Cardiorespiratory Interactions: The Relationship Between Mechanical

Ventilation and Hemodynamics

Ira M Cheifetz MD FAARC

Introduction
Mechanical Ventilation and Hemodynamics: An Overview
Cardiorespiratory Economics
Oxygen Delivery
Oxygen Consumption
Effects of Mechanical Ventilation on the Right Ventricle
Effects of Ventilator Manipulations on the Left Ventricle
Effects of Mechanical Ventilation on the Pulmonary Vasculature
Effects of Ventilator Manipulations on Heart Rate
Ventilator Approach for Patients with Congenital Heart Disease
Increased Oxygen Consumption
Oxygen Delivery
Conclusions

The overall goal of the cardiorespiratory system is to provide the organs and tissues of the body with
an adequate supply of oxygen in relation to oxygen consumption. An understanding of the complex
physiologic interactions between the respiratory and cardiac systems is essential to optimal patient
management. Alterations in intrathoracic pressure are transmitted to the heart and lungs and can
dramatically alter cardiovascular performance, with significant differences existing between the
physiologic response of the right and left ventricles to changes in intrathoracic pressure. In terms
of cardiorespiratory interactions, the clinician should titrate the mean airway pressure to optimize
the balance between mean lung volume (ie, arterial oxygenation) and ventricular function (ie, global
cardiac output), minimize pulmonary vascular resistance, and routinely monitor cardiorespiratory
parameters closely. Oxygen delivery to all organs and tissues of the body should be optimized, but
not necessarily maximized. The heart and lungs are, obviously, connected anatomically but also
physiologically in a complex relationship. Key words: cardiorespiratory interactions; cardiac output;
oxygen delivery; oxygenation; acidosis; neonate; pediatric; oxygen consumption; oxygen; mechanical
ventilation; hypoxia; hypoxemia. [Respir Care 2014;59(12):1937–1945. © 2014 Daedalus Enterprises]

Introduction quate supply of oxygen in relation to oxygen consumption

(V̇O2). Generally, this balance is easily maintained in the
The overall goal of the cardiorespiratory system is to healthy person. However, cardiorespiratory dysfunction
provide the organs and tissues of the body with an ade-

Dr Cheifetz is affiliated with the Department of Pediatrics, Children’s ogy in Critically Ill Patients of the AARC Congress 2013, held Novem-
Services, and Pediatric Critical Care Medicine, Duke University Medical ber 16–19, 2013, in Anaheim, California.
Center, Duke Children’s Hospital, Durham, North Carolina.
Correspondence: Ira M Cheifetz MD FAARC, Pediatric Critical Care
Dr Cheifetz has disclosed no conflicts of interest. Medicine, Duke Children’s Hospital, Duke University Medical Center,
Box 3046, Durham, NC 27710. E-mail: ira.cheifetz@duke.edu.
Dr Cheifetz presented a version of this paper at the New Horizons in
Respiratory Care Symposium: Back to the Basics: Respiratory Physiol- DOI: 10.4187/respcare.03486

RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12 1937

 CARDIORESPIRATORY INTERACTIONS

and/or abnormalities of oxygen utilization, as can occur pressure and PEEP) are generally minor and must be bal-
with sepsis, inborn errors of metabolism, and some toxic- anced by the effects of P៮ aw. Alterations in intrathoracic
ities, can lead to the failure of adequate oxygen delivery to pressure are transmitted to the heart and lungs and can
the cells of the body. The result can be metabolic acidosis, dramatically alter cardiovascular performance. As de-
hypoxic-ischemic injury, and ultimately organ dysfunc- scribed in more detail below, significant differences exist
tion/failure. From the perspective of the clinician, an un- between the physiologic response of the right and left
derstanding of the complex physiologic interactions be- ventricles to changes in intrathoracic pressure.3
tween the respiratory and cardiac systems is essential for The discussion that follows focuses on the effects of
optimal patient management. respiratory interventions and their associated effects on
Equally important is an understanding of each patient’s mean intrathoracic pressures on the heart and pulmonary
pathophysiology. An imbalance between oxygen delivery vasculature. The complex interplay between the ventilator
and V̇O2 will be managed differently depending on the and the right and left ventricles and pulmonary vasculature
specific pathophysiology. The following scenarios require is discussed. Adjunct therapies that may affect cardio-
differing management strategies: decreased arterial oxy- respiratory interactions are considered as well.3-6
gen content due to ventilation/perfusion mismatching,
poor cardiac output secondary to ventricular dysfunction, Cardiorespiratory Economics
altered oxygen utilization related to sepsis, decreased car-
diac output secondary to severe hypovolemia, and decreased Oxygen Delivery
arterial oxygen content related to profound anemia. Most
of these conditions can be managed by increasing the ar- Oxygen delivery (DO2) is the product of cardiac output
terial oxygen content and/or cardiac output. However, it and arterial oxygen content (CaO2). Oxygen content is af-
should be noted that septic shock1,2 is a specialized situ- fected by the dissolved and bound components of oxygen
ation characterized by mitochondrial dysfunction, im- in the blood as well as hemoglobin (Hb).
paired oxygen extraction capability, and/or abnormal tis-
sue oxygen utilization. Thus, augmenting cardiac output in DO2 共mL/min兲 ⫽ 10 ⫻ cardiac output 共L/min兲
a patient with septic shock should not be expected to cor-
rect tissue hypoxia, unless there is associated cardiac ⫻ CaO2 共mL O2/100 mL blood) (1)
dysfunction.
As oxygen delivery is a broad topic, the primary focus CaO2 ⫽ 共1.34 mL O2/g Hb) ⫻ Hb (g/100 mL)
of this article is the effects of positive-pressure ventilation
(PPV) on hemodynamics and cardiac output. The use of ⫻ oxygen saturation ⫹ [(0.003 mL/mm Hg) ⫻ PaO2] (2)
inotropes/vasoactive agents, intravascular fluid adminis-
tration, and mechanical support devices to augment car- where 1.34 is the amount of oxygen (mL) carried by 1 g of
diac output is beyond the scope of this paper. hemoglobin, and 0.003 is the solubility of oxygen in plasma.
In critically ill patients, especially those with severe
Mechanical Ventilation and Hemodynamics: lung injury, intrapulmonary shunt and ventilation/perfu-
An Overview sion mismatching can result in profound hypoxemia, which
can compromise arterial oxygen content and cause tissue
From both respiratory and cardiac perspectives, venti- hypoxia, especially if associated with decreased cardiac
lator management for critically ill patients should be aimed output, low hemoglobin concentration, and/or increased
at the specific needs of the individual patient, providing metabolic demand. As oxygen delivery is a function of
the highest benefit with the least risk of complications. cardiac output and arterial oxygen content, cardiac output
The criteria for initiating mechanical ventilation vary ac- augmentation with preload optimization, inotropic agents,
cording to the intended goals and pathophysiology of the vasodilators, optimization of mechanical ventilation, and,
clinical situation. It should be noted from the start that when indicated, mechanical cardiac support devices can
PPV can positively or negatively impact the cardiovascu- maintain adequate oxygen delivery, even if arterial oxygen
lar status, although most often, no overall effect is seen content is diminished. Thus, oxygen delivery may be aug-
due to the body’s ability to compensate for changes in mented by increasing cardiac output, oxygen saturation,
intrathoracic pressure. and/or hemoglobin content. The various components of
It should also be stressed that the key parameter affect- oxygen delivery are displayed in Figure 1.
ing cardiorespiratory interactions is mean airway pressure Accordingly, cardiac output, arterial oxygen content,
(P៮ aw), which directly influences mean intrathoracic pres- and hemoglobin concentration are physiologically inter-
sure. Any cardiovascular effects of the phasic change in dependent, and a decrease in one component may be bal-
airway pressure (ie, difference between peak inspiratory anced by a compensatory increase in another. Therefore,

1938 RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12

 CARDIORESPIRATORY INTERACTIONS

Fig. 2. Effects of mean intrathoracic pressure on systemic vascular

return. Systemic venous return to the right atrium is passive, with
blood flow occurring as a result of a pressure gradient between
the superior/inferior vena cava and right atrium. PSV ⫽ systemic
venous pressure; RAP ⫽ right atrial pressure; PPV ⫽ positive-
Fig. 1. Primary determinants of oxygen delivery. Hb ⫽ hemoglo-
pressure ventilation.
bin; SaO2 ⫽ arterial oxygen saturation.

right atrial pressure increases, resistance to systemic blood

tissue hypoxia may not occur during hypoxemia or anemia return increases, thus reducing venous return and the pre-
if cardiac output is adequately increased. Clinical studies load available to the right ventricle. The increased in-
have demonstrated that cardiac output increases substan- trathoracic pressure created by PPV is transmitted to the
tially in normal subjects during acute hypoxemia.7,8 right atrium, which, if excessive or if the patient is hypo-
volemic, can impede venous return and decrease right ven-
Oxygen Consumption tricular preload. This decrease in right ventricular preload
impairs right ventricular performance by decreasing stroke
In direct relation to the oxygen delivery needed to avoid volume and thus adversely affecting cardiac output.
end-organ injury is V̇O2, which is the product of cardiac During spontaneous breathing (ie, negative intrathoracic
output and the arteriovenous oxygen content difference pressure), right atrial pressure is low, impedance to blood
(C(a-v)O2). flow to the right ventricle is also low, and systemic venous
return is normal. Initiation of PPV (invasive or noninva-
V̇O2 共mL O2/min) ⫽ 10 ⫻ cardiac output (L/min) sive) increases intrathoracic pressure, which is transmitted
to the right heart (ie, increased right atrial pressure) and
⫻ C(a-v)O2 (mL O2/100 mL blood) (3) can result in a decrease in systemic venous blood return
(ie, decreased right heart preload). This effect is most pro-
V̇O2 can be decreased by appropriately managing pa- nounced in situations of significant increases in mean in-
tient-ventilator asynchrony, providing appropriate patient trathoracic pressure and/or decreases in intravascular vol-
work of breathing, and avoiding excessive agitation, shiv- ume (eg, septic, hemorrhagic, or hypovolemic shock). In
ering, and hyperthermia. summary, PPV increases mean intrathoracic pressure and
reduces right ventricular performance by decreasing right
Effects of Mechanical Ventilation on the ventricular preload.
Right Ventricle When increases in intrathoracic pressure are needed to
obtain adequate oxygenation and right ventricular output
Physiologic differences affect the interaction of PPV on is compromised, intravascular volume administration may
the right and left sides of the heart.3-6 The right ventricle be indicated. As shown in Figure 3, intravascular fluid
receives blood from beyond the thorax (ie, from the supe- administration shifts the systemic venous return line up-
rior vena cava and inferior vena cava) and pumps it within ward and rightward. Thus, for the same right atrial pres-
the thorax. As such, the flow of blood to the right ventricle sure (mean intrathoracic pressure), systemic venous return
is sensitive to alterations in mean intrathoracic pressure for is augmented. Although the amount of fluid required to
several physiologic reasons.3 obtain this physiologic benefit is based on the patient’s
Systemic venous return to the right atrium is passive, intravascular status and the magnitude of the increase in
with flow occurring as a result of a pressure gradient be- P៮ aw (mean intrathoracic pressure), in general, ⬃5 mL/kg is
tween the superior/inferior vena cava and right atrium. required. Subsequent fluid boluses should be administered
When the right atrial pressure is low, there is minimal as clinically indicated based on the pathophysiology and
resistance for venous return to the right atrium (Fig. 2). As the individual patient’s physiologic response.

RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12 1939

 CARDIORESPIRATORY INTERACTIONS

tricular output and global cardiac output are affected. These

physiologic principles are discussed in more detail in the
corresponding section below.

Effects of Ventilator Manipulations on the

Left Ventricle

In contrast to the right ventricle, the left ventricle re-

ceives blood from within the thorax (ie, right ventricle via
the pulmonary circulation) and pumps it outside the tho-
rax.3 Thus, the physiologic effects of mechanical ventila-
tion on the left ventricle are quite different from those on
the right ventricle as described previously. Ventricular in-
terdependence dictates that the left ventricle can only eject
Fig. 3. Effects of preload augmentation on systemic vascular re-
turn. Intravascular fluid administration shifts the systemic venous the quantity of blood it receives from the right ventricle.11
return line upward and rightward. Thus, for the same right atrial When right ventricular output is decreased (in relation to
pressure (mean intrathoracic pressure), systemic venous return is the effects of PPV or any other etiology), the blood flow
augmented. PSV ⫽ systemic venous pressure. to the left side of the heart (ie, left ventricular preload)
decreases. Furthermore, right ventricular afterload, and
Although changes in intrathoracic pressure do not di- thus right ventricular systolic pressure, can increase with
rectly affect myocardial contractility, indirect effects can elevation of the P៮ aw.4 An increase in right ventricular pres-
occur. Myocardial perfusion (ie, blood flow) is dependent sure can cause a conformational change in the inter-
on intrathoracic pressure, aortic pressure, and right ven- ventricular septum and a subsequent decrease in left ven-
tricular systolic pressure. An increase in intrathoracic pres- tricular compliance, preload, and potentially output.
sure, increase in right ventricular systolic pressure, and/or Historically, a ventilator approach used to improve left
reduction in aortic pressure will reduce right ventricular ventricular preload was thoracic augmentation of left ven-
myocardial blood flow. In most clinical situations, aortic tricular filling.3,5 This concept describes a phasic increase
pressure exceeds right ventricular and intrathoracic pres- in intrathoracic pressure similar to a proposed mechanism
sures, and right ventricular myocardial blood flow is not for increasing cardiac output with chest compressions
affected by PPV. When low aortic pressure exists, mod- during cardiopulmonary resuscitation. As the left ventricle
erate-to-severe right ventricular dysfunction occurs, and/or derives its preload from intrathoracic sources, the phasic
increased intrathoracic pressure is present, these complex increase in intrathoracic pressure is transmitted to the pul-
interactions can be clinically important. When these pre- monary vascular capacitance, resulting in an increase in
disposing conditions are present, the adequacy of right pulmonary vascular pressure above the left-sided filling
ventricular myocardial perfusion should be addressed, and pressure. As intrathoracic pressure increases, the pressure
interventions (eg, reducing mean intrathoracic pressure, in the pulmonary vasculature further exceeds left atrial
if possible, and/or increasing systemic systolic pressure) pressure, creating a pressure gradient that augments left-
considered to optimize right ventricular myocardial per- sided preload and cardiac output. Thus, left ventricular
fusion and thus right ventricular function. preload can be augmented by this thoracic augmentation
In addition to the primary effect of PPV on right ven- physiology. However, this traditional strategy is no longer
tricular preload and the potential secondary effect on right employed, as the tidal volume (VT) required to achieve a
ventricular contractility, the effects of mechanical ventila- clinically pertinent increase in cardiac output can approach
tion on pulmonary vascular resistance (PVR) and right 15–20 mL/kg, which violates the clinical approach of low
ventricular afterload must be considered.9,10 At low lung VT for lung-protective ventilation.
volume, PVR is elevated due to hypoxic vasoconstriction Although left ventricular contractility is unaffected
and tortuosity of the medium and large intrapulmonary by mechanical ventilation, left ventricular output can
blood vessels. As lung volume increases toward functional be dramatically altered by mean intrathoracic pressure-
residual capacity, PVR decreases as the larger blood ves- dependent effects on left ventricular afterload. Left ven-
sels expand and become less tortuous. At a point signifi- tricular afterload is dependent on left ventricular myocar-
cantly above functional residual capacity, increases in PVR dial wall tension,5 which is affected by the difference
occur due to compression of the intra-alveolar capillaries between the left ventricular systolic pressure and mean
by overexpanded alveoli. As intrathoracic pressure and intrathoracic pressure.5,6 The physiologic effects of PPV
lung volume are altered, variable changes in PVR and on left ventricular wall tension are complex, as the sys-
right ventricular afterload occur. Subsequently, right ven- temic arterial system has both intrathoracic and extratho-

1940 RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12

 CARDIORESPIRATORY INTERACTIONS

diomyopathy, myocarditis). These physiologic principles

should also be considered when deciding on the timing of
extubation in such a patient. As a secondary effect, a pa-
tient with cardiovascular compromise may benefit from a
reduction in V̇O2 by reducing his/her work of breathing. It
should be noted that the clinician must carefully weigh the
potential risks and benefits of this high-risk intubation for
each patient based on the individual physiology and patho-
physiology. Similarly, extubation of the patient with left
ventricular dysfunction/failure may lead to decompensa-
tion as left ventricular afterload and the stress on the left
heart is increased. In summary, the complex interaction
between intrathoracic pressure, myocardial wall tension,
and aortic autoregulation leads to decreased left ventricu-
Fig. 4. Effects of intrathoracic pressure on left ventricular (LV)
lar afterload and further augmentation of cardiac output in
afterload. An augmentation of mean intrathoracic pressure will
reduce left ventricular afterload. PPV ⫽ positive-pressure ventila- the setting of persistently elevated mean intrathoracic pres-
tion; AO ⫽ aorta. sure, as occurs with PPV.
It should be noted that under common clinical condi-
tions (ie, normal ventricular function), mean intrathoracic
racic components. When mean intrathoracic pressure is pressure is low compared with left ventricular pressure,
increased, the pressure is rapidly transmitted to the in- and inspiration occurs over a very limited number of car-
trathoracic arterial system. Left ventricular wall tension diac cycles. The result is only minor phasic changes in left
remains the same, as the left ventricular pressure and in- ventricular afterload, and thus, autoregulation may not oc-
trathoracic pressure generated are equally affected. The cur to a clinically relevant degree. However, if mean in-
extrathoracic arterial system also senses this increase in trathoracic pressure is high, a sustained increase in peak
arterial pressure due to the propagation of the pressure intrathoracic pressure occurs over multiple cardiac cycles,
waveform into the peripheral circulation. and/or the left ventricle is dysfunctional, left ventricular
When an increase in mean intrathoracic pressure results afterload and left ventricular output can be significantly
in a significant increase in arterial pressure, aortic pressure reduced.
will autoregulate secondary to baroreceptor stimulation.5,6
A reflexive decrease in aortic pressure occurs along with a
Effects of Mechanical Ventilation on the
compensatory reduction in left ventricular pressure. When
Pulmonary Vasculature
aortic pressure reflexively returns to baseline, left ven-
tricular systolic pressure decreases. as does the transmyo-
cardial pressure gradient. Therefore, a persistent elevation As briefly discussed previously, right ventricular after-
in mean intrathoracic pressure decreases left ventricular load is affected by lung volume (Fig. 5). When lung vol-
wall tension (ie, decreased left ventricular afterload) as a ume is low (ie, atelectasis, lung collapse), PVR is elevated
result of aortic pressure autoregulation. These physiologic secondary to hypoxic pulmonary vasoconstriction or the
principles are shown schematically in Figure 4. On the tortuous course of the large-to-medium-size pulmonary
left side of this example, the left ventricle of a spontane- blood vessels (Fig. 5A).9,10 As lung volume increases, these
ously breathing patient needs to generate a transmural pres- larger pulmonary vessels become linear with their capac-
sure of 120 mm Hg in response to a systemic systolic itance increasing, hypoxia improves, and PVR decreases.
pressure of 100 mm Hg and a mean intrathoracic pressure As lung volume continues to increase, hyperexpansion of
of ⫺20 mm Hg. When this patient is mechanically ven- the alveoli and compression of the pulmonary capillaries
tilated with a mean intrathoracic pressure of ⫹20 mm Hg, can occur, and vascular resistance increases (Fig. 5B).12
the left ventricle has to generate a pressure of only The total PVR becomes the sum of these two scenarios,
80 mm Hg to result in the same systemic systolic pressure and thus, PVR can be elevated at small or large lung
of 100 mm Hg. Thus, an augmentation of mean intratho- volumes and is lowest at an ideal lung volume. Of course,
racic pressure will reduce left ventricular afterload and the clinical challenge is determining this optimal point
potentially improve overall left ventricular function. (Fig. 5C).
The impact of PPV and its associated beneficial effect Thus, PPV can reduce right ventricular afterload in
on the left side of the heart can be an important consider- patients with low lung volume by expanding collapsed
ation in the decision to initiate mechanical ventilation in lung units and reducing PVR. PPV can increase right ven-
the setting of impaired myocardial performance (eg, car- tricular afterload in conditions of pulmonary overdisten-

RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12 1941

 CARDIORESPIRATORY INTERACTIONS

reduce right ventricular afterload. In a classic report by

Lyrene et al,19 the effects of alkalosis on lowering PVR
were similar regardless of whether the alkalosis was met-
abolic or respiratory in etiology. The use of respiratory
and/or metabolic alkalosis must be directed by the pa-
tient’s overall clinical status and the relative risks and
benefits of each approach.
In another classic study, Malik and Kidd16 demonstrated
that at a buffered pH (7.4), decreases in PaCO2 did not
affect PVR, whereas increases in PaCO2 increased PVR.
Thus, for those patients with pulmonary hypertension, hy-
perventilation without a related increase in pH will not
have a clinically beneficial effect. However, for those pa-
tients in whom an increase in PVR is desired (eg, mechan-
ically ventilated infants with single-ventricle physiology
and pulmonary overcirculation), an increase in PaCO2, ir-
relevant of a decrease in pH, will be effective.
Additionally, studies have shown that an increase in
alveolar oxygen and arterial oxygen (PaO2) by increasing
FIO2 can reduce PVR.15,16 Increasing the FIO2 improves
PaO2 in patients without a right-to-left shunt and reduces
pulmonary artery vascular resistance. Increasing inspired
oxygen in patients with intracardiac shunts produces little
change in PaO2; however, a reduction in PVR still results.
Thus, an increase in alveolar and/or arterial oxygen con-
tent can reduce PVR.
In animal studies, increasing FIO2 has been shown to be
a more potent pulmonary vasodilator in neonates than in
adults.16 The use of inspired oxygen to reduce PVR has
been useful in the ICU and is a common approach to
assessing the reactivity of the pulmonary vascular bed in
the cardiac catheterization laboratory.16
In conditions of increased afterload, a reduction in PVR
will decrease right ventricular afterload and lead to an
improvement of right ventricular function. These benefi-
cial cardiorespiratory interactions may be employed, es-
pecially after congenital heart surgery and in those with
pulmonary hypertension, to assist patients with right ven-
tricular dysfunction. Beyond the use of oxygen, inhaled
Fig. 5. Effects of mean lung volume on pulmonary vascular resis-
nitric oxide and other pulmonary vasodilators may be
tance (PVR). Right ventricular afterload is affected by mean lung employed.
volume. A: Component of PVR related to the larger pulmonary It should be noted that a reduction in PVR is not always
vessels. B: Component of PVR related to the smaller pulmonary beneficial, as can be seen in patients with large intra-
vessels. C: Total PVR. cardiac or extracardiac shunts. The addition of inspired
oxygen can reduce PVR and increase pulmonary blood
tion with excessive alveolar expansion and subsequent flow. In conditions of decreased pulmonary flow, this
compression of the perialveolar capillaries.12 may improve oxygen delivery by increasing arterial oxy-
In addition to optimizing lung volume, therapy directed gen content. However, in clinical situations of increased
at reducing pulmonary hypertension can consist of increas- pulmonary flow (eg, ventricular septal defect, large patent
ing pH, decreasing PaCO2, increasing PaO2 and alveolar ductus arteriosus, and hypoplastic left heart syndrome),
partial pressure of oxygen, and minimizing intrathoracic the associated increase in pulmonary blood flow can
pressures.13-16 Increasing pH has been shown to signifi- result in pulmonary overcirculation and pulmonary edema.
cantly reduce PVR in a variety of studies.15,17,18 Both an For those patients with single-ventricle physiology, the
increase in pH and a reduction in PaCO2 can independently resultant pulmonary overcirculation may occur at the ex-

1942 RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12

 CARDIORESPIRATORY INTERACTIONS

pense of systemic output with a reduction in global oxygen ate patient work of breathing, and avoiding excessive patient
delivery. agitation, shivering, and hyperthermia.20,21
In summary, PVR (and right ventricular afterload) can Patient-ventilator dyssynchrony22-24 can lead to exces-
be minimized when lung volume is optimal, pH is nor- sive V̇O2 by the respiratory muscles. Dyssynchrony can be
mal or increased, and PaCO2 is not increased. Additionally, flow-related or trigger-related and occurs when spontane-
pulmonary vasodilators (eg, inhaled nitric oxide, prosta- ous inspiratory effort is out of phase with the ventilator-
cyclin), whether inhaled or administered intravenously, can delivered breaths.22-24 When dyssynchrony is present, pri-
help to reduce PVR and potentially optimize cardiorespi- mary hypoxemia due to ventilation/perfusion mismatching,
ratory interactions. A detailed discussion of these pharma- mucous plugging, pneumothorax, and reactive airway dis-
cologic agents is beyond the scope of this article. ease must be eliminated as the etiology. When these causes
are eliminated, altering the mode (ie, inspiratory flow pat-
Effects of Ventilator Manipulations on Heart Rate tern), improving the trigger sensitivity, or increasing the
support provided by the ventilator may improve patient-
PPV generally causes minor changes in heart rate. Over- ventilator synchrony. Improving patient-ventilator syn-
distention of the lung can result in a reflex bradycardia; chrony can reduce V̇O2, especially for infants and small
however, changes in heart rate tend to be limited at the VT children.
commonly used in clinical practice. However, ventilation
with excessive VT can result in a reflex bradycardia that Oxygen Delivery
may become clinically important. Profound hypoxia with
depressed oxygen delivery can result in bradycardia as
well. Overall, the effects of PPV on heart rate tend to be Although an important clinical goal is to maintain an
less pertinent than the other physiologic changes discussed optimal balance between oxygen delivery and V̇O2 for the
throughout this article. critically ill patient, the clinician is often in a situation in
which this balance cannot be quantified. Pulmonary artery/
cardiac output catheters are less commonly used than in
Ventilator Approach for Patients With
the past, especially for infants and children. Thus, the
Congenital Heart Disease
clinician is often left with surrogate end points.
Poor oxygen delivery (or a mismatch between delivery
Patients with congenital heart lesions represent a spe- and consumption) can be seen clinically by decreased re-
cialized population. The principles of cardiorespiratory nal function, abnormal mental status, and poor right ven-
interactions apply to these patients in a similar fashion as tricular function. It must be noted that these clinical as-
the general population; however, the effects tend to be sessments can be altered by the use of diuretics and
more pronounced. Furthermore, the clinician should care- pharmacologic sedation. Additionally, right heart dysfunc-
fully consider the physiologic effects of mechanical ven- tion as a result of poor oxygen delivery is a late marker of
tilation on the systemic and pulmonary ventricle(s) rather a problem.
than on the more traditional right- and left-sided princi- Beyond clinical assessment, the clinician can investi-
ples. In those patients with single-ventricle physiology, the gate an imbalance between oxygen delivery and V̇O2 by
ventricle can have systemic and pulmonary physiologic evaluating laboratory markers of metabolic acidosis as an
components, which will likely differ in the preoperative indicator of global oxygen debt. Such studies to confirm
state compared with the palliated postoperative state. Be- that tissue oxygen supply is adequately maintained include
cause of the complex cardiorespiratory interactions that mixed venous oxygen saturation (Sv៮ O2), blood lactate, base
occur and the diversity of the conditions treated, a single deficit, and arterial pH.
standardized approach is not possible. Respiratory strate- Sv៮ O2 can be measured intermittently via repeated blood
gies should be designed to address the specific pathophys- gas analysis or continuously via a fiberoptic catheter.18,25
iologic condition present in each patient. Sv៮ O2 is probably the best single indicator of the adequacy
of oxygen transport, as it represents the amount of oxygen
Increased Oxygen Consumption remaining in the systemic venous blood after blood passes
through the organs and tissues throughout the body.26 Sv៮ O2
Although the clinician has multiple approaches for op- reflects the balance between oxygen supply and demand
timizing oxygen delivery, the options for minimizing/op- and can be a surrogate for cardiac output as a target for
timizing V̇O2 are more limited. In the situation in which goal-oriented hemodynamic therapy.27 It should be noted
V̇O2 is excessive in relation to oxygen delivery, the clini- that the oxygen extraction ratio may be preferable to Sv៮ O2
cian should consider treating patient-ventilator dyssyn- as an indicator of global tissue hypoxia because arterial
chrony, titrating ventilator support to provide an appropri- hypoxemia reduces Sv៮ O2 without necessarily indicating

RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12 1943

 CARDIORESPIRATORY INTERACTIONS

oxygen debt (ie, the oxygen extraction ratio can be normal,

whereas arterial oxygen saturation [SaO2] and Sv៮ O2 are pro-
portionally reduced).
It must be stressed that all markers of oxygen delivery
are global indicators. Regional areas of tissue/organ hyp-
oxia may exist. Specific examples include sepsis, necro-
tizing enterocolitis in the neonate, acute kidney injury in
the critically ill patient with shock, and intracranial injury.
Alterations of microcirculatory blood flow reduce the
sensitivity of global markers of tissue hypoxia to detect
regional abnormalities of tissue oxygenation.26,28 In sepsis,
tissue hypoxia is largely due to impaired oxygen utiliza-
tion, although Sv៮ O2 may be normal or high.29,30
Noninvasive monitoring of cerebral and splanchnic
oxygen saturation is being increasingly advocated as a
reliable assessment of oxygen delivery to the vital organs
of the body.31,32 However, it must be noted that the clinical
reliability of the use of near-infrared spectroscopy to mon- Fig. 6. Effects of oxygen delivery on tissue oxygenation. The dotted
vertical line represents the anaerobic threshold.
itor oxygen delivery remains controversial.33
Tissue oxygen delivery is determined not only by SaO2
but also by the patient’s hemoglobin content, cardiac out-
Conclusions
put, oxygen affinity of hemoglobin, oxygen extraction
ratio, and metabolic demands. SaO2 is not a sensitive index
of tissue oxygenation/oxygen delivery. It must be kept in Tissue and organ oxygen delivery, but not necessarily
PaO2 and SaO2, must be optimized in relation to V̇O2. This
mind that, to date, no randomized trial has assessed the
fine balance varies between patients based on the individ-
relationship between SaO2 and tissue oxygenation in criti-
ual’s pathophysiology, as well as within a given patient
cally ill patients.
over time as his/her clinical condition changes. The bot-
As the clinical goal is to provide the tissues and organs
tom line is that oxygen delivery to all organs and tissues
of the body with an appropriate supply of oxygen, one
of the body must be optimized (but not necessarily max-
must be aware that providing supranormal oxygen deliv- imized). In terms of cardiorespiratory interactions, the
ery is not necessarily beneficial. Randomized controlled clinician should titrate the P៮ aw to optimize the balance
trials have demonstrated that maintaining supranormal ox- between mean lung volume (ie, arterial oxygenation)
ygen delivery does not improve survival and may be det- and ventricular function (ie, global cardiac output), mini-
rimental.27,34-36 However, other clinical trials have reported mize PVR, and routinely monitor cardiorespiratory param-
favorable outcomes with supranormal oxygen delivery.37-39 eters closely. The heart and lungs are, obviously, con-
An optimal approach may be early goal-directed therapy, nected anatomically but also physiologically in a complex
which has been shown to reduce mortality for patients relationship.
with severe sepsis and septic shock.40 This approach in-
volves adjustments of cardiac preload, afterload, and con- REFERENCES
tractility to balance oxygen delivery with V̇O2. The phys-
iologic rationale for an oxygen delivery early goal-directed 1. Mtaweh H, Trakas EV, Su E, Carcillo JA, Aneja RK. Advances in
monitoring and management of shock. Pediatr Clin North Am 2013;
therapy approach is to augment systemic oxygen delivery 60(3):641-654.
to alleviate tissue hypoxia, which can progress to multi- 2. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal
organ dysfunction/failure. SM, et al. Surviving Sepsis Campaign: international guidelines for
Despite the conflicting data, a common clinical ap- management of severe sepsis and septic shock, 2012. Intensive Care
Med 2013;39(2):165-228.
proach is to augment oxygen delivery when the patient is 3. Robotham JL, Lixfeld W, Holland L, MacGregor D, Bromberger-
on the oxygen-dependent portion of the tissue oxygenation- Barnea B, Permutt S, Rabson JL. The effects of positive end-
oxygen delivery portion of the curve (Fig. 6), but not expiratory pressure on right and left ventricular performance. Am
augment oxygen delivery once the patient is on the flat Rev Respir Dis 1980;121(4):677-683.
4. Pinsky MR. Determinants of pulmonary arterial flow variation dur-
portion of the curve (ie, the point between which tissue ing respiration. J Appl Physiol 1984;56(5):1237-1245.
oxygenation is not oxygen delivery-dependent). The el- 5. Pinsky MR, Summer WR. Cardiac augmentation by phasic high
bow of this curve is often noted as the anaerobic threshold. intrathoracic pressure support in man. Chest 1983;84(4):370-375.

1944 RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12

 CARDIORESPIRATORY INTERACTIONS

6. Pinsky MR, Summer WR, Wise RA, Permutt S, Bromberger-Barnea 24. Branson RD, Blakeman TC, Robinson BR. Asynchrony and dys-
B. Augmentation of cardiac function by elevation of intrathoracic pnea. Respir Care 2013;58(6):973-989.
pressure. J Appl Physiol 1983;54(4):950-955. 25. Reinhart K, Rudolph T, Bredle DL, Hannemann L, Cain SM. Com-
7. Phillips BA, McConnell JW, Smith MD. The effects of hypoxemia parison of central-venous to mixed-venous oxygen saturation during
on cardiac output. A dose-response curve. Chest 1988;93(3):471-475. changes in oxygen supply/demand. Chest 1989;95(6):1216-1221.
8. Cargill RI, Kiely DG, Lipworth BJ. Left ventricular systolic perfor- 26. Jakob SM, Takala J. ARDS. Monitoring tissue perfusion. Crit Care
mance during acute hypoxemia. Chest 1995;108(4):899-902. Clin 2002;18(1):143-163.
9. Benumof JL. Mechanism of decreased blood flow to atelectatic lung. 27. Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Pesenti A,
J Appl Physiol 1979;46(6):1047-1048. Fumagalli R. A trial of goal-oriented hemodynamic therapy in crit-
10. Benumof JL, Rogers SN, Moyce PR, Berryhill RE, Wahrenbrock ically ill patients. N Engl J Med 1995;333(16):1025-1032.
EA, Saidman LJ. Hypoxic pulmonary vasoconstriction and re- 28. Ruokonen E, Takala J, Kari A, Saxén H, Mertsola J, Hansen EJ.
gional and whole-lung PEEP in the dog. Anesthesiology 1979;51(6): Regional blood flow and oxygen transport in septic shock. Crit Care
503-507. Med 1993;21(9):1296-1303.
11. Pinsky MR. Heart-lung interactions. Curr Opin Crit Care 2007;13(5): 29. Leach RM, Treacher DF. ABC of oxygen: oxygen transport-2.
528-531. Tissue hypoxia. BMJ 1998;317(7169):1370-1373.
12. Cheifetz IM, Craig DM, Quick G, McGovern JJ, Cannon ML, 30. Société de Réanimation de Langue Française, The American Tho-
Ungerleider RM, et al. Increasing tidal volumes and pulmonary racic Society, European Society of Intensive Care Medicine. Third
overdistention adversely affect pulmonary vascular mechanics and European Consensus Conference in Intensive Care Medicine. Tissue
cardiac output in a pediatric swine model. Crit Care Med 1998;26(4): hypoxia: how to detect, how to correct, how to prevent. Am J Respir
710-716. Crit Care Med 1996;154(5):1573-1578.
13. Custer JR, Hales CA. Influence of alveolar oxygen on pulmonary 31. Ostadal P, Kruger A, Vondrakova D, Janotka M, Psotova H, Neuzil
vasoconstriction in newborn lambs versus sheep. Am Rev Respir Dis P. Noninvasive assessment of hemodynamic variables using near-
1985;132(2):326-331. infrared spectroscopy in patients experiencing cardiogenic shock and
14. Drummond WH, Gregory GA, Heymann MA, Phibbs RA. The in- individuals undergoing venoarterial extracorporeal membrane oxy-
dependent effects of hyperventilation, tolazoline, and dopamine on genation. J Crit Care 2014;29(4):690.e11-690.e15.
infants with persistent pulmonary hypertension. J Pediatr 1981;98(4): 32. Marimón GA, Dockery WK, Sheridan MJ, Agarwal S. Near-infrared
603-611. spectroscopy cerebral and somatic (renal) oxygen saturation corre-
15. Drummond WH, Lock JE. Neonatal ‘pulmonary vasodilator’ drugs. lation to continuous venous oxygen saturation via intravenous oxi-
Current status. Dev Pharmacol Ther 1984;7(1):1-20. metry catheter. J Crit Care 2012;27(3):314.e13-314.e1318.
16. Malik AB, Kidd BS. Independent effects of changes in H⫹ and CO2 33. Nolan JP. Cerebral oximetry during cardiac arrest-feasible, but ben-
concentrations on hypoxic pulmonary vasoconstriction. J Appl efit yet to be determined. Crit Care Med 2014;42(4):1001-1002.
Physiol 1973;34(3):318-323. 34. Hayes MA, Timmins AC, Yau EH, Palazzo M, Hinds CJ, Watson D.
17. Koehler RC, Chandra N, Guerci AD, Tsitlik J, Traystman RJ, Rogers Elevation of systemic oxygen delivery in the treatment of critically
MC, Weisfeldt ML. Augmentation of cerebral perfusion by simul- ill patients. N Engl J Med 1994;330(24):1717-1722.
taneous chest compression and lung inflation with abdominal bind- 35. Heyland DK, Cook DJ, King D, Kernerman P, Brun-Buisson C.
ing after cardiac arrest in dogs. Circulation 1983;67(2):266-275. Maximizing oxygen delivery in critically ill patients: a methodologic
18. Cavaliere F, Zamparelli R, Martinelli L, Scapigliati A, De Paulis S, appraisal of the evidence. Crit Care Med 1996;24(3):517-524.
Caricato A, et al. Blood from the right atrium may provide closer 36. Alía I, Esteban A, Gordo F, Lorente JA, Diaz C, Rodriguez JA,
estimates of mixed venous saturation than blood from the superior Frutos F. A randomized and controlled trial of the effect of treatment
vena cava. A pilot study. Minerva Anestesiol 2014;80(1):11-18. aimed at maximizing oxygen delivery in patients with severe sepsis
19. Lyrene RK, Welch KA, Godoy G, Philips JB 3rd. Alkalosis atten- or septic shock. Chest 1999;115(2):453-461.
uates hypoxic pulmonary vasoconstriction in neonatal lambs. Pediatr 37. Lobo SM, Salgado PF, Castillo VG, Borim AA, Polachini CA,
Res 1985;19(12):1268-1271. Palchetti JC, et al. Effects of maximizing oxygen delivery on mor-
20. Manthous CA, Hall JB, Olson D, Singh M, Chatila W, Pohlman A, bidity and mortality in high-risk surgical patients. Crit Care Med
et al. Effect of cooling on oxygen consumption in febrile critically ill 2000;28(10):3396-3404.
patients. Am J Respir Crit Care Med 1995;151(1):10-14. 38. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of
21. Leach RM, Treacher DF. The pulmonary physician in critical care 2: the effect of deliberate perioperative increase of oxygen delivery
oxygen delivery and consumption in the critically ill. Thorax 2002; on mortality in high-risk surgical patients. JAMA 1993;270(22):
57(2):170-177. 2699-2707.
22. Gilstrap D, MacIntyre N. Patient-ventilator interactions. Implica- 39. Wilson J, Woods I, Fawcett J, Whall R, Dibb W, Morris C,
tions for clinical management. Am J Respir Crit Care Med 2013; McManus E. Reducing the risk of major elective surgery: random-
188(9):1058-1068. ized controlled trial of preoperative optimization of oxygen delivery.
23. Vignaux L, Grazioli S, Piquilloud L, Bochaton N, Karam O, Jaecklin BMJ 1999;318(7191):1099-1103.
T, et al. Optimizing patient-ventilator synchrony during invasive 40. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B,
ventilator assist in children and infants remains a difficult task. Pediatr et al. Early goal-directed therapy in the treatment of severe sepsis
Crit Care Med 2013;14(7):e316-325. and septic shock. N Engl J Med 2001;345(19):1368-1377.

RESPIRATORY CARE • DECEMBER 2014 VOL 59 NO 12 1945