses cardiac function moderately more than thiopentone and ketamine. The peak
concentrations required for induction of anaesthesia are, however, quite different between
agents. Peak plasma concentrations during induction of anaesthesia have been reported to
be approximately 0·5 μmol/1 for midazolam, 3 μmol/1 for etomidate, 60 μmol/1 for
ketamine, 50 μmol/1 for propofol, and 100 μmol/1 for thiopentone (Fig 10.7). Thus, under
clinical conditions midazolam and etomidate are basically not cardiodepressive, whereas
thiopental and propofol exert significant direct negative inotropic action.
In contrast to volatile anaesthetic agents,52 only little is known about the molecular
mechanisms responsible for the direct negative inotropic properties of intravenous
induction agents. Initial contradictory results about the direct action of ketamine on the
myocardium prompted some more detailed investigations. Ketamine is a myocardial
depressant in intact dogs, isolated dog heart preparations, and isolated rabbit hearts.53 54
Riou et al65 have demonstrated that ketamine has a dual opposing effect on the
myocardium:
1 A positive inotropic effect associated with increased Ca2+ influx.
2 A negative inotropic effect possibly because of an impaired function of the sarcoplasmic
reticulum.55
In an isolated ventricular papillary muscle model of a ferret, the positive inotropic effects of
ketamine could be blocked by bupranolol, indicating
Page 344
involvement of β-receptors.56 Moreover, depletion of noradrenaline stores by reserpine also
abolished the positive inotropic action of ketamine. It was concluded that inhibition of
neuronal catecholamine uptake is the predominant mechanism of the positive inotropic
effect of ketamine.56 The negative inotropic effect of ketamine was elucidated by
Kongsayreepong et al.57 using measurements of intracellular Ca2+ transients with aequorin.
Fig 10.7 Comparison of the direct negative inotropic effects of common intravenous induction anaesthetic
agents in an isolated working heart model. The effects of etomidate, ketamine, midazolam, propofol, and
thiopental were studied in an isolated working heart model. Percent change in the peak positive derivative of
left ventricular pressure (+ dLVP/dtmax) are shown. A wide range of concentrations from 0·5 to 1000 μmol/1
was investigated (symbols). All induction anaesthetic agents (including midazolam and etomidate) produce a
concentration dependent depression of myocardial contractility. In concentrations that are equivalent to peak
plasma levels during induction of anaesthesia (marked by symbols, name, and equivalent anaesthetic
concentration for each agent), however, propofol and thiopentone appear to depress cardiac function more
than ketamine, etomidate, and midazolam. , midazolam; , etomidate; , ketamine; , propofol; ,
thiopentone. (Data from Stowe et al.64)
Page 345
Ketamine decreased intracellular calcium availability, which also occurred when the
sarcoplasmic reticulum was blocked by ryanodine.
Gelissen et al58 were the first to compare the direct effects of different intravenous
hypnotics on the intrinsic contractility of isolated human myocardium. They recorded force
development during isometric contraction of atrial myocytes exposed to various
concentrations of etomidate, propofol, ketamine, and thiopental.58 All anaesthetic agents
produced dose dependent inhibition of isometric contraction. At clinically relevant
concentrations, however, the effects of thiopental and ketamine were pronounced, whereas
propofol and etomidate had no direct negative inotropic effect on the atrial myocyte. There
are some methodological limitations to this study. First, some patients were under chronic
therapy with β-adrenoceptor blocking agents or Ca2+ channel antagonists. Second,
experiments were performed under hypothermic conditions (30°C). Third, no control
experiments with the solvent alone were carried out. Moreover, as atrial but no ventricular
myocytes were studied, it remaines unclear whether the results can be extrapolated to
ventricular preparations.58 In contrast to earlier work, propofol produced no direct negative
inotropic effect, suggesting that the clinically observed cardiovascular depression is
predominantly the result of changes in cardiac loading conditions and sympathetic drive.58
The influence of volatile anaesthetic agents on myocardial contractility has been
extensively investigated in papillary muscle experiments, working heart models, and intact
animals.52 All volatile anaesthetic agents decrease myocardial contractility in a dose
dependent manner.59–64 Isoflurane seems to be less negative inotropic than enflurane and
halothane59 (Fig 10.8). Desflurane and sevoflurane also decrease developed contractile
force in different in vitro models.65 66The decreases are comparable to those caused by
isoflurane. The cardiodepressive effect of volatile anaesthetic agents is caused by an
alteration of calcium transients involving sarcolemmal calcium exchange processes, as well
as uptake and release of calcium by the sarcoplasmic reticulum.63 It has been suggested that
halothane and isoflurane depress cardiac function by different mechanisms.63 In papillary
muscle preparations, N2O exerted moderate negative inotropic effects (that is, 5–15%
depression of contractility by 50% N2O).67
The effects of opioids on cardiac muscle have been studied in only a very few isolated
papillary muscle experiments. Negative inotropy has been demonstrated for morphine,
meperidine, fentanyl, and alfentanil. These effects occurred, however, at concentrations one
hundred to several thousand times above those expected clinically.5 68
Assessment of myocardial contractility in intact animals or in patients requires
considerable methodological efforts. The crucial question is whether an observed decline in
blood pressure or stroke volume after application of an anaesthetic drug is the result of
decreased contractility or
Page 346
merely of changes in loading conditions. Several contractility indices have been suggested
in the literature (see Chapter 2). The end systolic elastance index69 and the preload
recruitable stroke work are considered relatively load independent (see Chapter 2). There is,
however, a paucity of clinical studies in which the effects of anaesthetic agents on
myocardial function are characterised based on end systolic elastance or preload recruitable
stroke work. A reason might be that transoesophageal echocardiography has only recently
become available as a research tool in anaesthesiology. A very interesting clinical study of
this type has been performed by Mulier et al,70 in which the haemodynamic effects of
thiopental and propofol were compared. In contrast to previous studies, propofol
significantly decreased myocardial contractility, and (in accordance with the in vitro results
of Stowe et al50) was at least as negative inotropic in vivo as thiopental. The authors even
suggested that the negative inotropic properties of propofol are more pronounced and more
prolonged than those of equipotent doses of thiopental.70 The differences were, however,
only marginal and statistically barely significant. It is questionable whether the observation
of Mulier et al is the result of direct negative inotropic effects of propofol alone. An
alternative explanation could be a marked decrease in sympathetic outflow,
Fig 10.8 Effects of halothane (HAL), enflurane (ENF), and isoflurane (ISO) on myocardial contractility in
isolated papillary muscles. All volatile anaesthetic agents exert a dose dependent negative inotropic effect
shown as percentage depression of peak developed tension. Halothane was found to be significantly more
depressant than isoflurane at 1·6 MAC, wheras enflurane caused intermediate depression. (Data from Lynch
and Frazer.59)
Page 347
which has been demonstrated during propofol anaesthesia7 (see Fig 10.2). As in other
studies, the induction anaesthetic agent with the least negative inotropic effects seemed to
be etomidate.
As expected from results in isolated models, volatile anaesthetic agents are
cardiodepressants in intact animals and patients67 71 72 during systole and early diastole
mainly via alterations of intracellular Ca2+ homoeostasis. In general, the inward Ca2+
current is diminished by inhibiting the function or reducing the number of Ca2+ channels in
the sarcolemmal membrane. Thus, both the availability of Ca2+ for contraction and the
amount of Ca2+ stored in the sarcoplasmic reticulum are decreased. In addition, the
concentration of intracellular Ca2+ is decreased by direct anaesthetic induced alteration of
the sarcoplasmic reticulum, combined with partial inhibition of Ca2+ uptake and enhanced
Ca2+ leakage. Whether inhalational anaesthetic agents modify the responsiveness of
contractile proteins to Ca2+ remains controversial. Pagel et al73 demonstrated, in chronically
instrumented dogs with and without autonomic blockade, that isoflurane and desflurane
produce less depression of myocardial contractility than halothane or enflurane. Without
autonomic blockade, however, desflurane preserved mean arterial pressure and cardiac
output to a greater degree than equipotent doses of isoflurane, especially at higher
concentrations.73 This finding can be explained by the enhanced sympathetic drive caused
by desflurane. Sevoflurane decreases contractility dose dependently by up to 40–45% of
baseline values in animal experiments with and without autonomic blockade. This
magnitude of contractile depression is comparable to that observed with isoflurane.74
The in vivo effects of N2O are very complex. As outlined above, N2O increases
sympathetic activity and in some patients increases blood pressure.9 11 12The weight of
currently available evidence suggests, however, that N2O is a myocardial depressant75 when
used as supplement during anaesthesia. Pagel et al75 investigated N2O in chronically
instrumented dogs in the presence of pharmacological blockade of the autonomic nervous
system. Based on preload recruitable stroke work, myocardial contractility decreased when
N2O was added to a baseline anaesthesia with isoflurane or sufentanil (Fig 10.9). The
evidence for a myocardial depressant effect of N2O is strong in patients with compromised
myocardial function, that is, in those patients in whom the cardiovascular properties of
anaesthetic agents are clinically relevant. In summary, N2O has a weak direct myocardial
depressant effect, which may be counterbalanced by an increase in sympathetic activity by
N2O in intact animals and healthy patients.75
Myocardial oxygen consumption

Measurements of myocardial blood flow and myocardial oxygen consumption in patients
are very cumbersome and can, therefore, be performed only for investigational purposes
(see Chapter 4). An estimate of
Page 348
myocardial oxygen demand from haemodynamic parameters would therefore be desirable.
The major determinant of myocardial oxygen demand is the energy required for
development and maintenance of systolic wall tension. As reliable measures of wall tension
are difficult to obtain, several derived indices of myocardial oxygen demand have been
suggested.76 Intuitively, most clinicians believe that an increase in both heart rate and blood
pressure will increase myocardial oxygen demand. This belief forms the rationale for using
the rate pressure product (RPP), which was first suggested by Rhode,77 and which is still
widely used as an index of myocardial oxygen demand. In animals as well as in patients,
however, only a poor correlation is found between RPP and myocardial oxygen uptake.78 79
Similarly, the tension time index (TTI) does not correlate very well with
Fig 10.9 Influence of nitrous oxide (N2O) on myocardial contractility in chronically instrumented dogs. Two
sets of experiments were performed in chronically instrumented dogs with pharmacological blockade of the
autonomic nervous system. Basic anaesthesia was performed with isoflurane (ISO, 1·25 MAC) or sufentanil
(SUE, 100–150 μg/kg–1h–1). Myocardial contractility was evaluated by the relationship of preload recruitable
stroke work (PRSW) to end diastolic length (EDL). The addition of nitrous oxide (N2O 30% and N2O 70%)
results in a decrease in myocardial contractility in both groups. (Data from Pagel et al.75)
Page 349
Fig 10.10 Correlation of myocardial oxygen uptake and modified pressure work index (PWImod) as
an estimate of myocardial oxygen consumption. PWImod and showed a correlation coefficient of 0·84 in
patients undergoing coronary artery bypass surgery. Haemodynamic variables required for calculation of
myocardial oxygen demand by PWImod are heart rate (Hf), arterial blood pressure (Psyst, Pdiast) and cardiac
index (CI: ml/m2). The modified pressure work index is currently the most reliable index for estimation of
myocardial oxygen demand derived from haemodynamic variables in patients. (Data from Hoeft et al.76)
measured myocardial oxygen uptake.76 78 A clinically useful alternative is the pressure work
index, which was first suggested by Rooke and Feigl.79 80 In addition to blood pressure and
heart rate, this index requires measurement of cardiac output, which is routinely performed
in many critically ill patients. But (as with all other haemodynamic indices of myocardial
oxygen demand) species specific differences exist between animals and humans. The
empirical constants derived by Rooke and Feigl for dogs cannot be used for estimation of
myocardial oxygen demand in humans.79 A modification of the pressure work index was
therefore developed and validated by Hoeft et al.76 It is based on data derived from
measurements of myocardial blood flow and myocardial oxygen uptake in patients using a
modified Kety–Schmidt technique (Fig 10.10). Currently, this modified pressure work
index seems to be the best alternative for the clinical estimation of myocardial oxygen
demand. This estimate seems to be applicable during the conscious state as well as during
anaesthesia.76 79
Effects of anaesthetic agents on myocardial oxygen consumption

Theoretically, it is possible that anaesthesia is associated with specific metabolic effects,
such as uncoupling of oxidative phosphorylation or
Page 350
Fig 10.11 Influence of thiopentone, methohexital, etomidate, and ketamine on heart rate, mean arterial blood
pressure, stroke volume index, and myocardial oxygen consumption. Induction of anaesthesia with thiopental
, and methohexitone is associated with a decrease in blood pressure (MAP) and stroke volume index
(SVI), while heart rate (HR) and myocardial oxygen consumption are increased. Ketamine
exerts the most pronounced effect on myocardial oxygen consumption because sympathetic activation leads to
an increase in heart rate and blood pressure. Stroke volume index is, however, decreased by ketamine, too.
The haemodynamic effects of etomidate are almost negligible and myocardial oxygen consumption
remains unchanged. (Data from Kettler and Sonntag.81)
uncoupling of myocardial energy turnover from myocardial haemodynamic performance.

Myocardial oxygen demand (estimated by the modified pressure work index) and measured
myocardial oxygen uptake do not, however, in general differ between the conscious state
and anaesthesia.76 Therefore, a specific metabolic effect of anaesthesia, for example
uncoupling of myocardial energy conversion, is very unlikely.
Under resting conditions, myocardial oxygen uptake is approximately 10–11 ml/min per
100 g (Figs 10.11, 10.12, and 10.13). As most anaesthetic agents are cardiodepressive either
by their direct negative inotropic effects or indirectly by decreasing sympathetic drive,
induction with virtually any anaesthetic agent leads to a decrease in arterial blood
Page 351
Fig 10.12 Effects of volatile anaesthetic agents and propofol on heart rate, mean arterial pressure, stroke
volume index, and myocardial oxygen consumption in patients undergoing coronary artery bypass surgery.
Induction of anaesthesia with volatile anaesthetic agents or propofol is associated with a decrease of blood
pressure and stroke volume index, and only minor changes of heart rate. As a result, myocardial oxygen
consumption is decreased as well. , enflurane; , propofol; , halothane; , isoflurane; —, pooled data.
(Data from Hoeft et al.82)
pressure and stroke volume index (Figs 10.11, 10.12, and 10.13). Ketamine (without
supplemental benzodiazepines) might be an exception (Fig 10.11); it causes sympathetic
stimulation, and increases in arterial blood pressure and heart rate, whereas stroke volume
is decreased (Fig 10.11). In general, a reflex increase in heart rate is seen with barbiturates,
but only a slight increase with etomidate (Fig 10.11). Heart rate, on the one hand, and
arterial pressure and stroke volume, on the other, have opposing effects on myocardial
oxygen demand. The net effect is an increase in myocardial oxygen consumption with
methohexital and thiopental, more so with ketamine, and mostly unchanged
haemodynamics and myocardial metabolic rate with etomidate81 (Fig 10.11).
Induction of anaesthesia with volatile anaesthetic agents as well as with propofol leads to
a decrease in blood pressure and stroke volume. As a result of inhibition of the baroreceptor
reflex, the increase in heart rate is
Page 352
less blunted. Myocardial oxygen consumption therefore decreases by almost one third,
which is more than seen with classic induction agents (Fig 10.12). Induction of anaesthesia
with high doses of opioids is associated with a decrease in blood pressure and stroke
volume, and also a considerable reduction in myocardial oxygen consumption82 (Fig 10.13).
Intense surgical stimulation (for example sternotomy) results in sympathetic activation,
which is difficult to block by anaesthesia. Blood pressure control is usually easier to
achieve with volatile anaesthetic agents than with high dose opioids (Fig 10.12 and
10.13).82 The poor control of blood pressure during opioid based anaesthesia can lead to up
to a twofold increase in myocardial oxygen consumption. Adjuvant measures (such as
vasodilators, β blockade, and other antihypertensive drugs) and supple-
Fig 10.13 The effects of opioid based anaesthesia on heart rate, mean arterial pressure, stroke volume index,
and myocardial oxygen consumption in patients undergoing coronary artery bypass surgery. As with volatile
anaesthetic agents (see Fig 10.12), induction of anaesthesia is associated with a decrease in blood pressure and
stroke volume index, whereas heart rate remains unchanged. Myocardial oxygen consumption is therefore
also decreased. Blood pressure control is more difficult to achieve during sternotomy. On average, an increase
in myocardial oxygen consumption is observed. , fentanyl–midazolam; , high dose fentanyl; , high dose
morphine; , high dose sufentanil; , sufentanil–nitrous oxide; —, pooled data. (Data from Hoeft et al.82)
Page 353
mentation with volatile anaesthetic agents are therefore often employed during opioid based
anaesthesia in order to maintain arterial pressure within normal limits.
Very high rates of myocardial oxygen consumption can develop during the recovery
period. A dramatic sympathetic activation occurs during this phase, and noradrenaline
(norepinephrine) levels by far exceed levels observed during anaesthesia and surgery (Fig
10.14).83 Although measurements of myocardial blood flow and myocardial oxygen
consumption are not available for this critical period, an estimation of myocardial oxygen
demand can be made based on the modified pressure work index and on data from the
literature. Shivering patients in particular may have an extremely high myocardial oxygen
demand,83 because the very high systemic metabolic rates will increase cardiac work load.
During anaesthesia, total body oxygen consumption is in the range of 80–120 ml/min per
m2. Thus, with a haemoglobin of 12 g% and an arteriovenous oxygen extraction of 25% the
required cardiac index is 1·5–2·3 1/min per m2. For a normal blood pressure (120/80 mm
Hg) and heart rate (80 beats/min), myocardial oxygen demand calculated by the modified
pressure work index
Fig 10.14 Noradrenaline (norepinephrine) levels during major abdominal surgery and during recovery from
surgery. Median and range (min., max.) of noradrenaline plasma levels during and after major abdominal
surgery are shown. The most pronounced sympathetic activation occurs during recovery from surgery. (Data
from Turner.83)
Page 354
is in the range of 10·4–10·9 ml/min per 100 g, which is consistent with values measured in
patients. Myocardial oxygen demand also depends on the haemoglobin content of the
blood. The lower the haemoglobin content, the higher the required cardiac output. During
the recovery period, systemic metabolic rates can be extremely high. On average, a
systemic metabolic rate of 260 ml/min per m2 is observed during this period in patients
after major abdominal surgery. Corresponding myocardial oxygen demand is 13·5 ml/min
per m2 at a haemoglobin of 12 g%, which increases to more than 19·2 ml/min per 100 g at a
haemoglobin of 8 g%. Values of up to 480 ml/min per m2 have been observed in shivering
patients.83 Thus, a decrease in haemoglobin content, as often observed after surgery in the
recovery room, would further increase myocardial oxygen demand and may cause problems
in patients with limited coronary blood supply.
Effect of anaesthesia on myocardial efficiency

In technical terms, efficiency is the ratio of work performed to energy required. In a
similar way efficiency of myocardial oxygen utilisation can be defined as the ratio of
external pressure–volume work to myocardial oxygen consumption. Pressure–volume work
is by definition the area under the left ventricular pressure–volume loop, which is
unfortunately difficult to assess in patients. For clinical purposes, however, a good estimate
of external pressure–volume work can be obtained from the product of stroke volume and
mean systolic pressure during the ejection phase.
It has been known for a long time that myocardial efficiency is not constant but varies
with preload, afterload, and contractility. Evans and Matsuoka84 demonstrated as early as
1915 that an increase in cardiac work caused by an increase in blood pressure requires a
higher increase in oxygen consumption than the same increase in cardiac work caused by
an increase in stroke volume.84 Recently, theoretical models have been developed that try to
relate the efficiency of myocardial oxygen utilisation to haemodynamics, that is, to indices
of preload, afterload, and contractility.85–87 Although the approaches for estimation of
myocardial efficiency differ in the way that myocardial oxygen demand is estimated from
haemodynamic indices, comparable results have been obtained. In general, myocardial
efficiency is improved by an increase in preload and/or by a decrease in afterload. The most
notable implication of the theoretical models is the hypothesis that myocardial performance
can be optimised with respect to myocardial efficiency: for a given preload and afterload a
maximal efficiency is achieved when contractility is neither too low nor too high, that is,
when contractility is ideally tuned to an optimal value. In fact, there is a growing body of
evidence that preload, afterload, and contractility are optimised under resting conditions at
the point of maximal efficiency.85 87 Induction of anaesthesia is associated with a decrease
in myocardial oxygen utilisation efficiency (Fig 10.15). Myocardial oxygen consumption is
not,
Page 355
however, decreased in proportion to the decrease of external myocardial work (Fig 10.15).
As a result, efficiency is decreased, regardless of the type of anaesthesia87 (Fig 10.15).
Theoretically, an increase in efficiency would be possible because of a decrease in afterload
(blood pressure). The decrease in contractility and stroke volume outweigh the former
effect, that is, contractility and afterload do not match preload conditions after induction of
anaesthesia. In terms of ‘‘ventricular–vascular matching” (see Chapter 2), stroke volumes
and contractility are too low. It is somewhat
Fig 10.15 Effects of anaesthesia on myocardial efficiency. Efficiency of myocardial oxygen use is calculated
from the ratio of external myocardial work (mean systolic pressure x cardiac index) to myocardial oxygen
consumption. Induction and maintenance of anaesthesia with any anaesthetic agent is associated with
decreased myocardial efficiency. Upper: , fentanyl-midazolam; , high dose fentanyl; , high dose
morphine; , high dose sufentanil; , sufentanil – N2O; — pooled data. Lower: , enflurane; , halothane;
, propofol; —, pooled data. (Data from Hoeft et al.87)
Page 356
surprising that this effect is seen not only with anaesthetic agents that are well known to
have direct negative inotropic properties (that is, halothane and enflurane), but also to the
same extent with anaesthetic agents that supposedly have no direct negative inotropic
effects (that is, opioids).87 This finding indicates that under in vivo conditions even opioid
anaesthesia is associated with a decrease in contractility, presumably because of decreased
or inadequate sympathetic drive.
There is evidence that sympathetic activation during anaesthesia is of a different quality
than that during conscious conditions. Usually, increases in heart rate are not very
pronounced, and sometimes heart rate even decreases (because of the attempt to block the
cardiovascular response by deepening anaesthesia). At times, however, significant increases
in blood pressure can be observed, in particular during opioid based anaesthesia. Although
sympathetic activation should enhance myocardial contractility, stroke volume indices and
myocardial efficiency of oxygen utilisation remain decreased. These findings also suggest
that anaesthesia significantly interferes with cardiovascular control mechanisms because
optimal tuning of the system is obviously not achieved during volatile as well as opioid
anaesthesia. Thus, there is substantial evidence that, under in vivo conditions, the direct
negative inotropic effects of anaesthetic agents might be of minor importance compared
with the central effects of anaesthesia on cardiovascular control mechanisms.
Another approach to quantify the interaction between the mechanical state of the left
ventricle and the vascular bed is based on sequential pressure–volume measurements and a
series of elastic chamber model of the circulation.88 89 The ratio of the elastances of the left
ventricle (end systolic elastance, Ees) and the arterial elastance (Ea) defines coupling
between the left ventricle and the arterial tree. This approach enables relatively load
independent studies on the effects of ventricular–arterial coupling in vivo. In addition, the
pressure–volume analysis permits evaluation of left ventricular efficiency as defined by the
ratio of stroke work (SW) to pressure–volume area (PVA). Hettrick and co-workers90–92
performed a series of experiments in which they studied the effects of isoflurane,
desflurane, sevoflurane, and propofol on left ventricular–arterial coupling and mechanical
efficiency in the open chest dog model anaesthetised with barbituates. All three inhalational
anaesthetic agents caused similar, dose dependent decreases in ventricular contractility and
afterload. Isoflurane, sevoflurane, and desflurane preserved optimum left ventricular–
arterial coupling at low anaesthetic concentrations (<0·9 MAC) (see Fig 10.18), whereas
coupling was impaired at 1·2 MAC. Mechanical efficiency was also altered at higher
anaesthetic concentrations (> 0·9 MAC for isoflurane and desflurane, >0·6 MAC for
sevoflurane) (Fig 10.16). These experimental data demonstrate that ventricular–arterial
coupling is impaired at higher anaesthetic concentrations, indicating that the reduction
Page 357
Fig 10.16 Effects of increasing doses of volatile anaesthetic agents on left ventricular–arterial coupling and
mechanical efficiency. Ventricular–arterial coupling was assessed by the ratio of left ventricular end systolic
elastance (Ees) to effective arterial elastance (Ea) (see text for details). Mechanical efficiency was examined
by the ratio of left ventricular stroke work (SW) and pressure volume area (PVA). A significant decrease in
both variables could be observed at 1·2 MAC of each volatile anaesthetic agent compared with control (C)
values. Thus, it can be concluded that left ventricular–arterial coupling and myocardial efficiency were
preserved at low concentrations of anaesthetic agents, whereas mechanical matching was impaired at higher
end tidal concentrations. No significant differences between the three agents could be observed. These
findings support in vivo results showing a dose dependent reduction of cardiac performance by volatile
anaesthics. , desflurane; , sevoflurane; , isoflurane. (Modified from Hettrick et al.90 Reproduced with
permission from Lippincott, William and Wilkins. Anesthesiology 1996;85:403–13.)
Page 358
Fig 10.17 Effects of increasing infusion rates of propofol on (a) left ventricular–arterial coupling and (b)
mechanical efficiency. A significant dose dependent decrease in left ventricular–arterial coupling (Ees/Ea) and
mechanical efficiency (SW/PVA) could be observed after infusion of propofol compared with control (C).
The effects were most pronounced at a dose of 20 mg/kg per h. Higher infusion rates did not lead to more
pronounced effects on either Ees/Ea or SW/PVA, demonstrating that the effects of propofol on ventricular–
arterial coupling are not strictly dose dependent. (Modified from Hettrick et al.92 Reproduced with permission
from Lippincott, Williams and Wilkins. Anesthesiology 1997;86:1088–93.)
Page 359
in myocardial contractility is not appropriately counterbalanced by simultaneous declines in
afterload.91
Propofol in a dose range of 5–40 mg/kg per h impairs ventricular–arterial coupling
significantly92 (Fig 10.17). The reduction of the ratio Ees/Ea is primarily caused by a
decrease in end systolic elastance. In contrast to volatile anaesthetics, Ees/Ea plateaus at
higher doses of propofol (20–40 mg/ kg per h) resulting from modest further reductions in
arterial elastance, indicating that the impairment of left ventricular–arterial coupling
associated with propofol is not strictly dose dependent.92 Mechanical myocardial efficiency
as derived from the SW/PVA ratio was decreased at the intermediate dose of propofol (5–
20 mg/kg per h), whereas at the 40 mg/kg per h dose, the decline in SW/PVA ratio was
reverted to some degree.
Fig 10.18 Coronary venous oxygen saturation (SCVO2)in conscious patients (A) and after induction of
anaesthesia with ketamine (K), methohexitone (ME), thiopental (TH), etomidate (ET), isoflurane (I),
halothane (HA), enflurane (EN), propofol (P), sufentanil (S), and fentanyl (FE).With the exception of
isoflurane, coronary venous oxygen saturation remains more or less unchanged compared to the conscious
state, that is, autoregulatory control of coronary blood flow is not affected by most anaesthetic agents.
Halothane, enflurane, and more so isoflurane are coronary vasodilators. (Modified from Hoeft et al.93)
Page 360
Effects of anaesthetic drugs on coronary blood flow

With the exception of isoflurane, anaesthetic agents do not effect autoregulation of
coronary blood flow. Myocardial blood flow is adjusted to match myocardial oxygen
demand. As a result myocardial blood flow decreases with induction of anaesthesia in
conjunction with the decrease in myocardial oxygen demand, and coronary sinus oxygen
saturation remains practically unchanged93 (Fig 10.18). In contrast, administration of
isoflurane causes coronary vasodilation, resulting in a concomitant increase in coronary
sinus oxygen saturation.93 Halothane, sevoflurane, and enflurane are coronary vasodilators
as well, but to a lesser degree.93 The coronary vasodilatory effects of volatile anaesthetic
agents are probably mediated by ATP sensitive potassium channels located on the surface
of vascular smooth muscle.94 95 A potassium leakage causes hyperpolarisation of vascular
smooth muscle cells, resulting in muscle relaxation.94 For instance, Crystal et al.94
demonstrated that the coronary vasodilating effects of halothane can be blocked by
administration of glibenclamide (a K+ATP channel inhibitor). Glibenclamide also inhibits
the vasodilatory effects of adenosine which strongly supports the fact that adenosine
receptors are involved in mediating vasodilation by K+ATP channels.94
A variety of studies has been performed to investigate whether the coronary vasodilator
properties of isoflurane cause coronary steal.95 In most studies that demonstrated
myocardial ischaemia during isoflurane, however, high doses were used with considerable
concomitant arterial hypotension. It is now commonly accepted that isoflurane can be used
safely in patients with coronary artery disease as long as high doses and arterial
hypotension are avoided.95 Thus, in cases of a hypertensive response to surgical stimuli,
isoflurane can safely be used for blood pressure control.
Effects of anaesthesia on ischaemia–reperfusion injury

In patients undergoing coronary revascularisation procedures, brief periods of ischaemia
are regularly observed and result in general or regional ischaemia–reperfusion injury.96 This
injury can cause moderate to severe myocardial dysfunction.96 Depression of myocardial
contractility associated with the use of volatile anaesthetic agents may be advantageous
during ischaemic episodes if the decrease in myocardial oxygen consumption is greater
than the regional decrease in myocardial perfusion caused by anaesthetic induced
hypotension.97–100 There is growing evidence that volatile anaesthetic agents protect the
myocardium by different cellular mechanisms during periods of ischaemia. First, volatile
anaesthetic agents alter the Ca2+ homoeostasis in the myocardium by inhibiting
transsarcolemmal calcium influx. This effect can be reversed by an increase in the
extracellular calcium concentration. In the presence of isoflurane and halothane, increased
leakage of Ca2+ from ryanodine sensitive Ca2+
Page 361
channels was observed,98 which reduces the available activator Ca2+ by depleting the Ca2+
content in the sarcoplasmic reticulum. There is experimental evidence that Ca2+
myofilament sensitivity is reduced during exposure to volatile anaesthetic agents.99 Second,
isoflurane, enflurane, and halothane decreased the size of the infarct zone after global as
well as regional ischaemia, whereas propofol, pentobarbital and ketamine/xylazine do
not.100 P-Sulphophenyl theophylline, an adenosine receptor antagonist, prevents the
protective effect of halothane, suggesting involvement of the adenosine receptor cascade in
this process.100
Second, there is evidence that the inhibition of the protein kinase C also limits the
protective effects of halothane. Thus, it can be hypothesised that factors related to protein
kinase C activation (for example, generation of inositol trisphosphate and diacylglycerol)
may be alternative targets for anaesthetic action in the myocardium.101 Third, Kersten et
al102 demonstrated that administration of 1 MAC isoflurane is beneficial in restoring
segment shortening after a myocardial ischaemic episode in dogs. The selective blockade of
the adenosine-1 receptor prevents the protective effects of isoflurane on ischaemia–
reperfusion injury. Blockade of K+ATP channels also inhibits anaesthetic induced
myocardial protection.103 Thus, it is possible that both mechanisms are related, resulting in
anaesthetic dependent stimulation of adenosine-1 receptors which, in turn, activate K+ATP
channels. The efflux of K+ through this sarcolemmal channel hyperpolarises the
myocardium. This may be the mechanism by which anaesthetic agents (as well as
myocardial preconditioning) limit the extent of myocardial stunning.99 Halothane increases
the potassium conductance and emphasises ransmembrane polarity,104 resulting in a
decrease in the duration of systole and a concomitant decrease in Ca2+ influx and ATP
consumption. These physiological changes preserve energy stores and transmembrane ion
gradients. The effects of 1–2·5 MAC sevoflurane seem to be similar to those of halothane.
Sevoflurane also decreases the Ca2+ influx into myocytes and causes a mild decrease in
contractility.105
As in other fields of research in anaesthesia, the interaction of drugs and reactive species
of oxygen are a matter of interest. Volatile anaesthetic agents may protect the ischaemic
myocardium by preventing the generation of reactive species of oxygen. Glantz et al106
demonstrated that halothane anaesthesia attenuates hydroxyl radical generation during the
reperfusion period after regional myocardial ischaemia in dogs.107 Furthermore, lactate
efflux into the coronary sinus was significantly reduced in dogs anaesthetised with
halothane compared with control animals.106
The exposure of isolated human neutrophils to halothane, enflurane, or sevoflurane
results in a decrease in oxygen burst and hydrogen peroxide production in response to the
bacterial peptide.107 In contrast, exposure to desflurane results in an increase in oxygen
burst and generation of hydrogen peroxide. Thus, anti-inflammatory effects of some
volatile anaesthetic
Page 362
drugs may be beneficial in limiting the extent of ischaemia–reperfusion injury.107
Halothane, isoflurane, and sevoflurane reduced the postischaemic adhesion of
polymorphonuclear neutrophils (PMNs) in the coronary system in an isolated Langendorff
preparation.108 The inhibitory effect of volatile anaesthetic drugs on adhesion of PMNs may
be beneficial for the heart under ischaemic conditions.
In contrast to the effects of volatile anaesthetic agents on ischaemia–reperfusion injury,
only a little is known about the effects of intravenous drugs. Ketamine/xylazine,
pentobarbital or propofol did not reduce the infarct size after ischaemia in rabbit
myocardium.100 In contrast, a marked enhancement of energy metabolism, as well as
mechanical function, was observed when propofol was given both pre-ischaemia and
during reperfusion.109 In summary, there is evidence that volatile anaesthetic agents, with
the possible exception of desflurane, as well as propofol, protect the myocardium against
brief periods of ischaemia by different mechanisms.
Central and peripheral circulation

Anaesthesia exerts direct and indirect effects on the peripheral circulation. As in the case
of the heart, it is difficult to differentiate between direct and indirect effects under in vivo
conditions. Decreased sympathetic drive, which is associated with virtually any anaesthetic
technique, will also affect peripheral vascular tone. This can be expected to alter regional
vascular resistances, blood flows, and the distribution of intravascular blood volume. In
addition, many anaesthetic agents exert direct effects on vascular smooth muscle, either by
direct vasodilation or by altering the response to other factors that modulate vascular tone.
Effects of anaesthetic agents on vascular smooth muscle
Volatile anaesthetic agents are known to cause mild vasodilation under in vivo
conditions. This is the result of both diminished vascular smooth muscle sensitivity to
circulating catecholamines and diminished neurovascular tone. Halothane selectively
attenuates α2-adrenoceptor mediated vasoconstriction.110 It was suggested that this is the
result of interference with calcium entry through smooth muscle membranes. More recent
investigations have demonstrated that volatile anaesthetic drugs might additionally
attenuate the response to α1-adrenoceptor mediated vasoconstriction, and that this effect is
not mediated by blockade of calcium influx through voltage dependent channels.111 The
attenuation of α1- and α2-responsiveness most probably contributes to in vivo effects of
volatile anaesthetic agents because stimulation of α1- and α2-adrenoceptors contributes to
basal vascular tone. Contractility of vascular smooth muscle
Page 363
is also regulated by intracellular concentration of Ca2+. The effects of halothane and
isoflurane/sevoflurane on myoplasmic Ca2+ transients of muscle cells from peripheral
arteries vary.112 The administration of halothane inhibits Ca2+ loading of the sarcoplasmic
reticulum and potentiates Ca2+ release from it after caffeine stimulation. In contrast,
isoflurane and sevoflurane administration do not lead to caffeine induced Ca2+ release
which may partly explain the different effects of halothane and isoflurane/sevoflurane on
peripheral arterial resistance.112
As outlined above, the vasodilator effects of volatile anaesthetic agents can differ within
the circulation. For instance, halothane, like nitroglycerine, is a more potent dilator of large
coronary arteries, whereas isoflurane is a more potent dilator of small coronary arteries.113
Only the second will cause a significant increase in coronary blood flow, because, under in
vivo conditions, dilatation of large coronary arteries is counteracted by the metabolic
control of coronary blood flow at the level of the arterioles and the microcirculation.
Many in vivo studies indicate that propofol has a vasodilating effect. It is, however, still
controversial whether this effect is caused by a direct vasodilating action of propofol or by
a reduction in sympathetic outflow. Propofol in concentrations of 1.1–4 mmol/1 causes
vasodilation of isolated vascular rings.114 115 Since 97–99% of Propofol is bound to plasma
proteins, clinically relevant concentrations do not lead to direct vasodilation.115
Recently, Robinson et al116 studied the effects of therapeutic doses of propofol (directly
infused into the brachial artery) on forearm vascular resistance (FVR) and forearm vein
compliance (FVC) in conscious human volunteers. Infusion of propofol does not lead to a
reduction in FVR or an increase in FVC, despite therapeutic plasma concentrations. In
contrast, infusion of sodium nitroprusside leads to a reduction in forearm vascular
resistance.116 In a second study, one sided stellate blockade was performed, and
subsequently general anaesthesia with propofol was initiated (Fig 10.19). Under these
circumstances a significant decrease in FVR and an increase in FVC were observed only in
the arm without sympathetic denervation. Changes in resistance and compliance were
comparable to those found with stellate blockade. These results suggest that the peripheral
vascular actions of propofol can primarily be attributed to the effects on sympathetic nerve
activity116 (Fig 10.19).
Anaesthesia and vascular tone
It is obvious that the mechanical properties of the arterial vascular tree oppose left
ventricular ejection. Quantitative evaluation of afterload in vivo remains, however,
difficult. In clinical as well as experimental studies, decreases in systemic vascular
resistance are most often used to describe reductions in left ventricular afterload after
exposure to anaesthetic drugs.117 It is, however, known that systemic vascular resistance
inade-
Page 364
Fig 10.19 Change in vascular resistance after infusion of propofol with and without autonomic blockade,
shown as percentage change in (a) compliance (FVC) and (b) forearm vascular resistance (FVR) in the left
arm after stellate blockade and in the right arm before and during propofol infusion . Stellate
blockade increased FVC and decreased FVR on the side of blockade. After infusion of propofol no further
changes were observed in the blocked arm, whereas a significant change in FVR and FVC in the unblocked
arm was observed, suggesting that the effects of propofol are primarily caused by inhibition of sympathetic
vasoconstrictor nerve activity. (Modified from Robinson et al.116 Reproduced with permission from
Lippincott, Williams and WilKins. Anesthesiology 1997;86:64–72.)
Page 365
quately describes afterload, because this calculated index does not take into account the
viscoelastic and frequency dependent properties of the arterial wall. Furthermore, the
dynamic phase nature of pressure and flow, as well as the effects of wave reflection in the
arterial tree, are not included in the clinically used calculation of systemic vascular
resistance. Thus, attempts have been made to characterise arterial mechanical properties by
the measurement of aortic input impedance spectra (Zin). Zm incorporates the viscoelastic
and resistive properties of the arterial system and has become a widely accepted
experimental description of left ventricular afterload (see Chapter 2). Many of the
characteristics of the aortic impedance spectra are difficult to quantify because of frequency
dependence and, therefore, a three element Windkessel model is often used to interpret
aortic impedance. This model consists of a resistor (characteristic aortic impedance, Zc) in
series with another resistor (total arterial resistance, R) and a capacitor (total arterial
compliance, C). Using the three element Windkessel model, Hettrick et al117 demonstrated
that halothane and isoflurane produced different effects on left ventricular afterload in the
chronically instrumented dog. Both drugs do not alter characteristic aortic impedance, but
halothane at 1·25, 1·5, and 1·75 MAC alters total arterial resistance, a property of arteriolar
vessels.
Thus, the effects of volatile anaesthetic agents are primarily located at the level of the
resistance vessels. Using the same study design, Lowe et al118 investigated the effects of
propofol on aortic impedance. Propofol alters left ventricular afterload by a decrease in
arteriolar tone and an increase in characteristic aortic impedance and total aortic
compliance.116 118 Etomidate increases left ventricular afterload and compromises left
ventricular systolic and diastolic performance in chronically instrumented dogs with pre-
existing left ventricular failure.119 These results indicate that the use of etomidate for
induction of anaesthesia may impair left ventricular function, especially in patients with
preoperative myocardial dysfunction.119
Effects of anaesthesia on blood volume distribution
From a physiological point of view, the vascular system can be divided into two parts.
One is the low pressure system that comprises all postarteriolar vessels, the right heart, the
pulmonary vascular system, and the left heart during diastole. The other is the high pressure
system, which includes the left ventricle during systole, systemic arteries, and arterioles.
The low pressure system holds about 85% of the total intravascular blood volume, one third
of which is in the intrathoracic compartment and two thirds in the extrathoracic low
pressure system. According to the definition by Gauer and Henry,120 the central blood
volume, which represents the intravascular blood volume between the pulmonary and the
aortic valve, is part of the intrathoracic blood volume. Intrathoracic and central blood
volumes hold a key position within the circulation because they serve as a
Page 366
reservoir for the left ventricle.120 121 It is common clinical practice to deduce changes in
intrathoracic and central blood volumes from simultaneous changes in central venous and
pulmonary capillary wedge pressures.122 123 There are, however, several problems that limit
the applicability of pressure measurements in the low pressure system for evaluation of
intravascular volume status during anaesthesia:
1 It was Gauer and Henry who demonstrated a considerable inter-individual variability of
low pressure compliance.
2 The compliance of the low pressure system is naturally affected by vascular smooth
muscle tone. When adrenergic drive is enhanced by infusion of noradrenaline, the
compliance decreases.124 Consequently, sympathetic tone influences central venous
pressure as a result of changes in effective compliance.125
3 Mechanical ventilation with positive airway pressure (in particular with positive end
expiratory pressure, PEEP) will alter the relationship between central venous pressure,
pulmonary capillary wedge pressure, and intrathoracic volume status.121
4 It has repeatedly been demonstrated that changes in central venous pressure do not
correlate with changes in pulmonary capillary wedge pressure in patients with impaired
left ventricular function.126
Direct measurements of intrathoracic or central blood volume by indicator dilution
techniques appear to be better suited for evaluation of intravascular volume status,
particulary in patients undergoing cardiac surgery, in whom myocardial dysfunction and a
critical dependence on adequate intravascular filling can be anticipated.127
Surprisingly, very few studies have been performed on the impact of anaesthesia on
intrathoracic blood volume and left ventricular preload. It is common practice to give
additional fluids with induction of anaesthesia in order to compensate for decreases in
blood pressure. It has not, however, been systematically investigated to what extent the
commonly observed decrease in blood pressure results from a volume shift between the
intra-and extrathoracic compartment or of a decrease in contractility. In general, both can
be expected with decreased sympathetic drive. Hedenstierna et al128 as well as Krayer et
al129 found that anaesthesia with muscle paralysis and mechanical ventilation leads to a
decrease in thoracic blood volume associated with a decrease in functional residual
capacity. Both groups, however, presented different results with respect to the intrathoracic
blood volume. Krayer et al129 measured total thoracic cavity volume by three dimensional x
ray computed tomography, and functional residual capacity by inert gas clearance
techniques. From the difference between both, Krayer derived an increase in intrathoracic
blood volume after induction and during maintenance of anaesthesia.129 In contrast,
Hedenstierna et al128 measured a decrease in intrathoracic blood volume using a double
indicator
Page 367
dilution technique. This result seems to be more plausible taking into account the known
effects of anaesthesia on the sympathetic system and the impact of mechanical ventilation
on intrathoracic pressures. A volume redistribution from the intrathoracic to the
extrathoracic compartment with induction of anaesthesia would support the clinical practice
in which hypotension after intubation is treated by intravenous fluids. Different anaesthetic
techniques could possibly explain the contradictory findings. Krayer et al used
fentanyl/thiopentone anaesthesia, whereas Hedenstierna and co-workers used halothane
anaesthesia.128 129 Currently, there are no systematic data available on how different types of
anaesthesia affect the expected redistribution of intravascular blood volume from the
intrathoracic to the extrathoracic compartment.
It is well known, that spinal and epidural anaesthesia cause a loss of sympathetic
innervation in the corresponding areas. The impact of regional sympathectomy by spinal
and epidural anaesthesia has been investigated by Arndt and co-workers130 in a very elegant
way using labelled erythrocytes and whole body scintigraphy.130 Regional anaesthesia
elicits a redistribution of blood to the denervated musculature and skin at the expense of
cardiac filling. The capacitance vessels of the remaining innervated muscles and skin areas,
usually of the upper extremities, might constrict in a compensatory manner. The data of
Arndt and co-workers130 also suggest that compensatory vasoconstriction might occur in the
splanchnic area, the mechanism of which is unknown. In patients with compromised
compensatory mechanisms, central blood volume might decrease dramatically. On the basis
of these findings, Lipfert and Arndt131 concluded that the poor outcome of resuscitation
attempts in cases of cardiac arrest during spinal anaesthesia132 results largely from
insufficient filling of the central vascular system.
In summary, there is evidence that regional as well as general anaesthesia is associated
with a redistribution of blood volume from the intrathoracic to the extrathoracic
compartment. Little is known about the time course of volume distribution during surgery
and especially during recovery from anaesthesia. This, together with the influence of
various anaesthetic techniques on intravascular volume distribution, is certainly a subject
that merits further investigations.
1 Sebel PS, Bovill JG. Cardiovascular effects of sufentanil anesthesia. Anesth Analg 1982;61:115–19.
2 Bovill JG, Sebel PS, Stanley TH. Opioid analgesics in anesthesia: With special reference to their use in
cardiothoracic anesthesia. Anesthesiology 1984;61:731–55.
3 Skarvan K, Schwinn W. Haemodynamic interactions between midazolam and alfentanil in patients with
coronary disease. Anaesthesist 1986;35:17–23.
4 Tomicheck RC, Rosow CE, Philbin DM, Moss J, Teplick RS, Schneider RC. Diazepam fentanyl
interaction-hemodynamic and hormonal effects in coronary artery surgery. Anesth Analg 1985;62:881–4.
Page 368
5 Strauer BE. Contractile response to morphine, piritramide, meperidine and fentanyl: A comparative study
of effects on the isolated ventricular myocardium. Anesthesiology 1972;37:304–10.
6 Hug CC Jr. Does opioid ‘‘anesthesia” exist. Anesthesiology 1990;73:l–4.
7 Ebert TJ, Muzi M, Berens R, Goff D, Kampine J. Sympathetic responses to induction of anaesthesia in
humans with propofol or etomidate. Anesthesiology 1992;76:725–33.
8 Wallin BG, Fagius J. Peripheral sympathetic neural activity in conscious humans. Annu Rev Physiol
1988;50:565–76.
9 Ebert TJ, Kampine JP. N2O augments sympathetic outflow: Direct evidence from human peroneal nerve
recordings. Anesth Analg 1989;69:444–9.
10 Sellgren J, Ponten J, Wallin BG. Characteristics of muscle sympathetic nerve activity during general
anaesthesia in humans. Acta Anaesthesiol Scand 1992;36:336–45.
11 Sellgren J, Ponten J, Wallin G. Percutaneous recordings of muscle nerve sympathetic nerve activity
during propofol, N2O, and isoflurane anesthesia in humans. Anesthesiology 1990; 73:20–7.
12 Ebert TJ, Kanitz DD, Kampine JP. Inhibition of sympathetic neural outflow during thiopental anesthesia
in humans. Anesth Analg 1990, 71:319–26.
13 Yli-Hankala A, Randell T, Seppala T, Lindgren L. Increases in hemodynamic variables and
catecholamine levels after rapid increase in isoflurane concentration. Anesthesiology 1993;78:266–71.
14 Ebert TJ, Muzi M. Sympathetic hyperactivity during desflurane anesthesia in healthy volunteers. A
comparison with isoflurane. Anesthesiology, 1993;79:444–53.
15 Muzi M, Ebert TJ, Hope WG, Robinson BJ, Bell LB. Site(s) mediating sympathetic activation with
desflurane. Anesthesiology. 1996;85:737–47.
16 Ebert TJ, Harkin CP, Muzi M. Cardiovascular responses to sevoflurane: a review. Anesth Analg
1995;81:11–22.
17 Ebert TJ, Muzi M, Lopatka CW. Neurocirculatory responses to sevoflurane in humans. A comparison to
desflurane. Anesthesiology 1995;83:88–95.
18 Griffiths R, Norman RI. Effects of anesthetics on uptake, synthesis and release of transmitters. Br J
Anaesth 1993;71:96–107.
19 Goldstein DS, McCarty R, Polinsky RJ, Kopin IJ. Relationship between plasma norepinephrine and
sympathetic nerve activity. Hypertension, 1983;5:552–9.
20 Honda T, Ninomiya I. Changes in AdSNA and arterial catecholamines to coronary occlusion in cats. Am J
Physiol 1988;255:H704–10.
21 Esler M, Jennings G, Korner P. Assessment of human sympathetic nervous system activity from
measurements of norepinephrine turnover. Hypertension 1988;11:3–20.
22 Wallin B, Sundlf G, Eriksson B. Plasma noradrenaline correlates to sympathetic muscle nerve activity in
normotensive man. Acta Physiol Scand 1981;111:69–73.
23 Pocock G, Richards CD. Cellular mechanisms in general anaesthesia. Br J Anaesth 1991;66:116–28.
24 Philbin DM, Rosow CE, Schneider RC, Koski G, D’Ambra MN. Fentanyl and sufentanil anesthesia
revisited: How much is enough. Anesthesiology 1990;73:5–11.
25 Sonntag H, Stephan H, Lange H, Rieke H, Kettler D, Martschausky N. Sufentanil does not block
sympathetic responses to surgical stimuli in patients having CABG. Anesth Analg 1989;68:584–92.
26 Liem TH, Booij LHDJ, Gielen MJM, Hasenbos MAWM, van Egmond J. Coronary artery bypass grafting
using two different anesthetic techniques: Part 3: Adrenergic responses. J Cardiothorac Vasc Anesth
1992;6:162–7.
27 Ebert TJ. Sympathetic hyperactivity during desflurane anesthesia in healthy volunteers. Anesthesiology
1993;79:444–53.
28 Skovstedt P, Price ML, Price HL. The effect of short acting barbiturates on arterial pressure,
preganglionic sympathetic activity and barostatic reflexes. Anesthesiology 1970;33:10–8.
29 Many J, Gauzit R, Lefevre P, et al. Effects of diazepam and midazolam on baroreflex control of heart rate
and on sympathetic activity in humans. Anesth Analg 1986;65:113–9.
Page 369
30 Kotrly KJ, Ebert TJ, Vucins EJ, Roerig DL, Stadnicka A, Kampine JP. Effects of fentanyl–diazepam–
N2O anaesthesia on arterial baroreflex control of heart rate in man. Br J Anaesth 1986;58:406–14.
31 Kotrly KJ, Ebert TJ, Vucins E, Igler FO, Barney JA, Kampine JP. Baroreceptor reflex control of the heart
rate during isoflurane anesthesia in humans. Anesthesiology 1984;60:173–9.
32 Muzi M, Ebert TJ. Randomized, prospective comparison of halothane, isoflurane, and enflurane on
baroreflex control of heart rate in humans. Adv Pharmacol 1994;31:379–87.
33 Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectrum analysis of heart
rate fluctuation: a quantitative probe of beat to beat cardiovascular control. Science 1981 ;213:220–2.
34 Pomeranz B, Macaulay RJ, Caudill MA, et al. Assessment of autonomic function in humans by heart rate
spectral analysis. Am J Physicl 1985;248:H151–3.
35 Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure
variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res 1986;59:178–
93.
36 Ikeda T, Doi M, Morita K, Ikeda K. Effects of midazolam and diazepam as premedication on heart rate
variability in surgical patients. Br J Anaesth 1995;73:479–83.
37 Michaloudis D, Kochiadakis G, Georgopoulou G, et al. The influence of premedication on heart rate
variability. Anaesthesia 1998;53:446–53.
38 Widmark C, Olaison J, Reftel B, Jonsson LE, Lindecrantz K. Spectral analysis of heart rate variability
during desflurane and isoflurane anaesthesia in patients undergoing arthroscopy. Acta Anaesthesiol Scand
1998;42:204–10.
39 Ireland N, Meagher J, Sleigh JW, Henderson JD. Heart rate variability in patients recovering from general
anaesthesia. Br J Anaesth 1996;76:657–62.
40 Komatsu T, Singh PK, Kimura T, Nishiwaki K, Bando K, Shimada Y. Differential effects of ketamine
and midazolam on heart rate variability. Can J Anaesth 1995;42:1003–9.
41 Galletly DC, Buckley DH, Robinson BJ, Corfiatis T. Heart rate variability during propofol anaesthesia. Br
J Anaesth 1994;72:219–20.
42 Latson TW, OFlaherty D. Effects of surgical stimulation on autonomic reflex function: assessment by
changes in heart rate variability. Br J Anaesth 1993;70:301–5.
43 Zickmann B, Hofmann HC, Potkamper C, Knothe C, Boldt J, Hempelrnann G. Changes in heart rate
variability during induction of anesthesia with fentanyl and midazolam. J Cardiothorac Vasc Anesth
1996;10:609–13.
44 Latson TW, McCarroll SM, Mirhej MA, Hyndman VA, Whitten CW, Lipton JM. Effects of three
anesthetic induction techniques on heart rate variability. J Clin Anesth 1992;4:265–76.
45 Komai H, Rusy BF. Differences in the myocardial depressant action of thiopental and halothane. Anesth
Analg 1984;63:313–8.
46 Kissin I, Motomura S, Aultmann DF, Reves JG. Inotropic and anesthetic potencies of etomidate and
thiopental in dogs. Anesth Analg 1983;62:961–5.
47 Roewer N, Proske O, Schulte am Esch J. The effects of midazolam on the mechanical and electrical
properties of the isolated ventricular myocardium. Anaesth Intensither Notfallmed 1990;25:354–61.
48 Azuma M, Matsumura C, Kemmotsu O. Inotropic and electrophysiological effects of propofol and
thiamylal in isolated papillary muscles of the guinea pig and the rat. Anesth Analg 1993;77:557–63.
49 Mattheusen M, Housmans PR. Mechanisms of the direct, negative inotropic effect of etomidate in isolated
ferret ventricular myocardium. Anesthesiology 1993;79:1284–95.
50 Stowe DF, Bosjnak ZJ, Kampine JP. Comparison of etomidate, ketamine, midazolam, propofol and
thiopental on function and metabolism of isolated hearts. Anesth Analg 1992;74:547–58.
51 Park WK, Lynch C III. Propofol and thiopentone depression of myocardial contractility. A comparative
study of mechanical and electrophysiological effects in isolated guinea pig ventricular muscle. Anesth
Analg 1992;74:395–405.
Page 370
52 Rusy BF, Komai H. Anesthetic depression of myocardial contractility. A review of possible mechanisms.
53 Dowdy EG, Kaya K. Studies of the mechanisms of cardiovascular responses to CI-581. Anesthesiology
1968;29:931.
54 Urthaler F, Walker AA, James TN. Comparison of the inotropic action of morphine and ketamine studies
in canine artery muscle. J Thorac Cardiovasc Surg 1976;72:142.
55 Riou B, Viars P, Lecarpentier Y. Effects of ketamine on the cardiac papillary muscle of normal hamsters
and those with cardiomyopathy. Anesthesiology 1990;73:910–18.
56 Cook DJ, Carton EG, Housmans PR. Mechanisms of the positive inotropic effect of ketamine in isolated
ferret ventricular papillary muscle. Anesthesiology 1991;74:880–8.
57 Kongsayreepong S, Cook DJ, Housmans PR. Mechanisms of the direct, negative inotropic effect of
ketamine in isolated ferret and frog ventricular myocardium. Anesthesiology 1993;79:313–22.
58 Gelissen HP, Epema AH, Henning RH, Krijnen HJ, Hennis PJ, den Hertog A. Inotropic effects of
propofol, thiopental, midazolam, etomidate, and ketamine on isolated human atrial muscle. Anesthesiology
1996;84:397–403.
59 Lynch C III, Frazer MJ. Depressant effects of volatile anesthetics upon rat and amphibian myocardium:
Insights into anesthetic mechanisms of action. Anesthesiology 1989;70:511–22.
60 Wolf WJ, Neal MB, Mathew BP, Bee DE. Comparison of the in vitro myocardial depressant effects of
isoflurane and halothane anesthesia. Anesthesiology 1988;69:660–6.
61 Kemmotsu O, Hashimoto Y, Shimosato S. Inotropic effects of isoflurane on mechanics of contraction in
isolated cat papillary muscles from normal and failing hearts. Anesthesiology 1973;39:402–15.
62 Kemmotsu O, Hashimoto Y, Shimosato S. The effects of fluroxene and enflurane on contractile
performance of isolated papillary muscles from failing hearts. Anesthesiology 1974;40:253–60.
63 Lynch C III. Differential depression of myocardial contractility by volatile anesthetics in vitro:
Comparison with uncouplers of excitation–contraction coupling. J Cardiovasc Pharmacol 1990;15:655–
65.
64 Stowe DF, Monroe SM, Marijic J, Bosnjak ZJ, Kampine JP. Comparison of halothane, enflurane, and
isoflurane with nitrous oxide on contractility and oxygen supply and demand in isolated hearts.
65 Hatakeyama N, Momose Y, Ito Y. Effects of sevoflurane on contractile responses and electrophysiologic
properties in canine single cardiac myocytes. Anesthesiology 1995;82:559–65.
66 Gueugniaud PY, Hanouz JL, Vivien B, Lecarpentier Y, Coriat P, Riou B. Effects of desflurane in rat
myocardium: comparison with isoflurane and halothane. Anesthesiology 1997;87:599–609.
67 Lawson D, Frazer MJ, Lynch C III. Nitrous oxide effects on isolated myocardium: a reexamination in
vitro. Anesthesiology 1990;73:930–43.
68 Bovill JG, Boer F. Opioids in cardiac anesthesia. In: Kaplan J, ed, Cardiac anesthesia. 3rd edn. WB
Saunders & Co, 1993:467–511.
69 Sagawa K. The ventricular pressure–volume diagram revisited. Circ Res 1978;46: 677–87.
70 Mulier JP, Wouters P, van Aken H, Vermaut G, Vandermersch M. Cardiodynamic effects of propofol in
comparison with thiopental: Assessment with transesophageal echocardio-graphic approach. Anesth Analg
1991;72:28–35.
71 Pagel PS, Kampine JP, Schmeling WT, Warltier DC. Reversal of volatile anesthetic induced depression of
myocardial contractility by extracellular calcium also enhances left ventricular diastolic function.
72 Heinrich H, Fontaine L, Fösel TH, Spilker D, Winter H, Ahnefeld FW. Echocardio-graphic assessment of
the negative inotropic effects of halothane, enflurane and isoflurane. Anaesthesist 1986;35:465–72.
73 Pagel PS, Kampine JP, Schmeling WT, Warltier DC. Influence of volatile anesthetics on myocardial
contractility in vivo: Desflurane versus Isoflurane. Anesthesiology 1991;74:900–7.
Page 371
74 Harkin CP, Pagel PS, Kersten JR, Hettrick DA, Warltier DC. Direct negative inotropic and lusitropic
effects of sevoflurane. Anesthesiology 1994;81:156–67.
75 Pagel PS, Kampine JP, Schmeling WT, Warltier DC. Effects of nitrous oxide on myocardial contractility
as evaluated by the preload recruitable stroke work relationship in chronically instrumented dogs.
76 Hoeft A, Sonntag H, Stephan H, Kettler D. Validation of myocardial oxygen demand indices in patients
awake and during anesthesia. Anesthesiology 1991;75:49–56.
77 Rhode E. Über den Einfluss der mechanischen Bedingungen auf die Tätigkeit und den
Sauerstoffverbrauch des Warmblüterherzens. Archiv für Experimentelle Pathologie und Pharmakologie
1915;68:401–34.
78 Bailer D, Bretschneider HJ, Hellige G. Validity of myocardial oxygen consumption parameters. Clin
Cordial 1979;2:317–27.
79 Rooke GA, Feigl EO. Work as a correlate of canine left ventricular oxygen consumption, and the problem
of catecholamine wasting. Circ Res 1982;50:273–86.
80 Rooke GA. Low dose halothane anesthesia does not affect the hemodynamic estimation of myocardial
oxygen consumption in dogs. Anesthesiology 1990;72:682–93.
81 Kettler D, Sonntag H. Intravenous anaesthetics: Coronary blood flow and myocardial oxygen
consumption. Acta Anaesthesiol Belg 1974;25:384–401.
82 Hoeft A, Sonntag H, Stephan H, Kettler D. The influence of anesthesia on myocardial oxygen utilization
efficiency in patients undergoing coronary bypass surgery. Anesth Analg 1994;78:857–66.
83 Turner E. Pathophysiologie der Aufwachphase. Anästhesiologie und Intensivmedizin (Bd 179). 1986,
Berlin: SpringerVerlag.
84 Evans CL, Matsuoka Y. The effect of various mechanical conditions on the gaseous metabolism and
efficiency of the mammalian heart. J Physiol (Lond) 1915;49:378–105.
85 Burkhoff D, Sagawa K. Ventricular efficiency predicted by an analytical model. Am J Physiol
1986;250:R1021–7.
86 Suga H, Igarashi Y, Yamada O, Goto Y. Mechanical efficiency of the left ventricle as a function of
preload, afterload, and contractility. Heart Vessels 1985;1:3–8.
87 Hoeft A, Koeb H, Hellige G, Sonntag H. Energetics and efficiency of cardiac pump work. Anaesthetist
1991;40:465–78.
88 Sunagawa K, Maughan WL, Burkhoff D, Sagawa K. Left ventricular interaction with arterial load studied
in isolated canine ventricle. Am J Physiol 1983;245:H773–80.
89 Sunagawa K, Maughan WL, Sagawa K. Optimal arterial resistance for the maximal stroke work studied in
isolated canine left ventricle. Circ Res 1985;56:586–95.
90 Hettrick DA, Pagel PS, Warltier DC. Desflurane, sevoflurane, and isoflurane impair canine left
ventricular–arterial coupling and mechanical efficiency. Anesthesiology 1996;85:403–13.
91 Hettrick DA, Pagel PS, Warltier DC. Isoflurane and halothane produce similar alterations in aortic
distensibility and characteristic aortic impedance. Anesth Analg 1996;83:1166–72.
92 Hettrick DA, Pagel PS, Warltier DC. Alterations in canine left ventricular–arterial coupling and
mechanical efficiency produced by propofol. Anesthesiology 1997;86:1088–93.
93 Hoeft A, Sonntag H, Stephan H. Coronary flow and myocardial oxygen balance in anaesthesia. AINS
1993;26:398–407.
94 Crystal GJ, Gurevicius J, Salem MR, Zhou X. Role of adenosine triphosphate sensitive potassium
channels in coronary vasodilation by halothane, isoflurane, and enflurane. Anesthesiology 1997;86:448–
58.
95 Priebe HJ. Isoflurane and coronary hemodynamics. Anesthesiology 1989;71:960–76.
96 Nicklas JM, Becker LC, Bulkley BH. Effects of repeated brief coronary occlusion on regional left
ventricular function and dimension in dogs. Am J Cardiol 1985;56:473–8.
97 Gurevicius J, Holmes CB, Salem MR, Abdel Halim A, Crystal GJ. The direct effects of enflurane on
coronary blood flow, myocardial oxygen consumption, and myocardial segmental shortening in in situ
canine hearts. Anesth Analg 1996;83:68–74.
98 DeTraglia MC, Komai H, Redon D, Rusy BF. Isoflurane and halothane inhibit tetanic contractions in
rabbit myocardium in vitro. Anesthesiology. 1989;70:837–42.
Page 372
99 Nader-Djalal N, Knight PR. Volatile anaesthetic effects on ischaemic myocardium. Curr Opin
Anaesthesiol 1998;11:403–6.
100 Cope DK, Impastato WK, Cohen MV, Downey JM. Volatile anesthetics protect the ischemic rabbit
myocardium from infarction. Anesthesiology 1997;86:699–709.
101 Shah AM, Silverman HS, Griffiths EJ, Spurgeon HA, Lakatta EG. cGMP prevents delayed relaxation at
reoxygenation after brief hypoxia in isolated cardiac myocytes. Am J Physiol 1995;268:H2396–404.
102 Kersten JR, Orth KG, Pagel PS, Mei DA, Gross GJ, Warltier DC. Role of adenosine in isoflurane induced
cardioprotection. Anesthesiology 1997;86:l128–39.
103 Liu Y, Gao WD, O’Rourke B, Marban E. Synergistic modulation of ATP sensitive K+ currents by protein
kinase C and adenosine. Implications for ischemic preconditioning. Circ Res 1996;78:443–54.
104 Hirota K, Ito Y, Momose Y. Effects of halothane on membrane potentials and membrane ionic currents in
single bullfrog atrial cells. Acta Anaesthesiol Scand 1988;32:333–8.
105 Park WK, Pancrazio JJ, Suh, CK, Lynch C III. Myocardial depressant effects of sevoflurane. Mechanical
and electrophysiologic actions in vitro. Anesthesiology 1996;84:1166–76.
106 Glantz L, Ginosar Y, Chevion M, et al. Halothane prevents postischemic production of hydroxyl radicals
in the canine heart. Anesthesiology 1997;86:440–7.
107 Frohlich D, Rothe G, Schwall B, et al. Effects of volatile anaesthetics on human neutrophil oxidative
response to the bacterial peptide FMLP. Br J Anaesth 1997;78:718–23.
108 Kowalski C, Zahler S, Becker BF, et al. Halothane, isoflurane, and sevoflurane reduce postischemic
adhesion of neutrophils in the coronary system. Anesthesiology 1997;86:188–95.
109 Kokita N, Hara A, Abiko Y, Arakawa J, Hashizume H, Namiki A. Propofol improves functional and
metabolic recovery in ischemic reperfused isolated rat hearts. Anesth Analg 1998;86:252–8.
110 Larach DR, Schuler HG, Derr JA, Larach MG, Hensley FA, Zelis R. Halothane selectively attenuates α2-
adrenoceptor mediated vasoconstriction, in vivo and in vitro. Anesthesiology 1987;66:781–91.
111 Kenny D, Pelch LR, Brooks HL, Kampine JP, Schmeling WT, Warltier DC. Calcium channel modulation
of α1 and α2-adrenergic pressor responses in conscious and anesthetized dogs. Anesthesiology
1990;72:874–81.
112 Yamamoto M, Hatano Y, Kakuyama M, et al. Different effects of halothane, isoflurane and sevoflurane
on sarcoplasmic reticulum of vascular smooth muscle in dog mesenteric artery. Acta Anaesthesiol Scand
1997;41:376–80.
113 Nakamura K, Toda H, Hatano Y, Mori K. Comparison of the direct effects of sevoflurane, isoflurane and
halothane on isolated canine coronary arteries. Can J Anaesth 1993;40:257–61.
114 Park WK, Lynch C III, Johns RA. Effects of propofol and thiopental in isolated rat aorta and pulmonary
artery. Anesthesiology 1992;77:956–63.
115 Nakamura K, Hatano Y, Hirakata H, Nishiwada M, Toda H, Mori K. Direct vasoconstrictor and
vasodilator effects of propofol in isolated dog arteries. Br J Anaesth 1992;68:193–7.
116 Robinson BJ, Ebert TJ, O’Brien TJ, Colinco MD, Muzi M. Mechanisms whereby propofol mediates
peripheral vasodilation in humans. Sympathoinhibition or direct vascular relaxation? Anesthesiology.
1997;86:64–72.
117 Hettrick DA, Pagel PS, Warltier DC. Differential effects of isoflurane and halothane on aortic input
impedance quantified using a three-element Windkessel model. Anesthesiology 1995;83:361–73.
118 Lowe D, Hettrick DA, Pagel PS, Warltier DC. Propofol alters left ventricular afterload as evaluated by
aortic input impedance in dogs.Anesthesiology 1996;84:368–76.
119 Pagel PS, Hettrick DA, Kersten JR, Tessmer JP, Lowe D, Warltier DC. Etomidate adversely alters
determinants of left ventricular afterload in dogs with dilated cardio-myopathy. Anesth Analg
1998;86:932–38.
120 Gauer OH, Henry JP. Circulatory basis of fluid volume control. Physiol Rev 1963;43:423–81.
Page 373
121 Arndt JO. The low pressure system: the integrated function of veins. Eur J Anaesth 1986;3:343–70.
122Henry JP, Gauer OH, Sieker HO. The effects of moderate changes in blood volume on left and right atrial
pressure. Circ Res 1956;4:91–4.
123Gauer OH, Henry JP, Behn C. Changes in central venous pressure after moderate hemorrhage and
transfusion in man. Circ Res 1956;4:79–90.
124Echt M, Düweling J, Gauer OH, Lange L. Effective compliance of the total vascular bed and the
intrathoracic compartment derived from changes in central venous pressure induced by volume changes in
man. Circ Res 1974;34:61–8.
125Bonica JJ; Kennedy WF, Nakamatsu TJ, Gerbershagen HU. Circulatory effects of peridural block: III
effects of acute blood loss. Anesthesiology 1972;36:219–27.
126Mangano DT. Monitoring pulmonary artery pressure in coronary artery disease. Anesthesiology
1980;53:364–70.
127Hoeft A, Schorn B, Weyland A, et al. Bedside assessment of intravascular volume status in patients
undergoing cardiac surgery. Anesthesiology 1994;81:76–86.
128Hedenstierna G, Strandberg A, Brismar B, Lundquist H, Svensson L, Tokics L. Functional residual
capacity, thoracoabdominal dimensions and central blood volume during general anesthesia with muscle
paralysis and mechanical ventilation. Anesthesiology 1985;62:247–54.
129Krayer S, Rehder K, Beck KC, Cameron PD, Didier EP, Hofman EA. Quantification of thoracic volumes
by three dimensional imaging. J Appl Physiol 1987;62:591–8.
130Arndt JO, Hock A, Stanton-Hicks M, Stühmeier KD. Peridural anesthesia and the distribution of blood in
supine humans. Anesthesiology 1985;63:616–23.
131Lipfert P, Arndt JO. Major conduction anaesthesia, Pathogenesis, prophylaxis, and therapy of circulatory
complications Anaesthesist 1993;42:773–87.
132Caplan RA, Ward RJ, Posner K, Cheney FW. Unexpected cardiac arrest during spinal anesthesia: A
closed claims analysis of predisposing factors. Anesthesiology 1988;68:5–11.
Page 374
This page intentionally left blank

Page 375
Index
acetylcholine 185, 220, 315

action potential 73, 78, 79–86
depolarisation 84–6
plateau 85–6, 109–10
resting membrane potential 80
active hyperaemia 236
acute respiratory distress syndrome 196, 199–200, 202, 203
adenosine 130–1, 140, 232
renal circulation 284–5
splanchnic circulation 290
adenosine triphosphate (ATP) 4, 9, 315
trapping 18
adrenaline 186, 290, 335
anaesthetic effects 337
and microcirculation 315
and pulmonary circulation 196
afterload 32–3, 38–43
aortic input impedance 40–2
arterial/ventricular systolic pressures 39–40
effective arterial elastance 43
peripheral vascular resistance 40
systolic wall stresses 42–3
afterload mismatch 55–6
ageing, and ventricular performance 65–7
air–blood barrier 204
alfentanil 308
almitrine, and pulmonary circulation 194, 197
alpha-adrenergic receptors 132–4
alpha-blockers, and pulmonary circulation 194
alveolar hypoxia 200
alveolar vessels 173–4
amino acids 12
amniotic fluid embolism 198–9
anaesthesia see anaesthetics
anaesthetics 331–73
see also individual anaesthetics
arrhythmic potential 112
and blood volume distribution 365–7
cardiac effects 342–62
and cerebral blood flow 260–2
circulatory effects
central and peripheral circulation 362–7
cutaneous circulation 295–6
microcirculation 317–27
muscle circulation 301–3
pulmonary circulation 194–6
splanchnic circulation 292–3
inhaled 260–2
intravenous 262
ischaemia–reperfusion injury 360–2
and myocardial oxygen consumption 349–54
and sympathetic system 332–42
and tissue hypoxia 308–9
and vascular resistance 321–3, 363–5
anaplerosis 8
anatomical re-entry 98–100
anatomy
cerebral circulation 240–4
coronary 119–22
angiotensin converting enzyme inhibitors 185
angiotensin II 184, 186, 290
anisotropic atrial conduction 100
Anrep effect 32
antiarrhythmics 106–12
desirable drug properties 108
ion channels as targets for 108–12
structure of myocardium 107–8
antidiuretic hormone 230, 283–4, 290
aortic input impedance 40–2
aortic valve 30, 31
arrhythmia
Page 376
see also antiarrhythmics
anaesthetic-induced 112
mechanisms for 91–2, 105–6
arterial chemoreceptors 226–7
arterial oxygen pressure 259
arterial systolic pressure 39–40
arterioles, and vascular resistance 310
ATP-sensitive potassium channels 131–2, 142
atrial booster pump 58
atrial contraction 58–60
atrial natriuretic peptide 232, 284
atrial septal defects 200
atrial/vena caval low pressure baroreceptors 225–6
atrioventricular node 220
auto PEEP 179
automated border detection 36
automatic transmembrane potential 78
automaticity 73, 80, 86–8
abnormal 92–3, 106
altered normal 107
modulation of 88
Purkinje fibre 88
sinus node 86–8
subsidiary (latent) pacemakers 88
autonomic hyperreflexia 228–9
autonomic insufficiency 225
autonomic nervous system 217–22, 260
circulatory effects 220–2
parasympathetic nervous system 219–20
sympathetic nervous system 217–19
autoregulation 137–44, 153, 234, 255–8
adenosine 140
ATP-sensitive potassium channels 142
left ventricular 143–4
of local blood flow 311–12
metabolic hypothesis of 310–11
myogenic control 142–3
nitric oxide 140–1
oxygen/carbon dioxide tension 141–2
pressure 234–6
right ventricular 144
AV nodal re-entry tachycardia 102
AV node re-entry 102–3
with accessory pathways 103–5
Bainbridge reflex 225

baroreceptor reflex 222–5
anaesthetic effects on 338–41
beta-adrenergic receptors 134
beta-agonists, and pulmonary circulation 194
Bezold–Jarisch reflex 226
blood volume distribution, anaesthetic effects on 365–7
blood–brain barrier 244–6
Bowditch’s positive staircase response 50
bradykinin 128, 184, 185, 290, 315
calcitonin gene-related peptide 135, 294
calcium channel blockers, and pulmonary circulation 194, 197
calcium homeostasis 7
capillaries 310
in gas exchange 313
recruitment 122–3, 242
capillary transit time 313
carbon dioxide tension 141–2, 258–9
cardiac control centre 215–17
cardiac output 40, 54, 64
cardiac pacing 50–1
carotid sinus 222
catabolism 9–10
catecholamines 229
see also individual catecholamines
cellular physiology/metabolism 1–26
catabolism of substrates 9–10
clinical relevance of myocardial metabolism 13–20
energy transfer in heart muscle 2–5
metabolic pathways and moiety conserved cycles 5–9
nutrition and myocardial protein turnover 10
substrate competition 11–13
thermodynamic aspects of energy transfer 5
cerebral blood volume 242–3
cerebral circulation 240–77
control mechanisms 268–70
in disease states 263–8
functional anatomy 240–4
measurement of cerebral blood flow 246–54
microcirculatory transport and blood–brain barrier 244–6
pharmacological modulation 260–3
Page 377
regional cerebral blood flow 255–60
cerebral perfusion, determinants of 244
chamber compliance 59
chamber stiffness 59
cholecystokinin 290
cholinergic receptors 134
chronotropism 220
circle of Willis 241
circulatory system, function of 309
circus movement 95
CNS ischaemic reflex 227–8
cold reflex 227
collateral circulation 123–4
computed tomography 37
conduction 89–90
conduction block 105
contractility 33, 43–50, 110
corner vessels 175–6
coronary anatomy 119–22
coronary blood flow regulation 124–58
autoregulation 137–44
coronary flow reserve 155–8
endothelial control 125–30
flow–function relationship 154–5
metabolic control 130–2
neurohumoral control 132–7
pressure–flow relationships 152–4
variations in myocardial perfusion 144–51
coronary flow reserve 155–8
coronary physiology 119–70
coronary anatomy 119–22
coronary collateral circulation 123–4
coronary microcirculation 122–3
coronary/myocardial blood flow 158–62
regulation of coronary blood flow 124–58
coronary vascular resistance 148–51
cranial nerves 219
critical closing pressure 153
Cushing reflex 228
anatomy 293–4
control of 294
cyclic GMP 185
cyclo-oxygenase inhibitors, and pulmonary circulation 194, 197
defibrillation threshold 110

delayed after depolarisations 95
delayed relaxation period 57–8
depressed fast response 84, 91, 106
depressor reflex 223
desflurane 262, 335, 336
and myocardial efficiency 357
and noradrenaline levels 338
diazepam, and baroreceptor reflex 338–9
distribution of ventilation 178, 179
dobutamine 46, 47, 53, 196
dopamine, and pulmonary circulation 196
Doppler ultrasound 159–60, 253–4, 318
dynamic X ray computed tomography 248
early after depolarisations 94–5

echocardiography 37
effective arterial elastance 43
eicosanoids 315
ejection fraction 30, 35
elasticity 33
electromagnetic flow meters 159
electrophysiology 73–118
abnormal 91–105
antiarrhythmic drug action 106–12
mechanisms for clinical arrhythmias 105–6
normal 73–91
electrotonic effects 84
emphysema 199
end diastolic pressure 31
end diastolic volume 31, 32, 35
end systolic pressure 48, 51–2
endothelin-1 290
endothelins 128, 186, 233–4, 315
endothelium-derived contracting factors 128–30
endothelium-derived hyperpolarising factor 315, 326
endothelium-derived relaxing factor see nitric oxide
endotoxin 186
endurance training 65
energy transfer 2–5
rate of 4
thermodynamics of 5
enflurane 260–2, 296, 308
and baroreceptor reflex 339, 340
and blood pressure 351
and heart rate 351
Page 378
and myocardial oxygen consumption 351
negative inotropic effect 346
and organ blood flow 321, 322
and stroke volume index 351
etomidate 262, 333
and baroreceptor reflex 338–9
and heart rate 342, 350
excitability 89
excitable gap 98, 99
exercise testing 64–5
extra-alveolar vessels 175, 177
extrapulmonary vessels 173
extrinsic reflexes 227–8
Fallot’s tetralogy 203

fast response fibre 80
fatty acids 12
fentanyl 352, 355, 359
Fick’s method 35
flow–function relationship 154–5
flow–metabolism coupling 255, 270
fluid balance in lung 204–7
Fourier analysis 40–1
fractional area change 36, 38
Frank–Starling law 35, 53–6, 61
external control mechanisms 55
internal control mechanisms 54–5
free fatty acids 11
functional hyperaemia 236
functional re-entry 100–2
functional residual capacity 177
garden hose effect 155

gas exchange 313–14
glibenclamide 142
glucagon 290
glucose 11, 17
glucose-insulin-potassium 19–20
glycogen loading 19
glycolysis 17
gravity, and pulmonary blood flow 173–6
alveolar vessels 173–5
corner vessels 175–6
extra-alveolar vessels 175
Gregg effect 140, 154–5
Guyton’s diagram 55, 56
haematocrit 259–60
haemorrhagic shock 307
halothane 260–2, 292, 307, 308
and heart rate 351
and microcirculation 318–21
and myocardial efficiency 355, 359
and organ blood flow 321, 322, 323
head injury 265–6
heart see coronary; myocardial
heart muscle
energy transfer 2–5
features of 3
length dependent activation 28
ventricular 27–9
heart rate 50–1, 60
variability 341
heart transplantation 221
hexobarbitone
and heart rate 350
hibernating myocardium 16
histamine 232–3, 290, 299–300, 315
Hodgkin–Huxley model 75, 77
homoeometric regulation 32
humoral control mechanisms 229–34
atrial natriuretic peptide 232
catecholamines 229
endothelium-derived nitric oxide 230–2
microcirculation 315
renin–angiotensin system 229–30
vasopressin 230
hypercapnia 315
hypertensive encephalopathy 266
hypovolaemia 314, 315
hypoxia 186, 315
hypoxic pulmonary vasoconstriction 183–4
imidazoline receptors 216

Page 379
imidazoline-like substances 216–17
impulse propagation 88–91
conduction 89–90
excitability 89
refractoriness 90–1
indicator dilution techniques 318
inert gas clearance technique 158–9
inertance 33
interferon 186
intra-alveolar vessels, resistance of 177
intrapulmonary vessels 173
intrinsic PEEP 179
intrinsic reflexes 222–7
ion channels
Hodgkin–Huxley model of Na+ channel gating 75, 77
ligand gated 75, 77–8
properties of 74
structure–function relationships 78–9
as targets for antiarrhythmics 108–12
voltage gated 74–5, 76
ischaemia 307
cellular consequences 16–18
cerebral 263–4
low flow 17
metabolic support 18–20
normal flow 17
ischaemia–reperfusion injury, anaesthetic effects on 360–2
isoflurane 260–2, 296, 308, 336
and heart rate 351
and organ blood flow 321
isolated perfused lung preparations 189
isoprenaline, and pulmonary circulation 196
isovolumic relaxation period 57–8
jugular venous oximetry 254
ketamine 263, 307

and organ blood flow 321, 322, 323
ketone bodies 11, 12
Kety–Schmidt technique 247, 248, 349
kinins 233, 285
kissing papillary muscles sign 38
lactate 11
Laplace’s Law 39, 42, 66
laser Doppler velocimetry 318
leading circle re-entry 100–1
left atrial pressure 62
left ventricular assistance 63
left ventricular autoregulation 143–4
length dependent activation 28
leukotrienes 186
ligand gated ion channels 75, 78
lignocaine 194, 197
lung volume, and pulmonary vascular resistance 176–82
luxury perfusion 119
magnetic resonance imaging 37, 207, 251

maximal left ventricular power 49
maximum diastolic potential 80
Mayer waves 227
membrane potential, loss of 91, 106
metabolic acidosis 300
metabolic pathways 5–9
in intact heart 14–16
metabolic regulation 311
methohexitone 334, 359
Michaelis constant 12
anatomy 309–10
capacitance 313
cerebral 242
coronary 122–3
exchange 313–14
local control 315–17
neural control 314–15
resistance 310–13
systemic humoral control 315
midazolam 338
Page 380
moieties
excess of 9
recycling of 8
moiety conservation, principle of 4
moiety conserved cycles 5–9
anatomy 296–7
control of 297–301
myocardial contrast echocardiography 162
myocardial efficiency, anaesthetic effects on 354–9
myocardial infarction 60
myocardial metabolism, clinical relevance 13–20
acutely ischaemic myocardium 18–20
adaptation and deadaptation 16–18
tracing metabolic pathways 14–16
myocardial oxygen consumption 347–9
myocardial perfusion, variations in 144–51
coronary vascular resistance 148–51
phasic myocardial perfusion 145–7
resting myocardial perfusion 145
transmural myocardial perfusion 147–8
myocardial protein turnover 10
myocardial remodelling 107
myocardial stiffness 60
myocardium, anaesthetic effects on 342–62
coronary blood flow 359–60
ischaemia–reperfusion injury 360–2
myocardial contractility 343–7
myocardial efficiency 354–9
myocardial oxygen consumption 347–54
neural control mechanisms 214–29

autonomic nervous system 217–22
CNS cardiovascular centres 214–17
reflex control of circulation 222–9
neurohumoral control 132–7
alpha-adrenergic 132–4
autonomic 132
beta-adrenergic 134
cholinergic 134
peptidergic control 135–6
neuropeptide K 135
neuropeptide Y 135
nifedipine 197, 311
nitric oxide 125–8, 129–30, 140–1, 185, 221, 230–2
as bronchodilator 203
and cerebral haemodynamics 267–70
inhibition of endothelin-1 formation 290
and pulmonary circulation 194
nitric oxide synthase 185, 315–16
nitroglycerine
dilatation of coronary arterioles 311
and pulmonary circulation 194, 196
nitrous oxide 324, 335, 346
and myocardial contractility 348
L-NMMA 316–17, 324–7
NMR spectroscopy 15
noradrenaline 214, 220, 283, 290, 299, 335
anaesthetic effects 337
obstructive sleep apnoea 200

oculocardiac reflex 227
opiates 262–3, 286, 292, 301
and heart rate 352
organ blood flow
brain 260–2
heart 124–58, 359–60
local, autoregulation of 311–12
lung 173–6
measurement of 158–62, 318
regional inhomogeneity of 176–7
overdrive suppression 88
oxidative phosphorylation 9–10
oxygen extraction fraction 251
oxygen tension 141–2, 180–2
pacemaker currents 108

pain reflex 227
paired pacing 50–1
parasympathetic nervous system 219–20
parasystole 92
pentobarbitone 307
Page 381
and organ blood flow 321
peptidergic control 135–6
peripheral vascular resistance 40, 41
phasic myocardial perfusion 145–7
phenylephrine 223
photosynthesis 5
platelet activating factor 124
positive end expiratory pressure (PEEP) 179
positron emission tomography 15–16, 161, 207, 248
postganglionic neurons 217
potassium ion channel
ATP-sensitive 131–2, 142
blockers 110–12
preganglionic neurons 217, 218
preload 31–2, 37–8
preload recruitable stroke work 48
pressor reflex 223
pressure autoregulation 234–6
pressure volume index 243, 244
pressure–flow curves 152–4, 188
pressure/volume area 52–3
primary hypoventilation syndrome 200
propofol 262, 295, 301, 333–4
and heart rate 351
propranolol 194
prostacyclin 185
prostaglandins 127–8, 186
renal circulation 284
splanchnic circulation 290
prostanoids 128–9, 186
anatomy and physiology 172–86
effects of drugs on 186–204
fluid balance in lung 204–7
pulmonary embolism 198–9
pulmonary hypertension 197–8
pulmonary oedema 206–7
pulmonary vascular resistance 176–82
and drug effects 187–92
flow distribution methods 192–3
pulmonary vasodilators 201–4
pulmonary vasomotor tone 182
pulmonary venous hypertension 201
pump function graph 34
pump function of heart 33–4
clinical evaluation 34–50
Purkinje fibre automaticity 88
radioactive microsphere technique 158, 318

random re-entry 101–2
rapid filling 58
rate pressure product 348
re–entry of excitation 95–105, 106, 107
anatomical re-entry 98–100
AV re-entry with accessory pathways 103–5
functional re-entry 100–2
SA and AV node re-entry 102–3
slow conduction 96–7
unidirectional conduction block 97–8
reactive hyperaemia 236–7
reciprocation 95
recovery from inactivation 75
rectification 84
reflex circulatory tone 222–9
extrinsic reflexes 227–8
intrinsic reflexes 222–7
spinal shock and autonomic hyperreflexia 228–9
refractoriness 90–1
regulation of cardiovascular system 213–39
humoral control of circulation 229–34
local regulation of circulation 234–7
long term regulation of circulation 237–8
neural control of heart and vasculature 214–29
anatomy 278–81
control of 281–6
effects of anaesthetics on 286–7
renal nerves 283
renin–angiotensin system 229–30, 283
reperfusion 16–18, 360–2
resting membrane potential 78, 80
resting myocardial perfusion 145
right atrial pressure 62
right ventricular autoregulation 144
right ventricular performance 60–4
SA node re-entry 102–3

semilunar valve 30
Page 382
septic shock 308, 313
serotonin 233, 290
sevoflurane 262, 336, 357
single photon emission tomography 248, 251
sinoatrial node 220
sinus node automaticity 86–8
sliding filament model of contraction 8
slow conduction 96–7
slow filling 58
slow response fibre 80
Smyth’s procedure 223
sodium ion channel 109
sodium nitroprusside 223
and pulmonary circulation 194, 196
spinal shock 228–9
spiral wave re-entry 101
anatomy 287–9
control of 289–91
stellate ganglia 221
stress failure 203
stretch-induced contraction 142
stroke volume 30, 32–3, 35, 51–2
anaesthetic effects on 350, 351, 352
stroke work 52–3
subarachnoid haemorrhage 266–7
subsidiary (latent) pacemakers 88
substance P 135
substrate catabolism 9–10
substrate competition 11–13
sufentanil 337, 352, 355, 359
sympathetic nervous system 217–19
systolic interdependence 62
systolic wall stresses 42–3
tension time index 348

thallium-201 scintigraphy 161
thermodilution technique 35, 160–1
thermodynamics, laws of 4
thiopentone 262
thoracic duct obstruction 203
thrombin 186
thromboxanes 185, 186
torsade de pointes 94–5, 111
transcranial Doppler ultrasonography 253–4
transmembrane potential 80, 82, 83
see also action potential
transmural myocardial perfusion 147–8
triggered activity 93–5, 106, 107
delayed after depolarisations 95
early after depolarisations 94–5
tumour necrosis factor 186
unidirectional conduction block 97–8
vagus nerves 219

vascular capacitance 313
vascular endothelium 315–17
vascular resistance 66, 310–13
anaesthetic effects on 321–3, 363–5
vasoactive intestinal peptide 135–6, 290
vasomotor centre 214–15
vasomotor tone 221–2
vasopressin see antidiuretic hormone
venous drainage, cerebral 242, 243
ventilation–perfusion ratio 178, 179
ventricular diastolic performance 57–60
atrial contraction 58–60
heart rate 60
isovolumic relaxation period 57–8
rapid filling 58
slow filling 58
ventricular function curves 53–6
external control mechanisms 55
internal control mechanisms 54–5
ventricular hypertrophy 60
ventricular performance 27–72
age-related changes 65–7
during exercise 64–5
global ventricular pump function 34–50
heart as pump 33–4
heart rate 50–1
isolated ventricle 29–31
right ventricle 60–4
ventricle as muscle 27–9
ventricle in situ 31–3
ventricular diastolic performance 57–60
ventricular function curves 53–6
ventriculoarterial coupling 51–3
ventricular reflexes 226
Page 383
ventricular septal defects 200
ventricular systolic pressure 39–40
ventriculoarterial coupling 43, 45, 51–3
venules 310, 311
vis a tergo 63
voltage gated ion channels 74–5, 76
Wagner inert gas technique 193

watershed infarctions 240
wedge pressure 62
Wolff–Parkinson–White syndrome 97, 103
Woodforth’s negative staircase response 50
xenon-133 wash-out 247, 248, 249
zero flow pressure 153

zona occludens 245

Physiology Fundamentals of Anaesthesia and Acute Medicine

Uploaded by

Copyright:

Available Formats

Physiology Fundamentals of Anaesthesia and Acute Medicine

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Physiology Fundamentals of Anaesthesia and Acute Medicine

Uploaded by

Copyright:

Available Formats

Page i

Fundamentals of Anaesthesia and Acute Medicine

© BMJ Books 2000

BMJ Books is an imprint of the BMJ Publishing Group

Second edition published in 2000

British Library Cataloguing in Publication Data

Typeset in Great Britain by

Ronald M Jones, Professor of Anaesthetics, St Mary’s Hospital Medical School, London,

Titles already available:

Cardiovascular Physiology (second edition)

Clinical Cardiovascular Medicine in Anaesthesia

Intensive Care Medicine

Management of Acute and Chronic Pain

Paediatric Intensive Care

Pharmacology of the Critically Ill

Anaesthesia for Obstetrics and Gynaecology

John Atlee III, MD

James E Baumgardner, PhD, MD

Andreas Hoeft, PhD, MD

Nguyen D Kien, PhD

Alex L Loeb, PhD

David K Menon, MD, FRCP, FRCA

Niraj Nijhawan, MD, MS

Hans-Joachim Priebe, MD, FRCA

Keith Sykes, MB BChir, FRCA

Anne B Taegtmeyer, BM BCh

Heinrich Taegtmeyer, MD, DPhil

David C Warltier, PhD, MD

1: Cardiac cellular physiology and

Principles of energy transfer in heart muscle

Salient features of heart muscle

Thermodynamic aspects of energy transfer in biological systems

Metabolic pathways and moiety conserved cycles

It is a characteristic property of all living cells, including heart muscle, to provide an

z to decrease the proton gradient

A convenient way of viewing energy transfer in heart muscle is as a “three-ring-circus’’,

Nutrition of the heart and myocardial protein turnover

Table 1.1 Metabolite concentrations in human plasma under various conditions

Glucose Lactate Free fatty acids Ketone bodies

Rest 5.0 0.5 0.1 0.1

Reproduced by permission from Taegtmeyer H. Basic Res Cardiol 1984;79:322–36.

and non-essential fuels, which provide only acetyl-CoA:

z fatty acids of all chain lengths

Clinical relevance of myocardial metabolism

Clinical forms of heart disease caused by altered energy metabolism

Tracing metabolic pathways in the intact heart

A detailed knowledge of the pathways of individual substrates for energy production is

nuclei (for example, 13C) leads to an increase in their sensitivity.

Positron emission tomography

Metabolic adaptation and deadaptation: the cellular consequences of

Metabolic support of the acutely ischaemic myocardium

From Gradinak et al.127

Preservation of optimal cardiovascular function represents one of the foremost goals of

characteristics on resting muscle length (preload) is a fundamental property of the

ventricular performance on preload is an expression of the Frank–Starling law of the heart

The heart as a pump

Clinical evaluation of global ventricular pump function

It can be measured non-invasively by radionuclide ventriculography or echocardiography.

Assessment Normal values

Mean arterial pressure 70–105 mm Hg