Orthopaedic Biomechanics Made

Easy
 Orthopaedic Biomechanics Made
Easy

Sheraz S. Malik, MRCS, MSc

trust-grade registrar in orthopaedics, Newham
University Hospital, Barts Healthcare NHS Trust, UK
and
Shahbaz S. Malik, MRCS, MSc
specialty registrar in orthopaedics, West Midlands
Deanery, UK
University Printing House, Cambridge CB2 8BS, United Kingdom

Cambridge University Press is part of the University of Cambridge.

It furthers the University's mission by disseminating knowledge in the pursuit
of education, learning and research at the highest international levels of
excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107685468
© Sheraz S. Malik and Shahbaz S. Malik 2015
This publication is in copyright. Subject to statutory exception and to the
provisions of relevant collective licensing agreements, no reproduction of
any part may take place without the written permission of Cambridge
University Press.

First published 2015

Printed in the United Kingdom by TJ International Ltd. Padstow Cornwall
A catalogue record for this publication is available from the British
Library
Library of Congress Cataloguing in Publication data
Malik, Sheraz S., author.
Orthopaedic biomechanics made easy / Sheraz S. Malik, Shahbaz S. Malik.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-107-68546-8 (Paperback)
I. Malik, Shahbaz S., author. II. Title.
[DNLM: 1. Musculoskeletal Physiological Phenomena.
2. Orthopedics–physiology. 3. Biomechanical Phenomena. WE 102]
QP303
612.7′6–dc23 2014021240
ISBN 978-1-107-68546-8 Paperback
Cambridge University Press has no responsibility for the persistence or
accuracy of URLs for external or third-party internet websites referred to in
this publication, and does not guarantee that any content on such websites is,
or will remain, accurate or appropriate.

Every effort has been made in preparing this book to provide accurate and
up-to-date information which is in accord with accepted standards and
practice at the time of publication. Although case histories are drawn from
actual cases, every effort has been made to disguise the identities of the
individuals involved. Nevertheless, the authors, editors and publishers can
make no warranties that the information contained herein is totally free from
error, not least because clinical standards are constantly changing through
research and regulation. The authors, editors and publishers therefore
disclaim all liability for direct or consequential damages resulting from the
use of material contained in this book. Readers are strongly advised to pay
careful attention to information provided by the manufacturer of any drugs or
equipment that they plan to use.
To our parents Muhammad S. Malik and Shahnaz Akhtar for their
prayers and blessings, and showing us the value of education
whether being taught or teaching others, and to our brother,
Shahzad S. Malik, for being there for us.
Sheraz S. Malik

Also to my wife Nadia, for her endless patience.

Shahbaz S. Malik
 Contents
Contributors
Preface
Acknowledgements

Part I: Orthopaedic biomaterials and their properties

1. Introduction to orthopaedic biomechanics
Introduction to orthopaedic biomechanics
Force
Moment of a force
Static analysis
Static analysis applied to the musculoskeletal system
Simple machine
Simple machines in the musculoskeletal system
Stress and strain
Stress–strain curve
Mechanical properties
Viscoelastic properties of materials
2. Orthopaedic biomaterials and their properties
Structure and properties of materials
Metals
Alloys
Metals in orthopaedics
Ceramics
Ceramics in orthopaedics
Polymers
Polymers in orthopaedics
Composites
Composites in orthopaedics
Bone I
Bone II
Part II: Engineering theory applied to orthopaedics
3. Modes of loading in the musculoskeletal system
Introduction
Compression and tension
Bending I
Bending II
Torsion
Material and geometric properties of long bones
4. Biomechanics of fracture
Fundamentals of fracture
Mechanism of bone fracture
Patterns of bone fractures I
Patterns of bone fractures II
Patterns of bone fractures III
Stress raisers
Corrosion
Biological process of bone fracture healing
Biomechanical process of bone fracture healing
5. Biotribology
Introduction to biotribology
Friction
Wear
Lubrication I
Lubrication II
Lubrication of synovial joints
Lubrication of prosthetic joints

Part III: Clinical biomechanics

6. Biomechanics of the hip and total hip replacement
Axes of the lower limbs
Hip joint reaction force I
Hip joint reaction force II
Total hip replacement: cemented fixation
Total hip replacement: cementless fixation
Total hip replacement: design and alignment of components
Total hip replacement: femoral head size
Total hip replacement: bearing surfaces
7. Biomechanics of the knee and total knee replacement
Knee: functional anatomy I
Knee: functional anatomy II
Knee: flexion–extension arc
Knee joint reaction force
Total knee replacement: design of components I
Total knee replacement: design of components II
Total knee replacement: alignment of components
8. Biomechanics of the shoulder
Shoulder: functional anatomy I
Shoulder: functional anatomy II
Shoulder joint reaction force
Shoulder replacement
Reversed shoulder replacement
9. Biomechanics of the elbow
Elbow: functional anatomy
Stabilisers of the elbow
Elbow joint reaction force
10. Biomechanics of the spine
Biomechanics of spinal components I
Biomechanics of spinal components II
Stability of the spinal column
The loads acting on the spinal column
11. Biomechanics of the ankle and foot
The ankle joint
Total ankle replacement
Three rockers of normal gait
The foot
12. Biomechanics of fracture fixation
Introduction to fracture fixation
Cast
Wires
Surgical screw I
Surgical screw II
Plate fixation I
Plate fixation II
Intramedullary nail I
 Intramedullary nail II
External fixation I
External fixation II
External fixation III
13. Trauma meeting: case-based discussions
Mechanics of trauma meeting
Fractures of the middle third of clavicle
Fractures of the proximal humerus
Fractures of the mid-shaft of humerus
Intra-articular fractures of distal humerus
Distal radius fractures
Intertrochanteric fractures of the proximal femur
Fractures of distal third of tibia
Pilon fractures
Ankle fractures

Index
 Contributors
Usman Ahmed MRCS, PhD
Specialty Registrar in Trauma & Orthopaedics
West Midlands Deanery
Bola Akinola MRCS, MSc, FRCS (Tr & Orth)
Specialty Registrar in Trauma & Orthopaedics
East of England Deanery
Chee Gan FRCR
Interventional Neuroradiology Fellow
The National Hospital of Neurology and Neurosurgery
Simon MacLean MRCS(Ed) FRCS (Tr & Orth)
Specialty Registrar in Trauma & Orthopaedics
West Midlands Deanery
Ravi Shenoy MRCS(Ed), MS(Orth), DNB(Orth), MD
Specialty Registrar in Trauma & Orthopaedics
Northeast (Stanmore) Rotation, London Deanery
Pritam Tharmarajah MRSC(Ed), MD
Specialty Registrar in General Practice
East Midlands Deanery
Shaheryar Malik
This type of surgery demands training in mechanical techniques,
which, though elementary in practical engineering, are as yet
unknown in the training of a surgeon.
Sir John Charnley

Everything should be made as simple as possible, but no simpler.

Albert Einstein
 Preface
Orthopaedic Biomechanics Made Easy introduces you to the fundamental
biomechanical principles in orthopaedics, and shows you how these relate to
the clinical practice. The book seeks to fulfil two objectives:

To bring together important biomechanical concepts relevant to surgical

practice.
To make these ideas simple and easy to learn.

Our efforts have been about taking you back to the first principles, and
making them more interesting and fun to learn. We have avoided point-by-
point references for this reason, as we feel that this might affect the reading
experience.
To help you explore the subject, the book is signposted into three parts:
Orthopaedic biomaterials and their properties; Engineering theory applied to
orthopaedics; and, Clinical biomechanics. Each concept is introduced and
explained in a discrete double-page spread. Consecutive sections are usually
related and follow a common theme. Naturally, some ideas are more difficult
than others, and we expect you to skip over them initially and to come back
to them after covering the simpler topics. You do not need to deal with
advanced maths to understand the presented biomechanical principles.
Mathematical explanations are provided in some sections only to
demonstrate how a particular biomechanical fact is derived. You may skip
over the mathematical workings without missing out on the learning points.
We hope this book helps to make your clinical practice easier and more
rewarding.

Sheraz S. Malik

Shahbaz S. Malik
 Acknowledgements
We are grateful to Miss Caroline Hing at St George’s Healthcare NHS Trust
for advice and help in setting up this project. We are in debt to two groups of
teachers: the faculty at Cardiff School of Engineering, Cardiff University,
where we read MSc in Orthopaedic Engineering, and our clinical trainers for
sharing their experience and wisdom. Thanks also to our colleagues at the
Engineering School and various hospitals for the group discussions that
helped to clarify and develop ideas.
 Part I Orthopaedic biomaterials and
their properties

Just as there are three states of matter: solid, liquid and gas, there are three
basic forms of solid materials: metals, ceramics and polymers. A composite
is formed when any of the types of materials combine in an insoluble state.
This radiograph shows all four types of solid material in function in a total
hip replacement. The internal structure of the materials produces their unique
physical properties, which are utilised in designing orthopaedic implants.
 1 Introduction to orthopaedic
biomechanics

Introduction to orthopaedic biomechanics

Orthopaedic surgery is the branch of medicine that deals with congenital and
developmental, degenerative and traumatic conditions of the musculoskeletal
system. Mechanics is the science concerned with loads acting on physical
bodies and the effects produced by these loads. Biomechanics is the
application of mechanics to biological systems. Therefore, orthopaedic
biomechanics is about the effects of loads acting on the musculoskeletal
system only or with the associated orthopaedic interventions.
Mechanics, and therefore biomechanics, is divided into two main domains:

Statics is concerned with the effects of loads without reference to time.

Static analysis is applied when the body is stationary or at one instant in
time during dynamic activity.
Dynamics addresses the effects of loads over time. It is further divided
into two main subjects:
Kinematics describes motion of a body over time and includes
analyses such as displacement, velocity and acceleration.
Kinetics is the study of forces associated with the motion of a
body.

The main functions of the musculoskeletal system are to support loads and to
provide motion of body segments. These two functions come together to
achieve the musculoskeletal system's third main purpose: to provide
locomotion, i.e. movement from one place to another. These are all
mechanical tasks and therefore mechanics can be applied to the
musculoskeletal system in the same way as to ordinary mechanical systems.
Biomechanics is, in fact, a fundamental basis of orthopaedic practice: the
mechanics of the body guide the principles of orthopaedic interventions.
Biomechanics is also central to the design and function of modern
orthopaedic devices. The orthopaedic surgeon therefore has the
responsibility to understand musculoskeletal biomechanics and materials and
structural limitations of orthopaedic devices and the principles of their
application in order to minimise failure.
This chapter introduces fundamental biomechanical concepts. It defines
different types of loads and material properties, and the relationships
between them. All the physical interactions between loads and materials can
be considered in terms of the two domains of biomechanics mentioned
above: statics and dynamics. This book, in fact, focuses on these basic
interactions; the different sections simply consider the fundamental statics,
kinematic or kinetic aspects of the musculoskeletal system and/or
orthopaedic interventions. The basic principles are introduced and explained
in the initial sections, and then integrated together in the latter sections.
Therefore, even if the biomechanical concepts become complex, they can
always be considered in terms of statics, kinematics and kinetics.
Fig. 1.1 Orthopaedic surgery is the branch of medicine that deals with
congenital and developmental, degenerative and traumatic conditions of
the musculoskeletal system. Although it is a surgical discipline, over two-
thirds of patients with orthopaedic issues are managed with non-surgical
treatments.
Fig. 1.2 Biomechanics merges together three sciences: anatomy,
physiology and mechanics

Fig. 1.3 The loads acting on the musculoskeletal system can be

considered in the same way as averages in mathematics. The average of a
data set can be represented by (a) mean, (b) median or (c) mode – the
average that is considered depends on the conditions of the data set. In the
same way, loads acting on the musculoskeletal system can be represented by
force, stress or strain – the load that is considered depends on the conditions
of the loading situation.
Fig. 1.4 Orthopaedic biomechanics is a fundamental basis of
orthopaedic practice. The understanding of biomechanics is as essential in
orthopaedics as the understanding of properties of the different types of
sutures and needles in surgical practice.

Force
Force is a simple way of representing load in biomechanics. Force is defined
as the action of one object on another. Therefore, there must be interaction
between two objects to produce a force. Force can have two effects on the
object it is acting upon: it can change the shape and/or the state of motion of
the object.
Force is a vector quantity, which means that it has a magnitude and a
direction. Force, in fact, has three characteristics: magnitude, direction and
point of application; and direction is further divided into ‘line of action’ and
‘sense’ of the force. All these factors determine the effect of a force on an
object. In diagrams, force is drawn as a vector arrow that represents these
four characteristics.
When there are multiple forces acting on an object, they can be resolved
into a single ‘resultant’ force that has the same effect as all the other forces
acting together. However, forces cannot just be added together, as their
direction must also be taken into consideration. A single force can also be
broken down into two component forces, which are usually taken
perpendicular to each other as ‘rectangular’ components.

Newton's laws of motion

A force can change the motion of an object. Newton's laws of motion explain
the relationship between force and motion:

Newton's first law states that a resultant force must act on an object to
change its state of motion. Therefore, a stationary object remains
 stationary and a moving object maintains its velocity, i.e. speed and
direction, unless a resultant force acts on it.
This law shows that objects have an inherent reluctance to change in
their motion. This built-in resistance to change in motion is known as
inertia. Inertia is directly proportional to the mass of an object.
Newton's second law states that a resultant force leads to a change in
momentum of an object. Therefore, a resultant force causes an object to
accelerate (or decelerate).
This law shows that force is directly proportional to acceleration.
Newton's third law states that, for every force, there is an equal and
opposite force. As force is basically an ‘action’, therefore every action
has an equal and opposite reaction.
This law shows that forces always act in pairs and that the two forces
are always equal in magnitude but opposite in direction. These forces
do not simply cancel each other out because they are acting on different
objects.

Newton's laws of motion are principles of interaction between different

forces and between forces and the material world. These are fundamental to
understanding the biomechanics of the musculoskeletal system (Table 1.1).
Fig. 1.5 Force is the action of one object on another. It is drawn as a
vector arrow that represents the four characteristics of force: (a) magnitude,
point of application, line of action and sense. The displacement of the
football is affected by a change in any of these variables. (b) Multiple forces
can be resolved into a single force, and (c) a single force can be broken
down into component forces, which are usually taken perpendicular to each
other as rectangular components.

Table 1.1. Forces acting on the musculoskeletal system

Two types of forces act on an object: internal and external forces.

Internal forces are the molecular forces found within the object.
These are expressed in all directions. Internal forces do not have a
specific action on the object, but instead hold it together. Internal
forces are considered to be unlimited, but with the overall effect
 of producing no resultant action on the object. Different
components of the musculoskeletal systems, e.g. bones and
muscles, all have internal forces acting within them.
External forces are the usual forces that represent the action of one
object on another. External forces acting on the musculoskeletal
system arise due to the action of another object on the
musculoskeletal system. External forces are produced in only
three ways in the musculoskeletal system:
action of a part of the body, e.g. muscle contraction
action of gravity of Earth, i.e. weight
reaction* from other objects, e.g. the ground.
Joint reaction* force, or joint force, is an unusual internal force. It
acts within a joint and holds it together, but its point of application
is at the contact point between two bones. Therefore, it acts within
a structure, i.e. a joint, as well as between two structures, i.e.
bones. As a result, although joint reaction force is an internal
force, it can also be considered as an external force. Therefore,
the fourth external force considered in the musculoskeletal system
analyses is:
joint reaction force.

* The term ‘reaction’ in this context simply means ‘contact’, so a reaction force is due to
contact between two objects.

Moment of a force
A force acting on an object can cause it to rotate. This turning effect of a
force is called the moment or torque. The moment of a force depends on the
magnitude of force and perpendicular distance from the force to the axis
(also known as the lever arm):
Moment [Nm] = Force [N] × Distance [m]

*
Moments are conventionally described as clockwise or anticlockwise.
Couple
A couple is formed when two forces acting on an object are equal in
magnitude and opposite in direction, but have different (but parallel) lines of
action. A couple produces no resultant force; it only produces a moment on
the object.

Conditions of equilibrium
In equilibrium, an object maintains its state of motion, i.e. a stationary object
remains stationary and a moving object maintains its velocity. An object is in
equilibrium only when:

there is no resultant force acting on it, i.e. the sum of all forces is zero
there is no resultant moment acting on it, i.e. the sum of all moments is
zero.

This is the application of Newton's first law of motion.

Levers
A lever is a simple machine that operates on moments. It consists of a rigid
bar that rotates about an axis, and has two forces acting on it: an applied
force that works against a resistance force. The lever amplifies either the
magnitude of applied force or the range and speed of motion it produces.
There are three types of levers:

First-class lever: Axis is between applied force and resistance.

Force amplification. When the axis is closer to resistance, the applied
force has a longer lever arm; therefore, less force is required to
overcome resistance.
Motion amplification. When the axis is closer to applied force,
resistance has a longer lever arm; therefore, greater force is required to
overcome resistance, but the resistance moves through a larger range of
motion or at greater speed than the applied force.
Second-class lever: Resistance is between axis and applied force.
Force amplification. Applied force has a longer lever arm; therefore,
less force is required to overcome resistance.
 Third-class lever: Applied force is between axis and resistance.
Motion amplification. Resistance has a longer lever arm; therefore,
greater force is required to overcome resistance, but the resistance
moves through a larger range of motion or at greater speed than the
applied force.

Levers in the musculoskeletal system

Levers are commonly found in the musculoskeletal system: bones represent
rigid bars, joints represent axes, and the external loads represent applied
force, e.g. muscle contraction, and resistance, e.g. body weight and external
contact forces. Most of the anatomical levers are third-class; this
arrangement promotes the musculoskeletal system to be designed for speed
and range of motion at the expense of force.

Fig. 1.6 Moment = Force × Distance

Fig. 1.7 A couple produces only a moment, but no resultant force, on an
object.
Fig. 1.8 A lever is a simple machine that operates on moments. The
illustrations show the arrangement of axis (A), applied force (F) and
resistance (R) in first-, second-, and third-class levers. Most of the levers in
the musculoskeletal system are third-class.

Static analysis
Static analysis is an engineering method of analysing forces and moments
produced when objects interact. There are a number of steps to applying
static analysis.
 Step one. A dynamic system is simplified to a static system at one instant
in time. Therefore, all interacting objects are taken to maintain their relative
positions.
This removes the need to deal with parameters of dynamic motion, such as
displacement, velocity and acceleration.
Step two. As static analysis of all the forces in three dimensions is too
complicated, the analysis is limited to the main (usually up to three) forces
and their moments in one plane only.
In this step, the relevant ‘co-planar’ forces are isolated from the complex
three-dimensional systems of forces.
Any ‘simplifying assumptions’ that have a bearing on the analysis are
stated and their rationale explained, e.g. forces acting at an angle could be
assumed to act in a vertical direction to make calculations manageable. A
static situation can justifiably have different solutions, provided the
supporting assumptions are valid.
Step three. A ‘free-body force diagram’ is drawn of the object under
consideration, and all the forces acting on it are identified.
This is a simple but carefully drawn diagram, so that the forces are
accurately represented in terms of their magnitude, direction and point of
application with reference to the object. Newton's third law of motion is
used to determine any ‘reaction’ forces acting on the object, e.g. ground
reaction force.
Normally, more than one free-body diagram can be drawn for the same
situation. The selection of free-body diagram for use often depends on the
information available.
Step four. The conditions of equilibrium are applied to the free-body
diagram, and any unknown forces and moments acting on the object are
calculated.
Since the object maintains its relative position, the sum of all the forces
and moments acting on it must be zero. This is the application of Newton's
first law of motion.
 The simplifying steps and assumptions applied mean that static analysis
provides an estimation of the minimum magnitude of forces and moments
acting on an object.

Free-body force diagram

A force is produced by interaction between two objects, and it acts upon one
of them. It is ineffective to represent all the forces on all the interacting
objects in a particular situation. Instead, a free-body force diagram is used to
show all the forces acting on just one object. The free-body force diagram
therefore isolates an object and the forces acting on it – this is an essential
requirement before applying the conditions of equilibrium to the object.

Fig. 1.9 A free-body force diagram is useful for isolating an object and
the forces acting on it. It helps to simplify interactions between several
objects. A simple example of a person standing on Earth involves four
forces, which can be represented on separate free-body force diagrams for
the (b) person and (c) Earth. It is first assumed that they are static with
respect to each other, and that there are no other forces acting on them.
Normally, only one of these free-body force diagrams is required, based on
the object under consideration and information available.
Fig. 1.10 This free-body force diagram shows the forces acting on a
person whilst running. Here, a dynamic situation is simplified to a static
situation at one instant in time. Although the force from air resistance acts
over the whole body, in this free-body force diagram it is assumed to act
collectively as a point load. In order to take a step forward, the person lifts
one foot off the ground and pushes off with the other foot. The supporting foot
exerts a contact force on the ground acting backwards. Friction is equal and
opposite to this force and prevents the foot slipping backwards. Friction
therefore acts forwards and provides motive force, i.e. a force that drives
something forwards. The ground reaction force is equal and opposite to body
weight. As this is a free-body force diagram of the person, it is not necessary
to represent the forces acting on the ground.

Static analysis applied to the musculoskeletal

system
Static analysis can be applied to the musculoskeletal system to estimate
unknown forces, i.e. muscle and joint reaction forces, produced during
normal everyday activities.
 A number of simplifying assumptions have to be made in order to apply
static analysis to a biological system in the same way as in general
mechanics. The following general assumptions apply to most static analyses
of the musculoskeletal system, and further specific assumptions must also be
applied to each situation.

Static analysis
The overall assumption in static analysis is that a limited two-
dimensional analysis can provide a realistic estimation of the actual
forces and moments.
Average measurements of the human body, e.g. length and weight of
limb segments, are usually taken from the reference anthropometric data.

Bones
Bones are rigid bars of a lever. They transmit forces, but are not
deformed by them.

Joints
Joints are frictionless hinges. Other joint movements, e.g. rotation and
translation, are ignored.

Forces
Static analysis is commonly applied to the musculoskeletal system to
estimate joint reaction force. Joint reaction force is an internal force in
absolute terms, but is considered as an external force in static analysis.
It is considered to be a compressive force that holds a joint together. All
other internal forces are considered to cancel each other out (see page 5
for further information).
The only external forces that can be applied to the musculoskeletal
system are weight of the body segments, reaction force from other
objects and muscle contraction.
Muscles are the only soft tissues that actively produce force. Forces
produced by other soft tissues, e.g. joint capsule and ligaments, are
ignored.
Muscles produce only tensile force, i.e. there is no compressive
component to their action.
The main group of muscles produces all of the force for a particular
movement, e.g. triceps contraction produces elbow extension, and there
is no other agonist or antagonist muscle action.
The line of application of force is taken to be along the centre of the
area of muscle cross-section. The point of application of force is where
the muscle inserts onto the bone.
All forces are taken to be as point loads, i.e. act at a specific point;
instead of as distributive loads, i.e. act over a large area. The weight of
an object or a body segment is taken to act at its centre of gravity. This
is the average position of an object's weight distribution. In simple solid
objects, e.g. a ball, the centre of gravity is located at the geometric
centre; in objects with non-uniform weight distribution, it is closer to
where most of the weight is located.
Fig. 1.11 Free-body force diagram of the upper limb showing forces
acting about the elbow joint during flexion. The elbow is held at 90° of
flexion. It is acting as an axis of a class III lever. The moment arm of the
muscles is shorter than that of the weight of the forearm.
Assumptions
In addition to the general assumptions, the following specific assumptions
are applied to this static analysis.

Brachialis and biceps provide all of the flexion force.

The force produced by brachialis and biceps is acting vertically
upwards.
The following physical measurements are based on anthropometric
data:
Forearm mass is approximately 2.25% of total body mass.
Therefore, in a 70 kg adult, forearm mass is assumed to be 1.5 kg.
Assuming gravitational force is 10 N/kg, the weight of forearm
equates to 15 N. This weight is assumed to act as a point load at
the centre of the weight of the forearm, which is located 130 mm
from the centre of rotation of the joint.
 At 90° elbow flexion, the insertions of brachialis and biceps are
50 mm from the centre of rotation of the joint.

Calculations
Applying the conditions of equilibrium.

1. Sum of all moments is zero.

2. Sum of all forces is zero.

i.e. Force J has a magnitude of 24 N and is acting vertically downwards

(hence the minus sign).
Elbow joint reaction force during flexion. This is an example of
application of static analysis to the musculoskeletal system. It shows that the
elbow joint reaction force is over one and a half times the weight of the
forearm. This example also highlights the magnitude of loads acting on the
joints of the musculoskeletal system. Most levers in the body are third-class,
so most muscles have a shorter lever arm than the loads they are supporting.
Therefore, in most joints, the reaction force is multiple times the weight of
the supported body segments.

Simple machine
The concept of machine in mechanics
A machine converts energy from one form into another. Energy is the ability
to do work. Work is done when a force moves an object:
Work [J] = Force [N] × Distance [m]
Therefore, in mechanics a machine converts energy from one form into
another by doing work, i.e. generating movement. This distinction is
important because different types of machines convert energy from one form
to another by different methods, e.g. a microphone converts sound energy into
electrical energy.

Functions of a simple machine

A simple machine is a device that changes the magnitude or direction of a
force. There are six classical simple machines, which are divided into two
groups.

Lever-based simple machines include lever, wheel and axle, and pulley.
These simple machines use rotational movement to redirect a force.
Incline plane-based simple machines include inclined plane, wedge and
screw. These machines also redirect a force, but in a perpendicular
direction to the original direction.

Simple machines can be combined together and with other devices to build
more complex machines. In the most basic arrangement, a simple machine
has two forces acting on it: an applied force that works against a resistance
force. A simple machine is used to achieve mechanical advantage for work
done by the applied force:

Work done by a force is the product of force and displacement of

object:
Work [J] = Force [N] × Distance [m]

Work done by both forces acting on a simple machine is the same, but a
trade-off between force and distance is used to gain mechanical
advantage for work done by the applied force. Therefore, a simple
machine does not reduce the amount of work that is done, but changes
the way in which it is done. Each simple machine has a specific
mechanism, i.e. movement, that modifies the applied force.
As force and motion are connected, a simple machine can also be thought of
as a device that modulates motion.

Sliding and rolling are examples of basic motions that displace an

object. Simple machines can move an object by various means, e.g. a
lever can produce rotational motion and a pulley can produce a lifting
motion, and a screw can convert a rotational motion into linear motion.
Simple machines, therefore, are the fabric of motion in complex
machines.

Fig. 1.12 The six classical simple machines and their mechanisms of
action.

Simple machine Mechanism of action

(a) A lever is a rigid bar that rotates about an

axis. Different arrangements of applied force,
axis and resistance lead to three classes of
levers. Examples: a bottle opener and a crow
bar.

(b) The wheel and axle are a type of lever, as

they also rotate about an axis. The axle passes
through the centre of the wheel and both turn at
the same time. Examples: a screwdriver and a
steering wheel.
Simple machine Mechanism of action

(c) A pulley is a variation of the wheel and

axle. In a pulley, the wheel and/or the axle
have grooves around the outside with a cord
running through. A pulley can change the
direction and/or magnitude of applied force. A
‘pulley system’ consists of two or more
pulleys working together. Examples: a flagpole
and blinds.

(d) An inclined plane is a flat surface that is

higher at one end. It makes work easier by
reducing the applied force needed to move a
resistance. However, as in all simple
machines, the smaller applied force has to
travel a longer distance. In this case the
applied force moves along the full length of
inclined surface, whereas the resistance moves
a shorter distance vertically. Examples: a ramp
and a slide.

(e) A wedge consists of two inclined planes

joined back to back. Therefore, a wedge is a
type of inclined plane. It is used to do work by
pushing the inclined plane into objects to split
them apart. Examples: the blade of an
axe/knife and a zip.

(f) A screw is really an inclined plane

wrapped around a cylinder. The sharp edges of
the inclined plane enable the screw to move in
a linear direction as it is rotated. Examples: a
light bulb and a jar lid.
Fig. 1.13 A screw is an inclined plane wrapped around a cylinder.

Simple machines in the musculoskeletal system

The musculoskeletal system is a set of simple machines working together to
support loads and produce movement. Only three types of simple machines
are found in the musculoskeletal system: lever, wheel and axle, and pulley.
These simple machines provide three functions:

force amplification
motion amplification
change in the direction of applied force.

Most simple machines in the musculoskeletal system are designed to amplify

motion rather than force.

Lever
Levers in the musculoskeletal system are discussed on pages 6–7.

Wheel and axle

This simple machine consists of two interconnected discs of different
diameters with a common centre of rotation. Therefore, when either the
wheel or axle turns, the other must turn as well, and both discs complete a
turn at the same time. The wheel and axle are basically a form of lever, and
their radii function as lever arms; usually, a wheel has a bigger radius than
the axle. Force is applied to one component to overcome resistance acting at
the other component. This arrangement provides mechanical benefits to the
applied force as follows:

Force amplification. Less force needs to be applied to the wheel to

overcome resistance at the axle.
 Motion amplification. Greater force needs to be applied to the axle to
overcome resistance at the wheel, but the wheel turns in a bigger arc at
a faster speed than the axle.

In the musculoskeletal system, the wheel and axle arrangements provide

amplification of both force and motion.

Pulley
A pulley is a device that supports the movement of a cord along its
circumference. A force is applied to one end of the cord to overcome
resistance acting at the other end. There is minimal friction between pulley
and cord, and the cord transmits forces without stretching. A pulley produces
mechanical benefits to the applied force as follows:

A single pulley changes the course of the cord, and therefore the
direction of the applied force.
Multiple pulleys in series can change the direction and magnitude of the
applied force: two pulleys in series halve the effort required to
overcome the resistance, but the cord needs to move twice the distance.
Similarly, as more pulleys are added in series, so the force required is
divided, but the distance travelled by the cord is multiplied by the same
number.

In the musculoskeletal system, a pulley is formed by bone, cartilage or

ligament, with the cord formed by a muscle tendon. The tendon is lubricated
to facilitate its gliding motion over the pulley.
Fig. 1.14 Wheel and axle arrangements in the musculoskeletal system
enhance range and speed of motion. The connection between the shoulder
and the hand forms a wheel and axle arrangement: when the elbow is flexed
to 90° degrees, the head of the humerus represents the axle, whereas the hand
represents a segment of the wheel. A small turn of the head of the humerus
produces a large corresponding movement of the hand, which increases the
speed of action. Similarly, spinal muscles acting upon the vertebrae also
operate in a wheel and axle arrangement: the spinal column represents the
axle, whereas the ribcage represents the wheel. Small rotations of the
vertebrae result in large corresponding movements of the ribcage. On the
other hand, the action of oblique abdominal muscles attached to the ribs
balances large forces produced by spinal muscles.
Fig. 1.15 Single pulley arrangements in the musculoskeletal system
change the effective direction and efficiency of the applied force. The
lateral malleolus and the peroneus longus tendon are an example of a single
pulley arrangement: the peroneus longus tendon runs from the muscle on the
lateral aspect of the leg, passes behind the lateral malleolus and inserts into
the plantar aspect of the base of the first metatarsal. The muscle contraction
therefore results in plantar flexion of the ankle and eversion of the foot. In the
absence of the lateral malleolus, the tendon would have passed anterior to
the ankle joint. The muscle contraction then would have produced
dorsiflexion of the ankle and eversion of the foot. Similarly, the patella
increases the angle of insertion of the patella tendon, thereby improving the
direction and efficiency of the quadriceps force.

Stress and strain

The first section in this chapter explained that a force may have two effects
on an object: it can change the shape and/or state of motion of the object. The
previous sections were related to its effect on the motion of an object. This
and the following sections relate to the effect of a force producing local
shape change, i.e. deformation.
Force is a direct indicator of load in biomechanics. It is taken to act at a
point when analysing its effect on motion of an object. However, it is usually
integrated with other physical parameters to obtain derived measures of load
when analysing its deforming effect.

Stress
Simply comparing two forces acting on two surfaces can be misleading
because it does not take into account the size of the cross-sections. Stress
relates the force to its area of application:
Stress [Nm−2 or Pa] = Force [N] / Area [m2]

Stress indicates the intensity of force acting on a section. It is therefore a

more fair comparison of loads acting on different surfaces. Although stress is
derived from the applied force, it is actually a property of the object, i.e.
stress is produced within an object by the applied force, and the same force
can produce different stresses in different objects.

Strain
Strain is the deformation of an object due to stress. Strain is the ratio of
change in length of an object to its original length:
Strain = Change in length [m] / Original length [m]

Strain has no units. It is usually expressed as a ratio or percentage.

An object under stress experiences strain in the transverse direction as
well as in the longitudinal direction. Transverse strain is usually opposite to
longitudinal strain, i.e. as an object gets longer, it gets thinner and vice versa.
Therefore, force can change the size of the cross-section it is acting upon.
Still, stress and strain are usually based on the original dimensions of the
object.
Although strain is a response of an object to internal stress, and therefore
to the applied force, it can also be considered as a load in itself. This is
further explained with the following example:

Adult cortical bone tolerates strains of only up to 2% before fracture.

Conversely, strains at a fracture site must be less than 2% to allow
 cortical bone to form. This is due to the material properties of cortical
bone, and is irrespective of its size. Therefore, a small bone, e.g.
scaphoid, or a large bone, e.g. femur, would both tolerate strains of up
to 2% before fracture. Therefore, biomechanical studies commonly
consider strain (and strain rate) as an independent load, correlating it
with other physical measurements, e.g. type of tissue formed at the
fracture site, without directly going back to the stress or force producing
the strain.

Fig. 1.16 The relationship between force, stress and strain. When
considering the deforming effect of force, it is important to take into account
the cross-sectional area over which the force is acting. Force produces stress
along the entire length of the object, which leads to strain in the material.
Note that the force changes the cross-sectional dimensions of the area it is
acting upon.
Fig. 1.17 Experimental studies have shown that an ankle fracture with
lateral displacement of talus greatly increases the ankle joint load. In
fact, a 1 mm lateral displacement of the talus can double the joint load in
experimental models. In the normal position, the talar articulating surface is
highly congruent with the tibial articulating surface, so the force on the talus
is spread over a large area. A slight lateral displacement of the talus makes
the articulation incongruent and dramatically reduces the talar contact area
receiving the force, which doubles the stress acting on it.

Fig. 1.18 Strain at a fracture site is described as the change in fracture

gap length (due to motion of the fracture fragments) divided by the
original fracture gap length.

Stress–strain curve
The stress–strain curve shows how a material subjected to an increasing
tensile load deforms until failure. The calculations for stress and strain take
into account the dimensions of the object, so the curve is characteristic of the
material, and is not affected by the size and shape of an object.
The mechanical properties describe how a material deforms under load.
The mechanical properties therefore determine the range of usefulness and
service life of a material. The stress–strain curve describes a number of
these mechanical properties:

The initial linear section corresponds to elastic deformation. Strain is

directly proportional to stress – this relationship is known as Hooke's
law. The bonds between the molecules of the material are stretched but
not broken.

Therefore, deformation is temporary, and if stress is removed,

strain is fully recovered and the material returns to its original
shape and size.
The gradient, i.e. steepness, of the straight line equates to the
material's resistance to deformation. This is the stiffness of the
material, and is technically described as ‘Young's modulus’ (or
more correctly as ‘Young's modulus of elasticity’) or elastic
modulus. The higher the Young's modulus of a material, the smaller
the deformation produced for a given load.

The yield point, or yield strength, is the maximum stress up to which a

material undergoes elastic deformation. It is the limit of proportionality
of Hooke's law, and beyond this point strain is not proportional to
stress. Therefore, the material begins to ‘yield’ past this point.
The non-linear section of the curve corresponds to plastic deformation.
The molecular bonds are broken and the molecules move too far apart
to return to their original positions. The deformation is permanent, and
if the stress is removed, strain is not completely recovered and the
material does not return to its original state.
 The ultimate tensile strength is the maximum stress that the material
can sustain before fracture.
The fracture point is where the material eventually fails. Stress at
the fracture point can be slightly less than the ultimate tensile
strength because the latter can cause the material to ‘neck’ (see
pages 62–63), which reduces its cross-sectional area and therefore
the force required to fracture.

The area under the stress–strain curve represents the energy absorbed
per unit volume of the material. It therefore indicates the energy
absorbed by the material to failure. This is described as ‘toughness’. A
tough material takes a lot of energy to break it.
Stiffness is an elastic property and strength and toughness are plastic
properties of a material.

Fig. 1.19 The stress–strain curve defines the stiffness, strength and
toughness of a material. These mechanical properties are determined by
different sections of the curve: stiffness is defined by the gradient of the
initial linear region, strength is the maximum stress that the material can
withstand before failure and toughness is indicated by the total area under the
curve. Different materials have different characteristics in these properties.
Fig. 1.20 Materials can be classified and compared with each other in
terms of their mechanical properties. The stress–strain curves of three
materials show that material A is the stiffest and strongest, material B is the
toughest and material C is the most flexible (but weak). (Material A
represents ceramics, material B represents metals and material C represents
unreinforced polymers – these are discussed in the next chapter.)

Mechanical properties
The most commonly considered mechanical properties of a material are
stiffness, strength, toughness, hardness and wear resistance. This section
explains the connections and differences between them.

Stiffness vs strength
Stiffness and strength are often used interchangeably, but these are distinct
mechanical properties. The commonest mistake is to refer to the strength of a
material, when actually considering its stiffness. Stiffness is the material's
resistance to change in shape and depends on elastic deformation. Strength is
the load required to break a material and depends on plastic deformation.
The stress–strain curve shows that a load that causes a material to fail must
also take it beyond the yield point (i.e. its initial stiffness). For most
applications, stiffness of a material is the more important mechanical
property, i.e. a material must be stiff enough to not plastically deform under
loads applied in its application. This is for two reasons, both of which relate
to the stress–strain curve:

The load range for elastic deformation is much greater than plastic
deformation. Therefore, once a material passes beyond its yield
strength, it is much closer to its failure point. It would be therefore
unsafe to apply a material where it could yield, as a further small
increase in load could cause it to fail. Therefore, in practical terms, the
material's ‘yield strength’ is more important than its ultimate tensile
strength.
A material's behaviour is easier to predict in the elastic region, due to
the linear relationship between load and deformation. Therefore, elastic
deformation can be taken into account in the design of structures made
from the material. However, it is more difficult to predict plastic
deformation due to the non-linear relationship between load and
deformation.

Hardness vs ductility
Hardness describes the material's resistance to localised surface plastic
deformation, e.g. scratch or dent. Hardness is not a basic mechanical
property, but instead is derived from a combination of other mechanical
properties, e.g. stiffness and strength. Hardness determines the wear
resistance of a material. Therefore, a harder material has a greater wear
resistance than a softer material under the same loading conditions. Ductility
describes the amount of deformation a material undergoes before fracture,
e.g. a copper wire is very ductile.
Hardness and strength both measure plastic deformation of a material.
Therefore, these are roughly proportional to each other, i.e. harder materials
are generally also stronger. However, hardness and strength are gained at the
expense of ductility, i.e. as the materials get harder and stronger, they
generally become less ductile.

Toughness
Toughness is the material's ability to absorb energy up to the fracture.
Toughness is derived from both strength and ductility of a material.
Therefore, a tough material is normally both strong and ductile.

Fig. 1.21 Materials are generally only suitable for application under
loads up to their yield strength. The elastic deformation range is therefore
also known as the working range of a material. A material that is only
elastically deformed returns to the original shape and size when unloaded.
There is zero net work done over a loading cycle. The work done elastically
deforming the material is stored as strain energy, which changes the material
back to the original form when the load is removed. A material that is
plastically deformed does not return to its original state when unloaded.
Fig. 1.22 Toughness is derived from both strength and ductility of a
material. The stress–strain curves of three materials show that only material
B is tough, Although materials A and C are both not tough, material A shows
little plastic deformation whereas material C shows extensive plastic
deformation. Material A is referred to as ‘brittle’ as it not very ductile.
Material C is the most ductile of three materials, but it is not as tough as
material B because it is not sufficiently stiff and strong.

Viscoelastic properties of materials

Mechanical properties describe how a material deforms under a load.
Mechanical properties are not always constant, and may vary with external
conditions, e.g. temperature. In other words, a material may show variable
deformation behaviour under the same load, depending on the external
conditions. In the same way, some materials also show variable deformation
under the same load over a period of time or under different loading rates.
Therefore, these materials display time-dependent, as well as load-
dependent, deformation. Viscoelastic properties describe how a material
deforms with time, and can be considered as extension of usual mechanical
properties.
A material that displays time-dependent deformation is described as
viscoelastic. Biological materials, e.g. bone, tendon and skin, are typically
viscoelastic. Different materials have different mechanisms for their
viscoelasticity.
Viscoelastic behaviour is commonly described in terms of four factors:
creep, stress relaxation, strain-rate dependent mechanical properties and
hysteresis.

Creep vs stress relaxation

If a viscoelastic material is placed under constant stress over a prolonged
period, it shows increasing strain with time – this is creep. The technical
definition of creep is that it is time-dependent deformation of a material
under constant load that is below its yield strength. Therefore, creep
produces plastic (i.e. irreversible) deformation of a material below its yield
strength. Creep can eventually cause the material to fail. Therefore, creep can
lead to failure of a material under loads significantly below its ultimate
tensile strength.
Creep can be managed by reducing the applied stress over time. Stress
relaxation is the decrease in stress required to maintain constant strain over
time, i.e. prevent creep.

Strain-rate-dependent mechanical properties

The above section explained that strain in a viscoelastic material depends on
not just stress, but instead is a product of stress–time combination. Time
increases deformation produced by a given load. Therefore, a given load
produces less deformation over a shorter time period than over a longer time
period. Since stiffness is the resistance of a material to deformation,
viscoelastic materials are stiffer when loaded over a short time period than
over a longer time period. This demonstrates why many mechanical
properties, e.g. stiffness and strength, of viscoelastic materials vary with the
rate of loading. Typically, viscoelastic materials are stiffer, stronger and
tougher when loaded at a faster rate, i.e. at a higher strain-rate, because there
is less time for them to strain.

Hysteresis
Hysteresis occurs when a viscoelastic material is cyclically loaded and
unloaded. It is the ability of the material to dissipate energy between the
loading and unloading cycles. The dissipated energy is used to change the
shape of the material during loading and unloading. As a result, further
energy input is required to continue the loading and unloading cycles.
Therefore, hysteresis allows viscoelastic materials to act as ‘shock
absorbers’, e.g. intervertebral discs and menisci.

Fig. 1.23 These stress–time and strain–time graphs show the

phenomena of creep and stress relaxation displayed by viscoelastic
materials. (a) Creep is the time-dependent deformation of a material under
constant stress that is below its yield strength. (b) Stress relaxation is the
decrease in stress required to maintain constant strain over time, i.e. to
prevent creep. Therefore, stress relaxation is the inverse of creep. A clinical
example of application of creep response is the conservative treatment of
clubfoot deformity in children. The plaster cast treatment applies constant
load to lengthen the soft tissue over a period of time. The deformity is
gradually corrected with sequential casting.
Fig. 1.24 The mechanical properties of viscoelastic materials vary with
the rate of loading. Viscoelastic materials typically are stiffer, stronger and
tougher when loaded at a faster rate, i.e. at a higher strain rate. Viscoelastic
materials have both elastic and viscous properties; the viscous properties are
influenced by the rate of loading.
Fig. 1.25 A viscoelastic material subjected to cyclical loading displays
two stress–strain curves: a loading curve and an unloading curve, which
form a ‘hysteresis loop’. This is due to the fact that viscoelastic materials
do not obey Hooke's law perfectly. The hysteresis loop shows that a
viscoelastic material is harder to deform when loading than unloading.
Hysteresis is the energy lost between the two cycles, and is represented by
the area between the curves.

* Square brackets show SI (International System) units of physical

measurements or chemical symbols of elements and compounds.
 2 Orthopaedic biomaterials and their
properties

Structure and properties of materials

Our world is full of objects. These objects are made of different materials,
which are composed of different internal microscopic structures. Therefore,
all objects can be analysed in terms of their molecular arrangement. The
properties and behaviour of materials is determined by the following
structural factors:

types of bonds between atoms, e.g. ionic or covalent

arrangement of atoms, e.g. short molecules or long chains
microstructure, e.g. crystalline or amorphous
macrostructure, e.g. single crystal or polycrystalline.

This structural hierarchy divides all materials into four groups: metals,
ceramics, polymers and composites. Materials in each group have similar
molecular structure and therefore exhibit a similar range of properties.
Material properties can be divided into chemical, physical, electrical and
mechanical. These properties define how different groups of materials can be
applied in the physical world.
A biomaterial is any substance, natural or engineered, that forms a part of
either a biological structure or a biomechancial device that augments or
replaces a natural function. All of the four groups of materials may be used
as biomaterials. However, the properties of a material must be compatible
with the body for it to be used as a biomaterial. Mechanical properties
determine if a material can withstand applied loads and therefore define its
practical functional capacity. Mechanical properties can be divided into two
groups: static and viscoelastic (time-dependent). Static properties include
stiffness, strength, ductility, toughness and hardness. Viscoelastic properties
include fatigue, creep, stress relaxation and hysteresis. All these variables
are just different measures of how the material deforms under load. This
chapter looks into how the molecular structure of different groups of
materials determines their mechanical properties and their applications in
orthopaedics.
As well as being mechanically suitable, orthopaedic biomaterials should
ideally also be biologically inert, easy to fabricate at reasonable costs and
have appropriate handling properties. The biological environment can be
highly corrosive, and orthopaedic biomaterials must also have high
resistance against corrosion. All these requirements mean that the
engineering materials that are practical for use in orthopaedic surgery are
only a small fraction of the vast numbers that are potentially available.
An orthopaedic biomaterial must also be of ‘medical grade’, i.e. the type
and quality of a material must meet a set of minimum requirements for
medical use. In addition, stringent global manufacturing standards ensure that
orthopaedic components are of high quality and have consistent mechanical
properties and dimensional specifications.

Fig. 2.1 All objects can be analysed in terms of their molecular

arrangement.
Fig. 2.2 A material is a physical substance used to make objects. There
are four basic groups of materials: metals, ceramics, polymers and
composites. Each group of materials has similar molecular and bulk
properties.

Fig. 2.3 This classical material science tetrahedron highlights that a

material's structure, properties, processing (including application) and
performance are all interrelated. Physical laws of biomechanics also have
a bearing on these aspects.
Metals
Molecular structure
Metals typically have the simplest internal structural arrangement. They
consist of atoms tightly packed together in a regular, repeating pattern to form
a lattice. The lattices grow into crystals or grains, which are randomly and
therefore evenly orientated in different directions. Metals therefore are
polycrystalline materials.
The primary bonds in metals are metallic bonds: the adjacent atoms share
an electron, and the whole structure consists of positive ions in a cloud of
valence electrons. Metallic bonds characteristically are symmetrical, non-
directional and moderately stiff and strong.
Metals initially appear to have a uniform, orderly structure. However, a
large assembly of atoms inevitably develops imperfections. The different
layers of atoms can become out of phase with each other and create defects
within the lattice. In addition, the crystals also vary in size and shape. The
junction between adjacent crystals is called a grain boundary. There is also
misalignment between layers of atoms in different crystals at grain
boundaries. All these imperfections affect the mechanical properties of
metals.

Mechanical properties
Most metals have a higher density than non-metals due to the densely packed
arrangement of their crystals. As a result, all metals, except mercury, are
solids at room temperature. As there are numerous stiff primary bonds in a
closely linked structure, metals as a group are stiff materials. The
symmetrical and non-directional nature of primary bonds mean that metals
have isotropic elastic properties, i.e. the elastic properties remain constant
regardless of the loading direction.
Although metals as a group are stiff, strong and tough, pure metals are often
too soft and weak for most practical applications. These plastic properties
reflect the ease with which lattice defects move in pure metals when the
primary bonds are broken. Most pure metals therefore require some form of
‘treatment’ to improve their strength, hardness and toughness to a useful
level. A common method for achieving this is to transform pure metals into
alloys.
The mechanical properties of metals (and other materials) change at
elevated temperatures. This characteristic is used in the manufacturing
process to prepare metals for fabrication work such as moulding, shaping
and machining. The process of changing the mechanical properties of a
material with heat treatment is called annealing. Annealing involves heating a
material to above a critical temperature, maintaining that temperature for a
set time and then cooling the material slowly. This makes the material softer
and more workable.

Fig. 2.4 Molecular structure of metals. Metals usually have atoms of (a)
normally one element bonded together with metallic bonds, (b) into groups of
lattices that (c) contain imperfections. (d) The lattices grow into crystals that
are randomly orientated in different directions.
Fig. 2.5 Mechanism of plastic deformation in metals. (a) Plastic
deformation in metals occurs when some of the primary bonds between the
atoms are broken and the lattice defects are sequentially advanced until they
reach the edges. (b) Otherwise, the movement of a whole plane of atoms over
each other would require breakage of all primary bonds between the planes,
which would require too much energy. (The three methods of strengthening
metals discussed in the next section all involve restricting movement of
lattice defects so that the yield stress (i.e. start of plastic deformation) is
greater.)

Alloys
Highly pure metals are often too soft, weak, ductile and chemically reactive
for practical use. Therefore, they are commonly mixed into alloys that have
enhanced mechanical properties compared with the base metals. Alloys can
be custom designed for a particular use more easily than sourcing a pure
metal that meets the same requirements. Alloys can also be given specific
properties during the development phase, e.g. resistance to corrosion.
Therefore, almost all metals used in orthopaedics (and engineering) are
alloys.

Molecular structure
An alloy is a substance that is composed of two or more elements, at least
one of which is a metal, united by dissolving in each other when molten, e.g.
brass is an alloy of zinc in copper. Therefore, an alloy is a ‘solid solution’ of
different elements. The process of converting a pure metal into an alloy is
known as ‘solid–solution strengthening’.
The addition of a few ‘impurity’ atoms to a pure metal slightly distorts its
internal structural arrangement. The host metal atoms still pack together
tightly in lattices. However, depending on the size of the impurity atoms,
these can replace or position themselves between the host atoms.
The primary bonds in an alloy are still the metallic bonds. An alloy also
contains imperfections in the same way as a pure metal. However, the
impurity atoms impose ‘lattice strains’ on the surrounding host atoms and
impede the movement of lattice defects when the primary bonds are broken.

Mechanical properties
Plastic deformation in metals occurs by movement of lattice defects. The
‘mis-fitting’ impurity atoms impose restrictions to this process. Therefore,
alloys are harder, stronger and tougher than pure metals. The fraction of
impurity atoms can be adjusted to produce the most optimum combination of
hardness, strength and toughness.

Other methods of strengthening metals

Pure metals (and alloys) can also be strengthened by two other methods.

Grain size reduction

Grain boundaries form natural barriers to the movement of lattice defects.
Therefore, smaller crystals produce more grain boundaries that halt the
movement of lattice defects. Smaller crystals are produced by a process
called ‘quenching’, in which the metal is heated to above a critical
temperature and then rapidly cooled down, so that there is less time
available for the grains to grow.

Work hardening
This involves applying stress to a material until it plastically deforms and
then removing the stress. Then, when the material is re-loaded, the
deformation occurs at a higher yield point. Therefore, the material has
effectively become stronger and harder. Plastic deformation works by
increasing the overall number of lattice defects, thereby creating ‘traffic
jams’ that obstruct movement of lattice defects.

Fig. 2.6 Mechanism of solid–solution strengthening. The impurity atoms

in an alloy distort the lattice and significantly increase the resistance to
movement of lattice defects. Possible mechanisms are shown above: (a)
missing atom; (b) small atom substituting for a metal ion; (c) small additional
atom; (d) large atom substituting for a metal ion.
Fig. 2.7 and Fig 2.8 Mechanism of strengthening by grain size
reduction. Grain boundaries act as natural boundaries to movement of lattice
defects due to misalignment of atoms in different grains. Smaller crystals
produce more grain boundaries, which make metals stronger.

Fig. 2.9 and Fig. 2.10 Mechanism of strengthening by work hardening.

A material that is plastically deformed and then unloaded has a higher yield
point when it is loaded again. Strengthening by work hardening can be
demonstrated using a wire or a paper clip. It is difficult to bend a straight
portion back and forth repeatedly at the same place, because the strength of
material increases with the work done. If this cycle is continued, the material
eventually breaks due to fatigue failure.

Metals in orthopaedics
Three conventional alloys have obtained a wide use in orthopaedics:
stainless steel, cobalt–chrome and titanium and its alloys.
Stainless steel
Stainless steel is used in making common surgical instruments and the
temporary implants applied in trauma surgery, such as plates, nails, screws
and wires. Stainless steel is easy to process, has good corrosion resistance
in the biological environment and is relatively cheap. There are different
grades of stainless steel and surgical devices are developed from the alloy
type that has the most suitable combination of properties, e.g. surgical drill
bits are made from stainless steel that is extremely hard and can be well
sharpened (but it is also relatively brittle and the drill bits are prone to
breakage without much deformation). The use of stainless steel alloys in
developing long-term joint replacement prostheses has been limited because
other alloys have better wear and corrosion resistance.

Cobalt–chrome
Cobalt–chrome alloys have the longest and broadest history of use in joint
replacement prostheses. They are now mainly used to develop the metal-type
bearing components of total hip replacements (i.e. the femoral head and
acetabular cup). Cobalt–chrome alloys have very good wear resistance,
fatigue strength and corrosion resistance. Therefore, these alloys are most
suitable for making articulating interfaces of long-term joint replacement
prostheses. However, they are not the best choice for making the stems of the
prostheses. The stiffness of cobalt–chrome alloys is much greater than that of
the cortical bone, which means that, if they are combined together, the
prosthesis would take up most of the applied load. This would lead to stress
shielding and resorption of surrounding bone, which in turn would lead to
early prosthesis loosening.

Titanium and titanium alloys

Titanium and its alloys are comparatively new engineering materials that
have an exceptional combination of properties. They are extremely strong,
have half the stiffness of cobalt–chrome alloys and also have the ability to
osteointegrate with bone. These properties mean that titanium-based implants
can achieve good fixation with bone. Therefore, titanium alloys are the most
appropriate materials for making stems of long-term joint replacement
prostheses. They are also used to make implants for trauma surgery;
however, these implants can sometimes be difficult to remove due to their
osteointegration with bone. Titanium alloys have an oxide layer on the
surface, which provides excellent resistance to corrosion.
The major limitations of titanium are its relatively poor wear resistance
and high surface coefficient of friction. Titanium alloys therefore are not
suitable for developing articulating components. Titanium-based implants are
also prone to notch sensitivity, and any surface defects significantly increase
the risk of fatigue failure. Titanium is also quite difficult to process during
certain stages of manufacturing, e.g. forging and machining (Table 2.1).

Table 2.1. Alloys in orthopaedics

Alloy Composition
Alloy Composition

Stainless Steel is an iron–carbon alloy. Stainless steel is iron–

steel carbon alloy that also contains chromium and
sometimes nickel. It may also contain minor quantities
of other elements, such as manganese, molybdenum
and sulphur. Stainless steel has a high resistance to
corrosion when compared with plain steel – this is a
function of chromium and nickel.
The different grades of stainless steel used in
orthopaedics are broadly divided into two groups:
Instrument stainless steel. These alloys usually
do not contain nickel. Therefore, their corrosion
resistance is relatively low. These are extremely
hard and can break relatively easily.
Implant stainless steel. These alloys contain
nickel and therefore have greater corrosion
resistance. These also have better fatigue strength
and toughness.
Newer stainless steel alloys containing high nitrogen
content are being used in joint replacement prostheses.
High nitrogen content makes stainless steel stronger
and more resistant to localised corrosion than normal
stainless steel.

Cobalt– Cobalt–chrome alloys consist principally of cobalt

chrome and chromium with trace quantities of other elements,
such as molybdenum, tungsten and nickel. Cobalt–
chrome alloys are very stiff and strong and are also
used in gas turbines and dental implants. Their high
corrosion resistance is a function of their bulk
composition and surface oxide.
 Alloy Composition

Titanium Almost all of the commercially produced titanium is

and utilised in the aviation industry to produce high
Titanium performance components for aircrafts and space
alloys vehicles. The quantity of titanium used in orthopaedics
is a very small fraction (≅2%) of the overall titanium
usage. Therefore, titanium used in orthopaedics is of
the same high standard as in the aviation industry.
Titanium-based materials used in orthopaedics are
commercially pure titanium* and its alloys contain
small amounts of aluminium and vanadium. Titanium-
based implants are comparatively very expensive
because of the high manufacturing costs. However,
these implants are compatible with modern MRI
scanners, and can be used in patients with nickel
allergy.

* It is not possible to obtain an absolutely pure metal, because these are extremely reactive and
form an oxide layer.

Ceramics
Molecular structure
Ceramics are non-metallic materials produced with the use of heat, e.g.
pottery made from clay. Common ceramics are compounds that consist of a
metal combined with oxygen, carbon, nitrogen or sulphur, i.e. metal oxides,
carbides, nitrides or sulphides, respectively. The combination of different
elements means that ceramics are a large family of materials with varied and
complex molecular structures.
The primary bonds in ceramics are ionic and/or covalent bonds. An ionic
bond is formed by electrostatic force between positive (metal) and negative
(non-metal) ions. Ionic bonds are symmetrical, non-directional and very stiff
and strong. A covalent bond is formed when electrons are shared between
non-metal atoms. Covalent bonds are variably spread (symmetrical/
unsymmetrical), very directional and very stiff and strong.
Ceramics can be crystalline, partially crystalline or amorphous.
Crystalline materials consist of atoms arranged in regular lattices, e.g. NaCl.
Amorphous materials have randomly packed molecules with no regularity in
their spatial arrangement (amorphous means ‘no structure’). Partially
crystalline materials consist of a mixture of crystalline and amorphous
phases.

Mechanical properties
Ceramics tend to be less dense than metals: ceramics with ionic bonds
contain non-metals, which are generally lighter than metals; and ceramics
with covalent bonds are less densely packed. Ceramics, like metals, have a
network of primary bonds in a closely packed structure. However, ionic and
covalent bonds are usually stiffer and stronger than metallic bonds.
Therefore, ceramics are typically stiffer, stronger and harder than metals.
They also have a greater resistance to wear. The strength of the primary
bonds also means that ceramics have a high resistance to chemically reacting
with other substances. Therefore, ceramics have excellent resistance to
corrosion.
Ceramics are extremely brittle and do not deform significantly before
breaking. This is because crystalline ceramics have a set structure that cannot
be changed easily because rearrangement of positive and negative ions in a
lattice or atoms joined through covalent bonds takes a considerable amount
of energy, and amorphous ceramics do not have crystal lattices in which
lattice defects can move. Therefore, ceramics have low fracture toughness
and are highly susceptible to fracture.
In common with other brittle materials, ceramics are strong in
compression, but weak in shear and tension. This is because ceramics
contain impurities and defects (like all materials), which act to amplify stress
and therefore reduce the overall strength (see pages 72–73 for further details
on the effect of stress raisers). Stress amplification occurs mainly in shear
and tension, but not in compression – compression, in fact, closes defects and
reduces the overall number of imperfections in the material. Therefore,
ceramics are about ten times as strong in compression as in tension (Table
2.2).

Table 2.2. Basic chemistry

A compound is a substance that is formed when two or more

elements chemically react together to form an ionic or covalent
bond. A compound usually has completely different properties
from its base elements. The base elements in a compound cannot
be separated by physical methods. This can only be done by
further chemical reactions or by using electric current.

Fig. 2.11 Ceramics are earthly materials that consist of metallic and
non-metallic elements. The primary bonds in ceramics can be ionic, as in
sodium chloride, or covalent, as in diamond. Ceramics are characteristically
hard and brittle materials.
Fig. 2.12 Ceramics are a diverse group of materials. This chart shows
the different classes of ceramics and examples of their uses in domestic,
industrial and building products.

Ceramics in orthopaedics
Ceramics used in orthopaedics are classified as bioinert or bioactive.
Bioinert ceramics show little or no chemical change in the biological
environment and usually generate only a small, non-specific fibrous reaction
in the local tissues. Bioactive ceramics chemically interact with the
biological tissues. They are biodegradable and are slowly replaced by the
natural tissues.

Bioinert ceramics
Bioinert ceramics are used for developing the bearing components of total
hip replacement. The two main bioinert ceramics are alumina and zirconia.
Alumina is used for both the femoral head and acetabular cup, whereas
zirconia is used for only the femoral head. Both materials have a high degree
of hardness, produce very low wear rates and have excellent corrosion
resistance. However, they are also susceptible to catastrophic brittle
fracture.
The lowest wear rates are produced when the femoral and acetabular
components are both ceramic, i.e. in alumina-on-alumina combination.
However, ceramics are much stiffer than metals, and in patients with
osteoporotic bones, a ceramic acetabular component can produce stress-
shielding and bone resorption around the acetabulum, which can lead to
aseptic loosening of the implant. Therefore, a combination of zirconia head
articulation with a polyethylene acetabular cup is also commonly used.

Bioactive ceramics
Bioactive ceramics have similar or identical composition to the mineral part
of bone. They bond with the bone tissue and act as scaffold to stimulate new
bone formation on their surfaces. Bioactive ceramics are osteoconductive,
which means that they can only stimulate new bone formation within an
osseous environment. In contrast, an osteoinductive substance (e.g. bone
morphogenetic proteins) can stimulate new bone formation even outside the
osseous environment. Bioactive ceramics are used as coatings on
uncemented implants to enhance fixation and as bone substitute grafts to fill
in bone defects until bone regenerates.
Ceramics used as bioactive coatings are calcium hydroxyapatite
[Ca10(PO4)6OH2] and β-tricalcium phosphate [Ca3(PO4)2]. They are applied
onto implants in the form of a plasma spray to produce a 50–150 microns
thick layer. They stimulate bone ingrowth and ongrowth to the implant
surface, and therefore create a biological union between bone and prosthesis.
Bone substitute grafts are developed from calcium hydroxyapatite, β-
tricalcium phosphate or bioactive glasses. Bioactive glasses are also
ceramics and mainly consist of oxides of silicon, sodium, calcium and
phosphate [e.g. Na2OCaOP2O3-SiO], and gradually release these metal ions
into bone. As ceramics are weak under tension and have low fracture
toughness, bone substitute grafts are not suitable for load-bearing
applications (Table 2.3).

Table 2.3. Bioinert ceramics as joint-bearing surfaces

Advantages The wear of bearing surfaces of replacement

joints is a big issue, as this leads to aseptic
loosening of the prostheses. The main advantage
of ceramics is that their very high hardness and
low surface friction leads to much lower wear
debris generation compared with other materials.
However, although the mass of debris is low, the
number of debris particles is high, i.e. the wear
debris is very fine. Ceramics also have good
biocompatibility and excellent corrosion
 resistance.
Alumina = Aluminium oxide [Al2O3]. Naturally
occurring alumina is better known as sapphire or
ruby. Zirconia = Zirconium oxide [ZrO2].
Alumina is harder than zirconia. Zirconia
femoral head was introduced because zirconia
has a superior wear resistance in a ‘hard-on-
soft’ bearing combination. Therefore, zirconia-
on-polyethylene combination produces far less
wear debris than alumina-on-polyethylene
combination. However, as zirconia has a lower
hardness than alumina, it also has a lower wear
resistance in a ‘hard-on-hard’ bearing
combination. Zirconia-on-zirconia combination
has a much higher wear rate than alumina-on-
alumina combination, and therefore zirconia is
not used to make the acetabular cup.

Disadvantages As ceramics are brittle in nature, their main

drawback is the risk of sudden and unpredictable
fracture of the femoral head. It is important that,
during the manufacturing process, ceramics have
minimum impurities and defects introduced in
order to minimise the impact on the ceramics'
fracture strength. Therefore, ceramic bearing
components are comparatively expensive to
manufacture. ‘Zirconia toughened alumina’
composites have been found to have better
fracture toughness than individual ceramics, and
are also being used to produce the bearing
surfaces.
Modern ceramic femoral heads are individually
tested to a load of about 25 times body weight
before distribution to ensure that the risk of in-
vivo failure is minimised. In 1998, when a
manufacturer of zirconia femoral heads altered a
single production step, it led to a surge in
 femoral head fractures (at least 343 reported
cases worldwide). The company in question
stopped making these implants. Studies show that
the overall risk of in-vivo ceramic failure is in
the order of 0.02%–0.7%.

Bioactive ceramics

Calcium hydroxyapatite is widely used as a bioactive coating for

uncemented total joint replacements to avoid the problems
associated with the use of cement. Calcium hydroxyapatite
degrades on implantation approximately at the rate at which the
new bone generates.
Bone grafts are routinely required in spinal fusion, revision joint
replacements, fracture repair and to fill bone defects, such as after
resection of bone tumours. Currently, the source of bone for these
applications is approximately 35% autografts, 60% allografts and
5% synthetic materials, i.e. bone substitute bioactive ceramics.
Bone substitute grafts can be loaded with antibiotics or growth
factors for local delivery at a controlled rate.

Polymers
Molecular structure
Polymers are large molecules that consist of long chains of repeating simple
molecules (monomers). The process of monomers joining together to form
long chains is called polymerisation. Copolymers are polymers made of two
or more different types of monomers. Most polymers are organic and based
on carbon atom monomers. As carbon atoms are usually also bound to
hydrogen atoms, most polymers are hydrocarbons.
There are two types of bonds in polymers. Monomers forming the long
chains are bound to each other by primary covalent bonds. Covalent bonds
are stiff and strong and a chemical reaction is required to form or break them.
The long chains stick together in groups due to secondary bonds formed by
intermolecular attractions between molecules in adjacent chains. These
secondary bonds (also known as van der Waal's bonds) are much less stiff
and strong and can be overcome by mechanical methods. Primary covalent
bonds may also link adjacent long chains to form a network of cross-links.
The long chains can have a variable structure. They can vary in length,
degree of branching and the overall three-dimensional arrangement. The
exact structure of chains depends on the conditions under which a polymer is
synthesised. They can be randomly tangled to produce an amorphous
structure or orderly arranged into a crystalline structure.
Polymers can be naturally occurring or synthetic. Naturally occurring
polymers include proteins, DNA, rubber, wool and silk. Synthetic polymers
produce a wide range of materials with many uses in everyday life, e.g. all
plastics are polymers.

Mechanical properties
The mechanical properties of polymers depend on their molecular structure,
loading mode and external factors such as time and temperature. In general,
the structure of long chains affects mechanical properties of polymers as
follows:

Length. As chains become longer, polymers become stronger. This is

because longer chains are usually more tangled, which allow more
secondary bonds to form between chains.
Branching. Linear, unbranched chains can pack more closely than
branched chains; therefore linear chains produce crystalline polymers
that are stronger.
Cross-linking. As long chains become more extensively linked by
primary covalent bonds, polymers become stiffer, harder and stronger.

As a group, polymers tend to be less dense than metals and ceramics. They
are also not as stiff and strong as the other materials. However, they are
extremely ductile and pliable and can be easily formed into complex
structures. They have a low coefficient of friction and good resistance to
corrosion.
 Unlike metals and ceramics, polymers show viscoelastic behaviour, i.e.
their stiffness varies with time. This is due to entanglement of the long
chains, so that when the polymers are loaded, the long chains straighten out
first before the secondary bonds are affected, which gives polymers variable
stiffness.

Fig. 2.13 Polymer = Many parts. A polymer is like a necklace made from
small beads. (a) Polyethylene is the most common polymer, and consists of
the ethylene monomer. Polymers are a large family of materials. Their
mechanical properties depend on the nature of the monomer(s) forming (b)
long chains and the number and type of bonds between the adjacent long
chains. (c) The long chains can be packed together in an ordered or tangled
fashion, producing a crystalline or an amorphous structure, respectively.
Fig. 2.14 Polymers show viscoelastic behaviour because of their
tertiary structure. The long chains are loosely packed together in clusters.
There are primary covalent bonds between monomer units forming long
chains, and further bonds between chains, which can be the weak secondary
bonds or strong primary bonds. When a load is applied, the polymer chains
uncoil before the bonds break, which produces the viscoelastic behaviour.

Polymers in orthopaedics
Polymers used in orthopaedics are divided into two groups: long-term
implantable and biodegradable polymers.

Long-term implantable polymers

Ultra-high molecular weight polyethylene
(UHMWPE)
The main applications of UHMWPE are in the form of articular surfaces in
joint replacements, e.g. acetabular component in total hip replacements,
tibial insert and patellar components in total knee replacements and as a
spacer in intervertebral disc replacements. It is suited to these roles because
of its mechanical properties of good wear resistance and low friction. The
main drawback of UHMWPE is the production of wear particles, which
initiate an inflammatory response that, over time, leads to osteolysis and
aseptic loosening of the prosthesis.
 Highly cross-linked UHMWPE is an enhanced variety in which the
polymer long chains are cross-linked extensively by primary covalent bonds.
The long chains are therefore fixed in their relative positions and produce a
greater wear resistance. Therefore, highly cross-linked UHMWPE is harder
and has much lower wear rate than standard UHMWPE. However, the
restriction in the movement of long chains also affects plastic deformation
and makes highly cross-linked UHMWPE brittle.

Polymethylmethacrylate (PMMA) bone cement

PMMA bone cement is used to fix implants into bone in cemented total joint
replacements. Although referred to as ‘cement’, it acts as a grout (space
filler), rather than as an adhesive. It provides excellent primary fixation of
the prostheses, but does not simulate secondary biological fixation.
Therefore, the quality of fixation degrades over time.
PMMA bone cement comes in two parts: a powder phase and a liquid
phase, in a ratio of 2:1 by weight. The main ingredients in the powder phase
are:

prepolymerised PMMA
an initiator (benzoyl peroxide) to catalyse the polymerisation process
+/− barium sulphate or zirconia to add radio-opacity
+/− colouring agent, e.g. chlorophyll
+/− antibiotics.

The main ingredients in the liquid phase are:

monomer methylmethacrylate
reaction accelerator (N-N-dimethyl-p-toluidine, DMPT).

The two phases are mixed together and a polymerisation reaction combines
the monomer molecules to produce the final PMMA polymer. The ingredients
initially form a paste – which is easily applied where required – and then
harden and set by about 12 minutes from the start of mixing. The
polymerisation reaction is exothermic and is associated with a sharp
temperature rise.
Biodegradable polymers
Biodegradable polymers include polyglycolide (PGA), polylactide (PLA)
and polydioxanone (PDS). These polymers have been developed into
sutures, staples, pins, screws and plates. They provide temporary structural
support and gradually undergo hydrolytic degradation. Biodegradable
implants have several advantages over standard implants: they have less
stress-shielding effect; they eliminate the need for surgical removal of the
implant; and compared with metal implants, they have less interference with
radiological investigations. The degradation times of devices are customised
to titrate the load transmitted through the tissues and fixation device as the
healing progresses (Table 2.4).

Fig. 2.15 UHMWPE consists of chains that are about 500 000 carbon
atoms long. It is used to develop the bearing surfaces of total joint
replacements. The newer highly cross-linked UHMWPE has much lower
wear rates than standard UHMWPE, but is a more brittle and less forgiving
material. (Photo reproduced courtesy of Cutting Tal Engineering/Bill
Kennedy.)
Fig. 2.16 The typical time–temperature curve for bone cement setting
reaction. ‘Dough time’ is the time it takes for cement to reach a dough-like
state. ‘Set time’ is the time it takes for cement to harden and set. ‘Working
time’ is the interval between dough time and set time, during which cement
can be applied and implants inserted and adjusted.

Table 2.4. Bone cement

Advantages Flexibility of the procedure.

Delivery of antibiotics to the site.
Immediate stability of the components: patients
can mobilise bearing full weight straight away.
The use of cement is well established, and long-
term outcome studies have been completed and
cementing technique has been refined.

Disadvantages Cement sets with a large exothermic reaction,

which leads to local bone necrosis.
Residual monomer can enter the blood supply,
producing pulmonary embolism.
Radiopacifier particles can reduce the
mechanical properties of cement by up to 10%
due to the discontinuities produced.
Aseptic loosening in the long term.
Composites
Molecular structure
Composites consist of two or more base materials that remain discrete after
they are combined together. The basic structure of a composite is that a
‘reinforcement’ material is dispersed in the volume of a ‘matrix’ material.
The reinforcement material adds its mechanical properties to the matrix
material, which supports the former by maintaining the relative arrangement
of its fibres. When the composite is loaded, the matrix phase accepts the load
over a large surface area and transfers it to the reinforcement phase.
There is usually no chemical bonding between the base materials. Instead,
the matrix material ‘melds’ with the reinforcement material. This can occur
in a number of ways, such as the matrix material simply solidifying from a
molten state or there can be chemical polymerisation within the matrix
material. This results in a mechanical bond between the two phases.
However, there is still a clear interface between the two phases, and the
materials can be readily distinguished if the composite is broken down.
The selection and composition of base materials depend on the
composite's final desired characteristics. Composites are classified into
three groups according to the type of material forming the matrix phase: metal
matrix composites, ceramic matrix composites and polymer matrix
composites. Most of the commonly used composites are polymer matrix
composites. The two phases of a composite can belong to the same class of
materials, e.g. a polymer can be the reinforcement phase within another
polymer as a matrix phase.

Mechanical properties
Composites have enhanced properties compared with the base materials.
This is based on the principle of combined action, so that the synergistic
effect of the base materials produces better overall properties of the
composite. Composites are considered in situations where the combination
of required properties cannot be met by a single base material on its own.
The biggest advantage of composites is that they can be stiff and strong but
still very light, i.e. have high strength to weight ratio. Their main downside is
the associated very high development costs.
The size and arrangement of reinforcement material have a huge influence
on the properties of the composite. The reinforcement material can be in the
form of particulates (filled composite), fibres (fibre composite) or
continuous sheets (laminated composite). Composites usually have
anisotropic mechanical properties that vary according to the orientation of
the applied load to the alignment of reinforcement material. The interface
between the base materials is particularly vulnerable under stress and
composites are prone to failure by ‘delamination’ (i.e. separation) of the two
layers.

Fig. 2.17 Concrete is a composite of mainly crushed stones and cement.

It is the most used synthetic material in the world. It is used extensively in the
construction industry because it is cheaper than other materials and does not
rust. Concrete is strong in compression but weak in tension. (a) An arch
bridge is under mainly compressive loads, and therefore can be safely built
from concrete. However, (b) a simple beam bridge is under compressive and
tensile loads, and if built from concrete, its under-surface would crack and
fail. (c) The solution is reinforced concrete, which is a composite of steel
bars and plain concrete. Steel is strong in tension and supports the bridge at
the tensile under-surface.
Fig. 2.18 A shelf-stable carton for storage of liquid contents is a
laminated composite. It contains on average 74% cardboard, 22%
polyethylene and 4% aluminium. The cardboard is a cheap, light and flexible
material with adequate strength. The polyethylene coatings form waterproof
barriers. The aluminium coating prevents oxygen diffusing through the
container and causing the contents to oxidise and decay. Another example of
laminated composite is plywood.

Composites in orthopaedics
Composites have a wide range of applications in orthopaedics. The
following are a few examples of the more prominent uses of composites in
orthopaedics.
Plaster of Paris is a composite of a fabric bandage (reinforcement phase)
impregnated with gypsum cement (matrix phase). The gypsum cement forms a
paste when mixed with water and hardens into a solid to form the required
splints and casts. The newer synthetic cast materials are also composites that
consist of fibreglass in a polymetric matrix.
Carbon fibre composites are gaining an increasing role in orthopaedics.
They are already widely used in two types of orthopaedic hardware: in
components of external fixators (pins, rings and bars) and in limb
replacement devices (artificial limbs and braces). Carbon fibre composites
typically consist of carbon fibres (reinforcement phase) in a polymer such as
polyether sulphone (matrix phase). These composites have high stiffness and
strength and low density. They also have excellent resistance to chemicals
and heat, which is a function of the matrix phase. Together, this makes the
devices produced lightweight, easy to handle, radiolucent and more durable.
Lighter-weight devices reduce the energy expenditure during mobilisation.
Therefore, limb replacement prostheses made from carbon fibre composites
are better fitting and produce fewer sores and abrasions on limb foundation
than the conventional heavier prostheses.
Carbon fibre composites also have an emerging role in the development of
various other orthopaedic apparatus, such as in components of the operating
table, instruments (e.g. screwdrivers, taps and drills), and in internal fixation
implants (e.g. plates and nails). Carbon fibre equipments are lighter and
radiolucent, and therefore are easy to handle and do not obstruct intra-
operative fluoroscopic imaging. Implants made from carbon fibre composites
have a stiffness similar to bone, and therefore are less likely to cause stress-
shielding of bone or produce a fracture from a stress raiser effect. Carbon
fibre fatigue resistance is also far superior to that of conventional materials.
In addition, as the implants are radiolucent, the anatomical structures can be
seen clearly, despite the intervening implants. However, radio-opaque marks
are added to the implants to define their outline in radiographs.
Another application of composites in orthopaedics is in the form of
synthetic sutures. As an example, FibreWire® is a suture that is a laminated
composite of UHMWPE core covered in a braided jacket of polyester and
UHMWPE. This combination of materials produces a suture with superior
strength and abrasion resistance superior to conventional materials. It is used
in situations where high strength repairs are required.
Finally, a composite is also formed when an implant, such as a plate, nail
or prosthesis, is applied to bone. This is a laminated composite where the
two materials are arranged in distinct layers. The bone–implant composite
has different mechanical properties to the individual components (Tables 2.5,
2.6).

Table 2.5. Plaster of Paris

 The chemical name of plaster of Paris is calcium sulphate; it is
commercially also known as gypsum plaster. The term plaster of
Paris comes from Paris, France, where it was first widely used in
construction, art and medicine. However, it has been in medical
use since 1852, when Mathysen, a Dutch military surgeon, used
plaster of Paris bandages to make splints. The plaster of Paris
bandage consists of a roll of muslin stiffened by dextrose or starch
(reinforcement phase) and impregnated with gypsum cement
(matrix phase). The gypsum applied to the bandages is anhydrous
(dry) and produced by heating crystalline gypsum to around
150°C:

This reaction is reversed when water is added back to anhydrous

gypsum. The average setting time for plaster of Paris is between 3
and 9 minutes. Movement of the material whilst it is setting
grossly weakens the cast. The setting time is increased with the
use of cold water or addition of sugar into water, and is reduced
with the use of hot water or addition of salt into water. The
average drying time for plaster of Paris is between 24 and 72
hours. The cast achieves its optimal strength when it is completely
dry.

Table 2.6. Fibre-reinforced polymer composites

Glass fibre-reinforced polymer (GFRP). This is popularly known

as fibreglass, and is one of the commonest composites. It consists
of glass fibres embedded in a polymeric matrix. Glass fibres are
stiff, strong and brittle, whereas the polymer is flexible, weak and
ductile. GFRP has a combination of these properties.
Carbon fibre-reinforced polymer (CFRP). This consists of carbon
fibres embedded in a polymeric matrix. CFRP is stiffer and
stronger than GFRP. CFRP in fact has one of the best strength to
 weight ratio of all engineering materials. It is also used in the
aerospace industry and in advanced sporting equipment. The main
drawbacks of CFRP are that it is relatively brittle and much more
expensive to produce than GFRP.

Bone I
Molecular structure
Bone is a highly specialised connective tissue. It is a biological composite,
and its structural arrangement is much more complex than that of the
engineering composites. Like all connective tissues, it consists of cells and
an organic extracellular matrix. However, in addition uniquely it also has a
high content of inorganic mineral salts.
In terms of a composite, bone consists of mineral salts (ceramics) as the
reinforcement phase and collagen proteins (polymers) as the matrix phase.
The mineral salts take the form of small crystals and are arranged in an
orderly pattern within a collagen protein network. The key mineral salts
present in the bone are calcium phosphate (hydroxyapatite) and calcium
carbonate, and the collagen proteins are organised into type 1 collagen
fibres. The mineral salts contribute stiffness and hardness to the strength and
toughness of collagen fibres. The hydroxyapatite-reinforced collagen fibres
form the basic ‘building block’ of bone.
At the microscopic level, bone has the structure of a laminated composite,
which provides it with an extraordinary resistance to fatigue failure. The
reinforced collagen fibres are organised into separate layers (laminae) of
sheets and cylinders (osteons with central Haversian canals), which must all
be individually broken to completely fracture the bone. As a result, although
bone is filled with microscopic defects, its laminated organisation prevents
bones from progressing into complete fractures under normal conditions.
At the macroscopic level, bone consists of two distinct types of tissues:
cortical (compact) bone and cancellous (trabecular) bone. These are also
organised into separate layers, which enhances the bone's laminated
structure. Cortical bone forms the walls of the bone. It is dense (5%–30%
porous) and stiff and is similar to very hard plastic. Cancellous bone
occupies the central space within the bone. It is much less dense (90%
porous) but considerably more malleable. The mechanical properties of the
two tissues are determined by their density.
Bone is a dynamic tissue and its base materials are in a state of constant
turnover. The osteoclasts dissolve the minerals and collagen, and the
osteoblasts lay down new minerals and collagen. This remodelling process
enables the bone to alter its configuration (composition and/or shape) in
response to the functional demands. This connection between bone
remodelling and the applied load is summarised by Wolff's law, which states
that the rate of bone turnover is proportional to the mechanical stresses
experienced by the bone. Therefore, bone adapts to produce the minimum-
mass structure to withstand any sustained applied load.

Fig. 2.19 The composition of bone. Bone consists of approximately 45%

mineral salts, 35% collagen proteins and 20% water by weight.
Fig. 2.20 The hierarchical structural organisation of bone gives it
exceptional characteristics. It is very light for a structure that has the
mechanical properties equivalent to that of reinforced concrete. In fact, the
entire human skeleton makes up only about 20% of the total body mass.

Fig. 2.21 The trabecular patterns in the proximal femur are the result
of bone remodelling according to the loads sustained. Wolff's law explains
the variation in bone density in different regions of the same bone, and the
changes in bone density after prolonged periods of disuse or increased use.
Bone II
Mechanical properties
Bone is the third hardest structure in the body, after only dentine and enamel
found in the teeth. However, the most essential mechanical property of bone
for its daily functions is the stiffness. The stiffness of bone enables it to resist
deformation under load and maintain the body's upright posture. The
composite arrangement of different materials gives bone its complex and
versatile mechanical properties.
In common with most engineering materials, stiffness and brittleness of
bone are interrelated, which in the case of bone are a measure of its ceramic
mineral contents. The dramatic effects produced by the change in mineral
contents on bone's overall properties are observed between immature and
mature bone. In children, immature bone consists mostly of collagen and is
relatively less mineralised, but in adults mature bone has the opposite
composition. As a result, immature bone is much less stiff but pliable,
whereas mature bone is very stiff but brittle. These are optimum mechanical
properties for the two different types of functional demands: children are
involved in frequent, low-energy injuries, whereas adults need more support,
speed and agility. The other important mechanical characteristic that bone
inherits from its mineral contents is its variable strength: just like ceramics,
bone is strong in compression but weak in tension and shear.
Bone is also anisotropic and viscoelastic. These properties are derived
mainly from its collagen composition. Reinforced collagen fibres are aligned
longitudinally. These fibres are essentially long chains of polymers. There
are primary covalent bonds between the molecules of the chains, but mainly
secondary bonds between adjacent chains. Therefore, a longitudinally
applied load is working against stiff and strong bonds, whereas a
transversely applied load is working against much weaker bonds. Bone is
therefore stiffer when loaded longitudinally than transversely (anisotropy).
Polymers are characteristically viscoelastic due to their molecular
arrangement. Polymers have complex tertiary structures so that long chains
are loosely tangled together. When a load is applied, these chains uncoil
before secondary and then primary bonds are broken. A longer period of
loading allows the chains to unravel further, producing more deformation
than a shorter period of loading. It follows that collagen fibres and therefore
bones become stiffer, stronger and tougher at higher rates of loading
(viscoelasticity).
The mechanical properties of bone change due to degenerative changes in
the ageing bone. The cortical and cancellous sections are less dense in the
ageing bone; therefore ageing bone becomes weaker, less stiff and more
brittle.
For all of the above reasons, there is no single value for stiffness, strength
and hardness of bone.

Fig. 2.22 These charts illustrate the mechanical properties of bone. (a)
Bone is a brittle material and is stronger in compression than in other loading
modes. (b) It is stiffer and stronger when loaded longitudinally than
transversely (anisotropy) and (c) at higher rates of loading (viscoelasticity).
(L = Longitudinal stress; T = Transverse stress)
Fig. 2.23 Bone mass life cycle. The mechanical properties of bone are a
function of its mass and therefore density. Stiffness and strength of bone vary
with the third power and square of its density, respectively.

Mechanical properties of common orthopaedic biomaterials

Ultimate
Elastic Yield
tensile
Material modulus strength
strength
(GPa) (MPa)
(MPa)

Alumina (Purity level 394 – 260–300

99.5%)

Bone cement 3.5 20 70

(Polymethylmethacrylate)

Cobalt–Chrome (F75*) 220–230 275– 600–1785

1585
 Ultimate
Elastic Yield
tensile
Material modulus strength
strength
(GPa) (MPa)
(MPa)

Cancellous bone 1–11 – 2–20

Cartilage 0.02 – 4

Cortical bone 7–30 80 50–100

Plaster of Paris 1 20 70

Polyethylene 1 20 40

Stainless steel (316L*) 205–210 170–750 465–950

Titanium (Grade 4*) 105 692 785

Titanium (Ti-4Al-6V*) 110 850–900 960–970

Zirconia 207 – 248

Notes:

The values provided are for comparison only; the actual properties of any material depend on a
number of factors, e.g. composition of constituents and processing conditions.

* This is the composition of alloy most widely used in orthopaedics.

Further reading
Ann KN, Hui FC, Murrey BF et al. (1981). Muscles across the elbow joint:
a biomechanical analysis. J Biomech. 14: 659–696.
Bucholz RW, Heckman JD, Court-Brown CM (eds) (2006). Rockwood and
Green's Fractures in Adults. 6th edn. Philadelphia: Lippincott, Williams
 and Wilkins.
Callister, WD (2007). Material Science and Engineering: An Introduction.
7th edn. New York: Wiley.
Curry JD (2006). Bones: Structure and Mechanics. London: Princeton
University Press.
Dobson K, Grace D, Lovett DR (1998). Collins Advanced Science –
Physics. London: Collins.
Floyd RT, Thompson CW (2011). Manual of Structural Kinesiology. 18th
edn. London: McGraw-Hill.
Golish SR, Mihalko WM (2011). Principles of biomechanics and
biomaterials in orthopaedic surgery. JBJS-A. 93(2): 207–212.
Hall, SJ (1999). Basic Biomechanics. 3rd edn. London: McGraw-Hill.
Johnson K (2001). Physics for You. London: Nelson Thornes.
Johnson K (2006). New Physics for You. London: Nelson Thornes.
Johnson K, Hewett S, Holt S, Miller J (2000). Advanced Physics for You.
London: Nelson Thornes.
Kutz M (2002). Standard Handbook of Biomedical Engineering and
Design. New York: McGraw-Hill.
McLester J, Pierre PS (2008). Applied Biomechanics: Concepts and
Connections. New York: Thomson Wadsworth.
Miles AW, Gheduzzi S (2012). Basic biomechanics and biomaterials.
Surgery. 30(2): 86–91.
Navarro M, Michiardi A, Castano O, Planell JA (2008). Biomaterials in
orthopaedics. J R Soc Interface. 5(27): 1137–1158.
Nordin M, Frankel VH (2001). Basic Biomechanics of the Musculoskeletal
System. 3rd edn. London: Lippincott, Williams and Wilkins.
Ramachandran M (ed) (2007). Basic Orthopaedic Sciences: The Stanmore
Guide. London: Hodder Arnold.
Rodriguez-Gonzalez FA (2009). Biomaterials in Orthopaedic Surgery. New
York: ASM International.
Watkins J (1999). Structure and Function of the Musculoskeletal System.
New York: Human Kinetics.
Part II Engineering theory applied to
orthopaedics
This diagram highlights that physical structures are non-uniformly loaded.
Bending and torsional loads produce most of the stress on the surface of a
structure. So, a hollow tube, such as a long bone, can be almost as stiff and
strong as a solid tube of similar size, but being composed of less material is
lighter. Also, structures naturally contain geometric flaws, such as voids and
notches, which can cause local concentration of stress to several times the
average stress level in the material, e.g. a hole can concentrate stress by a
factor of three. These factors influence the design, application and
performance of orthopaedic devices.
 3 Modes of loading in the musculoskeletal
system

Introduction
This chapter develops on from the force–material interactions considered in
the previous chapters. It looks into how solid structures withstand various
types of forces. There are four basic modes of loading: compression, tension,
bending and torsion. One mode usually dominates in a given loading
situation. Therefore, structures can be labelled to highlight their loading
mode:

Columns carry compressive forces.

Ties carry tensile forces.
Beams carry bending forces.
Shafts carry torsional forces.

The applied forces generate stress within the structure. The performance of a
structure then depends on the properties of the material that it is made from,
i.e. material properties, and how and where that material is distributed, i.e.
geometric properties.

Material properties
Material properties determine the overall characteristics or personality of a
structure. Material properties are the mechanical properties derived from the
stress–strain curve, e.g. stiffness, strength and toughness. These properties
determine how a structure performs under different types of load, e.g. ductile
materials withstand tension better than brittle materials. Material properties
are intrinsic to each material and are unaffected by how the material is used.
The structure fails when the stress produced within the material reaches its
failure limit. Therefore, material properties set the stress limit at which a
structure fails. These properties have been considered in detail in the
previous chapters.

Geometric properties
The size and shape of a structure determines how much stress (and strain) is
produced in the material under a given force. The orientation of the applied
force to the structure also affects the internal stress levels. The geometric
properties therefore manage the stress levels in a structure.
Geometric properties can be considered in relation to cross-sectional and
longitudinal profiles. Both profiles have a significant bearing on the load-
carrying capacity of a structure. However, cross-sectional geometry has a
more direct interaction with the applied force, because the cross-section of a
structure forms the plane to which the load is applied, i.e. the plane of
loading. Therefore, cross-sectional geometry is further considered in the
analyses of the mechanics of different modes of loading. It is assumed that the
cross-sectional geometry remains constant along the entire length of a
structure.

Fig. 3.1 There are four basic modes of loading. Structures can be labelled
to highlight their function.
Fig. 3.2 Material and geometric properties together determine a
structure's ability to carry loads. It is similar to how the physical
properties of a building are a product of the properties of the building
materials and the properties of the completed structure.

Compression and tension

Solid structures that carry compressive forces are referred to as columns,
and those that carry tensile forces are referred to as ties. When considering
the deforming effects of a force, it is important to define the alignment of the
force to the structure.

Axial vs transverse force

An axial force acts along the axis of the structure. A transverse force acts
perpendicular to the axis of the structure.

Normal vs shear force

A normal force acts perpendicular to the plane of loading. Stress produced
by a normal force is termed ‘direct’ stress.
A shear force acts parallel to the plane of loading. A shear force produces
shear stress. Direct stress changes the shape and volume of a structure,
whereas shear stress only changes the shape.

Compression vs tension
A compressive force is directed towards the centre of a structure and acts to
flatten or squeeze it. A tensile force is directed away from the centre of the
structure and acts to pull it apart. Compression and tension by arrangement
are axial and normal forces. Similarly, compressive and tensile stresses are
normal stresses. On the other hand, a transverse force by arrangement is a
shear force and produces shear stress.

Stresses produced by compression and tension

Both types of forces generate stresses within the structure. Just as orientation
of applied force to the plane of loading affects whether the force and stress
produced are normal or shear, changing the orientation of a plane within the
structure to the applied force also affects the type of force and stress acting
on that phase. Therefore, different planes within a structure experience
different combinations of forces and stresses.
These physical principles are put together to derive the following fact
about compressive and tensile loading: under normal tension or compression,
the plane of maximum shear stress is at 45° to the axis of loading. This is
because a force as a vector quantity can be broken down into its component
forces. The magnitude of a component force is highest at 45° to the applied
force. Therefore, the plane at 45° to the applied force experiences maximum
shear stress. Mathematically, maximum shear stress is always exactly half the
normal stress. However, physical structures resist shear least well and often
fail along the plane of maximum shear stress.

The effect of cross-sectional geometry on the

properties of a column or tie
The strength of a structure under tension or compression is proportional to its
cross-sectional area. Stress is calculated by dividing force over area;
therefore the larger the cross-sectional area, the less the stress produced
under a given force. A solid column is thus stronger than a hollow column of
a similar size; it has more area and mass to withstand the applied force.

Fig. 3.3 The alignment of force to the structure is important when

considering the deforming effect of the force. An axial force acts along the
axis of the structure, whereas a transverse force acts perpendicular to the
axis of the structure. A normal force acts perpendicular to a plane of loading,
whereas a shear force acts parallel to it. An oblique plane within a structure
experiences a combination of normal and shear forces and stresses.
Fig. 3.4 A free-body force diagram of an object on an inclined surface.
Reaction force from the surface on the object is a normal force, whereas
friction is a shear force. Weight always acts in a vertical direction. It can be
divided into its rectangular components, which here are taken to be
perpendicular and parallel to the inclined surface. The perpendicular
component of weight is equal and opposite to the reaction force. The parallel
component of force is equal and opposite to friction. All forces are balanced
and the object is in equilibrium. This example shows that a plane that is at an
oblique angle to a force, in this case the inclined plane in relation to weight,
is subjected to a combination of normal and shear component forces of that
force.
Fig. 3.5 This basic spring-ball model roughly shows how an applied force
produces different types of stresses along different planes within a
structure. Here, the applied force produces compression, tension and shear
stresses in springs in different planes. The structure would fail along the
plane of springs that resists the internally generated stresses least well; this
is not necessarily the plane of the applied force. Compression, tension and
shear are the basic stresses, i.e. deformation mechanisms. Other loading
modes act through a combination of these stresses.

Bending I
A beam is a solid structure that supports bending forces. The beam's length is
considerably longer than its cross-sectional dimensions; typically at least ten
times longer.
Beams are classified according to how they are supported. The basic types
are as follows:

Simple (or simply supported) beam: This spans two simple supports.
Cantilever beam: This has a fixed support at one end and no support at
the other end.
 Overhanging beam: This spans two simple supports with one or both
ends extending beyond the supports.

Mechanics of a beam
This section describes the internal and external forces acting on a beam. A
beam works by transferring the applied bending force as a compression force
to adjacent supports. The applied force is transverse and acts perpendicular
to the axis of the beam. It is therefore a shear force. It produces two types of
internal forces within the beam: shear forces and bending moments. These
internal forces are then transferred to the supports. Then, according to
Newton's third law, the supports also exert an external reaction force back to
the beam. Therefore, a beam is subjected to two types of external forces:
applied and reaction forces. When combined together, the reaction forces are
equal and opposite to the applied forces, and therefore the beam is in
equilibrium.
The internal forces are not uniformly distributed, but instead vary along the
length of the beam. These are usually represented in shear force and bending
moments' diagrams.

Simple beam theory

Simple beam theory describes the stresses produced by bending forces. A
simple beam with a single applied force is subjected to three points of
loading (also known as three-point bending). The applied force produces
compressive stress on its side and tensile stress on the opposite side of the
beam. As a result, the length of the beam decreases on the compression side
and increases on the tension side. The beam also contains a neutral axis,
where the length of the beam is unchanged. There are neither tensile nor
compressive stresses acting at the neutral axis. Therefore, beam material is
neutral, or unreactive, to the applied load. The beam is under tensile stress
on one side of the neutral axis, and compressive stress on the other side.
These stresses increase with increasing distance away from the neutral axis,
and are highest at the outer surface of the beam. Therefore, beam material
further away from the neutral axis deforms more, i.e. changes more in length.
The material at the neutral axis is unreactive to bending forces, and in theory
can be removed from the beam without affecting its strength.
Fig. 3.6 Basic types of beams: (a) simple beam; (b) cantilever beam; (c)
overhanging beam.

Fig. 3.7 Mechanics of a beam. Shear force and bending moments' diagrams
for a simply supported beam subjected to a single point load acting in the
middle. The applied force produces shear forces and bending moments
within the beam, which vary along its length. (L = length, F = force).
Fig. 3.8 Simple beam theory. A beam tends to change shape when
subjected to an applied force. It experiences compressive stress on the side
of the applied force and tensile stress on the opposite side. The neutral axis
within the beam experiences neither compressive nor tensile stresses. The
stresses in the beam increase with increasing distance away from the neutral
axis, and are highest at the outer surface of the beam.

Bending II
The effect of cross-sectional geometry on the
properties of a beam
Simple beam theory shows that, in bending, stresses are unevenly distributed
within a structure; stresses are higher at the outer surface than at the centre of
a structure. Therefore, although a structure's strength in axial compression
and tension is determined by its cross-sectional area, i.e. mass, a structure's
strength in bending is related to its cross-sectional area and how the material
is distributed about the neutral axis.
The cross-sectional distribution of material in a beam is described
mathematically as ‘area moment of inertia’. The area moment of inertia
quantifies the bending resistance (stiffness) of a given cross-section: the
larger the area moment of inertia, the greater the bending resistance and so,
the less the stress produced within a structure under a given bending force.
Different cross-sectional shapes provide different resistances to bending.
Therefore, two beams made of the same material with equal cross-sectional
areas, but different cross-sectional shapes would be equally strong in
compression and tension, but have different strengths in bending.
 Simple beam theory and the principle of area moment of inertia show that
structures are more resistant to bending when the cross-sectional material is
distributed further away from the neutral axis. Therefore, it is possible that,
with careful planning, material in the centre of a beam may be removed
without significantly affecting its ability to withstand bending forces. The
bending stiffness of a hollow beam is related to the thickness of its walls and
the outer diameter; the outer diameter usually has a much greater effect than
the thickness of the walls. The main drawback of a hollow beam is that, if it
is loaded beyond its strength, it fails more suddenly than a solid beam, as
there is less material for a fracture to propagate across.
The cross-sectional geometry of orthopaedic implants determines their
bending stiffness. For example, an intra-medullary nail has a cylindrical
cross-section, therefore its bending resistance is proportional to its radius to
the fourth power, whereas a plate has a rectangular cross-section and so its
bending resistance is proportional to its thickness to the third power. Also,
based on inherent moment of inertia, a hollow nail must be wider than a solid
nail for an equal bending resistance. The material properties of implants also
affect their bending stiffness, e.g. the stiffness of titanium alloy is about half
that of stainless steel, and therefore a titanium plate would have half the
stiffness of an identical plate made of stainless steel. However, in general
situations, the cross-sectional geometry dominates the material properties in
determining bending stiffness, and therefore a titanium plate could be made to
be as stiff as stainless steel plate by slightly increasing its thickness (although
this is normally not required).
Fig. 3.9 A beam's stiffness depends on the distribution of the cross-
sectional material relative to the applied load. A plastic ruler can be bent
more easily in one direction than in the other. Although the cross-sectional
area of the ruler is constant, its stiffness increases when more material is
aligned in the direction of applied force. I-beams in the construction industry
take advantage of this concept.
Fig. 3.10 This table summarises the relationship between common
cross-sectional shapes and area moment of inertia (bending stiffness).
The neutral axis for all these regular shapes is located at the geometric
centre. Structures become more resistant to bending when the cross-sectional
material is distributed further away from the neutral axis.

Torsion
A shaft is a solid structure that carries torsional forces. A torsional force is
produced when equal and opposite torques are applied to the ends of a
structure, causing it to twist along its axis and undergo angular deformation.
A torsional force produces complex effects on the shaft. The basic principles
of torsional loading are illustrated by a cylindrical shaft. It is assumed that a
cylindrical shaft does not experience ‘warping’ effect when subjected to a
torsional force, i.e. its circular cross-section remains circular (whereas the
shape of square/rectangular cross-section would change as the shaft is
twisted).

Stresses produced in torsion

The stresses produced by torsional forces are shown by drawing squares and
diamonds on the surface of the cylinder. A square becomes a parallelogram;
this is shear-type deformation (see pages 52–53) and shows that shear
stresses are produced in the planes parallel and perpendicular to the axis of
the loading. A diamond becomes a rectangle; this is compression- and
tension-type deformation and shows that compressive and tensile stresses are
acting at 45° to the axis of loading. In fact, torsion produces pure shear in the
planes parallel and perpendicular to the longitudinal axis, and tension and
compression in planes at other angles. Tensile and compressive stresses are
highest at 45° to the longitudinal axis. Tensile and compressive stresses are
driven by, and add up to, shear stresses produced by the torsional force.
In the same way as in a simple beam, magnitude of stresses in a shaft also
varies along its thickness. A shaft has a neural axis at the centre of its cross-
section where no stresses act on the material. Stresses then increase with
increasing distance away from the neutral axis, and are highest on the outer
surface of the shaft.

Effect of cross-sectional geometry on properties of

a shaft
Just as in a simple beam, stresses are non-uniformly distributed within a
shaft; stresses are higher at the outer surface than at the centre. Therefore, a
structure's torsional strength is also determined by the cross-sectional area
and how the material is distributed about the neutral axis. A structure
develops a higher resistance to torsion as the material is distributed further
away from the neutral axis.
The cross-sectional resistance to angular deformation is measured
mathematically in terms of ‘polar moment of inertia’: the larger the polar
moment of inertia, the greater the resistance to angular deformation and so
less stress is produced within a structure under a given torsional force. Polar
moment of inertia (J) is derived from inherent moment of inertia (I) with the
following equation:
J [m4] = 2 × I [m4]

Therefore, polar and inherent moments of inertia essentially represent the

same basic measurement: the distribution of material about its neutral axis.
Fig. 3.11a-c Basic principles of torsional loading are illustrated by a
cylindrical shaft. A cylindrical shaft maintains a circular cross-section when
subjected to a torsional force. (a) The squares drawn on the surface of the
cylinder show shear-type deformation. This indicates that torsion generates
shear forces and stresses in planes parallel and perpendicular to the axis of
loading. If the cylinder is prismatic, i.e. it has a uniform cross-section
throughout its length, then all squares along the length of the cylinder become
identical parallelograms, indicating that shear stress remains constant along
the length of a prismatic shaft. (b) The diamonds drawn on the surface of the
shaft show compression- and tension-types of deformation. (c) Torsion
therefore produces a complex series of interrelated stresses.

Material and geometric properties of long bones

The musculoskeletal system is subjected to widely variable loads. Bones
have evolved into perfect scaffolds for withstanding all four modes of
loading. As dynamic structures, bones can change with age and in response to
applied loads and have the ability to self-repair. The material and geometric
properties of bones are finely adapted to their mechanical functions.

Material properties
Bones consist of two distinct sections: cortical bone on the outside and
cancellous bone on the inside. Cortical bone is dense and stiff, and almost
wholly supports all the forces. Cancellous bone is much less dense and much
more ductile. It is excellent in resisting compressive and shear forces.
Overall, bone is a brittle material and is therefore much stronger in
compression than in tension.

Geometric properties
Some of the highest stresses in the bones are generated by bending and
torsion. The long bones have cross-sectional geometry of a hollow cylinder.
This optimises their resistance against bending and torsion, without
compromising their strength in compression and tension.
Bones of children and adults can be considered to be made of different
materials: the immature bone is much less mineralised and therefore ductile
and weak, whereas mature bone is much more mineralised and therefore
stiffer and stronger but brittle. Therefore, a change in bone material makes
bones stiffer and stronger with age. However, the difference in stiffness and
strength of immature bone and mature bone cannot be fully explained by a
change in bone composition only. The growing bones also develop more
resistant cross-sectional geometry: the expanding diameter of growing bones
considerably increases their resistance to bending and torsion. In fact, the
relationship of the bone's cross-sectional diameter to length is critical in
children. The long bones form lever arms for muscle function, and in
children, as the long bones increase in length, there is a compensatory
increase in the cross-sectional diameter to maintain strength.
In general, the size of bones in adults seems to be predetermined by
genetics and nutrition. However, the thickness of cortical walls is greatly
affected by applied loads. Exercise, especially weight training, can
significantly increase the cross-sectional area of bones. Muscle contractions
generate some of the highest stresses on the bones. Therefore, as resistance
exercises increase the size of muscles over time, the underlying bones
remodel to develop thicker cortices. Thicker cortices lead to a larger cross-
sectional area and area/polar moments of inertia, which strengthen bones for
the increased demands. The relationship between muscle function and bone
mass is so critical that, throughout the animal kingdom, an individual's total
bone mass is a function of its total muscle mass, i.e. lean (and not total) body
mass (Table 3.1).

Fig. 3.12 An illustration of the influence of cross-sectional geometry of

bone on resistance against the four modes of loading. It is assumed that the
bones have identical composition and length.
Fig. 3.13 This diagram illustrates the change on cortical diameter of the
femur with age. In the growing bone, cross-sectional diameter increases
with bone length to maintain strength. In the ageing bone, osteoporosis affects
the bone material – making it mechanically weaker – but a compensatory
increase in bone diameter preserves whole bone strength.

Table 3.1. Allometry

Allometry is the study of the relationship between the size of an

organism and the size of its body parts. Allometric studies show
that, in an animal, the size of bones is proportional to its body
mass as follows:

These correlations are confirmed by studies investigating bone

dimensions in animals with body mass differing by six orders of
magnitude.
 4 Biomechanics of fracture

Fundamentals of fracture
In mechanics, a fracture is the separation of an object into pieces due to an
applied load at a temperature below its melting point. Materials fracture
through one of two modes: brittle and ductile. Both modes involve two basic
steps of the fracture process: crack initiation and propagation. The main
difference between the two modes is in the mechanism of crack progression.
All materials are rough and contain defects and cracks at microscopic level.
Crack initiation occurs when the applied load leads to sufficient stress in the
material to increase the size of a crack. Crack progression in a brittle
fracture is associated with little plastic deformation, whereas ductile fracture
involves significant plastic deformation. Materials therefore can be labelled
as brittle or ductile to describe their performance under load.

Brittle fracture
In brittle fracture, a crack almost always initiates on the surface of the
material. It then propagates almost perpendicular to the applied stress,
leaving a flat fracture surface. The crack is referred to as ‘unstable’ as it
propagates spontaneously when the stress threshold for crack initiation is
reached. There is relatively little energy absorbed in the fracture process.
Brittle fracture therefore occurs rapidly with little or no pre-warning. It is
usually a catastrophic situation, as there is no opportunity to anticipate an
impending fracture or to repair accumulating damage to prevent complete
failure.
Cortical bone is a brittle material and sustains brittle fracture. Other brittle
materials include ceramics, ice, cast iron and metals at low temperatures.
Brittle materials display little plastic deformation before failure and have the
tendency to break into many pieces.

Ductile fracture
In ductile fracture, a crack initiates in the substance of the material. The
material undergoes extensive plastic deformation at the site of the crack
propagation. The crack is referred to as ‘stable’ as it resists further extension
unless increased stress is applied. There is substantial energy absorption in
the fracture process. The plastic deformation and incremental fracture
progression produce a ‘cup and cone’ fracture surface.
Most metals are ductile at room temperature. Ductile materials display
large deformation before failure and usually break into two main pieces.
Ductile materials are preferred for most engineering applications. Ductile
materials are ‘tough’ as they absorb more energy than brittle materials. They
also show warning signs of fracture and therefore present an opportunity to
intervene before complete failure. Ductile materials are considered to be
forgiving, since their toughness usually allows them to tolerate excessive
loads. The properties of ductile materials can also be enhanced through
strengthening mechanisms.
Brittle fracture is the commonest mode of fracture of engineering
components (even those made from ductile materials). This is because most
engineering components undergo fatigue failure, in which the fracture
propagates in brittle mode (Table 4.1).
Fig. 4.1 Ductile fracture involves a number of steps. In contrast, brittle
fracture occurs rapidly in a single step. Ductile fracture produces a ‘cup and
cone’ fracture surface, whereas brittle fracture creates a flat fracture surface.
Fig. 4.2 This stress–strain curve shows deformation produced in brittle
and ductile fractures. The area under the curves represents energy absorbed
by the material until failure. Ductile materials are ‘tough’ as they absorb
more energy than brittle materials.

Table 4.1. Ductile to brittle transition in metals

The Titanic was the largest and most advanced ship of its time in
1912. It sank after colliding with an iceberg during its maiden
voyage. Steel used to make the Titanic was tested to be adequately
tough to withstand all predictable damages. However, steel taken
from the wreck of the Titanic showed signs of brittle failure, i.e. a
flat fracture surface with very little plastic deformation. The steel
was ductile and tough in the tests performed at room temperature,
but brittle and weak at sub-zero sea temperatures. Metals show a
transition from ductile to brittle failure at low temperatures.
Brittle fracture is sudden and catastrophic, and engineering
designs now anticipate and prevent brittle failures in all
components.
Mechanism of bone fracture
Cortical bone is anisotropic and brittle, although immature cortical bone is
more ductile. In common with other brittle materials, cortical bone is
strongest in compression, weakest in shear and intermediate in tension. These
mechanical properties determine the mechanism and pattern of bone fracture.
The forces acting on the bone also determine the pattern and location of
fracture in the bone. As bone properties and the forces both determine
fracture pattern, the same forces can produce different fracture patterns in
mature and immature bones due to their different mechanical properties.

Mechanism of fracture
A bone fracture is a break in the continuity of bone. The bone typically
fractures in a brittle mode. There are two phases to the fracture process:
initiation and propagation. The fracture usually initiates on the outside
surface of the bone, because this is where stresses are usually highest under
most loading conditions. The bone surface is not smooth and polished, but is
irregular and full of microscopic defects and cracks. There are also localised
regions of relative osteopenia in the bone, where normal remodelling
process is under way. These factors together form stress concentration
points, one of which becomes a fracture initiation site when a crack begins to
increase in size. After initiation, the crack grows in size and propagates
rapidly until complete failure.

Acute vs fatigue fracture

Acute fracture occurs when the load applied to an object exceeds its ultimate
tensile strength. Most bone fractures are acute. A ‘pathological’ fracture
occurs when the strength of bone is reduced (e.g. due to osteoporosis or
malignancy) so that it fails below the usual stress threshold.
An object subjected to cyclical or fluctuating loads is susceptible to
fatigue or stress fracture. Fatigue fracture results from a load that is below
the material's yield or ultimate tensile strength under static load.* The term
‘fatigue’ describes that fracture that has resulted from a prolonged period of
repetitive loading cycles. Although this is an uncommon cause of bone
failure, it is the most common cause of failure of engineering structures.
 Fatigue failure builds up slowly but occurs suddenly and without warning.
The mechanism of fatigue fracture is similar to brittle fracture mode as it
occurs rapidly with relatively little plastic deformation, even in normally
ductile materials. The fracture similarly propagates perpendicular to the
applied tensile load.

The S–N curve

The S–N curve represents the material's performance under cyclical loads. It
plots stress amplitude (S) against the number of cycles to failure (N).
Materials fall into two categories, based on whether or not they have a
fatigue limit.

Fatigue limit. If stress amplitude is reduced below a particular

threshold, fatigue failure will not occur for an infinite number of cycles.
No fatigue limit. Fatigue failure will eventually occur irrespective of
stress amplitude level (Table 4.2).
Fig. 4.3 The S–N curve represents materials' fatigue properties. It
shows that the number of cycles a material can withstand without failure
depends on stress amplitude. Some materials show a fatigue (or endurance)
limit, e.g. iron and titanium alloys, whereas other materials eventually fail no
matter how low the stress amplitude. The study of fatigue failure, especially
high-cycle fatigue failure, is a relatively new field (since the age of the steam
engine), as in the distant past machines did not go fast enough to develop
millions of stress cycles. It is of increasing interest in orthopaedics, as the
patients receiving permanent implants are increasingly younger and more
active, so fatigue failure is becoming more important.

Table 4.2. Fatigue properties of bone

Bone has interesting fatigue properties. In-vitro experiments show

that bone does not have a fatigue limit and is susceptible to fatigue
failure at even small stress amplitudes, given an adequate numbers
of cycles. In-vivo bone, however, has the capacity to repair, and
 any fatigue-related micro-damage stimulates bone remodelling.
Fatigue failure of bone only occurs when the rate of micro-damage
exceeds the rate of remodelling. Therefore, fatigue failure of in-
vivo bone is related to stress amplitude and frequency of loading.

Patterns of bone fractures I

A load applied to bone produces stress in all directions. The fracture occurs
along the weakest plane, which is normally the plane of maximum shear or
tensile stress.
A tensile load produces transverse fracture that is perpendicular to the
plane of loading. This plane experiences the highest tensile stress, and a
crack on this plane separates and propagates into a fracture.
A compressive load on the bone produces oblique fracture. The cortical
bone is strong in compression, and fracture occurs along the plane of highest
shear stress. A compressive force, as a vector quantity, can be resolved into
its rectangular components; the component at 45° is the maximum fraction of
the applied force (see pages 52–53 for details). This plane in bone is under
the highest shear stress and therefore fails, producing a short oblique fracture
that is at approximately 45° to the plane of loading.
A bending load (i.e. three-point bending) can result in a simple transverse
fracture or a ‘butterfly’ fragment. The bending load produces tensile stress at
one cortex of the bone and compressive stress at the opposite cortex. The
bone is weaker in tension than compression; therefore fracture initiates at the
cortex under tensile stress and propagates across to produce a transverse
fracture. At the same time, the opposite cortex under compressive stress can
also fail independently, initiating an oblique fracture line that joins the
transverse fracture line. This leads to fracture comminution and a ‘butterfly’
fragment, which increases in size with the amount of axial loading (i.e.
compression). Four-point bending produces a segmental fracture of the bone.
A torsional load produces spiral fracture. The torsional load generates a
complex series of interconnected stresses in different planes: shear stress is
maximum at planes parallel and perpendicular to the longitudinal axis to the
bone, and tensile and compressive stresses are maximum at 45° (see pages
58–59 for details). Tensile and compressive stresses are driven by shear
stresses produced by the torsional load. A small surface crack on the plane
of maximum tensile stress begins to increase in size and propagates through
to the core and the circumference of the bone, in a spiral pattern following
the 45° planes of maximum tensile stress. Shear stress is ultimately
responsible for fracture initiation, since maximum tensile stress is driven by
shear stress.
The average angle of spiral fracture to the longitudinal axis of bone in
experimental studies is between 30° and 40°. In clinical experience, the
spiral fracture angle varies widely between 20° to 90°. The difference
between experimental and clinical observations is because in reality fracture
progression is distorted by bone anisotropy, moments caused by contraction
of muscles attached along the length of bone and bony appendages. In
addition, an actual torsional injury is usually associated with a bending
moment, which limits the progression of the fracture along the length of the
bone.

Fig. 4.4 Basic patterns of bone fracture according to loading mode.

Fig. 4.5 A butterfly fragment fracture pattern is produce by a
combination of bending and compressive loads. The bending load initiates
a transverse fracture at the cortex under tensile stress, and the compressive
load initiates an oblique fracture at the opposite cortex. The two fracture
lines meet and a further oblique fracture line breaks off the butterfly
fragment.

Fig. 4.6 A spiral fracture pattern is produced by a torsional load. The

fracture initiates on the plane of maximum tensile stress, and propagates
along the planes of maximum tensile stress. A bending moment finally
divides the bone.

Patterns of bone fractures II

Immature bones in children have mechanical properties different from those
of mature bones in adults. Therefore, there are variations in fracture patterns
in children. Immature bones are weaker, but more ductile than mature bones.
They have a thicker and firmer periosteum, which acts as a shield around the
bone. The bone-periosteum unit has considerably more strength than the
immature bone singly, and the periosteum also limits fracture displacement.
The periosteum supplies nutrients to the bone and is more metabolically
active in children; therefore paediatric fractures unite more rapidly. Immature
bones contain physes (growth plates) that determine bone growth and rate of
fracture remodelling. The physes can be involved in fractures, which can
potentially lead to long-term bone growth disturbances.
A compressive load can produce a torus or ‘buckle’ fracture in children.
This type of fracture typically occurs in the metaphyseal region of long
bones. Immature cortical bone is more porous and less dense than mature
bone, and so is weak under compressive stress. Therefore, bone cortex under
compressive load sustains a local fracture, which does not propagate to the
opposite cortex due to the ductile nature of the bone.
A bending load can lead to a greenstick fracture in children. A greenstick
fracture is where one bone cortex is broken and the fracture propagates to the
opposite cortex, but does not completely disrupt it. This fracture pattern
reflects the ductile nature of paediatric bone. The fracture initiates at the
cortex under tensile stress, but the opposite cortex under compressive stress
undergoes plastic deformation. After initiation, the fracture propagates
longitudinally along the bone until all the imposed energy is absorbed. The
resulting fracture pattern is similar to how an actual greenstick breaks.
A paediatric bone can be so ductile that it can undergo pure plastic
deformation, without sustaining a fracture. The adult bone is brittle and does
not experience significant plastic deformation; the energy transmitted to the
bone dissipates mainly in the form of fracture. However, immature bone is
much more pliable and can plastically deform to absorb the applied energy.
As plastic deformation absorbs much more energy than brittle failure, bones
that have undergone plastic deformation are uncommonly also fractured.
Plastic deformation typically occurs under bending load.
Paediatric fractures are managed differently to fractures in adults. The
periosteum contributes significantly to fracture stability. Most fractures in
children are managed conservatively. Fractures are usually surgically treated
only if there is: significantly mal-alignment; significant displacement in a
fracture that involves the physis or joint; or if the fracture is unstable. These
fractures are stabilised with temporary fixation devices wherever possible
(Table 4.3).

Fig. 4.7 Variation in fracture patterns in children. Fracture types only

seen in children are shown in an illustration of the distal radius.

Table 4.3. Factors that determine remodelling potential of fractures

in children

Factor Effect on remodelling of fracture

Skeletal The younger the child, the greater the remodelling

maturity potential of the fracture.
 Factor Effect on remodelling of fracture

Bone Different bones have different potential to remodel.

fractured Generally, bones in the upper limb have higher
remodelling potential than bones in the lower limb.
Upper limb fractures around the elbow have a limited
remodelling potential. Lower limb fractures around
the knee have the greatest remodelling potential.

Site of Fractures in the metaphyseal region have higher

fracture remodelling potential than fractures in the diaphyseal
region. This is because the metaphysis undergoes
osteogenesis and remodelling as part of normal bone
growth.

Plane of Angulation remodelling potential is inversely related

deformity to the distance between the fracture and the nearest
joint. Angulation remodelling is highest in the plane of
the greatest movement of the joint, e.g. remodelling is
highest for volar or dorsal angulation of fractures
around the wrist.
Remodelling potential for translation and shortening
is multifactorial and related to the first three factors.
Rotational remodelling potential is very limited.

Patterns of bone fractures III

The previous sections have looked into how bone fracture patterns are
determined by the hierarchical structure of the bone, and the applied forces.
Here is a summary.

Material properties. Cortical bone is anisotropic and brittle. Therefore,

it fractures through brittle mode. The fracture starts on the surface of the
bone on a plane of highest shear or tensile stress. It then propagates
rapidly, permitting little plastic deformation of bone. Immature cortical
 bone is more ductile and shows greater plastic deformation and
variations in fracture patterns.
Structural properties. Long bones are hollow cylindrical tubes.
Bending and torsional loads are mainly concentrated on the peripheral
sections. Therefore, fracture starts on the outer surface of the bone. As
the bones are hollow, there is less material for the fracture to propagate
across.
Loading mode. The type of forces applied determine fracture pattern.

There are two additional physical factors that further influence the pattern of
bone fractures.

Arrangement of bones
Bones are just one of the building blocks of the body. Forces are applied to a
person rather than to an individual bone. The internal arrangement of bones
and other structures also affects the sequence and patterns of fractures of
bones. The ‘polo mint concept’ illustrates the significance of this ‘higher-
order’ arrangement. According to this concept, a ring-like structure, such as a
polo mint, usually does not break in only one place but in at least two or
more places. Therefore, when bones are arranged into a ring-like structure,
e.g. the pelvic ring or the ring formed by the forearm bones in children, it is
uncommon for one bone to be fractured in isolation. Therefore, the other
bones, joints and soft tissues must be assessed for occult injuries such as
fractures, dislocations or ligament sprains. Many fracture classifications are
based on the sequence of fractures of bones in close arrangements, e.g.
Lauge–Hansen classification of ankle fractures. Therefore, the pattern of
bones fracturing is also determined by the relative arrangement of the bones.

Fracture mechanics
The surface of the bone is uneven and contains defects and cracks at
microscopic level. When a material is loaded and the stress in the material
reaches a specific threshold, one of the defects increases in size and the
fracture ensues. It is a fundamental physical occurrence that only one of the
defects increases in size to produce a complete fracture instead of multiple
defects growing into incomplete fractures. This is because bone is brittle and
a propagating fracture keeps the stress in the surrounding areas of the
material below the threshold for initiating another fracture. Therefore, energy
imparted to the bone propagates one failure point to the end before initiating
the next failure point. This concept is illustrated in a paper-tearing
experiment.

Fig. 4.8 The polo mint concept. (a) A polo mint usually does not break in
just one place but instead in two or more places. Similarly, ring-like
structures, e.g. the pelvic ring (b) or the ring formed by the forearm bones in
children (c), usually do not break in one place only. An obvious injury
disturbing the ring should alter to other associated occult injuries, such as
fractures of other bones forming the ring or ligament sprains. (In adults, the
‘polo mint concept’ is less relevant to the injuries of the bones of the
forearm.)
Fig. 4.9 When a strand of dry, uncooked spaghetti is bent, it almost
always breaks at two or more places. Uncooked spaghetti is a brittle
material, and breaks with flat fracture surfaces. When the spaghetti strand is
bent and a weak point breaks, there is a momentary surge in the stress levels
along the length of the strand that leads to ‘cascade fracturing’ at the next
weak point, and so on. It may be that a fracture of a bone that is part of a ring
also produces excessive stresses at other points in the ring. As the bones
forming the ring are relatively fixed in position, increased stresses lead to a
cascade injury.

Fig. 4.10 Paper-tearing experiment: Take a plain piece of paper and

partially tear it in two places. Then, holding the piece of paper at the edges,
try to tear it into three pieces – this would not be possible. The stress
produced by the tensile forces propagates one stress raiser to complete
failure, instead of simultaneously propagating both stress raisers. This
fundamental principle of fracture mechanics also determines patterns of
injuries in the musculoskeletal system.

Stress raisers
Stress raisers are geometric flaws normally present on the surface or within
a structure. The following are examples of stress raisers:
 Macroscopic discontinuities: Voids, notches, threads, sharp corners
and sudden changes in cross-section.
Microscopic flaws: Cracks, pores, pits and surface scratches.

When a structure is loaded, these irregularities raise stress in the material in

two ways.

The defects cause a reduction in the area over which the load is
distributed, which increases the average stress in the material.
Stress also concentrates around the tip(s) of defects. The scale of local
stress magnification depends on the size, shape and orientation of
defects, i.e. different defects have a different stress concentration factor.
In general, the sharper the defect, the more severe the stress
concentration. Stress raisers often become fracture initiation sites.

Stress raisers therefore significantly reduce the strength of a structure.

Fatigue life
A fatigue fracture always starts at a stress raiser. The number and geometry
of stress raisers in a material are critical in determining the load and number
of cycles to fatigue failure. Therefore, fatigue life of a structure can be
increased by reducing the number and size of stress raisers.
Orthopaedic implants are designed, manufactured and implanted into
patients with precautions to minimise stress raisers introduced:

Design factors: Implants are designed to have optimal geometry with

the aim to avoid sharp surface discontinuities.
Manufacturing aspects: Materials for implants are carefully selected in
terms of their fatigue tolerance and, where possible, further treated to
increase their fatigue resistance. Polishing implants at the end of
machining operations removes scratches and grooves introduced during
the manufacturing process.
Surgical technique: Implants are handled and inserted with techniques
that minimise surface scratches.
Orthopaedic procedures can also produce discontinuities in the bone that
affect its overall strength. These can be divided into two types:

Stress raisers: A defect, e.g. screw/pin hole, in the bone reduces its
strength. There is a direct relationship between the size of the defect and
reduction in bone strength. A hole that is one-third of bone diameter
reduces bone strength to about 50% of intact bone. The reduction in
strength is less if the defect is filled in with another material, e.g. screw
or bone graft, than if left unfilled. The stress raiser effect of the defect
reduces with time due to bone remodelling. The sharp change in the
cross-sectional stiffness at the junction of the end of a prosthesis and
bone also creates a stress raiser. A fracture is more likely to initiate in
this region if the bone is loaded unexpectedly.
Open section defects: These defects are longer than the bone's diameter,
e.g. a bone window for infection or biopsy, and also dramatically
reduce bone strength.
Fig. 4.11 Stress raisers cause a disruption to stress flow, similar to a
rock in a stream causing turbulence to water flow. (a) A structure is
strongest when stress is uniformly distributed. Stress raisers, such as (b)
holes, (c) notches, (d) sharp corners and (e) transition zones, cause local
concentration of stress to several times the average stress level in the
material.
Fig. 4.12 This graph shows that stress is magnified at the tip of a defect
and returns to baseline within a distance of three times the diameter of
the defect from the tip. Therefore, larger defects affect a wider local area.
The impact of a stress raiser on a structure's strength is highlighted when
trying to tear open, for example, a bag of peanuts. It is very difficult to split
an intact thin sheet of plastic, but a small ‘tear here’ defect makes the bag
much easier to open. The weakening effect of a stress raiser is more
significant in a brittle material than a ductile material, and is particularly
marked in torsion. (Diagram adapted from Callister, WD (2007). Material
Science and Engineering: An Introduction. New York: Wiley, with
permission from John Wiley & Sons, Inc.)

Corrosion
Corrosion is the deterioration of a structure by chemical reaction with its
environment. It is a natural process that occurs because highly reactive
materials want to achieve a more stable state as a compound. However,
corrosion is a problem, as it removes surface material and creates stress
raisers, and therefore reduces the load-carrying capacity of the structure.
Corrosion can also induce fatigue failure of a structure under cyclical
loading. The biological environment can be very corrosive to foreign
material, and corrosion is one of the major processes that lead to implant
failure.
 As corrosion is a chemical process, it is not described as a mechanical
property of a material. Instead, materials are usually described in terms of
being immune, resistant or susceptible to corrosion. Metals are particularly
susceptible to corrosion. Deterioration of ceramics and polymers is usually
analysed as degradation rather than as corrosion.

Types of corrosion
Corrosion is commonly classified according to the appearance of the
structure undergoing corrosion or the circumstances in which corrosion
occurs. The following are clinically important types:

Uniform/ general corrosion evenly affects the whole surface area of the
structure.
Localised corrosion is concentrated over a small surface area. This is
further divided according to the appearance of the corroding area.
Pitting corrosion produces cavities (pits and cracks) on the
surface.
Crevice corrosion occurs in isolated areas (crevices) that are
usually shielded from the environment, e.g. under screw heads if
screws are not fully tightened into a plate.
Filiform corrosion affects in a random thread-like pattern.
Galvanic corrosion occurs when two metals with significantly different
electrochemical potentials have a physical or electrical contact with a
common environment.
Erosive corrosion occurs where a flowing corrosive fluid damages the
structure.
Fretting corrosion occurs between two surfaces in relative motion. The
rate of damage is accelerated because the relative motion between the
surfaces removes the corroded layer of material, exposing fresh
material for further corrosion (see pages 84–85).

Protection against corrosion

Corrosion prevention is a key consideration when designing engineering
components. Materials that have less tendency to corrode are utilised
wherever possible. Metals usually form a layer of oxide on the surface,
which protects them against corrosion. The protective oxide is ineffective in
steel (rust), as it is non-cohesive and peels off, exposing underlying layers to
the environment. It is very effective in other metals, such as stainless steel,
cobalt–chrome and titanium. Externally used components can be painted or
have other surface treatments (e.g. coated with a less reactive material) to
prevent water getting to the layer susceptible to corrosion (Table 4.4).

Table 4.4. Basic chemistry

Corrosion occurs when the combination of a material (usually a

metal) and its environment results in the formation of an
electrochemical cell. Corrosion involves an electrochemical
reaction that consists of two chemical half-reactions: oxidation
and reduction. In the oxidation process, an atom gives up
electrons, and in the reduction process, an atom gain electrons.
The site in a material where oxidation takes place is referred to as
the anode and where reduction occurs is referred to as the
cathode. The positive ions at the anode leave the surface for the
surrounding environment as corrosion products, e.g. as oxides or
hydroxides. The free electrons left behind travel to the site of the
cathode through a conductive pathway. At the cathode, the free
electrons react with positive ions available in the environment,
which restores the electric balance of the circuit. Corrosion
therefore is the electrochemical oxidation of a material.
Fig. 4.13 Corrosion reaction requires four essential conditions: an
anode, a cathode, a direct electrical connection between the anode and
cathode, and an electrolyte (containing dissolved positive ions). In nature,
air is the most common electrolyte. In the biological environment, salt
solutions (e.g. NaCl) act as the electrolyte medium.

Fig. 4.14 An experiment to show that corrosion requires oxygen and

water. Each test tube contains a nail made of iron. Nails in test tubes A and
B do not rust. In test tube A, boiled water contains no oxygen, and an oil
layer prevents the diffusion of new oxygen. In test tube B, calcium chloride
absorbs moisture from the air. The nail in test tube C rusts at a normal rate,
due to the presence of air containing oxygen and moisture. The nail in test
tube D rusts at an accelerated rate – salt water is an electrolyte (NaCl)
solution that conducts ions and therefore speeds up corrosion.

Biological process of bone fracture healing

The aim of the fracture healing process is to reconstruct the broken cortical
bone. This healing process begins immediately after a fracture and there are
two types: primary (cortical) and secondary (callus) bone healing. Both
healing methods involve three basic phases: inflammation, repair and
remodelling. Both methods have a common inflammatory phase, but they then
diverge in the repair and remodelling phases.
The inflammatory phase begins with the formation of fracture haematoma.
This releases inflammatory and repair cells needed to produce new bone.
Granulation tissue bridges the fracture site, providing limited mechanical
stability.

Primary (cortical) bone healing

The requirements for primary bone healing are that bone fragments are in
direct opposition or have only a small gap between them and are rigidly
stabilised. This is referred to as ‘anatomical reduction and absolute
stability’. These conditions are usually achieved by surgical reduction and
internal fixation, e.g. with screws and plates, of fracture.
Bone fragments in direct opposition undergo ‘contact’ healing – this is
where osteoclasts from one bone fragment make tunnels, known as cutting
cones, across the fracture into the other bone fragment. New blood vessels
and osteoblasts arrive in these tunnels and bridge the fracture with new
lamellar bone.
Bone fragments with a small gap (≤ 0.5 mm) undergo ‘gap healing’ – here
the gap size is too big for osteoclasts to cross, so osteoblasts work from one
end of the fracture and deposit lamellar bone layer by layer until the gap is
bridged.
Bone remodelling gradually restores the normal mechanical properties at
the fracture over a period of months and years.

Secondary (callus) bone healing

Secondary bone healing occurs where there is a bigger gap between bone
fragments and some controlled motion at the fracture site. This is referred to
as ‘relative stability’. These conditions exist when a fracture is managed
conservatively, e.g. in a cast, or internally fixed with implants that provide
relative stability, e.g. an intramedullary nail.
There are two stages to the repair process:

Soft callus formation. The granulation tissue is replaced by soft callus

consisting of fibroblasts and cartilage, which stabilises the fracture.
New bone tissue is laid at the periphery of the fracture site – this is
referred to as intramembranous ossification.
Hard callus formation. Intramembranous ossification continues as
before. The soft callus within the fracture gap is converted into rigid
calcified tissue, i.e. woven bone – this is referred to as enchondral
ossification. This process starts at the ends of the fracture and
progressively moves towards the centre. Intramembranous ossification
bridging the fracture on the outside completes first and enchondral
ossification bridging the bone cortices completes afterwards.

Bone remodelling begins after the fracture is fully united with hard callus. It
gradually replaces woven bone with lamellar cortical bone over a period of
months to years.

Fig. 4.15 Relative proportion of time involved in the three stages of

fracture healing. Fracture healing is a continuous process and there is an
overlap between the stages.
Fig. 4.16 Primary (cortical) bone healing consists of osteoblasts directly
depositing lamellar bone. Lamellar bone is the normal bone that consists of
parallel layers (lamellae) of collagen fibres.

Fig. 4.17 Secondary (callus) bone healing consists of intramembranous

and enchondral ossifications. In intramembranous ossification, osteoblasts
directly deposit woven bone (without an initial cartilage precursor). In
enchondral ossification, osteoblasts form an initial cartilage precursor,
which calcifies into woven bone. Long bones normally grow in width by
intramembranous ossification, and in length by enchondral ossification.
Woven bone consists of randomly organised collagen fibres.

Biomechanical process of bone fracture healing

The biomechanical conditions at the fracture site determine the time involved
in fracture healing. The two prerequisites for fracture healing are adequate
blood supply and adequate mechanical stability. Primary bone healing can
take longer than secondary bone healing if an implant fixed to the bone, in
order to rigidly stabilise the fracture, compromises its blood supply. On the
other hand, fracture repair through secondary bone healing can also be
affected if fracture stability is not optimal.
There is a complex interaction between the mechanical environment and
cellular repair processes at the fracture site. The interfragmentary strain
theory relates fracture ‘strain’ to the different types of fracture healing
tissues. Strain quantifies a material's deformation in terms of its original
length; therefore, interfragmentary strain is described as:

Interfragmentary strain therefore describes the amount of motion at the

fracture site. According to the interfragmentary strain theory, different types
of fracture healing tissues tolerate different amount of motion at the fracture
site. Therefore, interfragmentary strain:

above 100% would lead to non-union;

Between 10% and 100% would lead to granulation and fibrous tissue
formation;
Between 2% and 10% would lead to cartilage and enchondral
ossification;
Less than 2% would lead to cortical bone formation.

This explains the biomechanics of primary and secondary bone healing as

follows.

Primary (cortical) bone healing

Bone fragments are directly opposed, or are in close contact, so the initial
fracture gap is very small. Therefore, bone fragments must be rigidly fixed to
prevent any significant motion, which would equate to high strain, since the
initial fracture gap is so small. The rigid fixation ensures that
interfragmentary strain is less than 2%, leading to primary bone healing.

Secondary (callus) bone healing

A fracture with a wider gap can tolerate more motion at the fracture site and
still maintain interfragmentary strain to a reasonable level. Therefore, such
fractures only require relative stability.
Fracture stiffness is proportional to the amount of mineralised tissue at the
fracture site. The amount of mineralised tissue at the fracture site increases
with healing time. When a bone fracture occurs, fracture haematoma provides
limited stability to the fracture site. Once the interfragmentary strain is less
than 100%, granulation tissue bridges across the fracture gap. This stiffens up
the fracture and reduces interfragmentary strain. This, in turn, allows
formation of soft callus. The increased cross-sectional diameter from soft
callus further increases fracture stiffness and lowers interfragmentary motion.
The soft callus is then mineralised into hard callus, and so on. Therefore, a
cycle of progressively stiffer tissues forming at the fracture site reduces
interfragmentary strain from 100% to less than 2%, and the cortical bone
forms.

Fig. 4.18 According to the principles of moment of inertia, callus

formation at the periosteal surface significantly increases stiffness and
strength of the healing bone.
Fig. 4.19 Interfragmentary strain theory. Interfragmentary strain theory
states that strain at the fracture site determines the type of fracture healing
tissue formed. Primary bone healing takes place when the fracture gap is
anatomically reduced and the bone fragments are rigidly stabilised so that
interfragmentary strain is less than 2%. Secondary bone healing occurs when
the fracture gap is wider and when the bone fragments are relatively
stabilised. Secondary bone healing occurs in a cycle of progressively stiffer
tissues forming at the fracture site. It requires controlled interfragmentary
motion. An unstable fracture (interfragmentary strain >100%) may not heal,
despite good callus formation and may develop hypertrophic non-union. A
fracture fixed too rigidly or with too wide a fracture gap may not have
sufficient interfragmentary strain and may develop delayed union or atrophic
non-union.

Fig. 4.20 Secondary bone healing

* The other mechanical process that can cause a material to fail below its
yield strength is creep – see pages 22–23.
 5 Biotribology

Introduction to biotribology
A force may change the shape and/or state of motion of an object. The
deforming effects of force considered so far have been related to the
interactions between force and the entire object and mechanisms that result in
the failure of the whole structure. This chapter looks into the deforming
effects of force on the surface of an object, which in turn are linked with
motion parameters of the object. Therefore, in this chapter, deformation and
motion of an object are considered together.
Tribology is the study of the interactions between two solid surfaces in
relative motion. It deals with friction, wear and lubrication aspects of the
interface formed between the solid surfaces. A force acting at the interface
produces friction when the surfaces slide over each other. Excessive friction
wastes energy and leads to wear of the surfaces. A lubricant acts to reduce
friction and wear by creating a slippery film between the surfaces. The
thickness of the lubricant film formed between the surfaces determines the
effectiveness of the lubrication.
Different interfaces require a different balance of friction, wear and
lubrication. High friction is required between foot and ground for walking
and in brakes, whereas low friction is desirable between the components of a
car engine. Similarly, wear facilitates brushing teeth and is useful when
writing with a pencil, but can also be harmful and lead to the breakdown of
mechanical components such as gears. The principles of tribology are used to
optimise the performance of interacting surfaces in mechanical systems.
Biotribology is the application of tribology to biological surfaces. It is
therefore concerned with friction, wear and lubrication aspects of biological
systems in sliding contact. Biotribology also includes the study of the
interface conditions in prostheses.
Synovial joints are much more complex than general engineering
articulations. In addition, biological loading conditions are intense and
hugely variable. It is estimated that in a young person, a typical synovial joint
undergoes approximately 4700 to 5400 loading cycles everyday and that a
typical synovial joint undergoes more than 108 loading cycles in an 80-year
lifespan. Biotribology provides an insight into the functional conditions of
synovial joints, and how they maintain their remarkable performance over
prolonged periods of time. This understanding helps to guide the design and
development of bearing surfaces of prosthetic joint replacements. This
chapter looks into principles of biotribology as applied to synovial joints
and prosthetic joints.
Fig. 5.1 Biotribology deals with friction, wear and lubrication aspects of
the sliding interface formed between biological surfaces. A key interface
in the body is between the skin and the physical world. A sebum (lubricant)
film present on the surface reduces friction and wear of the skin. Friction
blisters develop when lubrication is not sufficient for the interface
conditions. These are almost exclusively confined to humans, and are most
common on hands and feet, which are involved in performing repetitive
motions. Friction blisters occur when temperature and moisture around the
skin are excessively raised (e.g. due to socks on the feet). Higher local
temperatures lead to excessive sweat, which is a less effective lubricant than
sebum. This leads to higher friction between the skin and the external
surface. Higher friction at the interface limits movement at the skin surface,
whilst the normal movement between the deeper layers of the skin is
preserved. The shear force produced between the layers of the skin creates a
void that is filled with fluid to form a blister. The fluid cushions the deeper
layers and protects them from trauma. Epidermal blisters typically are filled
with serous fluid, whereas dermal blisters are usually blood filled due to the
associated injury to blood vessels.

Friction
Friction is a force that resists relative movement between two surfaces in
contact. Friction occurs because no surface is absolutely smooth at the
microscopic scale, and asperities (microscopic projections) on one surface
interact with asperities on the opposite surface to resist motion. Static
friction is the force required to initiate motion, and dynamic friction is the
force required to maintain motion between contacting surfaces. Friction is
always parallel to the contacting surfaces and opposite to the direction of
motion. It can produce deformation, wear and heat, which can change the
properties of the contacting surfaces, e.g. when polishing a surface.

Friction equation
Friction is directly proportional to the load applied to the surfaces:

μ is the ‘coefficient of friction’ of the interface and is the ratio between

friction and load. It have no units. It indicates how much force per unit load
is required to initiate sliding motion between two surfaces. The coefficient of
friction depends on the roughness of interacting surfaces, and different
interfaces have a different coefficient of friction. Friction is directly
proportional to the coefficient of friction.
Friction does not depend on the contact area and sliding speed of the two
surfaces. Therefore, the contact area between the surfaces can be increased
to reduce pressure without increasing the friction.
Frictional torque
Linear motion between surfaces leads to friction; however, rotational motion
between surfaces leads to friction and frictional torque:

Frictional torque increases with the diameter, i.e. the moment arm, of the
rotating object, but friction is unaffected by an increase in the contacting
surface area. The wear of surfaces in rotational relative motion is directly
related to frictional torque.
Friction and frictional torque equations assume that there is no lubricant
present between the contacting surfaces.

Friction in fluids
Friction also resists movement between different layers in a fluid. The
resulting internal resistance to flow of fluid is described as viscosity, i.e. the
thickness of fluid. Low viscosity fluids are thin, e.g. water, and high viscosity
fluids are thick, e.g. honey. Fluid viscosity provides lubrication between
surfaces.

Friction and joints

In terms of biotribology, the functional components of a synovial joint are
bone, articular cartilage and synovial fluid. Healthy synovial joints
experience very low friction because the cartilage-on-cartilage interface has
an extremely low coefficient of friction and synovial fluid provides
lubrication. Therefore, synovial joints experience minimum wear throughout
life.
Friction and wear are significantly higher in the replacement joints,
because the coefficient of friction of replacement materials is a magnitude
higher than the cartilage-on-cartilage interface (Table 5.1).
Fig. 5.2 Friction is a force that resists relative movement between two
solid surfaces. It only exists when the surfaces are in contact. All surfaces
are rough at the molecular scale, so two interacting surfaces are actually in
contact at only a very few points, i.e. the ‘real’ contact area between surfaces
is a very small fraction (0.0001%) of the ‘apparent’ contact area. Therefore,
the force applied to the surfaces is acting over a very small area. The high
pressures generated deform the contacting points (asperities) and weld them
together. Frictional force arises from sliding surfaces breaking and forming
bonds between asperities. As the force acting at the interface increases, more
opposing asperities make contact, which increases friction.

Fig. 5.3 In total hip replacement, for the same bearing combination, a
larger diameter femoral head produces a higher frictional torque than a
smaller diameter femoral head. Friction between the bearing surfaces is
unaffected by the increasing contacting area. Wear of the bearing surfaces is
directly related to frictional torque; therefore a larger diameter femoral head
produces more wear.
 Table 5.1. Coefficients of friction of different interacting surfaces

Contacting surfaces Typical coefficient of friction

Metal on metal (dry) 0.41

Metal on metal (lubricated) 0.06

Teflon on Teflon 0.04

Ice on ice 0.03

Cartilage on cartilage 0.005

Wear
Wear is the progressive removal of material from a contacting surface due to
relative motion between two surfaces. It can be measured in terms of depth,
i.e. linear wear, or as volume, i.e. volumetric wear, which is more accurate.
Wear can also be measured in terms of weight, but volume provides a truer
picture when contacting surfaces are made of materials of different densities.
Wear is usually detrimental, as it leads to increased mechanical loading and
also fatigue failure of the bearing surfaces. The rate of wear is strongly
influenced by the interface conditions.

Wear equation
Volume of wear is directly proportional to load and sliding distance between
two surfaces.
 K is the ‘coefficient of wear’ and describes the volume of wear per unit
load and sliding distance, i.e. ‘wearability’ of a given combination of
materials. Therefore, the volume of wear is also directly proportional to the
coefficient of wear of the interacting surfaces.
In addition to the load and motion variables, wear is also related to the
properties of the contacting surfaces. Wear increases with surface roughness
and decreases with surface hardness. If the two surfaces are of different
materials, wear increases with the difference between the hardness of two
surfaces, i.e. the softer material wears more. In other words, when the two
surfaces have different hardness, wear depends on the hardness of the softer
surface. In the wear equation, the coefficient of wear reflects the hardness of
the softer material.

Wear and joints

In synovial joints, wear leads to failure of articular cartilage and
development of osteoarthritis. The term ‘osteoarthritis’ is a misnomer, as it
implies that the primary process is inflammatory, whereas there is only
degenerative wear of the joint.
The following are some examples of methods of reducing articular
cartilage wear:

A reduction in body weight reduces the overall joint load.

After an intra-articular fracture, restoration of the joint surface to
minimise intra-articular step, i.e. reducing surface roughness.
An osteotomy around the knee shifts the load from one compartment to
another and therefore modifies wear rate of the compartments.
An osteotomy around the hip can:
improve joint contact area/congruency
improve femoral head cover
shift normal articular cartilage to a weight-bearing area
reduce excessive joint reaction force.

Joint replacement prostheses are produced with highly polished bearing

surfaces. Polishing reduces roughness of the bearing surfaces, which
minimises excessive wear and also removes any stress raisers that can
potentially initiate a fracture.

Fig. 5.4a-e Wear of the contacting surfaces can be divided into

different types according to the underlying mechanism producing wear
particles. Mechanical mechanisms include adhesive, abrasive, fatigue and
erosive wear. Chemical mechanisms lead to corrosive wear. These
mechanisms may occur singly or in combination. (a) Adhesive wear:
opposing asperities bond to each other and shear off as one surface slides
over the other. (b) Abrasive wear: asperities on the harder material cut the
asperities on the softer material. The new particles become ‘third bodies’.
(c) Fatigue wear: cyclical loading causes accumulation of micro-damage
that breaks off as asperities. (d) Erosive wear: relative motion in a fluid
containing hard particles erodes the asperities. (e) Corrosive wear: surface
damage due to chemical reactions with environment.
Fig. 5.5 Wear rate is particularly high when a mechanical type of wear
occurs in combination with corrosive wear. Mechanical wear efficiently
removes wear particles produced by corrosive wear, which continually
reveals fresh material beneath that in turn corrodes rapidly, and the cycle
continues.
Fig. 5.6 According to McKellop's classification, wear of artificial joints
is divided into four modes. These four modes of wear are illustrated in
bearing surfaces of total hip replacement. Mode 1 represents normally
functioning implants, whereas modes 2–4 represent malfunctioning implants.
Modes 2–4 can occur singly or in combination, causing massive wear and
rapid implant failure.

Lubrication I
A lubricant is a material introduced between two solid surfaces in relative
motion in order to reduce friction. Movement between the layers of lubricant
and/or a surface and lubricant produces less friction, and therefore surface
wear, than movement between the two surfaces alone. The most important
property of a lubricant is viscosity. Different interacting surfaces require
lubricants with different viscosity. A lubricant may sometimes increase wear,
e.g. if it changes mode of wear of the surfaces or traps abrasive particles.

Modes of lubrication
When a lubricant is present between two surfaces, three types of lubrication
are possible, depending on the thickness of the fluid film formed during
relative motion between the surfaces.

Boundary lubrication. The fluid film between the interacting surfaces is

about the same thickness as their surface roughness, so the asperities on
the opposing surfaces are in contact. The load on the interface is carried
by surface asperities rather than by the lubricant. The interface has a
high coefficient of friction, so there is high friction and wear between
the surfaces; these can be reduced if a lubricant can adhere to the
asperities instead of simply pooling between the asperities.
Mixed lubrication. The fluid film between the two surfaces is slightly
thicker than their surface roughness, but there is some contact between
the longer asperities on the opposing surfaces. Therefore, the two
surfaces are partly in contact and partly separated by the fluid film. The
load is shared by the asperities and the fluid film. In comparison with
boundary lubrication, this interface has a lower coefficient of friction,
so there is lower friction and wear between the surfaces.
Fluid film lubrication. The fluid film completely separates the
interacting surfaces. Fluid film lubrication is further divided into two
main subtypes.
Elastohydrodynamic lubrication. The relative motion between the
two surfaces maintains a thin layer of fluid film between them. The
surfaces, however, are still close enough to cause elastic
deformation of the asperities.
Hydrodynamic lubrication. The relative motion between surfaces
maintains a sufficiently thick layer of fluid between them to allow
one surface to float above the other. The load is fully supported by
the fluid film.
Motion between the surfaces is essential for achieving and maintaining
fluid film lubrication. A change in motion parameters affects the type of
fluid film lubrication achieved, if at all. As an example, a machine can
have moving parts that experience fluid film lubrication at full function,
but not at start-up or shut-down.

The motion conditions between the two solid surfaces determine the type of
lubrication regime achieved. The three modes of lubrication can be achieved
by a number of different mechanisms, e.g. fluid film lubrication in the
synovial joints.
Fig. 5.7 Boundary lubrication. Thickness of fluid film is not sufficient to
separate the contacting surfaces. Relative motion between the surfaces
results in extensive contact between asperities. However, friction and wear
are reduced because the lubricant coats the asperities and the interaction
between the chemical compounds in the lubricant on asperities produces less
resistance than interaction between ‘dry’ asperities.

Fig. 5.8 Mixed lubrication. Fluid film between the surfaces is slightly
thicker than surface roughness; however, relative motion between the
surfaces still results in some contact between longer asperities on
opposing surfaces.
Fig. 5.9 Fluid film lubrication. A continuous layer of fluid film
completely separates the interacting surfaces. Fluid film lubrication is
divided into two types. (a) Elastohydrodynamic lubrication produces a thin
layer of fluid film that separates surface asperities. However, opposing
surface asperities can still be elastically deformed; (b) Hydrodynamic
lubrication achieves a sufficiently thick layer of fluid film to prevent any type
of interaction between opposing asperities. The layers of the fluid film
immediately adjacent to each surface travel at the same speed and direction
of each surface. The change in direction of relative motion occurs between
the layers of fluid, so the surface wear is negligible. Hydrodynamic
lubrication occurs when there is sufficient pressure within the lubricant to
produce a non-compressible fluid film that is three times thicker than the
surface roughness.

Lubrication II
Boundary lubrication is generally the least effective as there is still a
significant solid-on-solid contact, and friction and wear at the interface are
high; fluid film lubrication is the more desirable as the solid surfaces are
fully separated, and friction and wear at the interface are negligible.
However, fluid film lubrication is very complex to attain, and interface
conditions have to be optimum to achieve this low friction and low wear
state of relative motion between surfaces.

The Stribeck curve

The mode of lubrication achieved is determined by three variables:

viscosity of the lubricant

 speed of relative movement between surfaces
load acting at the interface.

The Stribeck curve shows the relationship of coefficient of friction to these

variables. It shows that boundary lubrication and high coefficient of friction
exist when the lubricant is too thin and/or when relative motion between
surfaces is too slow for the load acting at the interface. As the lubricant
viscosity and/or speed of motion increases or the load decreases, the
surfaces begin to move apart, and mixed and fluid film modes of lubrication
progressively take effect.
The Stribeck curve also shows that, if the lubricant viscosity remains
constant, then the speed of motion must increase with the load to achieve
fluid film lubrication.
The Stribeck curve further shows that the coefficient of friction can
increase again within fluid film lubrication. This is because drag occurs
within the lubricant when the speed of motion between the two surfaces is
high.

Synovial fluid as a lubricant

Synovial fluid provides nourishment to the articular cartilage, removes waste
products and provides joint lubrication.
Composition. Synovial fluid is produced by the synovial membrane of the
joint, and is an ultrafiltrate of blood plasma. It mainly consists of water,
hyaluronic acid and proteins, such as albumin.
Lubricant properties. Synovial fluid is a ‘shear-thinning’ type of fluid –
this means that it has a higher viscosity, i.e. is thicker, when still and has
lower viscosity, i.e. becomes runnier, when stirred. The Stribeck curve
shows that, as the lubricant's viscosity increases, it is able to support more
load.
The viscosity of synovial fluid is also proportional to hyaluronic acid
concentration, which is essential for its lubrication function. In addition to its
lubricant role in the ‘dissolved’ state, hyaluronic acid also forms complexes
with glycoproteins on the surface of the articular cartilage, thereby providing
a lubricant coating to the interacting surfaces.
Articular cartilage as a bearing material
Hyaline cartilage lines the bones in a synovial joint. It is a highly organised
tissue that consists of cells and extracellular matrix. It forms a low-friction,
low-wear bearing material that is viscoelastic, flexible and durable. The
different components of articular cartilage help to absorb and distribute the
compressive forces experienced by the synovial joint (Table 5.2).

Fig. 5.10 The Stribeck curve. The Stribeck curve shows how the operating
conditions at an interface determine the type of lubrication and therefore the
coefficient of friction and wear properties of the interface. In boundary
lubrication, the coefficient of friction is mainly determined by the properties
of interacting surfaces; as in a normal situation of two solid surfaces in direct
contact, the coefficient of friction is independent of sliding speed and
directly proportional to load. In mixed and fluid film lubrication, the
coefficient of friction varies with sliding speed. Just as there is a ‘friction
equation’ and a ‘wear equation’, the Stribeck curve can be considered to
show the ‘lubrication equation’.

Table 5.2. Characteristics of normal synovial fluid in an adult knee

joint
 Volume <3.5 ml

Clarity Clear

Colour Colourless/Straw colour

White cell count <2000/mm3

Polymorphs <25%

Gram stain Negative

Fig. 5.11 The composition of articular cartilage. Chondrocytes secrete the

extracellular matrix components. The relative proportion of different
components varies with depth from the articular surface. The thickness of
articular cartilage is highly variable; however, its basic composition remains
quite constant.

Lubrication of synovial joints

From a biotribology point of view, normal synovial joints have the following
functional conditions:

Articular cartilage forms the two bearing surfaces in relative motion.

Synovial fluid acts as the lubricant and is present only in small
quantities.
Synovial joints experience large and variable joint reaction forces.
 Synovial joints intermittently are mobile with variable speed of motion.

These are tough operating conditions for lubrication, and are generally
opposite to the requirements for fluid film lubrication. However, the articular
cartilage and synovial fluid function together to provide a very low
coefficient of friction at the interface.
A synovial fluid film exists between the two articular cartilage surfaces at
practically all times. The intermittent and relatively low speed of motion of
the joints mean that fluid film lubrication cannot always be produced by
motion dynamics alone, i.e. the standard mechanisms for elastohydrodynamic
and hydrodynamic lubrication. The articular cartilage has the capacity to
exude and absorb synovial fluid, much as a sponge can hold and release
water. The cartilage-on-cartilage interface therefore has additional
mechanisms for achieving fluid film lubrication under different conditions:

Weeping lubrication. If a synovial joint is simply loaded with little or

no relative motion between the articular surfaces, the fluid is squeezed
out of the cartilage into the joint cavity, thereby maintaining a fluid film
between the surfaces. There is a natural limit to this process when a
balance between hydrostatic and osmotic pressures is achieved between
the two compartments. The fluid returns into the cartilage as the load is
eased off.
Squeeze film lubrication: An applied load can also squeeze synovial
fluid out from between the two cartilage surfaces. The generated fluid
pressure keeps the two surfaces apart. This is considered to be the main
lubrication mechanism in shock absorption during high impact
situations, e.g. landing on ground after a fall from height.
Boosted lubrication: After prolonged loading, the cartilage absorbs
synovial fluid. Cartilage pores are too small for hyaluronic acid.
Therefore, the synovial fluid remaining in the joint is more concentrated
and has a higher viscosity and load-bearing capacity.
Other mechanisms: Further theories are emerging to explain the
maintenance of fluid film lubrication in the synovial joint.

Mixed and boundary lubrication regimes take effect progressively under

severe and/or prolonged loading conditions, i.e. when synovial fluid film
decreases. The wear of cartilage then depends on its surface properties. The
cartilage surface is covered in glycoproteins, which are large molecules and
equivalent to surface asperities. The interaction between the glycoproteins on
the opposing cartilage surfaces produces a very low coefficient of friction
because these are coated with hyaluronic acid or synovial fluid. In addition,
cartilage is viscoelastic and can undergo significant deformation before
permanent wear.

Fig. 5.12 Mechanisms of fluid film lubrication in synovial joints.

Different modes of lubrication act at different stages of joint motion.
(a) Hydrodynamic lubrication; (b) Elastrohydrodynamic lubrication; (c)
Weeping lubrication; (d) Squeeze film lubrication; (e) Boosted lubrication;
(f) Other mechanisms.

Lubrication of prosthetic joints

The prosthetic joints face the same biotribological challenges as their
respective synovial joints, but without the ultra-efficient, variable-mode
articular cartilage–synovial fluid lubrication system. The prosthetic bearings
do not have a built-in lubricant and only have the articular fluid available for
this. The design of bearing surfaces is constrained by functional
requirements, such as anatomical, stability and range of motion
considerations. Furthermore, the wear particles from prosthetic bearings can
interpose between the surfaces and increase wear. Therefore, it is very
difficult and complex to achieve satisfactory lubrication between prosthetic
bearings, and wear remains a major limitation in their performance.
Wear of total hip and knee replacement prostheses is a particularly
significant issue, as these are the two most common joint replacements. Fluid
film lubrication is the ultimate in minimising wear between bearing surfaces.
In total hip replacement, the ball and socket configuration and motion
parameters of bearing surfaces create the possibility of attaining fluid film
lubrication. However, a number of other factors also determine if fluid film
lubrication is actually achieved.

Material of the bearing surfaces. Polyethylene has a very coarse

surface, which does not support fluid film lubrication. In metal on
polyethylene combination, wear of polyethylene acetabular cup is
managed by using a smaller diameter femoral head that produces a
lower frictional torque, e.g. Charnley's ‘low frictional total hip
arthroplasty’ with a 22.5 mm diameter metal head on polyethylene cup,
and/or using highly cross-linked polyethylene that has a higher wear
resistance. Metals and ceramics have much lower surface roughness;
therefore metal on metal and ceramic on ceramic bearing combinations
support fluid film lubrication.
Diameter of the femoral head. The likelihood of achieving fluid film
lubrication increases as the femoral head diameter increases. Therefore,
larger diameter metal on metal and ceramic on ceramic bearing
combinations have the highest possibility of achieving fluid film
lubrication. The smaller diameter articulations are more likely to
experience mixed lubrication.
Radial clearance between femoral head and acetabular cup. The
polar bearing articulation allows fluid movement between the bearing
surfaces and therefore is the most receptive of fluid film lubrication.
However, if the radial clearance is too high, the contact area of bearing
surfaces becomes too small, which leads to high wear rates. The
congruent bearing articulation has more restricted flow of fluid and
equatorial bearing articulation ‘locks out’ fluid. Therefore, these effects
impede support for fluid film lubrication.
The design features and interface conditions of total knee replacement
components rule out any significant lubrication between the bearing surfaces.
The main biotribological issue in total knee replacement is the wear of the
polyethylene tibial insert. This is managed by using components of
appropriate design (see pages 120–124) and avoiding using a too thin (<8
mm) tibial insert. In addition, a tibial insert made of highly cross-linked
polyethylene has a higher wear resistance than one made of plain
polyethylene (Table 5.3).

Table 5.3. Wear rate of total hip replacement bearing surfaces

Bearing Typical wear rate per million cycles

combination (mm3)*

Metal on 55
polyethylene

Metal on metal 0.6

Ceramic on ceramic 0.16

* The femoral head diameter and joint load are similar in these bearing combinations.
Fig. 5.13 The Stribeck curves for different bearing combinations in
total hip replacement. The x-axis contains an additional element of the
interface: the radius of the femoral head.* The curves show that metal on
polyethylene (MoP) bearing combination does not support fluid film
lubrication. Metal on metal (MoM) and ceramic on ceramic (CoC) bearing
combinations experience a significant reduction in the coefficient of friction
as the radius of the femoral head increases. These therefore support fluid
film lubrication when femoral head size increases. Metal on polyethylene is
a ‘hard on soft’ bearing combination that experiences high wear rates despite
a very low coefficient of friction. Metal on metal and ceramic on ceramic are
‘hard on hard’ bearing combinations that experience significantly lower wear
rates despite having a much higher coefficient of friction.
*The Stribeck curve in Fig. 5.10 (page 89) is for two surfaces in linear
motion. When the two surfaces are in rotational relative motion, the x-axis
has an additional variable, the radius of the rotating object.
Fig. 5.14 Fluid film lubrication requires radial clearance between the
femoral head and acetabular cup. Radial clearance is defined as the
difference between the inner radius of the cup and radius of the femoral head.
As the radial clearance increases, the contact area between the bearing
surfaces decreases. Therefore, the optimal radial clearance provides polar
bearing articulation and high conformity between bearing surfaces.

Further reading
Ashby M, Messler RW, Asthana R et al (2009). Engineering Materials and
Processes Desk Reference. Oxford: Elsevier. 55–56.
Berrien LSJ (1999). Biotribology: Studies of the effect of biomechanical
environments on the wear and damage of articular cartilage. PhD thesis.
Virginia Polytechnic Institute and State University: USA.
Bucholz RW, Heckman JD, Court-Brown CM (eds) (2006). Rockwood and
Green's Fractures in Adults. 6th edn. Philadelphia: Lippincott, Williams
and Wilkins.
Callister, WD (2007). Material Science and Engineering: An Introduction.
7th edn. New York: Wiley.
Craig Jr RR (2011). Mechanics of Materials. 3rd edn. New York: John
Wiley & Sons. 237–275.
Gayon J (2000). History of the concept of allometry. Amer Zool. 40: 748–
758.
Johnson K (2001). Physics for You. London: Nelson Thornes.
Lee JY, Kim SY (2010). Alumina-on-polyethylene bearing surfaces in total
hip arthroplasty. Open Orthop J. 4: 56–60.
Lucas GL, Cooke FW, Friis EA (1999). A Primer of Biomechanics. New
York: Springer. 67–78.
Madihally SV (2010). Principles of Biomedical Engineering. Norwood:
Artech House. 189–190.
Mansour JM (2008). Biomechanics of cartilage. In Oatis CA (ed):
Kinesiology: the Mechanics and Pathomechanics of Human Movement.
2nd edn. 69–83.
Morgan EF, Bouxsein ML (2008). Biomechanics of bone and age-related
fracture. In Bilezikian JP, Raisz LG, Martin TJ (eds): Principles of Bone
Biology. 3rd edn. Vol 1. London: Elsevier. 29–52.
Neu CP, Komvopoulos K, Reddi AH (2008). The interface of functional
biotribology and regenerative medicine in synovial joints. Tissue Eng
Part B Rev. 14(3): 235–247.
Nordin M, Frankel VH (2001). Basic Biomechanics of the Musculoskeletal
System. 3rd edn. London: Lippincott, Williams and Wilkins.
Petit MA, Beck TJ, Kontulainen SA (2005). Examining the developing bone:
What do we measure and how do we do it? J Musculoskelet Neuronal
Interact. 5(3): 213–224.
Van der Meulen MCH, Jepsen KJ, Mikic B (2001). Understanding bone
strength: size isn't everything. Bone. 29(2): 101–104.
 Part III Clinical biomechanics

The Earth has an axis and orbit of motion. It rotates about its axis and orbits
the Sun, with the orbital motion being superimposed on its spinning motion.
The Earth's spinning axis is tilted 23.5° from the perpendicular to the plane
of its orbit around the Sun. The rotation of the planet about its axis forms day
and night. The tilt of the axis and the Earth's rotation around the Sun are
responsible for the planet's seasons.
Similarly, in the musculoskeletal system, synovial joints also have axes and
arcs of motion. The observable, main arc of motion of a joint is
superimposed on top of smaller but essential motions in other planes. The
understanding of orientation of normal axis and finer motions of a joint, and
their significance to its overall functions is essential in joint reconstructive
surgery.
 6 Biomechanics of the hip and total hip replacement

Axes of the lower limbs

In the musculoskeletal system, an axis describes a plane within a structure, e.g. the axis of rotation of a joint
and neutral axis in relation to bending and torsional loading of structures, or a plane of relative alignment of
structures, e.g. axes of the lower limbs. It can sometimes be difficult to appreciate the three-dimensional
orientation of the axes in the musculoskeletal system through descriptions and diagrams, and these are often
better appreciated in a three-dimensional set-up, e.g. in computer/actual models or in a clinical/theatre
setting.
The alignment of different bones and joints in the lower limbs is described by a number of axes. The
comparison of these axes with each other and with different elements of the bones and joints can produce a
constellation of angles and analyses. The alignment of the axes in the coronal plane is considered routinely,
as lower limb deformities are more common in this plane. Therefore, the axes of the lower limb are
discussed here in the coronal plane only, to simplify details and discuss important points. However, the
alignment of these axes in the sagittal plane is also important and must also be considered where
appropriate.

Anatomical axis
The anatomical axis describes the longitudinal anatomical alignment of the femur and tibia (i.e. the bones).
This is conventionally represented by a line drawn down the centre of the diaphysis of each bone. The
anatomical axes of the tibia and femur intersect at the knee at an average of 6° to each other.

Mechanical axis
The mechanical axis describes the alignment of the centres of the hip, knee and ankle joints. Conventionally,
this is also described in terms of the bones, so that the mechanical axis of the femur is defined by a line
connecting the centre of the femoral head to the medial tibial spine, and the mechanical axis of the tibia is
defined by a line connecting the medial tibial spine with the centre of the ankle. The normal mechanical axis
is not vertical, but is 3° valgus to the vertical, because the centre of the femoral head is in a valgus position
with respect to the other joints.

Weight-bearing axis
The weight-bearing axis represents the path of load transmission to the ground relative to the lower limbs. It
is represented by a line connecting the centre of the femoral head to the centre of the ankle.

The relationship between the axes

The anatomical axis of the femur is on average 6° valgus to its mechanical axis. The anatomical axis of the
tibia is about 2–3° varus to its mechanical axis.
The normal mechanical axis overlaps the weight-bearing axis of the lower limbs; the mechanical axis is
therefore commonly referred to as the weight-bearing axis.
Fig. 6.1 The alignments of the anatomical and mechanical axes of the lower limbs. The anatomical
axis of the femur is 6° valgus to its mechanical axis. The anatomical axis of the tibia is 2–3° varus to its
mechanical axis, but the two tibial axes are considered to be identical for practical purposes, as the
difference between them falls within the acceptable margin of error (<3°) in mechanical axis realignment
during lower limb procedures, e.g. total knee replacement or deformity correction.

Fig. 6.2 The relationship between the mechanical and load-bearing axes of the lower limbs. The
normal mechanical axis overlaps the load-bearing axis. In varus malalignment, the mechanical axis deviates
lateral to the weight-bearing axis and in valgus malalignment it deviates medial to the load-bearing axis.

Hip joint reaction force I

Functional anatomy
The hip is a stable ball and socket joint, formed by the femoral head and pelvic acetabulum. The abductor
muscles are the main stabilisers of the pelvis in the coronal plane. The total compressive force acting on the
hip joint is the resultant of forces due to body weight, tension in the abductor muscles and any impact loads
transmitted upwards through the body from the foot during everyday activities. A static analysis can be used
to estimate the magnitude of hip joint reaction force under different circumstances.

Double-leg stance
The force acting on the hip joint during double-leg stance can be estimated from the proportional
distribution of body weight. The legs comprise about 1/3 total body weight (TBW), so the weight of the
upper body supported by the hips is approximately 2/3 TBW. Therefore, during a simple double-leg stance,
each hip is subjected to a compressive force of about 1/3 TBW.

Assumptions
During double-leg stance, abductor muscles are relaxed; any minimal tension (and therefore force) in
abductor muscles is ignored.
In a static situation, impact load transmitted from the ground is zero.

Single-leg stance
During a single-leg stance, the abductor muscles of the supporting leg contract to stabilise the pelvis. The
hip joint acts as the axis of a class I lever and the pelvis acts as a rigid horizontal lever, supporting upper
body weight on one side and abductor muscles force on the opposite side.
The proportional distribution of body weight means that the weight of the supported upper body is
approximately 5/6 TBW. The abductor muscles generate a force to balance the moment produced by the
upper body weight. According to Newton's third law, the joint reaction force must be equal and opposite to
the sum of these two forces. The minimum hip joint reaction force is therefore estimated to be
approximately 1.5 x TBW.

Assumptions
In addition to the general assumptions, the following specific assumptions are made in this analysis:

The abductor muscles are the only active muscle group generating force, and there is no antagonistic
muscle action.
All forces are acting in a vertical direction. The line of action of the abductor muscles is actually 70°
to the horizontal and therefore the abductor muscles actually produce a bit more force than estimated
here.
The pelvis is horizontal.
In a static situation, impact load transmitted from the ground is zero.

Further exercise
The book cover shows a free-body force diagram showing forces acting on the right hip during a single-leg
stance. Here it is assumed that the centre of gravity lies behind the public symphysis (this assumption is
commonly made to simplify calculations). Could you calculate the abductor muscles force and hip joint
reaction force in this example? Assume body weight = 600 N, A = 70 mm and B = 125 mm. (Answer is
abductor muscles force = 1671 N and joint reaction force = 1671 N)
Fig. 6.3 The anatomical relationships between the hip joint, abductor muscles and the pelvis.
Measurements provided are typical for an adult and are based on anthropometric data. During a double-leg
stance, the centre of gravity of the supported upper body is in the midline and passes behind the pubic
symphysis. (TBW= Total body weight.)

Fig. 6.4 Free-body force diagram of the pelvis showing the forces acting on the hip during a single-
leg stance. The centre of gravity of the upper body shifts closer to the supporting hip to overlie the area of
support, i.e. the foot, and is taken to be approximately halfway between the pubic symphysis and the centre
of the hip joint. Each leg comprises 1/6 RV. The body weight below hips is 1/6 + 1/6 = 1/3 TBW. During
single-leg stance weight of supported upper body is 2/3 + 1/6 = 5/6 TBW. (Force W = Weight of the upper
body; Force A = Abductor muscles force; and, Force J = Joint reaction force.)
Calculations
Applying the conditions of equilibrium:

1. Sum of all moments is zero.

2. Sum of all forces is zero.

The hip joint reaction force during single-leg stance. The joint reaction force is estimated to be about 1.5
× TBW.
Hip joint reaction force II

Single-leg stance with a cane support in the opposite hand

A cane support in the hand transmits force applied to it to the ground, and receives an equal and opposite
ground reaction force, which it transmits back to the body. During a single-leg stance, when a cane is held in
the opposite hand to the supporting hip, the class I lever system has an additional upwards-acting load, the
ground reaction force, which has a longer lever arm than the supported upper body weight. This produces a
moment in the same direction as the abductor muscles force. Therefore, the moments produced by abductor
muscles force and ground reaction force both balance the moment produced by upper body weight, which
reduces the force requirements of the abductor muscles. Therefore, the joint reaction force is reduced to
about 1.3 x TBW during single-leg stance.

Assumptions
As discussed in the static analysis of hip joint reaction force during single-leg stance, plus:

The weight of the cane is relatively small and is therefore ignored.

Single-leg stance with a 100 N load in the opposite hand

If a load is carried in the hands in normal everyday activities, this generates a downwards-acting force of
weight. During a single-leg stance, when the load is held in the opposite hand to the supporting hip, the
class I lever system has an additional downwards-acting force, the weight of the load, which has a longer
lever arm than the upper body weight. This produces a moment in the same direction as the supported upper
body weight. Therefore, the abductor muscles have to generate a force to balance two opposing moments,
which significantly increases the force requirements of the abductor muscles. When a load of 100 N is
carried in the opposite hand, the hip joint reaction force is estimated to be about 2.5 x TBW during single-
leg stance.

Assumptions
As discussed in the hip joint reaction force analysis during single-leg stance.

Clinical implications
The hip joint reaction force during single-leg stance is more than the total body weight. The magnitude of
the force is determined by the supported body weight, abductor muscles force and the ratio of their lever
arms. The use of adjuncts can greatly influence the joint reaction force through their effect on these factors.
The cane should be held in the opposite hand and any load, e.g. a shopping bag, should be carried in the
same hand as the symptomatic hip, e.g. arthritic hip or during rehabilitation after a hip operation. In total hip
replacement, ‘medialising’ the acetabular component increases the ratio of abductor muscles lever arm to
the supported body weight lever arm, and therefore reduces the joint reaction force.
Fig. 6.5 Free-body force diagram of the pelvis showing the forces acting on the hip during a single-
leg stance with a cane support in the opposite hand. The cane support provides stability through an
additional point of contact with the ground, and improves the body's centre of gravity to a wider, stable
range. Therefore, the centre of gravity of the supported upper body lies behind the pubic symphysis (as in
the double-leg stance). The ground reaction force is limited by the force that can be applied to the cane by
the upper limb, which in an adult is typically about 100 N (or 1/7 TBW), taken to act at 400 mm from the
supporting hip. (Force W = Weight of the upper body; Force A = Abductor muscles force; Force J = Joint
reaction force; and Force G = Ground reaction force acting on the cane.)
Calculations
Applying the conditions of equilibrium:

1. Sum of all moments is zero.

2. Sum of all forces is zero.

The hip joint reaction force during single-leg stance with cane support in the opposite hand. The joint
reaction force is about 1.3 × TBW.
Fig. 6.6 Free-body force diagram of the pelvis showing the forces acting on the hip during a single-
leg stance with a 100 N load in the opposite hand. The centre of gravity of the supported upper body
weight shifts closer to the supporting hip as in the simple single-leg stance. In an adult, a 100 N load is 1/7
TBW and is taken to act at 400 mm from the supporting hip. (L = Weight of the load.)
Calculations
Applying the conditions of equilibrium

1. Sum of all moments is zero.

2. Sum of all forces is zero.

The hip joint reaction force during single-leg stance with a 100 N load in the opposite hand. The joint
reaction force is about 2½ x TBW.

Total hip replacement (THR): cemented fixation

Fixation of total hip replacement implants to bones with cement provides immediate stability to the
construct. Cement functions as grout, and not as an adhesive, to provide a mechanical interlock between
implant and bone. As cement does not stimulate new bone formation and there is no renewal of bonding at
the cement–bone interface, the quality of cemented fixation degrades with time.

Femoral stem implant design

There are two basic designs of cemented fixation femoral stem implant, which interact with cement
differently to achieve durable stability.

Composite beam
This type of stem has a small protrusion, a collar, at the level of the femoral calcar; a pre-coated, roughened
fixation surface; and a cylindrical profile throughout its length. These features optimise the stem for a strong
bond with cement. The proximal collar prevents distal sinkage of the stem, and the rough fixation surface
ensures maximum bonding between stem and cement.
The proximal collar also increases the load transferred from stem to the femoral calcar, aiming to
replicate natural load transmission in the proximal femur. The extensive bonding between stem and cement
maintains the stem in its position without any slip within the cement mantle; therefore, it is considered a ‘sit
up and stay’ prosthesis. The firm fixation achieved does not accommodate creep within the cement mantle.
The load is transferred to the femur by shear stress at the bone–cement interface. Due to all these
characteristics, the stem is described as having a ‘shape closed’ design.

Taper slip
This type of stem is collarless; has a highly polished fixation surface; and has a tapered profile from
proximal to distal. These features prevent the stem from bonding with cement. The stem therefore settles in
cement, re-engaging its taper, and so the fixation becomes progressively more stable; therefore it is
considered a ‘slip and slide’ prosthesis. The re-engagement of the taper converts shear stress at the
interface into radial compression of cement. The load is therefore transferred to the femur by compressive
stress at the bone–cement interface. As the stem is mobile, it can accommodate creep within the cement
mantle. Due to all these characteristics, the stem is described as having a ‘force closed’ design.

The effect of surface finish

The polished surface finish is an essential feature of the taper slip stem. The stem has been trialled with a
matte surface finish (as polishing is a labour-intensive and expensive process), and was found to have a
significantly high failure rate (revision rate 10% at 10 years, compared with 2.5% at 10 years for a
polished surface finish). The matte surface finish leads to the stem bonding with cement, although its
features are not suited for this. The matte taper slip stem is unable to settle in cement and convert the shear
forces to compressive forces and therefore has a higher failure rate.
Fig. 6.7 Comparison of design features of the two types of cemented fixation femoral stem implant.
(a) The composite beam stem is also referred to as the ‘Charnley’ stem, as it was developed by Charnley
and Harris in the 1950s. (b) The taper slip stem is also referred to as the ‘Exeter’ stem, as it was developed
in Exeter by Ling and Lee in the 1970s. (Image (a) is reproduced with permission of Smith & Nephew.
Image (b) is reprinted from Choy GG et al. (2013) Exeter short stems compared with standard length Exeter
stems: experience from the Australian Orthopaedic Association National Joint Replacement Registry. J
Arthroplasty. 28(1): 103–109. Copyright (2013) with permission from Elsevier.)

Fig. 6.8 Comparison of load transfer between the two types of cemented fixation femoral stem
implant. The stem–bone construct is a composite, and is subjected to the same loads as the natural hip. The
implant is much stiffer than bone and therefore accepts more load, which it then transfers to the bone.

Total hip replacement: cementless fixation

Total hip replacements with cementless fixation are being increasingly used in younger patients to preserve
bone stock and avoid problems associated with cemented fixation. The ‘press fit’ placement of implant in
the bone cavity achieves initial stability, and implant osteo-integration provides long-term stability.
Cementless fixation is therefore a biological fixation, which is dynamic because of bone turnover and
therefore the quality of fixation is maintained with time.

Femoral stem implant design

A cementless fixation femoral stem has a porous and/or a biologically active fixation surface. A
biologically active interface is formed by a textured surface finish or coating the implant with a biologically
active material. A textured surface is very finely rough, e.g. like sandpaper, and is produced by grit-blasting
the stem. The commonly used bioactive materials for application to the implants are commercially pure
titanium or ceramics, i.e. hydroxyapatite. The biological fixation of implant is achieved by bone ingrowth
into the porous surface or ongrowth to the biologically active surface. During the bonding process, new
bone can tolerate only small movements of the stem (up to 28 μm), and does not form if there is excessive
motion (more than 150 μm); in which case a layer of connective tissue forms between the stem and bone.
The early cementless fixation femoral stems (e.g. the ‘fully’ porous coated stems) were cylindrical and
prepared up to distal diaphyseal section to biologically bond with the bone. The extensive bonding
achieved in the femoral diaphysis produced problems of stress shielding of proximal femur (i.e. stress
protection osteopenia), thigh pain and difficulty removing the implant at revision. Therefore, most
uncemented femoral stems are now tapered and prepared for osteo-integration in the proximal part only.
Distal fixation stems are mainly used in revision surgery, e.g. for periprosthetic fractures. These revision
stems have different cross-section designs and may also have the option of interlocking screw fixation.

Cemented fixation vs cementless fixation

Clinical studies comparing the results of total hip replacements with cemented and cementless fixations
show that cemented fixation has better overall long-term results (i.e. lower revision rates), but cementless
fixation appears to have advantages in certain situations. Cementless fixation has potentially better
resistance to aseptic loosening in the younger patients, and the long-term outcome of total hip replacements
with cementless fixation does not appear to be affected by varus or valgus mal-alignment of the femoral
stem.
Cementless fixation is associated with a higher rate of intra-operative fractures. The common mechanism
of intra-operative fractures is generation of excessive hoop stresses during preparation of bone and
insertion of prosthesis. The impaction of prosthesis to achieve press fit fixation with the bone in cementless
fixation leads to a higher occurrence of excessive hoop stresses generation; whereas cemented fixation does
not require such compact fixation (Tables 6.1 and 6.2).
Fig. 6.9 Design features of a typical cementless fixation femoral stem implant. (Image is reproduced
with permission of Smith & Nephew.)

Fig. 6.10 Load transfer from stem to bone in cementless fixation femoral stem. The stem is designed
to transfer load mainly at the metaphyseal section as shear stress at the implant–bone interface.

Table 6.1. Hybrid fixation

Aseptic loosening of the acetabular component is commonly the limiting factor to the long-
term survivorship of cemented total hip replacement. Therefore, sometimes hybrid fixation is
used so that there is cementless fixation of the acetabular component and cemented fixation of
 the femoral component.
Cementless fixation acetabular component is formed of a metal cup with a separate inlay of
the bearing surface. The component is press fixed to the acetabulum. The fixation surface is
bioactive for bonding with the bone, but the component can also be further secured with
screws.
A ‘reverse hybrid’ fixation may also be used which consists of a cemented acetabular (usually
polyethylene) component and a cementless fixation femoral stem.

Table 6.2. Percentage of primary total hip replacements (THR) performed in the UK over a 10-
year period (Data taken from National Joint Registry report 2014)

Year

Fixation
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
type

Cemented 60.5 54.1 48.6 42.8 39.7 34.3 31.9 31.4 38.3 32.9 33.2
fixation

Cementless 16.8 21.4 25.6 30.1 33.3 39.4 43.2 45.8 44.9 44.9 42.5
fixation

Hybrid 12.3 13.3 14.1 15.2 15.0 15.1 15.8 16.2 17.2 17.7 20.2

Reverse 0.6 0.9 1.1 1.2 1.7 2.5 2.7 2.8 3.1 3.1 3.0
hybrid

Resurfacing 9.8 10.2 10.7 10.8 10.2 8.8 6.5 3.8 2.5 1.4 1.1

Actual 14 27 40 47 60 66 67 69 72 76 76
number of 413 993 150 523 460 707 547 891 835 607 274
THR
performed

Can you spot the general trends?

Total hip replacement: design and alignment of components

The main biomechanical goals of total hip replacement are to: restore the centre of rotation of the joint;
maintain mechanical axis of the lower limb; preserve leg length; and achieve correct soft tissue balance.
The design and alignment of components are key factors in determining the performance of the replacement
hip joint. It is important to undertake pre-operative planning, which helps to establish the correct size,
design and alignment of components.

Design
In the femoral component, neck length and off-set determine the neck-shaft angle and abductor muscles lever
arm. Correct restoration of these variables is important for proper soft tissue balancing of the hip. A change
in neck length has a greater effect on leg length than the abductors muscles lever arm, whereas a change in
off-set affects the abductor muscles lever arm more than leg length. Therefore, to optimise the abductor
muscles lever arm, it is better to increase the off-set than the neck of the implant.
A polyethylene acetabular cup can be used with or without a 20° elevation ‘posterior lip augmentation
device’. The elevated posterior lip reduces the risk of total hip replacement dislocation; however, it also
decreases the primary arc of motion. Dislocations can still occur, especially in very active persons, due to
femoral neck impingement on the prominent acetabular cup. Other ‘constrained’ liners are also available to
further enclose the femoral head, which are used most commonly in revision surgery for instability.

Alignment
Component alignment is a function of its position within the bone cavity and orientation with respect to the
body. The femoral stem should be positioned valgus in the coronal plane and parallel in the sagittal plane
with respect to the bone cortices; varus and ‘back to front’ placement must be avoided, especially in
cemented fixation, as the normal eccentric loading produces bending forces that are pushing the stem in
these directions.
Normally, the optimum position of the acetabular cup is ‘medialised’ within the acetabulum. This
achieves the best combination of bone/cup coverage for fixation interface and cup/femoral head coverage to
reduce contact stresses at the bearing interface. Medialising the acetabular cup also reduces joint reaction
force, by increasing the lever arm of the abductor muscles and decreasing the lever arm of body weight. The
position of the acetabular cup within the acetabulum is decided at the time of the operation.
The safe zone of orientation of components to minimise the risk of impingement and dislocation is as
follows:

Femoral stem: Axial plane (anteversion angle): 0°–20°

Acetabular cup: Coronal plane (theta angle): 40° +/− 10°

Sagittal plane (anteversion angle): 15° +/− 10°.

Practically, the acetabular cup normally is aligned with the transverse acetabular ligament during the
operation. This automatically ensures the correct orientation of the acetabular component, and fine
adjustments can then be made as required.

Fig. 6.11 A 20° elevation posterior lip augmentation device provides a greater femoral head cover
and helps to reduce dislocations, but also reduces the primary arc of motion.
Fig. 6.12 The effect of increasing the neck length and off-set of femoral component on leg length
and abductor muscles tensioning. The off-set required in the femoral component is determined by taking
measurements on the pre-operative radiographs.

Fig. 6.13 The safe zone of orientation of femoral and acetabular components.

Total hip replacement: femoral head size

The femoral head diameter is an important determinant of the range of motion, stability and wear properties
of the replacement joint. Range of motion and stability are related and are a function of the head–neck ratio.
The head–neck ratio increases with a larger diameter femoral head or with a tapered (narrower) neck. A
greater head–neck ratio permits a wider primary arc of motion before impingement of components.
Therefore, a greater head–neck ratio offers better range of motion and stability. In addition, a larger
diameter femoral head also requires a greater displacement for dislocation, i.e. ‘jump distance’, and
therefore offers further stability.
Wear of the bearing surfaces of total hip replacement can be measured in two ways:

Linear wear. The thickness of the acetabular cup decreases as it wears with use. Linear wear is the
change in the thickness of the acetabular cup with time, and is measured as follows:
Linear wear [mm] = Original thickness of acetabular cup [mm] – New shortest thickness of acetabular
cup [mm], as measured on a plain AP radiograph.
Volumetric wear. This describes the actual volume of wear of the acetabular component. Volumetric
wear is related to linear wear with this simple geometry-based equation:
Volumetric wear [mm3] = π x (radius of femoral head [mm])2 x linear wear [mm]
Therefore, a larger diameter femoral head produces more volumetric wear for the same linear wear.

Wear between different femoral head sizes and bearing combinations

Volumetric wear is proportional to the frictional torque of the replacement joint. The formula for frictional
torque is:
See pages 82–83 for further explanation.
Therefore, an increase in femoral head size increases frictional torque and related volumetric wear.
However, different bearing materials have different coefficients of friction and therefore produce different
amounts of wear for the same femoral head size.

Example
Metal head on polyethylene acetabular cup is the most commonly used bearing combination. In the past, a
28 mm diameter femoral head was considered to provide the optimum balance between stability and wear.
Now, other bearing materials with better wear properties are used to develop larger diameter femoral
heads.
The primary arcs of motion for a 28 mm and a 36 mm diameter femoral head are 123° and 136°,
respectively. In metal on polyethylene bearing combination, this increase in femoral head size increases
frictional torque by 1.29 times and volumetric wear by 2.6 times. The coefficient of friction of ceramic on
ceramic bearing combination is 1.33 times less than metal on polyethylene bearing combination. Therefore,
for the same increase in femoral head size, it experiences practically no increase in frictional torque, and
the increase in volumetric wear is 100 times less.

Fig. 6.14 The effect of head–neck ratio on primary arc of motion. A greater head–neck ratio permits a
wider primary arc of motion before impingement of components. Note that the outer diameter of the
acetabular cup remains the same.

Fig. 6.15 A larger diameter femoral head requires a bigger jump distance before dislocation. It also
reduces the empty space around the acetabular component into which it can dislocate. These factors make it
more difficult to dislocate a larger size femoral head.
Fig. 6.16 The relationship of wear and stability with femoral head size and bearing combinations. A
larger diameter femoral head provides more stability, but in a metal on polyethylene (MoPE) bearing
combination produces excessive wear that leads to osteolysis and implant loosening. Other bearing
combinations, i.e. metal on highly cross-linked polyethylene (MoXLPE), metal on metal (MoM) and
ceramic on ceramic (CoC), have much lower coefficients of friction and permit the use of larger diameter
femoral head with less wear. In fact, wear in MoM and CoC combinations decreases with an increase in
femoral head diameter, because fluid film lubrication takes effect as the femoral head diameter increases
(see pages 92–93).

Total hip replacement: bearing surfaces

Two of the main reasons for revising a total hip replacement are aseptic loosening and instability, i.e.
dislocations. Different combinations of bearing surfaces present various solutions to these problems, but
also have their own limitations. The governing factors that connect different bearing surfaces combinations
to aseptic loosening and instability are as follows:

(a) Osteolysis. This is the main cause of aseptic loosening. Wear particles from mainly the
acetabular cup produce a biological reaction that leads to osteolysis and implant loosening.
(b) Femoral head diameter. Although a larger diameter femoral head offers a greater range of
motion and is more stable, it also leads to increased wear.

Metal on polyethylene
Metal femoral head on polyethylene acetabular cup is the most commonly used bearing combination, and
has a well-established clinical track record. Femoral heads of 22 mm and 28 mm diameter have been used
traditionally, as a bigger femoral head increases polyethylene wear rate. The newer ‘highly cross-linked’
polyethylene is harder and produces about 60% less wear, and is encouraging the use of larger diameter
femoral heads. This is the only combination where a posterior lip augmentation device can be applied to the
acetabular cup to reduce the risk of dislocation.

Metal on metal
In this combination, femoral head and acetabular cup are both made of metal. As metals are hard materials,
this is a described as a ‘hard on hard’ bearing combination. In addition, metals can be polished to reduce
surface roughness, which further reduces wear. Therefore, larger size (≥36 mm) femoral heads can be
used, as even the increased wear rate is still a magnitude less than in metal on polyethylene combination.
Metal on metal combination can also retain fluid between the bearing surfaces and therefore can achieve
fluid film lubrication, which further minimises wear.

Ceramic on ceramic
In ceramic on ceramic bearing combination, the femoral head and acetabular cup are both made from
ceramic material. As ceramics are harder than metals, these can be highly polished. Ceramic on ceramic
combination is therefore very hard and smooth. Ceramics are also hydrophilic and very ‘wettable’, which
means that ceramic bearing surfaces can more easily achieve fluid film lubrication. Therefore, ceramic on
ceramic bearings have the lowest clinical wear rate of any bearing combination.

Other bearing combinations

Oxinium™ is a relatively new bearing surface that consists of a metal femoral head with a ceramic surface
coating, for use with a ceramic acetabular cup. It combines the benefits of both materials: metal femoral
head cannot fracture and the ceramic surface has a very low wear rate with no issues regarding the
accumulation of metal ions. It is, however, very expensive, and there are limited long-term clinical data
(Table 6.3).

Table 6.3. A summary of main advantages and disadvantages of different bearing combinations

Advantages Disadvantages

Metal on Predictable performance and failure Osteolysis and aseptic loosening:

polyethylene mechanisms. does not support fluid film
Local biological effects only. lubrication
Forgiving materials: performance is high wear rates with larger
not affected in a wide range of femoral head sizes.
alignment of implants. Highly cross-linked polyethylene is:
Polyethylene acetabular cup brittle, and susceptible to
available: fracture
in a large range of sizes less forgiving of implant mal-
with posterior lip augmentation alignment.
device, and other constraint
designs
in highly cross-linked
composition.
Revision options are not limited.
Relatively cheap.
 Advantages Disadvantages

Metal on Virtually no risk of fracture (although Effects of metal ions wear particles:
metal the components are still prone to immunological effects:
scratches). Avascular lymphocytic vascular
Low wear rate. associated lesions (ALVAL) and
‘Hip resurfacing’ implants are only Lymphocyte dominated
available in metal on metal immunological answer (LYDIA).
combination, as only this combination chromosomal aberrations.
can endure the design in the long toxicity.
term. pseudotumours.
Long history of use. potentially carcinogenic.
Requires long-term patient follow-up
to monitor for issues relating to metal
ions.
Absolute contraindications:
renal failure (as metals ions
accumulate in kidneys)
women of child-bearing age
(unknown effect of metal ions on
the foetus).

Ceramic on Lowest wear rate. Performance very sensitive to

ceramic Biological inert wear particles, implant alignment; susceptible to
therefore low inflammatory reaction catastrophic fracture if components
and osteolysis, and no issues of ion slightly out of alignment.
toxicity. Very fine wear debris, which is
difficult to remove at revision
surgery, therefore revision options
often restricted to the same
combination.
Can sometimes produce audible
noise when walking, e.g. squeaking
or clicking.
More expensive than other bearing
options.
 7 Biomechanics of the knee and total knee
replacement

Knee: functional anatomy I

The knee is the largest and most complex joint in the body. It is an
articulation among the distal femur, proximal tibia and patella, surrounded by
a soft tissue envelope of capsule, ligaments, tendons and muscles. The
specific anatomical features and interactions of these structures determine the
stability and movements of the knee joint.

Geometry and alignment of the bones

Tibiofemoral joint
The medial tibial plateau is larger than the lateral tibial plateau. It also has a
concave depression whereas the lateral tibial plateau has a convex surface.
Both plateaux have a posterior tilt of about 7°, but the joint surface tilt is
minimised to 3° by the menisci. The plateaux also have a 3° varus tilt, i.e. the
joint surface is slopped medially.
On the femoral side, the medial condyle is larger, more curved and extends
more distally than the lateral condyle.
This asymmetry between the medial and lateral compartments of the knee
determines the relative mobility of each compartment. The concave medial
tibial plateau offers more constraint and the convex lateral tibial plateau
offers less constraint to the movements of their respective femoral condyles.
The posterior and varus tilts of the tibial plateaux have implication for the
alignment of bone cuts for total knee replacement.

Patellofemoral joint
The patella is a sesamoid bone that articulates with the femur only. It has two
articular facets: the lateral facet is slightly larger than the medial facet. The
quadriceps tendon and patellar tendon insert at the anterior aspect of the
patella, and its thickness increases the extension lever arm of the quadriceps
muscle. This effect is maximum in full knee extension and reduces with knee
flexion, when the patella sinks into the intercondylar groove. Due to the
mechanical advantage provided by the patella, the quadriceps needs to
generate about 20%–30% less force for a certain extension torque. The
medial and lateral facets of the patella also increase the surface area over
which the compressive stress is applied to the femur. The patella also
protects the tibiofemoral joint from direct trauma.
The Q-angle (quadriceps angle) describes patella alignment. It is formed
by lines connecting the centre of patella with the anterior superior iliac spine
proximally and the centre of tibial tubercle distally. The normal Q-angle
range is 10°–14° for males and 15°–17° for females; the Q-angle is greater in
females, as they generally have a wider pelvic girdle. The greater the Q-
angle, the greater the lateral force on the patella, and a Q-angle more than
20° is a risk factor for patella subluxation and patellofemoral joint pain. The
restoration of the normal Q-angle is important in total knee replacement for
normal patella tracking.
Fig. 7.1 Geometry of articular surfaces of femoral condyles and tibial
plateaux. (LAP = Lateral anteroposterior diameter; MAP = Medial
anteroposterior diameter.)
Fig. 7.2 The Q-angle is formed by lines connecting the centre of patella
with the anterior superior iliac spine proximally and the centre of tibial
tubercle distally. The Q-angle is greater in females because they generally
have a wider pelvic girdle. The greater the Q-angle, the greater the lateral
force on the patella.
Fig. 7.3 There are three axes of the knee joint: flexion–extension and
axial rotational axes of tibiofemoral joint and flexion–extension axis of
the patella. The flexion–extension axis of the tibiofemoral joint is also
referred to as the transepicondylar axis or the surgical epicondylar axis. It is
defined by a line that connects the lateral epicondylar prominence to the
medial epicondylar sulcus. There is a parallel and orthogonal relationship
between the axes.

Knee: functional anatomy II

Soft tissues of the knee
The cruciate ligaments
The anterior and posterior cruciate ligaments are the primary restraint to
anterior and posterior translation of tibia, respectively. The cruciate
ligaments function in a special configuration, which in mechanical terms is
known as the ‘four-bar linkage mechanism’. In this model, the cruciate
ligaments form the two crossed bars and the bones effectively form the other
two bars. As the knee joint flexes and extends, the cruciate ligaments are able
to pivot about their insertion points. The length of the four bars remains
constant, but the angle between them changes during joint motion. In addition,
the centre of rotation of this four-bar linkage hinge is at the crossover point
of the cruciate ligaments, which also changes with knee position. The
cruciate ligaments are not rigid and experience some change in length, which
allows internal–external rotation of the joint. The ligaments set-up is
therefore described as ‘modified’ (as opposed to rigid) four-bar linkage
mechanism.

The menisci
The medial and lateral menisci increase the congruency between the tibial
and the femoral articular surfaces. The medial meniscus is attached to the
medial joint capsule, and the lateral meniscus is attached to the femur by the
meniscofemoral ligament. Therefore, during knee flexion, the lateral
meniscus translates posteriorly much more than the medial meniscus. The
menisci contribute to stability and facilitate movements of the knee joint.
The menisci also help to distribute knee joint load over maximum contact
area. The compressive load from the femur is transmitted to the ‘menisco-
tibial’ interface, with the menisci carrying approximately 70% of the load
across the knee joint. The removal of menisci results in load transmitted
through a limited, central contact area, which leads to a three- to five-fold
increase in the stress applied to the articular surfaces. The menisci therefore
protect the articular surfaces from excessive stresses.
A further function of the menisci is as ‘shock absorbers’ of the knee joint.
They reduce the actual load transmitted across the knee joint by dissipating
axial stress through generation of hoop stresses. The menisci deform
elastically when loaded, and experience radial and circumferential stresses.
The circumferential, or hoop, stresses are in a different plane to joint loading
planes, which reduces the stress transmitted across the menisci.
The menisci are viscoelastic and also display hysteresis, which is the
ability to absorb energy when subjected to repeated loading and unloading
cycles. This energy is dissipated in changing the shape of the menisci during
loading and unloading.
Fig. 7.4 The cruciate ligaments function as a four-bar linkage hinge to
stabilise the joint whilst permitting motion.
Fig. 7.5 The biomechanical functions of the menisci are to (a) stabilise
the articulating surfaces and facilitate joint motion; (b) distribute joint
load over maximum contact area; and (c) act as shock absorbers by
generating hoop stresses.

Knee: flexion–extension arc

The main arc of motion of the knee joint is flexion and extension in the
sagittal plane. However, the knee is not a simple hinge joint and small
motions in other planes are essential to enable the joint to flex and extend.
The flexion–extension arc of the knee can be divided into three ranges.

Screw home arc

−5° to 10°: This is the range where the knee is ‘locked’. The femur is in
neural alignment with the tibia when the knee is at 10° of flexion. It begins to
internally rotate as the knee extends. In full knee extension: the femur is
internally rotated with reference to the tibia; the cruciate ligament four-bar
linkage mechanism is tightened; and the joint is locked. The centre of weight
of the upper body lies anterior to the centre of rotation of the joint, and the
resultant moment is balanced by passive resistance of the posterior capsule
and ligaments. This allows the quadriceps to stop contracting and the
position is maintained passively with little energy expenditure.
The knee is ‘unlocked’ by the action of the popliteus located at the
posterolateral corner of the knee, and the femur begins to externally rotate as
knee flexion is initiated.

Functional arc
10° to 120°: This range is associated with further external rotation of the
femur. This is because, as the knee flexes, the smaller lateral femoral condyle
also begins to slide posteriorly, whereas the larger medial femoral condyle
maintains its relative position on the tibia, which therefore has the overall
effect of externally rotating the femur. This is considered an ‘active’ range,
because the ligaments and muscles are controlling the rate of motion.

Deep flexion arc

120° to 145°/160°: In this range, the femur continues to externally rotate and,
in addition, both femoral condyles translate posteriorly. This ‘rollback and
slide’ of the femur allows the femoral condyles to clear the tibia to achieve
deep flexion. This is considered a ‘passive’ range, as it can only be achieved
with manual assistance or by squatting.

Patellofemoral contact zones

The patella articulates with the lateral femoral ridge when the knee is in full
extension, and slides into the intercondylar notch at full knee flexion. This
sliding of patella along the femur brings specific zones of femur and patella
in contact during different ranges of the flexion–extension arc. The contact
area becomes more proximal on the patella and more distal on the femur with
increasing knee flexion. The size of the patellofemoral contact area also
increases with knee flexion, i.e. with increased force of the quadriceps.
When the knee is at 30° of flexion, patellar contact area is about 2 cm2, and
when it is at 90° of flexion, the contact area triples to about 6 cm2. This
maintains stress at the patellofemoral joint at a relatively constant level
despite an increase in quadriceps force with increasing knee flexion (Table
7.1).

Fig. 7.6 The flexion–extension arc of the knee can be divided into three
ranges, which are associated with specific motions of the femur over the
tibia.
Fig. 7.7 Patellofemoral contact zones during the flexion–extension arc.
The contact area becomes more proximal on the patella and more distal on
the femur with increasing knee flexion.

Table 7.1 A note on the flexion–extension arc

The flexion–extension arc of the knee is described here in terms of

the alignment of the femur with respect to the tibia with increasing
knee flexion. It can also be described in terms of the alignment of
tibia with respect to femur and/or in terms of increasing knee
extension. The description of the relative motion between the
bones will change accordingly in these cases.

Knee joint reaction force

Static analysis can be used to estimate the compressive forces acting on
tibiofemoral and patellofemoral joints. These forces increase with knee
flexion angle, and also depend on whether the leg is in the stance or swing
phase of the gait cycle. The stance phase is further divided into double-leg
and single-leg support periods. The knee joint reaction forces are greatest at
maximum knee flexion during the single-leg stance period of the gait cycle.
However, maximum knee flexion during the single-leg stance period varies
according to the activity undertaken, e.g. it is approximately 20° walking on
flat surface, 60° ascending stairs and 85° descending stairs (Table 7.2). As a
result, the joint reaction forces produced in the knee compartments are very
variable, and any analysis must take into account the exact loading
conditions.

Table 7.2. Typical functional range of motion (degrees) required for

some common everyday activities

Activity Hip joint Knee joint

Level walking −5–40 0–67

Ascending stairs 0–67 0–83

Descending stairs 0–35 0–90

Sitting down 0–105 0–110

Tying a shoelace* 0–125 0–106

Squatting** 0–117 0–130

* This has to be modified after total knee (and hip) replacement.

** Deep squatting is not possible/advisable after total knee (and hip) replacement.

Tibiofemoral joint reaction force

During a simple double- or single-leg stance: the knee is in full extension;
there is no resultant moment due to the supported upper body weight; and the
quadriceps muscles are not contracting. The tibiofemoral joint reaction force
is roughly equal to the supported upper body weight. The weight of the leg
below the knee is about 5% of total body weight and can be ignored;
therefore, each tibiofemoral joint is subjected to a compressive force
equivalent to one-half of total body weight during double-leg stance and total
body weight during single-leg stance.
 The tibiofemoral joint reaction force increases sharply with knee flexion.
It is the resultant vector of the ground reaction force and quadriceps force. A
commonly used example to illustrate the peak forces produced in the
tibiofemoral joint is during ascending stairs. The typical height of a step of
stairs is 20 cm, which requires about 60° of knee flexion during the single-
leg stance period of the gait cycle. The static analysis shows that
tibiofemoral joint reaction force is about five times total body weight.

Assumptions
In addition to the general assumptions, the following specific assumptions
are applied to this static analysis:

The quadriceps provides all of the force to extend the knee and exerts a
tensile force through the patella tendon.
The ground reaction force is equal and opposite to total body weight,
and acts below the centre of weight of the body.
The following physical measurements are based on anthropometric
data:
Moment arm of patella tendon at 60° of knee flexion is 45 mm.
Moment arm of ground reaction force at 60° of knee flexion is 180
mm.

Patellofemoral joint reaction force

The compressive force in the patellofemoral joint is the resultant vector of
the patella tendon and quadriceps tendon forces, and also increases with
knee flexion. At 60° of knee flexion during single-leg stance in ascending
stairs, the patellofemoral joint reaction force is estimated to be about four
times total body weight.
Fig. 7.8 Free-body force diagram of the knee joint during ascending
stairs at the instance of opposite foot toe-off. The tibiofemoral joint is
acting as a class I lever. The actual degree of knee flexion required to ascend
stairs is determined by the height of the step as well as the height of the
person. For a standard 20 cm rise and 30 cm run step, on average 60° of knee
flexion is required. This is associated with approximately 35° of hip flexion.
Moment arm of ground reaction force is usually four times the moment arm of
patella tendon at any degree of knee flexion. (Force G = Ground reaction
force; Force P = Patella tendon force; and Force J = Joint reaction force.)
Calculations
Applying the conditions of equilibrium:

1. Sum of all moments is zero.

2. Sum of all forces is zero.

Tibiofemoral joint reaction force during ascending stairs. The joint
reaction force is about five times total body weight (TBW).

Total knee replacement (TKR): design of

components I
Total knee replacement involves resurfacing the femoral condyles and tibial
surfaces with or without resurfacing the patella. Modern total knee
replacement implants have the following design features in common:

The metal-based femoral component has two symmetrical runners that

are bridged at the anterior aspect to form an articular surface for the
patella.
The polyethylene tibial insert has two concave articular surfaces.
The polyethylene patellar component is circular and has a convex
articular surface.
The tibial component is metal or polyethylene based and has a stem for
fixation to bone.

Single-radius vs multi-radius
The total knee replacement implants are designed to match the geometry of
bones as much as possible. In the past, the femoral component had a variable
radius of curvature, on the basis that the posterior femoral condyles are
elliptical when viewed from the side. However, the modern total knee
replacements have a femoral component with a single radius of curvature in
the sagittal plane, as the posterior femoral condyles are found to be circular
when viewed from the side along the slanted flexion–extension
(transepicondylar) axis of the knee. The single-radius total knee replacement
design has a similar axis of rotation to that of the natural knee joint, which
optimises the function of the collateral ligaments and other soft tissues of the
knee. Therefore, it is considered to better replicate the normal motions of the
natural knee joint.
Stability vs mobility
Total knee replacement prostheses can be classified according to their
degree of ‘constraint’ to natural knee motion. This, in turn, is related to the
design of the implants and depends on a number of factors, such as
conformity of the bearing surfaces and features that substitute the function of
supporting ligaments.
The range of motion of a natural or replacement joint is interconnected to
its stability. In general, the more mobile a joint is, the less relatively stable it
is and the more it relies on the surrounding soft tissues for stability. For
example, the shoulder is the most mobile joint in the body, but is also most
prone to dislocations and relies on its labrum and other soft tissues for
stability; in comparison, although the hip is also a ball and socket joint, its
motion is more constrained and the natural hip joint is far less prone to
dislocations.
As a rule in total knee replacement, as the bearing surfaces become more
conforming (i.e. matching) the prostheses gain more stability with less
reliance on soft tissues. However, this is at the cost of increased constraint to
motion. The selection of the type of implant design is therefore a compromise
between its stability and mobility.
Fig. 7.9 Single-radius and multi-radius designs of femoral component.
Medial and lateral collateral ligaments of the knee are inserted on the femur
along the transepicondylar axis. The single-radius femoral component
maintains natural tension in these ligaments throughout the flexion–extension
arc. Clinical studies indicate that single-radius knee replacement provides
better range of motion and improved overall functional results than multi-
radius knee replacement. (Note: The transepicondylar axis is slanted in the
axial plane, but is straight in the coronal plane as shown on page 113. It
appears straight on page 125, because the whole femur is rotated.)
Fig. 7.10 Total knee replacement prostheses can be classified according
to their degree of ‘constraint’ to natural knee motion. (ACL = anterior
cruciate ligament; PCL = posterior cruciate ligament.)

Total knee replacement: design of components II

Fixed bearing vs mobile bearing
Total knee replacements have excellent long-term results, but a major
limitation in their long-term survival is the wear of the polyethylene tibial
insert. All designs of total knee replacement utilise metal (femoral
component) on polyethylene (tibial insert) bearing combination. The tibial
insert is the softer of the bearing surfaces and with time starts to wear,
producing wear particles that lead to osteolysis and aseptic loosening at the
bone–implant interface. Polyethylene wear is directly related to contact
stresses generated on the tibial insert during relative motion between the
bearing surfaces. Therefore, wear can be minimised by reducing contact
stresses.
In the ‘fixed bearing’ total knee replacement design, the tibial insert is
fixed to the tibial base plate. This prevents wear of the inferior surface of the
tibial insert. The wear of the superior surface of the tibial insert is minimised
by increasing its conformity to the femoral component, which increases the
contact surface area and reduces contact stresses. An increase in conformity,
however, also increases constraint to motion. This, in turn, leads to increased
stresses at the bone–implant interface, which contributes to aseptic
loosening.
The ‘mobile bearing’ knee replacement design offers an alternative
solution to address the competing issues of wear, conformity and constraint.
In this design, the tibial insert is mobile and not fixed to the tibial base plate.
The wear of the superior surface of the tibial insert is minimised by the
highly conforming interface with the femoral component, which distributes
load over maximum contact area and therefore minimises contact stresses.
The high conformity means that the motion at the interface is limited to
flexion and extension only. However, axial rotation occurs at the interface
between the inferior surface of the tibial insert and tibial base plate.
Therefore, this design achieves high conformity without restricting axial
rotation. A better range of motion at the bearing surfaces also reduces
stresses at the bone–implant interface.
There are a number of limitations to the mobile bearing knee replacement
design. Although the wear of the top surface of the tibial insert is minimised,
there is added wear of the under-surface of the tibial insert, known as
‘backside’ wear. The wear particles are finer, and can more easily
accumulate between the bone–implant interface. Therefore, the mobile
bearing knee replacement design may even be at an increased risk of
osteolysis and aseptic loosening. In addition, an increased motion between
the interfaces makes it more prone to instability and dislocations. It is also
not clear yet if the design produces a better range of motion clinically.

Patellar component
The patellar resurfacing component is circular and has a non-anatomical
dome-shaped bearing surface. However, the convex shape is more tolerant of
small degrees of patella mal-alignment. This design also eliminates the need
to orientate the component.

Fig. 7.11 In fixed bearing total knee replacement design, the tibial
insert is fixed to the tibial component. All of the motion occurs at the
interface between the femoral component and tibial insert. The superior
surface of the tibial insert conforms to the shape of the femoral component.
Although this increases the bearing surfaces' contact area, it also restricts
axial rotation between them. (Image (a) is reproduced with permission of FH
Orthopaedics.)
Fig. 7.12 In mobile bearing total knee replacement design, the tibial
insert is not fixed to the tibial component. Flexion–extension occurs at the
interface between the femoral component and tibial insert. The superior
surface of the tibial insert is highly conforming to the shape of the femoral
component. This ensures that the load is distributed over a maximum area.
Axial rotation takes place at the interface between the tibial insert and the
base plate.

Total knee replacement: alignment of components

The main biomechanical goals of knee replacement are to: maintain the
mechanical axis of the lower limb; restore joint line and load-carrying
capacity of the joint; and provide a functional range of movement with a
well-fixed and durable prosthesis. The correct size and alignment of
components and soft tissue balance are essential for a successful total knee
replacement.

Femoral component
During flexion and extension of the knee, two aspects of the femoral condyles
articulate with tibia: the distal condylar surface in knee extension and the
posterior condylar surface in knee flexion. Both surfaces of the femur require
correctly orientated cuts to achieve correct alignment of the femoral
component in knee flexion and extension.
The distal femoral cut is made perpendicular to the mechanical axis. This
ensures that the component is neutral with the mechanical axis and aligned
with the centre of the femoral head. This is actually achieved by basing the
cut on the femoral anatomical axis. As the mechanical axis is 5°–7° valgus to
anatomical axis, a distal femoral cut 5°–7° valgus to the femoral anatomical
axis is perpendicular to the mechanical axis.
The mechanical axis cannot be used to guide the posterior femoral cut, as
the knee is in flexion. The posterior femoral cut is made parallel to the
surgical epicondylar axis, which is defined as the line connecting the lateral
epicondylar prominence with the medial epicondylar sulcus. A cut in this
alignment results in 3° external rotation of the femoral component, which
corresponds to the posterior slope of the tibial surface. The external rotation
of the femoral component therefore ensures an equal gap between the bones
in knee flexion. It is also important in restoring the correct Q-angle.

Tibial component
The proximal tibial cut is made perpendicular to mechanical axis of the tibia
(which in practical terms is the same as its anatomical axis) to match with the
distal femoral cut and ensure an equal gap between the bones in knee
extension. The tibia is cut with about 3° posterior slope, unless the design of
the implants accounts for the posterior slope.
The rotational alignment of the tibial component determines patella
tracking. The internal rotation of the tibial component leads to external
rotation of the tibial tubercle, which increases the Q-angle and the associated
risk of patella subluxation and patella–femoral joint pain. The most reliable
method for correct orientation of the tibial component is to align it with the
medial third/border of the tibial tubercle.

Patellar component
Most of the complications after total knee replacement are related to the
patellofemoral joint. The correct alignment of the femoral and tibial
components is essential for normal patella tracking. The patellar implant is
placed in a slightly medialised position, which ensures that the apex of the
patella is in normal alignment. In some cases, the femoral implant can also be
placed in a slightly lateral position, which relatively medialises the patella
and improves patella tracking (Table 7.2).

Fig. 7.13 The three main cuts in total knee replacement that determine
the alignment of femoral and tibial components are distal and posterior
femoral cuts and a proximal tibial cut. The other cuts involved in total knee
replacement are: anterior femoral cut and anterior and posterior chamfer cuts
– these shape the femur to match the geometry of the femoral component; and
patellar cut to resurface the patella. Therefore, total knee replacement
involves seven bone cuts.
 8 Biomechanics of the shoulder

Shoulder: functional anatomy I

The shoulder links the upper limb to the trunk. It is a complex of four joints:
glenohumeral, acromioclavicular, sternoclavicular and scapulothoracic
joints. The glenohumeral joint is the main joint of the shoulder complex, and
is usually referred to as the shoulder joint. This chapter considers the
biomechanics of the glenohumeral joint.
The glenohumeral joint is a ball and socket joint formed by the articulation
of the head of the humerus with the glenoid of scapula.

Geometry and alignment of bones

The humeral head is retroverted 20°–30° with respect to the intercondylar
plane of the distal humerus. This provides some restraint to anterior
dislocation. The head–shaft angle is 130°–150°. Adjacent to the humeral
head are the greater and lesser tuberosities, which are the attachment sites for
ligaments and rotator cuff muscles.
The scapula is angled 30° anterior to the coronal plane – this is the ‘plane
of the scapula’. The scapula is also superiorly inclined 3°. The glenoid is
retroverted 5° from the plane of the scapula and is superiorly inclined 5°.
The superior inclination of the glenoid provides some restraint to inferior
displacement of the humeral head.
The humeral head is approximately four times the size of the glenoid; this
relationship is similar to that of a golf ball on a tee. As the humeral head is
not that well ‘covered’ (constraint) by the glenoid, it can have a wide range
of motion before structures on the opposite sides of the joint come into
contact. As a result, the shoulder joint is the most mobile joint in the body.
However, this lack of bony constraints also means that the joint is mainly
dependent on other factors for stability. The key factors for shoulder stability
are discussed below and in the next section.
Glenoid labrum
The glenoid is augmented by a rim of fibrocartilage, the glenoid labrum,
which effectively doubles the depth of glenohumeral joint. The glenoid
without the labrum encompasses about a quarter of the humeral head, but
with the labrum this increases to a third of the humeral head. Although the
overall joint contact area is relatively small, the articulating surfaces are
highly congruent. The motion of the glenohumeral joint is therefore almost
purely rotational. There is normally limited translation between the
articulating surfaces. However, if the labrum is missing, abnormally formed
or traumatically detached from the glenoid, there is loss of congruency
between the articulating surfaces. This increases relative translation between
the surfaces and therefore the risk of shoulder dislocation.
The three common structural abnormalities of the glenoid and labrum that
can increase the risk of shoulder instability are as follows:
Bankart lesion: This is a traumatic detachment of anterior labrum from the
glenoid.
Bony bankart lesion: This describes loss of a portion of glenoid with the
anterior labrum.
SLAP tear: This is a superior labral tear from anterior to posterior
(SLAP).
Fig. 8.1 Geometry and alignment of bones forming the glenohumeral
joint.

Fig. 8.2 The glenohumeral joint is like a golf ball on a tee. It is therefore
a ‘shallow’ ball and socket joint. The humeral head is naturally inclined to
‘fall off’ i.e. dislocate. Most of the shoulder dislocations (90%–97%) are
anterior.
Fig. 8.3 The glenoid labrum is an important stabiliser of the
glenohumeral joint. It increases the congruency between the articular
surfaces. Bankart lesion, bony bankart lesion and SLAP tears are defects of
the glenoid–labrum interface and can increase the risk of shoulder instability.

Shoulder: functional anatomy II

Joint capsule
The glenohumeral joint capsule is inherently lax and has twice the surface
area of the humeral head. It is therefore quite flexible and permits a wide
range of motion. However, different sections of the capsule tighten in
different joint positions. Therefore, the capsule also contributes to joint
stability. In frozen shoulder (adhesive capsulitis), there is contracture of the
anterior aspect of the capsule, which reduces the overall motion of the joint,
especially external rotation.
The capsule provides joint stability through a second mechanism. It
maintains a negative intra-articular pressure, which pulls the capsule and the
surrounding ligaments inwards. The closely adhered soft tissues prevent
excessive translation between the articular surfaces and therefore reduce the
risk of joint dislocation. A defect in the capsule or significant joint effusion
greatly increases translation between articular surfaces.

Ligaments
There are three glenohumeral ligaments, which resist translation of the
humeral head in different directions.

The superior glenohumeral ligament is the primary restraint to inferior

translation.
The middle glenohumeral ligament is a restraint to anterior translation,
being most effective in mid-abduction.
The inferior glenohumeral ligament is the most dominant in maintaining
joint stability. It consists of three portions: anterior band, axillary pouch
and posterior band. The anterior band provides restraint to anterior
translation when the arm is abducted and externally rotated. The
posterior band provides restraint to posterior translation when the arm
is abducted and internally rotated.

In addition, the coracohumeral ligament also resists inferior translation and

external rotation of the humeral head.

Rotator cuff muscles

Rotator cuff muscles provide joint motion and stability. There are four
rotator cuff muscles:

supraspinatus assists deltoid in shoulder abduction

subscapularis internally rotates the shoulder
infraspinatus and teres minor externally rotate the shoulder.

The supraspinatus works with deltoid to produce a force couple (see pages
6–7) to abduct and forward flex the shoulder. The force of deltoid is directed
mainly upwards and the force of supraspinatus is directed mainly medially.
The resulting force couple rotates the humeral head and draws it medially
into the glenoid.
Rotator cuff muscles provide stability by passive muscle tension and
dynamic contraction. The importance of dynamic stability provided by the
rotator cuff muscles is highlighted by the following example: it is difficult to
abduct the arm fully with the palm facing towards the back but much easier
with the palm facing forward. In the first case, supraspinatus and deltoid
abduct the arm, but the impaction of greater tuberosity against acromium of
the scapula halts the motion. In the second case, infraspinatus and teres minor
rotate the humeral head externally, and together with supraspinatus and
subscapularis draw the humeral head medially and inferiorly, allowing it to
clear the acromium and achieve high abduction. The full arc of motion also
requires rotation of the scapulothoracic joint.

Fig. 8.4 The shoulder is the most mobile joint in the body. It allows up to
180° of forward flexion, 60° of extension (posterior elevation), 180° of
abduction and 90° of internal and external rotation. The range of motion of
shoulder normally decreases as part of the ageing process, although physical
activity can counteract this process.
Fig. 8.5 There are four rotator cuff muscles: supraspinatus,
infraspinatus, teres minor and subscapularis. The supraspinatus pulls
mainly in a horizontal direction. The infraspinatus pulls approximately 45°
and the teres minor pulls approximately 55° to the horizontal. Similarly, the
subscapularis pulls approximately 45° to the horizontal.

Fig. 8.6 The deltoid and supraspinatus produce a force couple to abduct
and forward flex the shoulder. However, high shoulder abduction also
requires obligatory external rotation of the humeral head. The rotator cuff
muscles work together to draw the humeral head medially and inferiorly,
allowing it to clear the acromium.
Shoulder joint reaction force
The shoulder joint reaction force depends on the position of the glenohumeral
joint and the elbow joint. It is usually greatest when the shoulder is abducted
to 90°. The calculations of exact joint reaction force are complex due to the
large number of muscles active in any particular joint position. The
following simplified static analyses highlight the effect of elbow extension
and flexion on shoulder joint reaction force.

Shoulder abduction with elbow in extension

When the shoulder is in 90° of abduction, the deltoid contracts to maintain
the position of the arm. The shoulder acts as the axis of a class III lever
system; the muscles have a shorter lever arm than the weight of the arm.
According to the static analysis, the minimum joint reaction force in this
position is about one-half of total body weight.

Assumptions
In addition to the general assumptions, the following specific assumptions
are applied to this static analysis:

The deltoid provides all of the abduction force to stabilise the arm.
The deltoid force is directed medially in a horizontal plane.
The following physical measurements are based on anthropometric
data:
The mass of the arm is approximately 5% total body mass.
Therefore, in a 70 kg adult, the mass of the arm is assumed to be
3.5 kg. This equates to a weight of 35 N. This is assumed to act as
a point load at the centre of weight of the arm, taken to be at the
elbow joint, at 300 mm from the centre of rotation of the
glenohumeral joint.
At 90° shoulder abduction, the moment arm of deltoid is 30 mm
from the centre of rotation of the glenohumeral joint.

Shoulder abduction with elbow in flexion

Flexing the elbow moves the centre of weight of the arm closer to the body.
Therefore, the abductor muscles have to produce less force to hold the arm in
the abducted position. The static analysis shows that the joint reaction force
now is about a quarter of total body weight.

Assumptions
As above, except:

when the elbow is flexed, the centre of the weight of the arm is 150 mm
from the centre of rotation of the glenohumeral joint.

Clinical implications
The deltoid muscle force and therefore shoulder joint reaction force are
considerably less when the elbow is flexed than when it is extended.
Therefore, shoulder rehabilitation exercises can be adapted according to the
patient's functional level by changing the degree of elbow extension.
Fig. 8.7 Free-body force diagram of the upper limb showing the forces
acting about the shoulder joint. The shoulder is abducted to 90° and the
elbow is extended. (Force D = Deltoid force; Force W = Weight of the arm;
and, Force J = Joint reaction force.)

Calculations

1. Sum of all moments is zero.

2. Sum of all forces is zero.

Fig. 8.8 Shoulder joint reaction force in abduction with elbow in
extension. The joint reaction force is one-half of total body weight.
Free-body force diagram of the upper limb showing the forces acting
about the shoulder joint. The shoulder is abducted to 90° and the elbow is

flexed.
Calculations

1. Sum of all moments is zero.

2. Sum of all forces is zero.

Shoulder joint reaction force in abduction with elbow in flexion. The joint
reaction force is a quarter of the total body weight.
*The arrow for Force J represents the magnitude and direction of the joint
reaction force.
Shoulder replacement
Shoulder replacement is the third most common joint replacement. There are
three forms of shoulder replacement:

Hemiarthroplasty i.e. replacement of the humeral head only.

Total shoulder replacement, i.e. replacement of the humeral head and
glenoid.
Reversed shoulder replacement, i.e. the bearing surfaces have an
opposite profile to the normal anatomy.

This section takes a look at design and alignment of implants used in

hemiarthroplasty and total shoulder replacement. The essential prerequisite
for these ‘conventional’ procedures is that the rotator cuff is intact and
functional. However, isolated small rotator cuff tears are compatible with
these procedures. Non-functioning deltoid or rotator cuff are absolute
contraindications to hemiarthroplasty and total shoulder replacement.

Humeral component
There are two main designs of humeral component: stemmed implant and
resurfacing implant. The stemmed implant relies on intramedullary fixation
and therefore has a cylindrical stem, which can be fixed to bone, either with
or without cement. Cemented fixation immediately stabilises the component
in the applied position. Therefore, cemented fixation is advantageous: when
the bone quality is poor; when there is pre-existing bone deformity; or, when
joint replacement is being performed for proximal humeral fractures, where
the cement helps to stabilise fracture fragments as well as the implant. On the
other hand, cementless fixation is also used widely as it avoids the problems
associated with cemented fixation, especially the challenges of removing the
implant and cement in revision surgery. Both types of stem are textured, and
not smooth, for better interlocking with the surroundings. Cementless fixation
stem in addition is porous coated in the proximal section to stimulate bone
ingrowth for biological fixation. It is usually not coated along the whole
length as this could make implant removal difficult in revision surgery. After
the stem is satisfactorily fixed in the humerus, the appropriate size humeral
head is applied to complete the component.
 The resurfacing implant (shoulder resurfacing) only replaces the humeral
joint surface and does not have an intramedullary stem. It therefore preserves
bone stock and is easier to revise than the stemmed implant.

Glenoid component
This comprises an all-polyethylene implant, which has a concave articular
surface and multiple pegs at the back for insertion into the bone. Most
glenoid components are fixed with cement.
The glenoid is usually not affected by arthritis as much as the humeral
head, and therefore does not always need to be replaced. It is usually only
replaced when the chondral surface is completely worn out. The glenoid is
usually well preserved and therefore not replaced in isolated osteonecrosis
of the humeral head and proximal humeral fractures. Another reason for
replacing glenoid selectively is that loosening of the glenoid component is
one of the main causes of failure of the shoulder replacement. Therefore,
there are inherent reservations about replacing the glenoid (Tables 8.1, 8.2).
Fig. 8.9 Types of implants used in shoulder replacement. The humeral
component may be (a) stemmed or (b) resurfacing type. Both types of
humeral component can be implanted without replacing the glenoid, as a
hemiarthroplasty, or with the glenoid component, as a total shoulder
replacement. Image (a) courtesy of Arthrex. Image (b) is reproduced with
permission of Biomet. Biomet is the owner of the copyrights and all other
intellectual property rights in relation to the image. Other than providing
permission to use the image, this publication is not financially supported by
Biomet.

Table 8.1. Shoulder resurfacing

Indications Glenohumeral joint arthritis in younger, more active

patients – they are more likely to require a revision
procedure.
Proximal humerus deformity.
Rotator cuff tears.
Previous fractures of the humerus that have been
internally fixed with a nail or a plate – it may be
possible not to disturb previously implanted
metalwork by performing shoulder resurfacing.
 Advantages Does not involve humeral neck osteotomy, therefore
head–shaft angle does not need to be addressed.
Involves minimal bone resection.
Shorter operation time.
Lower risk of periprosthetic fractures.
Easier revision to a conventional shoulder
replacement.

Contra- Severe osteoporosis – as the bone would not be

indications able to support the component.
Proximal humeral fracture – this would require a
stemmed prosthesis.

Table 8.2. Alignment of components

The humeral head is implanted in 25–45° retroversion and with a

head–shaft angle more than 130°. Varus placement must be
avoided as the loads acting on the component are pushing the
implant in this direction. The glenoid component is usually
implanted in neutral alignment with respect to the plane of the
scapula.

Reversed shoulder replacement

Reversed shoulder arthroplasty is used in patients with non-functional rotator
cuff muscles. The two most common indications for reversed shoulder
replacement are rotator cuff tear arthropathy and massive, irreparable rotator
cuff tear (even without established arthritis). Other indications include:
rheumatoid arthritis, which is usually associated with rotator cuff
dysfunction; revision for failed hemiarthroplasty or total shoulder
replacement associated with rotator cuff damage; and displaced comminuted
proximal humerus fractures where anatomical alignment of standard
components may be difficult.
Rotator cuff tear arthropathy
Rotator cuff tear arthropathy is the glenohumeral joint arthritis secondary to
chronic massive rotator cuff tear. The function of rotator cuff muscles is to
stabilise the humeral head, which includes preventing it from translating
superiorly during abduction (see pages 128–129). However, a massive tear
in the rotator cuff tendon leaves the muscles effectively non-functional. As a
result, the superiorly directed deltoid force acts unopposed. In addition, the
long head of biceps is usually also torn in massive rotator cuff tear, and its
normal humeral head depressing effect is also lost. The humeral head
therefore migrates superiorly and wears against the superior aspect of the
glenoid and the under-surface of the acromium. This also leads to erosion of
the superior aspect of the glenoid.

Design of components
The humeral component is a stemmed implant with a polyethylene cup. The
stem can be fixed to bone with the cemented or cementless method. The
humeral component has a non-anatomical head–shaft angle of 155°.
The glenoid component consists of a large metal hemispherical ball
(glenosphere) attached to a metal base plate. The base plate is fixed to the
glenoid with screws, i.e. the cementless method. The humeral cup covers
less than half of the glenosphere.

Mechanics of reversed shoulder replacement

The reversed relationship of the bearing surfaces has two important
mechanical effects:

It translates the centre of rotation medially and distally. A medial centre

of rotation lengthens the lever arm of deltoid, so it works more
effectively in the absence of assistance from supraspinatus. It also
allows more deltoid fibres to be recruited for abduction and elevation.
Therefore, a larger fraction of deltoid works more effectively to
produce shoulder motion. The centre of rotation is, in fact, at the glenoid
bone–prosthesis fixation interface. This reduces the torque on glenoid
component. Therefore, the risk of glenoid component loosening in
 reversed shoulder replacement is less compared with conventional total
shoulder replacement.
It lowers the humerus relative to the glenoid. This increases tension in
deltoid and further improves its performance.

As a result, reversed shoulder replacement reduces shear force and increases

compressive forces across the joint.
The main limitation of reversed shoulder arthroplasty is that it is generally
unable to restore the lost external rotation, which is caused by infraspinatus
and teres minor deficiency.

Fig. 8.10 Rotator cuff tear arthropathy is the glenohumeral joint

arthritis secondary to chronic massive rotator cuff tear. In the absence of
normally functioning rotator cuff muscles, the deltoid force acts as a shear
force with respect to the glenoid, and draws the humeral head superiorly.
This leads to wear of the humeral head and superior glenoid.
Fig. 8.11 Reversed shoulder replacement utilises a convex glenoid
component and concave humeral component. Conventional shoulder
replacement prostheses are unconstrained and rely on rotator cuff muscles for
stability. However, reversed shoulder replacement prosthesis is semi-
constrained and is further stabilised by non-anatomical head–shaft angle of
the humeral component, and therefore does not rely on rotator cuff for
stability.

Fig. 8.12 Reversed shoulder arthroplasty translates the centre of

rotation medially and distally, so that it lies at the glenoid bone–
prosthesis interface. This provides the following mechanical advantages:
lengthening of deltoid lever arm; recruitment of more deltoid fibres for
elevation and abduction; and reduced torque on the glenoid component. It
also lowers the humerus compared with the glenoid, which tensions the
deltoid.
 9 Biomechanics of the elbow

Elbow: functional anatomy

The elbow increases the flexibility of the upper limb. It also transmits forces
between the arm and the forearm and acts as the axis for the forearm lever
system. The elbow is a complex of three joints of humerus, ulna and radius:
humeroulnar, humeroradial and proximal radioulnar joints. All three joints
are enclosed within the same capsule.
The distal humerus is divided into medial and lateral columns, which are
tilted anteriorly approximately 40° from the humeral shaft. The columns form
two articulating surfaces at the elbow joint: capitellum and trochlea.
The humeroulnar joint is a hinge joint formed by the hourglass-shaped
trochlea articulating with the saddle-shaped trochlea notch of the ulna. This
is an inherently stable configuration, and restricts undue relative motion
between the articulating surfaces. The humeroradial joint is a ball and socket
joint. It is an unconstrained joint formed between capitellum, which is an
almost perfect hemisphere, and radial head, which has little contact with the
capitellum. The proximal radioulnar joint is a pivot joint formed by
articulation between the adjacent surfaces of the radius and ulna. It is a
relatively constrained joint.

Range of motion
The elbow joint complex allows two types of motion: flexion and extension
occur at the humeroulnar and humeroradial joints; and pronation and
supination occur at the humeroradial and proximal radioulnar joints, and also
require simultaneous motion at the distal radioulnar joint. The two types of
motion are independent of each other. The normal range of flexion–extension
is 0°–140°, and pronation–supination is 75° pronation – 85° supination. The
functional range of flexion–extension is 30°–120°, and pronation–supination
is 50° pronation – 50° supination.
 The primary flexors of the elbow are brachialis and biceps brachii. The
brachialis is the main flexor of the elbow, and is also described as the
‘workhorse’ in elbow flexion. The biceps brachii is most effective as a
flexor when the forearm is in supination; its main function is supination of the
forearm. The brachialis and biceps brachii together produce more than 60%
of elbow flexion force. The main extensor of the elbow is triceps brachii.
Elbow pronation is produced by pronator teres and pronator quadratus and
supination by biceps brachii and supinator.

Cubitus angle
When the elbow is in full extension and supination, the longitudinal axis of
the forearm is valgus to the longitudinal axis of the arm. The angle formed
between these two axes is called the cubitus or carrying angle. The carrying
angle allows the forearm to clear the hip when the upper limb is swinging,
such as during walking. The normal range for the carrying angle is between
10° and 15°. It is less in children than adults, and gradually increases with
age. The carrying angle is generally greater in females than males, because
females on average have smaller shoulders and wider hips. The carrying
angle decreases with elbow flexion.
Fig. 9.1 The elbow is a complex of three joints of humerus, ulna and
radius. The flexion–extension axis lies in the trochlea. In the coronal plane, it
is 4°–8° valgus to the plane perpendicular to the longitudinal axis of the
humerus. In the sagittal plane, it lies at the centre of the trochlea. The
flexion–extension axis varies slightly at the extremes of motion. The elbow
joint therefore is not a simple hinge but instead is described as a modified
hinge.

Fig. 9.2 Mechanically, the distal humerus has the configuration of two
columns joined together by a tie arch (trochlea).
Fig. 9.3 The carrying angle is clinically measured as the angle between
the longitudinal axis of the arm and forearm. It results from the fact that the
trochlea extends distal to the capitellum, and therefore the longitudinal axis
of the ulna is, on average, 6° valgus to the longitudinal axis of the humerus.
Fig. 9.4 Supination–pronation axis passes through the centre of the
capitellum of the distal humerus, the radial head and the distal ulna head.
Therefore, it is not parallel to the longitudinal axis of the forearm.

Stabilisers of the elbow

The elbow joint complex derives its stability equally from bony and soft
tissues components. The arrangement of three joints together producing the
two main arcs of motion is a unique way of balancing motion with stability;
although each joint is relatively constrained, the joint complex as a whole
still produces a generous range of motion. As a result, the elbow joint
complex is overall inherently stable.
The elbow, however, is the second most dislocated joint in the body (after
the shoulder). This is an indication of the magnitude of forces transmitted to
the elbow joint in falls onto the outstretched hand and in direct trauma.
Elbow dislocation is associated with two contrasting mechanical
complications: stiffness and instability (recurrent dislocations). Elbow
stiffness usually develops after a prolonged period of immobilisation to
prevent instability. Therefore, a proper assessment of elbow stability is
important at the outset in order to plan the correct treatment.
The structures that provide elbow stability can be classified as static or
dynamic stabilisers.

Static stabilisers
Static stabilisers are further divided into primary or secondary stabilisers.
The primary static stabilisers are the humeroulnar joint and medial and
lateral collateral ligament complexes. The humeroulnar joint is a constraint
joint due to the congruency of the articulating surfaces. The stable nature of
the large humeroulnar joint stabilises the whole elbow unit.
The ligaments provide further stability to the elbow joint complex. The
medial collateral ligament complex consists of three bands: anterior,
posterior and transverse. The anterior band is the primary restraint to valgus
stress. If the anterior band is disrupted, the radial head, i.e. humeroradial
joint, acts as a secondary stabiliser against valgus stress. A radial head
fracture in the presence of an intact anterior band usually does not cause
instability.
The humeroulnar articulation (especially coronoid process) provides the
main resistance to varus stress. The lateral collateral ligament complex is the
main soft tissue restraint to varus stress; however, it has a minor role
towards varus stability. The main function of the lateral collateral ligament
complex is restraint to posterolateral rotatory instability. It consists of four
individual ligaments: radial collateral ligament, lateral ulnar collateral
ligament, annular ligament and accessory collateral ligament. The lateral ulna
collateral ligament and radial collateral ligament are the primary restraints to
posterolateral rotatory instability.
The joint capsule acts as a secondary stabiliser to limit excessive motion
in all directions. The tendons of the common flexors and extensors in the
forearm are also secondary stabilisers. The interosseous membrane between
the ulna and radius is also a stabiliser of the elbow joint complex, as it
maintains the relative position of the two bones and prevents longitudinal
instability.

Dynamic stabilisers
The muscles that cross the joint, e.g. triceps, biceps and brachialis, generate
stabilising compressive forces that provide dynamic stability to the elbow
joint complex (Table 9.1).
Fig. 9.5 Anatomy of the medial collateral ligament complex. The anterior
band is the main restraint to valgus instability; it is the stiffest and strongest
of all the ligaments of the elbow joint. The anterior band is tightened in
flexion and the posterior band is tightened in extension. The transverse
bundle does not span a joint and therefore does not contribute to elbow
stability.

Fig. 9.6 Anatomy of the lateral collateral ligament complex. The lateral
ulna collateral ligament and radial collateral ligament are the primary
restraints to posterolateral rotatory instability. The annular ligament
encompasses the radial head and holds the proximal radioulna joint together.

Table 9.1. Elbow stability: clinical perspective

A primary stabiliser is a structure which, if deficient, leads to

instability. A secondary stabiliser is a structure which, if deficient,
on its own does not cause instability. However, a deficiency of a
 secondary stabiliser when the primary stabiliser is also
compromised worsens the resultant instability.
Elbow dislocations are classified as simple, when there is no
associated fracture, or complex, when there are associated
fractures around the elbow. A simple elbow dislocation causes
injury to the ligaments (and joint capsule), usually in a circular
pattern from lateral to medial. Despite the ligamentous injury,
elbow stability is usually maintained by the dynamic structures.
Therefore, most patients with simple elbow dislocation are treated
functionally with protected early active motion exercises whilst
the injured ligaments heal. However, a ligamentous injury that
results in gross instability during functional arc of motion or
chronic instability (recurrent dislocations) usually requires
surgical repair of the injured ligaments. The management of a
complex elbow dislocation depends on the associated fracture and
the resultant instability. The term ‘terrible triad’ describes the
combination of elbow dislocation and fracture of the radial head
and coronoid process. This is associated with a high risk of
complications.

Elbow joint reaction force

Flexion
When the elbow is held at 90° of flexion by the side of the body, the
brachialis is the main muscle that maintains this position of the forearm.
According to the static analysis, the elbow joint reaction force in this
position is more than one and a half times the weight of the supported
forearm (see pages 10–11 for full analysis).

Extension
In this case, the elbow is held at 90° of flexion with the forearm positioned
above the head. Now, the triceps brachii is contracting to maintain the
position of the forearm. The static analysis shows that the elbow joint
reaction force is more than five times the weight of the supported forearm.
Assumptions
In addition to the general assumptions, the following specific assumptions
are applied to this static analysis.

The triceps brachii provides all of the extension force.

The triceps brachii generates a downward force only.
The following physical measurements are based on anthropometric data.
Forearm weight is 15 N, and is assumed to act as a point load at
the centre of weight of the forearm, 130 mm from the centre of
rotation of the joint (see pages 10–11 for details).
At 90° elbow flexion, the insertion point of triceps brachii is 30
mm from the centre of rotation of the joint.

Discussion
The elbow joint reaction force during extension is more than three times
during flexion. This is because the lever arm of the triceps brachii (30 mm)
is shorter than that of brachialis (50 mm). Therefore, muscular and joint
reaction forces produced during elbow extension are much greater than
during flexion.
The moment arms of the muscles acting at the elbow joint change with
elbow flexion angle. The moment arm of the brachialis increases with elbow
flexion. It is the longest at approximately 100° of flexion, and shortens
slightly with further flexion. At the same time, the moment arm of the weight
of the forearm also increases with elbow flexion, and starts to decrease after
90° of flexion. The brachialis is more effective during the second half of the
flexion–extension arc than during the first half. Similarly, the moment arm of
the triceps brachii is shortest when the elbow is in full flexion, and
progressively increases with extension.
Although the elbow joint is considered to be a non-weight-bearing joint, it
still experiences significant forces during everyday activities. The elbow
joint force is estimated to be 300 N when eating and dressing, 1700 N when
using the arms to rise from a chair and 1900 N when pulling a table across
the floor. The joint force can reach up to three times total body weight during
certain activities.
Fig. 9.7 Free-body force diagram of the upper limb showing forces
acting about the elbow joint during extension. The elbow is held at 90° of
flexion. It is acting as the axis of a class I lever. The moment arm of muscles
is shorter than that of the weight of the forearm. (Force T = Triceps force;
Force W = Weight of the forearm; and, Force J = Joint reaction force.)
Calculations
Applying the conditions of equilibrium:

1. Sum of all moments is zero.

2. Sum of all forces is zero.

Fig. 9.8 Elbow joint reaction force during extension. The joint reaction
force is more than five times the weight of the forearm.
The moment arms of brachialis and weight of the forearm vary with
elbow flexion angle. Similarly, the moment arm of triceps also varies with
elbow flexion angle. (Force B = Brachialis force; Force W = Weight of the
forearm.)
 10 Biomechanics of the spine

Biomechanics of spinal components I

The spinal column is designed to carry loads and provide a wide range of
movements, whilst also protecting the spinal cord and the related
neurovascular structures.
The vertebrae provide stiffness to the spinal column.

Vertebrae
A typical vertebra has the following common key features: vertebral body,
facet (intervertebral) joints, and spinous and transverse processes.

Vertebral body
The vertebral body is the main load-bearing part of the vertebra. It is
designed to support mainly compressive loads, which are produced by the
weight of the body above the vertebra. A typical vertebral body has the
shape of a short cylinder. It has a thin shell of cortical bone surrounding a
core of porous cancellous bone. The cancellous bone is organised into
vertical and horizontal trabaculae, which is the most effective arrangement to
resist compressive force. The compressive force is resisted mainly by the
vertical ‘columns’, and the horizontal ‘cross-beams’ prevent the columns
from bowing under stress. This arrangement of cancellous bone converts an
axial compressive force into a transverse tensile force. The fact that
vertebrae are not solidly filled with bone is also mechanically important.
Solid vertebrae would be significantly heavier and less effective at
withstanding dynamic loads. This is because solid structures are less
‘springy’ and provide less ‘cushion’ when loaded suddenly (hence the reason
why empty cardboard boxes can be used to cushion a person falling from a
height). The porous cancellous bone enables the vertebrae to be lightweight
and still be adequately stiff and strong to withstand different types of loads.
 The most common injury of the spinal column is compression fracture of
the vertebral body. It usually occurs in osteoporosis, where supporting
framework of cancellous bone is not as extensive as normal. It can also occur
in high-energy injuries where the applied compressive force exceeds the
strength of the normal vertebral body.
The vertebral bodies have smooth surfaces and therefore cannot control
relative motion between each other. They are completely dependent on other
structures for stability in the transverse plane.

Facet joints
The facet joints guide motion between adjacent vertebrae. The orientation of
the facet joint changes between the spinal segments; therefore, different
segments have different profiles of motion. The facet joints also have a load-
bearing function. They share the load transmitted between the vertebrae with
the intervertebral disc. The proportion of load shared by facet joints and
intervertebral disc varies with the position of the spine. They normally carry
up to 30% of the transmitted load.

Spinous and transverse processes

The spinous and transverse processes provide areas for attachment of
ligaments and muscles. They provide lever arms to ligaments and muscles,
and therefore reduce the force required from them to maintain function.

Fig. 10.1 The anatomy of a lumbar vertebra.

Fig. 10.2 Bucking strength of a structure is proportional to the number
of columns supporting it and the length of column between supports. The
vertebral body consists of a thin outer layer of cortical bone supported by an
extensive scaffold of cancellous bone. Osteoporosis reduces the bone density
and weakens the supporting framework.

Fig. 10.3 Humans are the only mammals that have a vertebral column
with four curves in the sagittal plane. These curves are a feature of bipedal
locomotion, and increase the flexibility and load-bearing capacity of the
spinal column. The orientation of facet joints of the vertebrae changes
between the spinal segments; therefore, different segments have different
profiles of motion.

Biomechanics of spinal components II

The discs and ligaments provide flexibility to the spinal column.

Intervertebral disc
An intervertebral disc consists of three structures: nucleus pulposus in the
centre, annulus fibrosis on the outside and vertebral end plates at the top and
bottom. The nucleus pulposus has a high water content and generates
hydrostatic pressure. The annulus fibrosis consists of concentric layers of
collagen fibres. Hydrostatic pressure in the nucleus pulposus creates tensile
stress in annulus fibrosis, which provides stiffness to the disc.
The vertebral end plate is a cartilage structure that connects the disc to the
adjacent vertebrae. It allows diffusion of nutrients and waste products
between the disc and blood vessels in the vertebral bone marrow.
The intervertebral disc acts as a shock absorber and provides movement in
the ‘functional spinal unit’. When the disc is loaded, hydrostatic pressure
increases in the nucleus pulposus. The nucleus pulposus exerts this pressure
against the surrounding annulus fibrosus, in accordance with Pascal's law,
which states that pressure within a fluid is transmitted equally in all
directions. The annulus fibrosus expands in the transverse plane and shortens
in the vertical plane. This change in shape of the intervertebral disc converts
axial stress into circumferential stress, which is also known as tensile hoop
stress. This process reduces the load transmitted between the vertebrae.
Under prolonged loading, the intervertebral disc experiences further
deformation due to outflow of fluid from the vertebral end plates. In
mechanical terms, this is described as creep, i.e. time-dependent deformation
under constant load. The intervertebral disc therefore is a viscoelastic
material. Disc deformation is recovered when the applied load is removed.
This leads to diurnal variation in a person's height: a person is usually
slightly taller in the morning compared with the evening.
Disc degeneration leads to mechanical failure of the intervertebral disc. It
starts with loss of hydration of nucleus pulposus. As the nucleus pulposus
becomes less turgid, collagen layers of annulus fibrosis delaminate, collapse
inwards and develop concentric tears. This is associated with loss of disc
height, which leads to subluxation of the facet joints. The facet joints begin to
transmit increased loads (up to 70% of total transmitted load) and as a result
develop degenerative changes. Disc degeneration therefore affects load
transfer and movements of the functional spinal unit.

Ligaments and muscles

The ligaments hold the vertebrae and intervertebral discs together. The
stiffness of ligaments increases up to 75 times outside the normal range of
motion. Therefore, the spinal column has high flexibility in the ‘elastic zone’
of ligaments, but has to overcome much higher resistance to exceed the safe
limits of motion. The muscles act to stabilise and create movement in the
spinal column.
Fig. 10.4 An intervertebral disc acts as a shock absorber and provides
movement in the functional spinal unit. It is the largest avascular structure
in the body, and relies on vertebral end plates for metabolic support. (a)
When it is loaded, the nucleus pulposus exerts uniform hydrostatic pressure
on the surrounding annulus fibrosis. (b) The collagen fibres of the annulus
fibrosis are arranged 30° to the vertebral end plate, alternating in direction
between layers. This is the optimal arrangement for withstanding both the
tensile forces generated by nucleus pulposus and the shear forces produced
by relative motion between the vertebrae. (c) The fibres flatten out when the
disc is loaded. The change in the shape of the disc produces circumferential
‘hoop’ stress. This process reduces the load transmitted to the underlying
vertebrae.
Fig. 10.5 The functional spinal unit is the shortest segment of spine that
has the same biomechanical properties as the whole vertebral column. It
is also described as the spinal motion segment. (a) It consists of two adjacent
vertebrae, an intervertebral disc and the associated ligaments. Motion in the
functional spinal unit occurs at three joints: the two facet joints and the
intervertebral joint.
(b) The relationship of the intervertebral disc with exiting and traversing
spinal nerves. A nerve exits above the disc, therefore a disc prolapse would
normally affect the traversing nerve and not the exiting nerve. However, if a
disc prolapsed into a ‘far lateral’ position, it would affect the exiting nerve.

Stability of the spinal column

The spinal column has a wide range of motion: flexion, extension, lateral
flexion and axial rotation. This large overall flexibility is achieved by small
displacements at each spinal motion segment. The vertebrae have six degrees
of freedom motion: they can translate along and about in the axial, sagittal
and coronal planes. Therefore, the overall motion of the spinal column is
produced by a combination of translation and rotation of individual
vertebrae. It is important that the displacement at each motion segment does
not exceed the normal limits, as this could potentially result in a neurological
injury, significant deformity or intolerable mechanical pain.
The spinal column is considered to be mechanically stable when normal
loading does not cause excessive displacement of the spinal motion segment.
In the setting of a spinal injury, one of the most important clinical tasks is to
determine whether the spinal column is stable. Two conventional
classification systems of spinal stability are based on the integrity of spinal
column structures in the sagittal profile. Both classification systems are
related to the thoraco-lumbar regions.

Two-column classification (Holdworth, 1960)

According to this classification, the spinal column is made of two individual
columns:

The anterior column consists of anterior longitudinal ligament, vertebral

body, intervertebral disc and posterior longitudinal ligament.
The posterior column consists of the pedicles, facet joints, transverse
and spinous processes and the associated ligaments.

A traumatic disruption of both columns leads to spinal instability and a

significant risk of injury to neurovascular structures. If only one column is
disrupted, then spinal stability depends on the integrity of the ligaments
supporting the affected column. Any associated subluxation (to any degree)
indicates disruption of supporting ligaments and therefore instability. The
classification proposes that spinal stability relies mainly on the posterior
column structures.

Three-column classification (Denis, 1983)

This classification considers the spinal column in terms of three individual
columns:

The anterior column consists of the anterior longitudinal ligament and

the anterior half of the vertebral body and intervertebral disc.
The middle column consists of the posterior longitudinal ligament and
the posterior half of the vertebral body and intervertebral disc.
The posterior column, as in the two-column classification, consists of
the pedicles, facet joints, transverse and spinous processes and the
associated ligaments.

Isolated disruptions of either the anterior or posterior columns (i.e. one

column) are mechanically stable. Injuries that affect the anterior or posterior
columns and the middle column are mechanically unstable, as are the injuries
that affect all three columns. The classification proposes that spinal stability
relies mainly on the middle column. Biomechanical studies confirm that the
middle column is indeed the primary determinant of spinal stability.
However, this is not an anatomically distinct zone, and isolated injuries of
the middle column are rare. In practical terms, the middle column is
important mainly for distinguishing between a purely compressive fracture
and a burst fracture: the compressive fracture affects only the anterior column
whereas the burst fracture involves the anterior and middle columns (Table
10.1).

Fig. 10.6 The two-column and three-column classifications of spinal

injuries. The stability of the spinal column relies on the interplay between
bony and soft tissue structures. A column is disrupted when any of the
structures in the column are affected.

Table 10.1. Spinal stability: clinical perspective

In the setting of an acute spinal injury, the Advanced Trauma Life

Support (ATLS) protocol should be used to stabilise, examine,
investigate and manage the patient. The stability of the spinal
 column is evaluated by clinical assessment and appropriate
radiological investigations. It is essential to perform a full
neurological examination. Focal signs around the spinal column
(e.g. swelling, bruising and tenderness) may indicate underlying
ligamentous or bony injury. Plain radiographs are useful to check
for obvious structural abnormalities. Computer tomography (CT)
scan provides further details on the extent of bony injuries and any
compromise of the spinal canal. Magnetic resonance imaging
(MRI) scan is useful to assess the soft tissues, such as the
ligaments, intervertebral disc and spinal cord. Sometimes,
dynamic radiographs may be required to evaluate spinal stability,
e.g. standing radiographs after application of a brace for a
vertebral fracture that initially appears stable.
The spinal column is considered to be mechanically stable when
the initial injury is not likely to cause excessive displacement of
the spinal motion segment when the patient bears weight.
However, a mechanically stable spinal column may not be
neurologically stable. It may be neurologically unstable due to
initial injury from fracture fragments, a herniated intervertebral
disc, epidural haematoma or a compromise in blood supply to the
spinal cord.
The management of a patient with spinal instability depends on the
underlying cause and whether there is associated neurological
compromise. A patient with a mechanically and neurologically
unstable spine may require surgical intervention, e.g.
decompression of neurological structures with or without surgical
stabilisation. On the other hand, a patient with vertebral fractures
and no neurological compromise may be treated with bed rest until
the fractures are healed. Generally, it is difficult to improve a
neurologically stable patient or to reverse neurological deficit in a
patient who has developed flaccid paralysis.

The loads acting on the spinal column

The cervical and lumbar segments of the spinal column are the most mobile.
Therefore, degenerative changes are also most common in these segments.
The loads acting on the spine increase at each lower level, making the
lumbar segment subject to the greatest loads. This section considers the loads
produced in the lumbar segment, to provide an appreciation of the peak loads
found in the spinal column.
The loads acting on the spinal column are affected by body posture. In an
upright standing position, the centre of body weight lies anterior to the fourth
lumbar vertebra. This produces a forwards moment on the spinal column,
which is balanced by spinal ligaments and muscles. In this position, the load
acting on the intervertebral disc between the third and fourth vertebrae is
estimated to be about twice the weight of the body above this level. The
material and geometric properties of the intervertebral disc mean that the
pressure within the disc is approximately 1½ times the applied stress.
Forward flexion increases the moment arm of the upper body weight and the
resultant loads on the functional spinal unit. The vertebrae are subjected to
mainly compressive loads even in variable positions. However, the
intervertebral discs are subjected to a combination of compressive, bending
and torsional loads, which increase with movements of the spinal column.
In a classical experiment, invasive pressure monitoring was used in
healthy volunteers (medical students) to measure pressures in the disc
between the third and fourth vertebrae in various body positions. The study
found that intradiscal pressures were greater when sitting than when standing.
In both cases, the intradiscal pressures increased on forward flexion and
carrying external loads. Another similar study found that intradiscal
pressures in the sitting position were reduced with the use of a lumbar
support (‘supported sitting’) and also in a reclined position.
The vertebral column, in evolutionary terms, is not designed for prolonged
periods of sitting down. However, this is quite common in modern life, and
is considered to contribute to lower back complaints. A person with acute
back pain may find that symptoms exacerbate when sitting, and should lie flat
to offload the intervertebral discs. The loads acting on the spinal column are
minimised when lying flat because loads produced by the body weight are
eliminated. However, tension in the psoas muscles – which originate from
transverse processes of thoracic and lumbar vertebrae – causes some
residual loading of the lumbar segment. This can be minimised by lifting the
legs on a support so that the hips and knees are flexed, which relaxes the
psoas muscles.
Fig. 10.7 Relative pressure in the disc between third and fourth
vertebrae measured in different body positions. Intradiscal pressure in
standing position is taken as 100%. (These findings were reported in the
following textbook: White AP, Panjabi MM (eds) (1990). Clinical
Biomechanics of the Spine. Philadelphia: Lippincott, Williams & Wilkins.)

Fig. 10.8 In manual handling, when lifting a load, it is important to bend

the knees, keep the upper back straight and hold the load close to the
body. These measures reduce the forward moments produced by the body
weight and external load, which reduces the force produced by spinal
muscles and the resultant force acting on the lumbar intervertebral discs.
 11 Biomechanics of the ankle and foot

The ankle joint

Functional anatomy
The ankle is a unit of two joints: the talocrural joint, which is usually
referred to as the ankle joint, and the talocalcaneal joint, which is usually
referred to as the subtalar joint. The ankle joint consists of tibiotalar,
fibulotalar and distal tibiofibular articulations. It resembles the mortise and
tenon joint used in carpentry.
The talus is central to mechanics of foot and ankle, as it forms a series of
joints that connect the foot to the leg:

The body of the talus (talar dome) articulates in the ankle joint.
The inferior aspect of the talus articulates in the subtalar joint.
The head of the talus articulates in the talocalcanonavicular joint.

Through these articulations, the talus transmits the entire weight of the body
to the foot. The talar dome has a wedge-shaped profile: it is wider at the
front than at the back; it is also wider at the lateral aspect than at the medial
aspect; furthermore, it is vertically longer at the lateral aspect than at the
medial aspect. Therefore, the talus has the shape of a frustum of a cone.

Ankle dorsiflexion–plantarflexion arc

The main arc of motion of the ankle joint is dorsiflexion and plantarflexion in
the sagittal plane. Normal walking requires a motion between 10°
dorsiflexion and 15° plantarflexion, with a total motion of 25°. Ascending
stairs requires 37° and descending stairs requires 56° of ankle motion.
The difference in the size of the medial and lateral aspects of the talar
dome means that, during dorsiflexion, the talus also externally rotates 5° and
slides backwards in the joint – these motions are reversed during ankle
plantarflexion. In addition, ankle dorsiflexion is coupled with eversion and
plantarflexion with inversion of the subtalar joint.

Loads acting at the ankle joint

The forces acting at the ankle joint vary during the gait cycle. It is estimated
that, during walking, the ankle joint reaction force reaches three times the
total body weight in the stance phase and more than five times the total body
weight at heel-off. The joint reaction force can be as high as 13 times total
body weight during running and jumping. The tibiotalar articulation carries
about 5/6 of the force and the fibulotalar articulation carries the remaining
1/6 of the force.
The contact area of the joint also changes during the gait cycle. During
ankle dorsiflexion, the joint contact area moves from posterior to anterior,
and vice versa during plantarflexion. The joint contact area is greatest, and
the average stress is lowest, in dorsiflexion; this is reversed in
plantarflexion. The fibula also moves 1–3 mm distally and rotates medially
when the ankle joint is loaded.
Fig. 11.1 The ankle is a unit of two joints: the talocrural joint, which is
usually referred to as the ankle joint, and the talocalcaneal joint, which
is usually referred to as the subtalar joint. The ankle joint resembles the
mortise and tenon joint used in carpentry to join two pieces of wood together
without glue or nails – the joint is stabilised by the congruency of the mating
surfaces. The ankle joint is congruent through 96% of its dorsiflexion and
plantarflexion arc. The supporting ligaments are also essential for joint
stability.

Fig. 11.2 The talus has the shape of a frustum of a cone. The asymmetry
between the medial and lateral aspects allows the talus to rotate and slide
within the joint during dorsiflexion and plantarflexion. Ankle dorsiflexion
results in the wider anterior aspect of the talus engaging with the mortise,
whereas ankle plantarflexion ‘brings out’ the talus and causes the narrower
posterior aspect of the talus to engage within the mortise. Therefore, the
ankle joint is close-packed and more stable in dorsiflexion, and is freer and
less stable is plantarflexion (e.g. when wearing high-heeled shoes).
Fig. 11.3 The axis of motion of the ankle joint. (a) In the axial plane, the
axis is angled posterolaterally. (b) In the coronal plane, the axis is angled
inferolaterally in dorsiflexion, and (c) inferomedially during plantarflexion.
Therefore, the ankle joint does not have a constant axis of motion, but instead
has ‘instant centres of rotation’ that vary with joint position.

Total ankle replacement

Total ankle replacement is an alternative to arthrodesis for treatment of
severe arthritis of the ankle.

Biomechanical considerations in total ankle

replacement
The ankle joint presents unique biomechanical challenges for total ankle
replacement:

During the gait cycle, the ankle joint experiences large forces over a
changing joint contact area. Therefore, stresses produced are non-
uniformly distributed in the joint.
The ligaments and other soft tissues surrounding the ankle joint are
critical in maintaining the stability of the joint. These soft tissues also
influence the distribution of stress in the joint.
The articulating bones have a variable strength. The talus is almost one
and half times stronger than the distal tibia, and the quality of both bones
decreases significantly with the distance from the articular surface.
Therefore, bone resection should ideally be less than 4 mm from the
 articular surface on each side, so that the resection surfaces can still
support the prostheses under large compressive loads.

Design and alignment of components

The early total ankle replacement designs consisted mostly of two
components: a polyethylene tibial component and a metal-based talar
component. These components were fixed to the bone with cement. The
bearing surfaces were either fully constrained or unconstrained; and both
combinations produced very poor results.
Modern total ankle replacement designs consist of metal-based tibial and
talar components with a polyethylene insert in the middle; the polyethylene
insert is either fixed to the tibial or talar component or is ‘mobile’ between
the surfaces. Most of the designs are cementless, with the components coated
with hydroxyapatite to encourage bone growth on the prostheses for a stable
bone–implant interface. Most designs also have a keel or a post at the back
of the prostheses, which increases the bone–implant contact surface area and
reduces pressure and micromotion between the surfaces. The cemented
designs mostly have been abandoned because the use of cement requires
much larger bone resections, due to the extra volume required to fit both the
implants and the cement.
The bearing surfaces in a total ankle replacement need to be congruent to
achieve as large a contact area as possible for load distribution, which
increases joint stability and reduces wear of the polyethylene insert.
However, bearing surfaces that are too congruent restrict the axial and
translation motions required between the surfaces to increase the range of
dorsiflexion–plantarflexion arc. Therefore, bearing surfaces in modern total
ankle replacements achieve a balance between congruency and mobility.
Correct alignment of components is critical to prevent overloading of the
joint. The most demanding step of the total ankle replacement operation is the
correct positioning of the talar component. One of the most common
complications after total ankle replacement is mal-position of the talar
component in the sagittal plane (Table 11.1).
Fig. 11.4 Modern total ankle replacement designs consist of metal-
based (usually cobalt–chrome) tibial and talar components and a
polyethylene insert in the middle. In the example shown, the polyethylene
insert is not fixed to the tibial insert and its motion is constrained only by
friction and conformity with the opposing surfaces. The current total ankle
replacement designs do not reproduce the convex–concave articulation of
normal ankle joint, and therefore change the normal joint kinematics. (Image
of Zenith™ Total Ankle Replacement – courtesy of Corin Group plc.)

Table 11.1. Clinical considerations in ankle arthrodesis and total

ankle replacement

Ankle
Total ankle replacement
arthrodesis
 Ankle
Total ankle replacement
arthrodesis

Advantages Ankle Total ankle replacement

arthrodesis is a restores joint range of
reliable pain- motion. It is suitable for
relieving patients who do not expect to
procedure. At participate in vigorous
present, it is activity.
considered to be Relative contraindications:
the surgical gold pre-operative ankle range of
standard for motion <20°, subtalar joint
end-stage ankle arthritis, poor condition of
arthritis. skin and soft tissues, laxity of
It can be collateral ligaments.
performed Absolute contraindications:
arthroscopically, severe deformity of the joint,
which reduces neuropathic joints and talar
the risk of necrosis affecting >25% of
complications talar dome.
and time to
achieve joint
fusion.
Patients
maintain a good
level of activity
after ankle
fusion. It is
durable, and
recommended
for patients who
expect to
continue with
high-impact
activity and
workload.
 Ankle
Total ankle replacement
arthrodesis
Limitations Ankle Total ankle replacement is a
and arthrodesis does technically demanding
complications not restore range procedure with a longer
of motion. learning curve. Improvement
Therefore, in the ankle range of motion
patients may not is typically between 0°–14°.
be able to It is unclear whether the
perform increased motion actually
activities that leads to a better level of
require ankle activity.
movement, e.g. Main complications are
standing on infection and aseptic
tiptoe, jumping, loosening, with associated
squatting or higher risk of major revision
lunging. surgery. Survival rate of
Main current designs at 5 years is
complications between 67% and 94%, and
are infection and at 10 years is 75%. Other
painful non- complications include delay
union. There is in wound healing and
also a risk of periprosthetic fractures,
mal-union. especially around the
Failure rate of malleoli.
ankle
arthrodesis is
between 5% and
10%.
In the long term,
it can also lead
to arthritis in the
adjacent
subtalar joint,
which can
 Ankle
Total ankle replacement
arthrodesis
require subtalar
joint fusion.

Three rockers of normal gait

The gait cycle describes how the musculoskeletal system achieves
locomotion. The gait cycle extends from one heel-strike to the next heel-
strike of the same leg. It consists of stance phase (65%) and swing phase
(35%). The stance phase extends from heel-strike to toe-off on the same foot,
and swing phase extends from toe-off to heel-strike on the same foot. In the
stance phase, the leg is in contact with the ground and supports the body,
whilst in the swing phase, it creates a new step forward.
In the stance phase, the mechanics of the foot and ankle is described in
terms of three rockers.

The first (heel) rocker is the very brief period between heel-strike and
foot-flat, where the landing foot prepares to receive the load. The heel
exerts a contact force on the ground and receives an equal and opposite
ground reaction force. The point of application of ground reaction force
is posterior to the ankle joint, which causes the ankle to move from
neutral to 10° plantarflexion. The dorsiflexors of the ankle (mainly
tibalis anterior) contract eccentrically to control the rate of this motion.
Therefore, this is a deceleration rocker.
The second (ankle) rocker is the much longer period between foot-flat
and heel-off, where the firmly placed foot allows the supporting leg and
the rest of the body to move forwards. Although the foot is on the
ground, the change in relative position of the leg alters the ankle attitude
from 10° plantarflexion to 10° dorsiflexion. The force for this change in
ankle position is produced effectively by the point of application of the
ground reaction force vector moving anterior to the ankle. The
plantarflexors of the ankle (gastrocnemius and soleus) contract
 eccentrically to control the rate of ankle dorsiflexion. Therefore, this is
also a deceleration rocker.
The third (forefoot) rocker is the short period between heel-off and
toe-off, where the supporting foot prepares to take off. The ankle
position rapidly changes from 10° dorsiflexion to 20° plantarflexion. As
the heel rises, the toes also undergo progressive extension, to a
maximum just before toe-off. The ground reaction force vector is still
acting anterior to the ankle joint, producing a dorsiflexion moment.
However, the plantarflexors of the ankle contact concentrically to
produce ankle plantarflexion. Therefore, this is an acceleration rocker.

In the swing phase, the leg is not in contact with the ground, and therefore
does not experience a ground reaction force. The ankle dorsiflexors bring the
ankle to a neutral position to permit toe clearance.

Fig. 11.5 The gait cycle extends from one heel-strike to the next heel-
strike of the same leg.
Fig. 11.6 The stance phase is divided into three rockers. The centre of
gravity, and therefore the ground reaction force, moves forwards with each
rocker. The muscles are contracting eccentrically in the first and second
rockers and concentrically in the third rocker. An eccentric muscle
contraction is in the opposite direction to the movement of the limb, i.e. the
muscle lengthens as it contracts; a concentric muscle contraction is in the
same direction as the movement of the limb, i.e. the muscle shortens as it
contracts; and the isometric muscle contraction maintains the position of the
limb, i.e. the muscle remains the same length as it contracts.
Fig. 11.7 The ankle joint angle during the gait cycle. In the stance phase,
the ankle briefly plantarflexes during the first rocker, dorsiflexes during the
second rocker, and then rapidly plantarflexes during the third rocker. In the
swing phase, the ankle comes back to the neutral position.

The foot
The foot has two main mechanical functions during walking: it acts as a
shock absorber and mobile adaptor to adjust to uneven terrain; and as a rigid
lever for forward propulsion. The foot functions in only one mode at any
particular time. Each function is linked to a specific period of stance phase
of the gait cycle: the foot acts a shock absorber and mobile adaptor during
the first and second rockers and as a rigid lever during the third rocker. The
subtalar joint is central to how the foot performs these contrasting functions,
switching between them and timing them with the gait cycle.

The subtalar joint

The subtalar joint is the articulation between talus and calcaneus. The
subtalar joint axis is half-way between the vertical axis of the leg and
horizontal axis of the foot. This unique orientation of the subtalar joint allows
it to function as what in engineering terms is described as a ‘mitred hinge’.
The subtalar joint can therefore convert rotation of the leg in the vertical
plane to rotation of the foot in the horizontal plane. The foot rotation is
produced in the transverse tarsal (midfoot) joints, i.e. the talonavicular and
calcaneocuboidal joints, which are immediately adjacent to the subtalar
joint.

Foot as a shock absorber and mobile adaptor

During the first rocker, the ankle joint plantarflexes (and rotates internally)
briefly to bring down the foot. The soft tissues of the foot absorb energy in
this period. The body has double-limb support for most of this time, so the
foot has relatively low loads applied to it.
During the second rocker, the ankle joint dorsiflexes and rotates externally
(i.e. the superimposed leg rotates internally). There is coupled eversion of
the subtalar joint and the hind foot. The eversion of the subtalar joint rotates
the transverse tarsal joints into congruent (parallel) alignment with each
other. This makes the foot flexible and soft to absorb energy, as the single
limb supports the body during this period.

Foot as a rigid lever

During the third rocker, the ankle joint plantarflexes and rotates internally
(i.e. the superimposed leg rotates externally). There is coupled inversion of
the subtalar joint and the hind foot. The inversion of the subtalar joint rotates
the transverse tarsal joints into incongruent alignment with each other. This
turns the foot into a rigid lever arm that can now be used for forward
propulsion of the body.
In addition, as the heel rises, the toes also undergo progressive extension,
to a maximum just before toe-off. The plantar fascia extends from the
calcaneum to the base of the toes. Extension of the toes tightens the plantar
fascia, which raises the foot arch and contributes to the rigid lever structure
of the foot. This is known as the ‘windlass’ effect.
Fig. 11.8 The subtalar joint has an oblique axis in all three planes. The
axis of the subtalar joint runs from a posterior, lateral, plantar position to an
anterior, medial and dorsum position.

Fig. 11.9 Movement at the subtalar joint produces triplane motion. In

addition, movements of the subtalar joints are always coupled with
movements at the ankle joint. The observable effect of these coupled
movements depends on whether (a) the foot is free or (b), (c) resting on the
ground.
Fig. 11.10 The subtalar joint acts as a mitred hinge. (a) A mitred hinge
has an oblique axis and acts as a torque convertor. During the second rocker,
the leg rotates internally, the ankle dorsiflexes, the subtalar joint everts and
the transverse tarsal joints become congruent (unlocked). This makes the foot
flexible and soft. During the third rocker, the leg rotates externally, the ankle
plantarflexes, the subtalar joint inverts and the transverse tarsal joints
become incongruent (locked). This makes the foot a rigid lever.
 12 Biomechanics of fracture fixation

Introduction to fracture fixation

The principles of fracture fixation are as follows:

1. Reduce the fracture

This can be achieved by: indirect method (i.e. closed reduction), which does
not require exposure of the fracture; or direct method (i.e. open reduction),
which involves formal exposure of the fracture. A fracture can also be
reduced by a percutaneous method, which involves a limited direct access to
the fracture to aid closed reduction.
Intra-articular fractures ideally require ‘anatomic’ reduction, to restore the
articular surface. In contrast, extra-articular fractures usually only require
‘adequate’ (i.e. non-anatomical) reduction, as the goal is to restore the
length, rotation and alignment of the bone, which usually does not require
perfect reduction of the fracture.

2. Hold the fracture reduced until it is united

This can be achieved by external fixation or by internal fixation methods. The
fracture does not heal any quicker whether treated by one method or the
other. However, different fixation methods have different mechanisms for
maintaining fracture reduction; some methods permit closed/adequate
reduction, whilst other methods enable open/anatomic reduction. A fixation
device is redundant once the fracture has healed.

3. Rehabilitation
This helps to reduce tissue oedema, preserve joint motion, improve muscle
power and restore normal function of the limb. Internal fixation provides
immediate skeletal stability and generally permits early rehabilitation and
return to function.
 It is useful to consider the different fracture fixation methods in terms of
stiffness of the device and stability of the bone–device construct:

Stiffness is the ability of a device to resist its deformation. It is also

referred to as the rigidity of the device, and depends on the material and
geometric properties of the device.
Stability is the ability of the bone–device construct as a whole to
maintain fracture reduction. The fracture and the device both contribute
to stability. If the fracture provides less stability to the construct, the
fixation device has to provide more.

The distinction between stiffness and stability is important because a device

with low stiffness can still provide high fracture stability, and vice versa;
stability is dependent on fracture configuration and reduction and on how the
fixation device is applied. This is why, although a plate is usually less stiff
than an intramedullary nail, plate fixation usually provides absolute stability,
whereas an intramedullary nail fixation usually provides relative stability. In
clinical practice, stability is often implied in term of the stiffness/rigidity of
the bone–device construct. Therefore, this chapter refers to both the stability
and stiffness of the bone–device construct (Tables 12.1 and 12.2).

Table 12.1. Absolute stability vs relative stability

The term ‘absolute stability’ implies maximum fracture stability

with minimal relative motion between fracture fragments.
Absolute stability is essential for primary bone healing, which is
typically required in intra-articular fractures. Therefore, the
principle of fixing intra-articular fractures is ‘anatomic reduction,
absolute stability’. The term ‘relative stability’ implies that there
is some relative motion between fracture fragments. Relative
stability is usually adequate for extra-articular fractures.
Therefore, the principle of extra-articular fracture fixation is
‘adequate reduction, relative stability’.
Fig. 12.1 This chart shows the spectrum of stiffness and stability in
fracture fixation techniques. The stiffness of a device can be modified to
meet the stability requirement of the fracture.

Table 12.2. Load-bearing vs load-sharing device

Load-sharing and load-bearing are commonly used terms in

clinical practice. These are used to make a distinction between
fixation devices based on distribution of load between bone and
device. A load-bearing device carries most of the load applied to
the bone–device construct and may lead to localised bone
resorption and osteopenia due to stress-shielding effect. On the
other hand, a load-sharing device leads to a more equal
distribution of load. Conventionally, plates are considered to be
load-bearing implants. However, any transient osteopenia
associated with plate fixation is more likely to be due to
associated vascular compromise from its application rather than
from its load-bearing properties. On the other hand, intramedullary
nails are considered to be load-sharing devices as these readily
permit axial loading of the fracture.
Traction I
Traction is the method of realigning a fractured bone by applying a tensile
force to the affected limb. It works on the principle of ligamentotaxis,
whereby a longitudinally applied tensile force causes the intact soft tissues
surrounding the fracture, such as periosteum, ligaments and joint capsule, to
guide the fracture fragments back together. The function of the applied force
is to maintain tension in the soft tissues.
Traction is usually applied to reduce and stabilise fractures temporarily
whilst awaiting definitive fixation. It is widely used in the lower limb,
mainly for femoral fractures as well as for acetabular and tibial fractures. It
can also be used as a long-term treatment in cases where an invasive surgical
treatment cannot be performed e.g. when a patient is unfit for surgery.
Traction is more commonly used as a long-term treatment in children. There
are two main reasons for this: children have a greater potential for bone
remodelling even if traction does not align the bones perfectly, and fracture
healing time is much shorter in children than in adults.
Traction can be applied to the limb by a tie secured to the skin (skin
traction) or connected to a metal pin fixed into the bone (skeletal traction).
The tensile force is produced by attaching a weight to the other end of the tie,
or by securing the tie to a fixed post and using the weight of the body as a
distracting force. When a weight is used to produce the tensile force, it is
common to include a pulley (or multiple pulleys) in the construct.
The load that can be applied with skin traction is limited by skin and soft
tissues concerns, such as: risk of friction blisters; skin necrosis over bony
prominences; and constriction leading to nerve palsy, tissue oedema or
vascular complications. Therefore, skin traction may not reduce a fracture
completely, but is usually adequate to provide pain relief and maintain
alignment. Skin traction may need to be readjusted periodically to prevent
slacking. Skin traction typically is used in the following cases: in children;
when only a small load (<5 kg) is required to maintain fracture reduction; or,
when it is required for the short term (a few days) only. Otherwise, it is more
appropriate to apply skeletal traction. Skeletal traction may also be required
if the conditions of the soft tissues do not permit application of skin traction,
e.g. skin lacerations. The usual sites for insertion of the pins for skeletal
traction in the lower limb are: distal femur and proximal tibia for femoral
fractures; and calcaneum for tibial fractures. For a femoral fracture, a load of
up to 10% of body weight can safely be applied through skeletal traction
(Table 12.3).

Fig. 12.2 Ligamentotaxis is the technique of reducing and stabilising a

fracture by applying tension across the fracture using the intact soft
tissues.

Fig. 12.3 Traction by gravity is used routinely for conservative

management of proximal humerus fractures. The collar and cuff support
the wrist above the elbow, and the weight of the unsupported arm creates a
gentle traction that realigns the fracture. The arm can be taken out of the
support for physiotherapy and personal care.

Table 12.3. Traction: clinical perspective

Traction is commonly used to achieve fracture reduction in a

 number of treatments. Manual traction is used to ‘dis-impact’ the
fracture fragments during closed reduction of fractures, such as
with an ankle or distal radial fracture. The surgeon usually
requires the help of an assistant to apply manual traction; the
surgeon applies traction to the limb on one side of the fracture,
whilst the assistant stabilises the limb by applying ‘counter-
traction’ on the opposite side of the fracture. Manual traction is
also used for closed reduction of joint dislocations.
Traction is also used routinely in theatre to temporarily stabilise a
fracture during a definitive fixation procedure. In this setting,
traction can be achieved manually or by using special equipment,
e.g. a traction table or a Chinese finger trap.
An external fixator also relies on traction (which in this context is
also referred to as distraction) and the principle of ligamentotaxis
to stabilise a fracture.
Traction by gravity is utilised for reducing certain fractures (and
dislocations). The principle of traction by gravity is to place the
limb in a gravity-dependent position so that the weight of the limb
provides the traction.

Traction II
Traction is commonly used to stabilise femoral fractures. The following are
the most common set-ups of traction for femoral fractures:

Straight traction: This is the simplest type of traction, and can be used
for femoral fracture in any age group. It involves application of
longitudinal traction using either skin or skeletal traction.
Thomas splint: This is a portable traction splint that provides straight
traction to the fractured femur. It involves application of skin traction to
the injured leg by placing it in an appropriate size splint. The tie of the
skin traction is secured to the distal end of the splint to provide a
traction force. The Thomas splint is used routinely in the pre-hospital
setting to transfer patients with femoral fractures. It may also be used
with additional equipment, e.g. additional ties and pulleys, to provide
long-term conservative treatment of the fracture.
Gallow's traction: This is used for an infant with a femoral fracture.
Both the fractured and the uninjured leg are placed in skin traction and
suspended in air from an overhead frame. The hips are flexed to 90° and
the legs are pulled vertically upwards. The correct amount of traction
allows the buttocks to only slightly lift off the bed (to allow change of
nappy). The child's weight provides counter-traction to splint the
fracture. This traction method can be used for infants up to about 12 kg
or 18 months of age.
The 90°–90° traction: This is used typically in an older child with a
subtrochanteric fracture of the femur, where the proximal fragment tends
to flex and abduct due to the action of the attached psoas and gluteal
muscles. The hip and knee are both flexed to 90°, which relaxes the
muscles acting on the femur. In addition, an upwards force is applied to
the distal fragment to stabilise the fracture. This usually needs to be
applied through skeletal traction.
Balanced (Hamilton–Russell) traction: This involves application of
two perpendicular forces to the leg, which is slightly flexed at the knee:
an upwards force and a longitudinal force. The resultant vector of the
two forces is a tensile force that acts in line with the femur. Balanced
traction can also be used to control the rotation of the leg, by directing
the upwards force medially or laterally. Generally, the limb has the
tendency to rotate externally in traction, and the upwards force can be
adjusted to ensure that the patella is pointing directly upwards.
Fig. 12.4 Straight traction: this example shows the use of the skin
traction technique to apply longitudinal traction to the limb. The padding
under the leg prevents the heel from pressing into the bed, which could cause
pressure sores. The end of the bed is raised slightly to prevent the patient
from slipping down.

Fig. 12.5 A Thomas splint combines skin traction with a portable frame.
Fig. 12.6 Gallow's traction is used for an infant with a femoral fracture.
It may be continued until the fracture unites or is converted to a hip spica cast
during the treatment.

Fig. 12.7 The 90°–90° traction involves flexing the hip and knee to 90°.
This can be achieved by raising the leg on a support or by applying a sling
under the calf and suspending it from the frame.
Fig. 12.8 Balanced traction involves application of an upwards and a
longitudinal force to the leg. The leg is slightly flexed at the knee. The
resultant vector of the two forces is a tensile force that acts in the line of the
femur.

Cast
A cast is a non-invasive external splint used to stabilise fractures. It can be
used in combination with other fixation methods to provide additional
support.

Biomechanics of cast
The two most commonly used materials with which to make casts are plaster
of Paris and fibreglass, although plastic and metal are also used routinely.
Plaster of Paris is weak in tension and much stronger in compression, and so
is prone to break down in regions under tension. Fibreglass is a lighter,
stiffer and stronger material than plaster of Paris. In addition, unlike plaster
of Paris, fibreglass is water resistant. However, as fibreglass is stiffer, its
casts are less accommodating of soft tissue swelling and are more difficult to
mould, which can result in formation of sharp edges. The stiffness of cast
made from both materials increases with the number of layers used (i.e.
thickness of the cast).

Biomechanics of bone–cast construct

Different types of casts are used for treating different fractures, and the
stability of bone–cast construct depends on the bone fractured, the fracture
pattern and the form of cast. There are four main groups of cast:

Back slab: This is a splint that only partly encircles the limb. It is
usually applied in the acute setting to stabilise the fracture temporarily,
whilst accommodating soft tissue swelling.
Full cast: This encircles the full circumference of the limb. In principle,
the joint above and the joint below the fracture are usually also
immobilised with the full cast to increase stability of the construct and
therefore reduce the risk of fracture displacement.
Spica: This encircles a part of the body, e.g. hip spica and thumb spica.
Functional brace: This encircles the full circumference of the limb, but
does not impede motion of the adjacent joint. A functional brace may be
developed from a range of materials. The mechanics of a functional
brace are considered on pages 193–194.

It is essential that the first three types of cast be moulded to provide three-
point fixation of the fracture, which increases construct stability. The
traditional orthopaedic saying is that ‘a straight (unmoulded) cast leads to
bent bones (i.e. loss of fracture reduction) and a bent (moulded) cast leads to
straight bones’. Three-point fixation is provided by a combination of
mechanisms:

direct buttressing effect of the moulded cast on the fracture fragments

principle of ligamentotaxis – provided by the ‘soft tissue hinge’ around
the fracture
hydraulic compression forces of the muscles surrounding the fracture.

It is also important that a cast is of adequate length to obtain proper fracture

stabilisation.
Fig. 12.9 A fracture often requires closed reduction before application
of a cast. (a) An angulated fracture, e.g. of distal radius, is reduced by
applying a force opposite to the initial deforming force. (b) The fracture is
stabilised with three-point fixation applied by a moulded full cast. The
stabilised bone may be considered as a beam subjected to three-point
loading (see pages 54–55). The geometry of the forces determines the loads
applied to each bone fragment. An important inference from this analysis is
that the shear force acts uniformly along the entire length of a bone fragment
involved in three-point fixation up to the point of application of the next force
(c), (d).
Fig. 12.10 Reduction of an off-ended fracture may require
exaggeration of deformity as part of the reduction technique. (a) The off-
ended fracture usually has an intact soft tissue hinge on the side that the distal
fragment has translated to, and ruptured soft tissues on the opposite side.
Simple traction or a bending force may not be sufficient to reduce the
fracture. (b) Therefore, fracture deformity needs to be increased to
approximate the opposing ends of the fracture, after which (c) the fracture
can be reduced and three-point fixation applied. This technique is
particularly important in children, in whom the periosteum is thick and
resilient.

Wires
Wires have a wide array of uses in fracture fixation. Wires are typically
made from stainless steel and are available in a range of sizes and finishes.

Biomechanics of wires
A wire's stiffness is proportional to its radius to the fourth power. Wires
generally are relatively flexible and can be easily bent. Wires can be broadly
divided into two types:

Kirschner (k)-wires: These are relatively stiff, straight wires with

sharp, pointed tips. However, there are many variations to this basic
design.
Cerclage wires: These flexible wires often come in a coil, and can be
wound around fracture fragments to reduce and hold them in position.

Biomechanics of bone–wire construct

Fracture fixation with k-wires
The following are important considerations in fracture fixation with k-wires.

The fracture should be reduced satisfactorily. A wire can sometimes be

used to ‘lever’ a fragment into position.
The diameter of a wire should be appropriate for the size of the bone
and fracture fragments.
Construct stability increases with the number of wires fixing the
fracture, but two to three wires are optimal for most fractures.
If multiple wires are used, they must not cross each other at the fracture
site, as this may splint the fracture apart.
In most cases, a wire should ideally ‘bite’ the far cortex to prevent
backing out.
In most cases, the fixation requires further stabilisation with a plaster or
splint until the fracture is sufficiently united for the wires to be
removed. The fracture typically heals with callus formation.

Tension band principle

This is a classical engineering concept. Principles of tension band wire
fixation are as follows:

It is typically applied to bones that are eccentrically loaded, e.g. for

olecranon and patella fractures. A cerclage wire band applied to the
 tension side of a fracture neutralises the distracting force and can even
convert it into a uniform compressive force.
Fractures suitable for tension band wire are typically transverse or
short oblique with no significant comminution at the far cortex.
The fracture is reduced anatomically and usually two parallel k-wires
are also passed across the fracture to act as rails along which the
fragments compress. k-Wires should ideally cross the fracture close to
the far cortex for maximum mechanical leverage.
The fracture typically unites by primary bone healing, which is the goal
since most fractures requiring tension band fixation are intra-articular.

Fig. 12.11 There are two groups of wires: k-wires and cerclage wires.
Thicker wires are conventionally referred to as pins; a wire is usually
between 0.9 and 1.5 mm in diameter, whereas a pin is usually between 1.5
and 6.5 mm in diameter. (Image (a) is reproduced with permission of Narang
Medial Ltd. Image (b) is reproduced with permission of Disposable
Instrument Co.)
Fig. 12.12 Tension band principle. A structure acts as a beam when
subjected to eccentric loading – the load can be compressive or tensile. If
there is a break in the continuity of the structure, the eccentric load causes it
to rotate as well as translate. A band applied to the tensile surface takes up
the tensile component of the load, thereby allowing stabilising compressive
component to act on the structure. Wires are the most common means of
achieving tension band effect, but sutures, screws and plates can also be used
to produce this effect. The tension band principle is less effective when there
is significant comminution of the fracture, especially at a compressive
surface.
Fig. 12.13 Tension band wire fixation of olecranon fracture. This can
also be applied to fractures of the patella, medial malleolus of the ankle and
tuberosities of the proximal humerus.

Surgical screw I
The screw is a simple machine that converts a torque into an axial force. The
screw is commonly used to hold materials together for assembly and
construction, although the ‘screw thread mechanism’ has a variety of
applications, e.g. corkscrew, screw top container lid and the Archimedes'
screw to move materials. A surgical screw may be made of metal, e.g.
stainless steel or titanium, or a biodegradable material, e.g. biodegradable
polymers.

Biomechanics of screw
The screw thread mechanism converts rotational motion into linear motion.
The torque applied to the screw head turns the cylindrical shaft and the outer
helical threads. The helix couples rotational motion to linear motion. The
linear motion in turn generates axial force. Importantly, an axial force
applied to the screw shaft cannot make the screw turn back the other way.
Although there are many different designs of screw, the basic mechanical
properties of any screw are determined by the following key characteristics:

Head: This permits attachment of the screwdriver and halts screw

advancement into a material. The coupling interface for the screwdriver
is typically hexagonal or star shaped. This design prevents the
screwdriver from losing grip, whilst spreading the stress from the
screwdriver over a large surface area. The under-surface of the screw
head is usually round to also maximise the area over which the stress
from the screw head is spread across the underlying surface.
Neck: This is the section between the screw head and threads. It is
usually the weakest section of the screw, and where a screw usually
breaks. A fully threaded screw has a very short neck, whereas a
partially threaded screw has a longer neck.
Core diameter: This is the diameter of the cylindrical shaft. It
determines the stiffness of the screw and its resistance to fatigue failure.
It also determines the diameter of the drill that should be used to create
the hole for the screw.
Outer diameter: This is the diameter of the screw threads. This is also
the minimum hole diameter through which the screw will glide without
its threads engaging. A surgical screw is usually referred to by this
diameter, e.g. a 3.5 mm cortical screw has a thread diameter of 3.5 mm.
Pitch: This is the distance between screw threads. It corresponds to the
linear distance a screw travels for each complete turn. The screw
amplifies a small torque into a large axial force; the force amplification
effect is inversely proportional to the pitch. In other words, the smaller
the linear distance a screw travels for each complete turn (i.e. output
distance < input distance), the greater the axial force produced for a
given torque (i.e. output force > input force) – this is due to the force–
distance relationship in simple machines, which is explained on pages
12–13 (Table 12.4).
Fig. 12.14 The functional anatomy of a screw. The screw's hold in the
bone is referred to as purchase. The axial force required to remove a screw
is referred to as its pullout strength. The pullout strength of a screw can be
increased by increasing the contact surface area (interference) between
screw threads and bone. This can be achieved by increasing the outer
diameter or reducing the core diameter (which effectively increases width of
the threads) or the pitch.

Table 12.4. The different types of surgical screws

Surgical screws come as fully or partially threaded. A fully

threaded screw has threads along its whole length, whereas a
partially threaded screw has threads only part-way along from the
screw tip.
A screw designed to gain purchase mainly in the cortices of the
bone is referred to as a cortical screw. Similarly, a screw
designed to gain purchase mainly in the cancellous bone is
referred to as a cancellous screw. The cortical bone is usually
dense but has limited thickness, whereas cancellous bone is much
less dense and more spread out. Therefore, to maximise purchase,
cortical screws have a small pitch and their tips are designed to
cut into dense cortical bone, whereas cancellous screws have
wider threads and larger pitch and their tips are designed to press
cancellous bone aside, much like a snow plough presses aside
spread-out snow.
A self-tapping screw has a cutting flute at the tip, which cuts a
channel into the bone for screw threads. It can be inserted directly
after a hole is drilled into the bone. A screw that is not self-
tapping has a plain tip and requires the drilled hole to be
separately tapped before it can be applied.
A standard screw has a solid central core, whereas a cannulated
screw has a canal through the central core, which can be used to
guide the screw over a guide wire. A cannulated screw has a
bigger core diameter than a standard screw of similar size. This
maintains the stiffness of the screw, but since the threads' span is
reduced, the cannulated screw has a lower pullout strength.
A locking screw is a special type of screw that has an additional
set of threads around the head. It is used in combination with a
‘locking’ plate, which has reciprocal grooves around the plate
holes. The locking screw therefore locks into the plate. This
makes the plate–screw construct much more rigid than a non-
locking construct, which provides greater stability to fixation.
Fig. 12.15 Examples of the different types of surgical screws. (a)
Cortical screws. (b) Cancellous screws. Left to right show the following
properties in each set: plain, self-tapping, partially threaded, cannulated and
locking. A screw may have different combinations of these properties, e.g. a
cortical screw may be locking, self-tapping and cannulated.

Surgical screw II
Biomechanics of a bone–screw construct
A surgical screw can be used to hold bone fragments together or to affix an
implant, e.g. plate, to the bone.
Fracture fixation with lag screw
A lag screw is used to affix bone fragments together – ‘lag’ is the mode in
which a screw is applied, rather than a specific type of screw. Therefore, in
principle any screw can function as a lag screw. The lag screw is applied
after the fracture has been reduced anatomically. It can be applied
independently or through a plate hole.
Fully and partially threaded screws are applied as a lag screw using
different techniques.
A fully threaded screw achieves inter-fragmentary compression when
applied using the lag technique.

The lag technique is most effective in oblique fractures that are

approximately twice as long as the bone diameter.
As before, it is essential that the screw is applied perpendicular to the
fracture line.
The near fragment is over-drilled to the size of the outer diameter of
screw (glide hole).
The far fragment is drilled to the same size as the core diameter of the
screw (pilot hole).
The cortex of near fragment is ‘countersunk’, so that stress from the
screw head is distributed over a large surface area.
As the screw is applied, its threads engage in the far fragment but not in
the near fragment.
With further tightening, the screw head contacts the cortex of the near
fragment and compresses it against the far fragment.
Fully threaded screws are typically used in diaphyseal fractures, where
the cortex is thick.

A partially threaded screw automatically achieves inter-fragmentary

compression due to its design.

As before, it is essential that the screw is applied perpendicular to the

fracture line.
The near and far fragments are drilled to the same size as the core
diameter of the screw.
 As the screw is applied, the screw threads engage in the far fragment,
but the unthreaded proximal section of the screw does not engage in the
near fragment. It is therefore essential that all the screw threads are
beyond the fracture and that no threads are in the near fragment or
across the fracture.
With further tightening, the screw head contacts the cortex of the near
fragment and compresses it against the far fragment.
Partially threaded screws are typically used in metaphyseal fractures,
where the cortex is thin. Here, a washer may also be used under the
screw head to prevent the screw head from being pushed through the
cortex.
Partially threaded screws are always applied as a lag screw.
Most of the partially threaded screws used in fracture management are
cancellous screws, and partially threaded cortical screws are not
commonly used.

Fig. 12.16 Lag screw by technique: a fully threaded cortical screw acts
as a lag screw when applied using the lag technique. It is essential that the
screw is applied perpendicular to the fracture line. More than one screw can
be applied, depending on the fracture pattern, fracture length and bone
quality. A lag screw is usually not adequate to resist shear and torsional
forces whilst the fracture heals. Therefore, it is usually also necessary to
protect the lag screw fixation. In the small bones, e.g. in the hand, a second
screw is applied across the fracture with the same technique as before except
that the screw is applied perpendicular to the long axis of the bone. This is a
‘neutralising’ screw, as this is the optimal position for resisting shear forces.
In larger bones, a neutralising plate is required to protect the lag screw
fixation.
Fig. 12.17 Lag screw by design: a partially threaded cancellous screw
automatically functions as a lag screw. This example shows fixation of a
medial malleolar fracture of the ankle.

Fig. 12.18 In order for the partially threaded cancellous screws to

achieve inter-fragmentary compression, it is essential that all the screw
threads are beyond the fracture and that no threads are in the near
fragment or across the fracture.

Plate fixation I
Plates are the most ubiquitous of fracture fixation implants. Plates are
available in a variety of sizes and designs, and can be applied in a number of
different modes.
Biomechanics of plate
The plate has the structure of a rectangular beam. Therefore, its stiffness is
proportional to its width, and to its thickness to the third power. Changing the
thickness of a plate has a greater effect on its stiffness than changing the base
material.

Biomechanics of bone–plate construct I

Conventional plate vs locking plate
The conventional plate relies on friction between plate and bone for
construct stability. This inevitably leads to compromise in the periosteal
blood supply. The screws affix plate to bone, and also rely on friction to
maintain hold in the bone. The quality of fixation is therefore limited by the
bone quality. The screws are prone to a ‘toggle’ effect, i.e. the screws can
rotate away from the applied position, and can also pull out on their own,
which can weaken the fixation.
The locking plate has threaded holes, which allow the screw heads to lock
into the plate. The locked screws are fixed in their position, which prevents
them from toggling and backing out of the plate. The screws and the locking
plate together form a ‘fixed-angle’ appliance, which is inherently much more
stable than a non-locking interface between plate and screws. The bone–
plate contact is not critical, and the periosteal blood supply is preserved.
The plate therefore acts as an ‘internal’ external fixator. Fracture fixation
with a locking plate is particularly desirable when there is less support
provided by the bone, e.g. when bone quality is poor, as in osteoporosis, or
if there is extensive fracture comminution.

Stiffness of bone–plate construct

The stiffness of bone–plate construct increases with the number of screws
stabilising the bone fragments. The construct stiffness is also influenced by
the distance between the screws closest to the fracture on either side; these
screws experience the most force and construct stiffness decreases as the
distance between them increases.
 The stiffness of bone–plate construct also varies with the direction of
loading. This is because the plate has different cross-sectional properties in
different planes. Therefore, the construct's stiffness is determined by the
plate's orientation to the plane of loading. Based on this concept, double-
plate fixation of a fracture creates an even stiffer construct. Although the two
plates can be applied in any arrangement, orthogonal plating, i.e. two plates
applied at 90° degrees to each other, in principle improves construct stiffness
uniformly in all planes. However, modern plates and fracture fixation
techniques mean that most fractures do not routinely require double-plate
fixation. Elbow supracondylar fractures and tibial pilon fractures are
examples of where double-plate fixation is still used routinely.

Fig. 12.19 The stiffness of a plate is proportional to its width, and to its
thickness to the third power. A small increase in the thickness of the plate
greatly increases its stiffness. (Reproduced with permission of Panchal
Meditech.)

Fig. 12.20 The stiffness of bone–plate construct is influenced by the

distance between the screws closest to the fracture on either side. This
is also referred to as the working length of the plate. There is an inverse
relationship between working length and fixation length. The stiffness of the
construct decreases as the working length increases.
Fig. 12.21 The stiffness of bone–plate construct depends on the
orientation of the plate to the plane of loading.

Plate fixation II
Biomechanics of bone–plate construct II
Modes of application
A plate can be used in a number of different modes, based on fracture
location and pattern and the type of fracture stability required:

Neutralisation
Lag screws produce excellent inter-fragmentary compression, but are often
inadequate to resist fracture displacement under functional loads. A plate can
be applied across the fracture to provide further stiffness to the construct.
The plate therefore ‘neutralises’ the forces and protects lag screw fixation
from failure. A lag screw can also be applied through the plate hole, in
which case it affixes the plate to the bone and compresses the fracture.

Compression
A plate can be applied to produce compression across a fracture. This is
achieved by applying the first screw in a neutral position in a plate hole on
one side of the fracture and then by applying the second screw in an eccentric
position in a plate hole on the other side of the fracture. This screw
configuration pulls the fragments together, producing inter-fragmentary
compression. This technique can be used up to twice if required, but the first
eccentric screw needs to be slightly loosened just before tightening the
second eccentric screw to permit movement of fragments, after which it is
tightened up again. Further screws can then be inserted as normal.

Bridging
It is not always necessary to fix every fragment in a multi-fragmentary
fracture. A plate can be applied across a comminuted fracture and fixed to
the bone at intervals. The aims of a bridging plate are to: minimise soft tissue
disruption at the fracture zone and therefore preserve fracture blood supply;
and splint the fracture in correct length, rotation and alignment. A plate
applied in bridging mode is at an increased risk of fatigue failure because the
bone–plate construct is likely to be less stable than in other modes due to
fracture characteristics, e.g. comminution, bone loss or poor-quality bone,
which may also lead to impaired bone healing.

Buttress/Antiglide
The term buttress means to support or reinforce. A plate is applied in
buttress mode when a fracture has an apex/axilla; a plate applied to this
surface of the fracture prevents sliding motion between fragments. The plate
therefore resists shear forces between fragments. A buttress plate supports an
intra-articular fracture and an antiglide plate supports a diaphyseal fracture –
both modes work on the same principle.

Tension band
Most long bones in the body are loaded eccentrically, so that the adjacent
sides of the bones are in tension and compression. A plate applied to the
tension side of the bone neutralises tensile force and can even produce
compression at the fracture site simply due to the tension band effect, i.e.
when the tension side of a bone is splinted (banded), eccentric loading leads
to fracture compression (this is further discussed on pages 166–177).
Fig. 12.22 Neutralisation mode. Primary fracture fixation is provided by
the lag screw. The function of the plate is to neutralise (withstand) the
applied load and protect lag screw fixation. The lag screw may be applied
through the plate or independently of it. A lag screw is often also referred to
as the ‘working’ screw, to distinguish it from the ‘holding’ screws in the
plate.

Fig. 12.23 Compression mode. The plate provides compression across the
fracture. This is achieved through a specific method of screw placement. The
first screw is inserted as normal. The second screw is inserted on the
opposite side of the fracture and on the side of the plate hole away from the
fracture. As the screw is inserted, the plate moves to accommodate the
screw. The first screw has fixed the plate to the first bone fragment, which
moves with the plate towards the fracture.

Fig. 12.24 Bridging mode. The plate acts as an extramedullary splint. It is

usually inserted away from the fracture using minimally invasive technique
and passed across the fracture without directly exposing it. The screws are
inserted with short ‘stab’ incisions under fluoroscopy control. The bridging
plate therefore allows the fracture to be stabilised with minimal disturbance
to the soft tissue envelope around the fracture site. This is the only mode of
plate application where the fracture is expected to heal by secondary bone
healing with callous formation.
Fig. 12.25 Buttress/ Antiglide mode. The application of plate to the apex
surface of the fracture inhibits shear motion between the fragments. A
buttress plate supports an intra-articular fracture and an antiglide plate
supports a diaphyseal fracture.
Fig. 12.26 Tension band mode. Plate fixation can also achieve
compression across the fracture by the way the bone is loaded. An
eccentrically loaded bone, e.g. femur, acts like a beam with its one side
under compression and the other under tension. A plate applied to the tensile
surface takes up the tensile component of the load, and allows the stabilising
compressive component to act at the fracture site.

Intramedullary nail I
An intramedullary (IM) nail acts as an internal splint to stabilise bone
fragments. It is contoured typically to the profile of the bone.

Biomechanics of intramedullary nail

The IM nail is a cylindrical tube; therefore its stiffness is proportional to its
radius to the fourth power. Therefore, a small increase in the radius has a big
effect on the stiffness of the implant.
A solid nail is stiffer than a hollow nail of the same diameter. However, a
hollow nail has a number of benefits.

Less material is required to make a hollow nail than a solid nail. It is

therefore also lighter.
The central canal of the hollow nail can be used to guide the nail into
position over a guide wire.

Hollow nails are therefore used routinely in the management of long bone
fractures. The size of the central canal is carefully determined to limit the
reduction in stiffness; although the outer radius of the nail has a far greater
impact on stiffness than the thickness of the wall (see pages 56–57).

Biomechanics of bone–nail construct

Proximal and distal interlocking screws, referred to as ‘bolts’ in this role,
provide rotational stability to the construct. Bolts are placed closer to the
ends of the nail, which extends the zone of fractures that can be treated with a
nail, but this is at the expense of construct stability. The rotational alignment
of fragments is set when the first bolt is introduced into the second fragment
after interlocking the first fragment. Bolts can be used in ‘static’ or ‘dynamic’
settings to control motion at the fracture.
The working length of the nail is the segment of nail around the fracture
that is unsupported by bone. This is inversely related to construct stability:

The nail is supported by bone where it is press-fixed to bone cortices,

and the nail and bone both support the applied load. This is mainly in
the diaphyseal region, where bone has the narrowest diameter, which
therefore limits the maximum diameter of nail that can be used.
The nail is not supported by bone at the fracture site, and the load is
supported by the nail only. The nail is ‘off-loaded’ at the first point of
contact between nail and bone on either side of the fracture – the
distance between these two points is the working length of the nail.
 The working length of a nail press-fixed to the bone is determined by
fracture pattern and location.
Transverse diaphyseal fractures lead to a working length of a few
millimetres.
Comminuted diaphyseal fractures produce a longer working length.
In metaphyseal fractures, working length is between the bolt
closest to the fracture and the first point of contact between nail
and bone at the diaphysis.
The working length of a nail with no surface contact with bone is
between the bolts closest to the fracture on either side, regardless of
fracture pattern and location.

Fig. 12.27 The stiffness of an IM nail is proportional to its radius to the

fourth power. Therefore, for the same wall thickness, a hollow nail with a
16 mm outer diameter is three times stiffer than one with an 11 mm outer
diameter. Proximal and distal bolts provide rotational stability. IM nails are
loosely referred to as IM rods. However, in mechanics, a rod is solid
whereas a tube is hollow. Therefore, it is more appropriate to refer to a solid
nail as an IM rod and a hollow nail as an IM tube. (Image courtesy of
MEDIN.)
Fig. 12.28 The working length of nail is the segment of nail around the
fracture that is unsupported by bone.
Applying conditions of equilibrium:

1. Sum of all moments is zero.

2. Sum of all forces is zero.

Fig. 12.29 External loads and muscles forces produce bending stress
between IM nail and bone. The bending stresses are greatest at the fracture
site. An IM nail fixation of a fracture of the distal third of a long bone
generates high stresses on the adjacent interlocking bolts. This may lead to
failure of these bolts.

Intramedullary nail II
Flexible IM nails are a subgroup of the IM nail family, and function with
different biomechanical principles. A flexible IM nail combines elasticity
and stability in one construct and is also referred to as an ‘elastic stable IM
nail’.

Biomechanics of a flexible IM nail

These are made from stainless steel or titanium. The stiffness of the nail is
proportional to its radius to the fourth power.

Biomechanics of bone–nail construct

Flexible IM nails are used most commonly in the management of paediatric
femoral and forearm bones fractures. These provide a number of surgical
advantages over other fracture fixation methods. These are minimally
invasive, relatively safe to insert and remove, present low risk of physeal
injury and allow early mobilisation. The nails are usually removed after
fracture union, and the risk of re-fracture is low.

Femoral fractures
The strain energy stored in the construct provides fixation stability:

Two nails are applied retrograde to prevent injury to the proximal and
distal femoral physes.
Both nails have the same diameter, which ideally should be 40% of the
internal diameter of the medullary canal at its narrowest (diaphysis).
Both nails are bent together, and the apex of the bend lies at the fracture
site.
The nails are introduced from opposite cortices to form diametrically
opposite curves at the fracture site.
Each nail provides ‘trifocal buttressing’ (i.e. three-point fixation), to the
fracture.
The construct has excellent stability under axial and bending loads, but
is still relatively weak under rotational load. It is stable enough for the
patient to mobilise bearing partial weight on the limb.
The fracture heals by secondary bone healing.
Flexible IM nails are typically used to manage femoral fractures in
children between the ages of 5 and 14 years with body weight up to 50
kg. They may be used in even older, heavier children, but additional
protection, e.g. with a cast brace, may be required to maintain adequate
stability.

Forearm bone fractures

Flexible IM nails are sometimes used to internally splint radius and ulna
fractures:

Generally, only unstable, displaced or open fractures require fixation.

Therefore, if both bones are fractured, it is not always necessary to fix
them both. If one fracture is displaced and the other is not, then fixing
one fracture may provide adequate stability to the forearm as a unit.
A single nail is used for each forearm bone, and is usually inserted
anterograde in the ulna and retrograde in the radius.
The diameter of the nail is determined by the size of the medullary canal
of the bone and is usually between 2.0 and 2.5 mm.
The construct requires additional protection with a cast.
The fracture heals by secondary bone healing.

Fig. 12.30 These radiographs show an example of the use of flexible

nails to stabilise a femoral fracture. Flexible nails are ideal for mid-
diaphyseal fractures with a transverse, short oblique or short spiral pattern,
although these may also be considered for more peripheral or multi-
fragmented fractures in selected cases.
Fig. 12.31 Principles of elastic stability. The nails are bent (i.e. plastic
deformation) so that the height of the apex of the curve is three times the
diameter of the femoral diaphysis. As a pre-contoured nail is inserted into
the relatively straight medullary canal, it is forced to straighten out (i.e.
elastic deformation). This creates a bending moment that tends to angulate the
fracture. This is balanced by the moment produced by the opposite nail.
Fig. 12.32 These radiographs show an example of the application of
flexible nails to stabilise radius and ulna fracture.

External fixation I
External fixation is a method of stabilising a fracture by applying implants to
the bone that extend outside the soft tissues and are externally linked by an
adjustable beam system. The components of an external fixator are away
from the fracture environment. This is advantageous in situations such as
open fractures with risk of infection and in fractures associated with massive
soft tissue injury where the external fixation does not disturb the zone of
injury.
External fixators may be used for temporary or definitive fracture
stabilisation, and are divided into pin-to-bar, ring or hybrid types. Pin-to-bar
fixators are applied in routine trauma surgery, e.g. open fractures, peri-
articular injury and polytrauma, whereas ring fixators are usually applied in
more complex reconstruction situations, e.g. fracture mal-union or non-union,
bone infection and limb deformity correction. Temporarily applied external
fixators are converted to definitive internal fixation when the fracture
environment is suitable.
 As an external fixator consists of different units, its rigidity depends on the
properties of its components and their spatial configuration. An external
fixator provides ‘flexible’ stiffness to the healing fracture because its
components and their arrangement can be adjusted to change the construct
stiffness. Although in general an external fixator may provide less rigid
fixation than internal fixation, its capacity for adjustment in stiffness is more
advantageous in certain fractures.

Biomechanics of pin-to-bar external fixator

A pin-to-bar external fixator has three components:

Pins: These are passed through the skin and fixed into the bone. There
are two types of pins:
Half-pins emerge on one side of the limb.
Trans-fixation pins come through both sides of the limb.
The bending and torsional stiffness of a pin are proportional to its
radius to the fourth power. Therefore, in comparison to a 4 mm
diameter pin, a 5 mm diameter pin is about one and a half times
stiffer, and a 6 mm diameter pin is about five times stiffer. A stiffer
pin places less stress at the bone–pin interface. However, pinholes
greater than one-third of bone diameter dramatically increase the
risk of fracture due to the stress raiser effect. This limits the pin
size that can be used in a particular bone.
Pins also come in a variety of profiles and thread section designs.
These mainly affect the pins' purchase in the bone. The main types
are discussed in the next section:
Bars: These form the frame of the fixator and stabilise the fracture by
spanning it between pins. Modern bars are made from carbon fibre, so
are radiolucent and lightweight and still adequately stiff. The stiffness
of a bar is also proportional to its radius to the fourth power.
Clamps: These connect a pin to bar or a bar to bar. The clamps must
securely hold the two components to maintain stiffness of the external
fixator. However, they inevitably lose some grip with time, and require
re-tightening periodically (Table 12.5).
Fig. 12.33 A pin-to-bar external fixator has three components: pins,
bars and clamps. A pin is technically a screw that protrudes out of the skin.
It is the least stiff of the three components. The weakest point within the pin
is the thread–shank junction. A bar stabilises the fracture between the pins. A
simple pin-to-bar clamp connects one pin to bar, whereas a modular pin-to-
bar clamp connects multiple pins to bar. An assembled externally fixator is a
fixed angle device.
Fig. 12.34 A ‘monobody’ external fixator is a type of pin-to-bar
external fixator, which consists of pre-assembled combined unit of
clamps and bar. It has considerable inherent stiffness due to its bulky and
rigid design. The pins are inserted into the bone fragments as normal and then
the monobody external fixator is applied as a whole. The external fixator can
be adjusted as required to facilitate fracture reduction. It is available in
variable sizes for application to different bones. (Image is reproduced with
permission of Orthofix.)

Table 12.5. Guidelines on suitable pin diameter for different bones

on the principle that pin diameter should not exceed a third of the
diameter of the bone

Bone Pin diameter (mm)

Humerus 5/ 6

Radius and ulna 4

Bones in the hand 3

 Bone Pin diameter (mm)

Femur 5/6

Tibia 5/6

Bones in the foot 3

External fixation II
Biomechanics of bone–frame construct
Bone–pin interface
The pins are inserted into bones in the ‘safe zones’ of the limbs to prevent
injury to important soft tissue structures, e.g. nerves and arteries. The bone–
pin interface is central to stability of the construct. The two main
complications of an external fixator are infection and loosening at the bone–
pin interface. The pin diameter is the main determinant of bone–pin interface
stability. A pin with too small a diameter experiences excessive micromotion
at the interface, which can lead to infection and/or fixation failure. However,
the pin diameter must also not exceed one-third of the bone diameter to
minimise the risk of pinhole-induced fracture.
A hydroxyapatite coating of pin threads improves pin purchase with time.
Pins with conical (tapered) threads are radially pre-loaded and also maintain
better purchase in bone over time. These especially ‘resistant’ pins are
usually necessary when an external fixator is needed to definitively fix a
fracture, and therefore are applied for a prolonged period of time.

Spatial configuration
External fixators can be assembled into a variety of constructs. Implant
position and orientation with respect to the bone determine bone–frame
stiffness:
 Position of implants with respect to bone.

Increasing the number of pins in each bone fragment distributes load

over more supports and increases construct stiffness. However, this also
increases the risk of infection at the bone–pin interface. Therefore, as a
compromise, usually two pins are inserted in each bone segment.
The ideal arrangement of pins in bone fragments is for the first to be
close to the fracture and for the second to be as far away from the
fracture as possible. This ‘near–near, far–far’ pin arrangement provides
maximum ‘control’ of bone fragment and improves construct stiffness.
The closer the bar is to the bone, the more stable the construct.
An additional bar between the same set of pins further increases
stiffness. The two bars should be arranged so that one is close to bone
and the other further away from it. This ‘near–near, far–far’ bar
arrangement further enhances construct stiffness.

Orientation of implants with respect to bone.

External fixators are classified as unilateral and bilateral. A unilateral

frame encircles up to 90° arc of area around the limb, and a bilateral
frame encircles more than 90° of the limb. Both types of frames are
divided further into uniplanar and biplanar. A uniplanar frame has all
the pins in one plane, and a biplanar frame typically has pins at 90° to
each other.
A biplanar frame is much more stable than a uniplanar frame. It controls
bending in different planes and provides high resistance against torsion.
A uniplanar frame provides the most stability to the fracture when
applied in the plane of bending forces.
A triangular (also known as delta) configuration is the most stable of all
the shapes.
Fig. 12.35 The stiffness of a bone–frame construct is improved when
pins and bars are applied in the ‘near–near, far–far’ arrangement. In
other words, shorter distances a and x and longer distances b and y lead to a
stiffer construct. However, the rigidity of fixation also depends on fracture
pattern, accuracy of fracture reduction and the amount of external loading. It
is imperative that the fracture is appropriately reduced to achieve adequate
construct stability.
Fig. 12.36 The four main spatial configurations of pin-to-bar external
fixator. Bilateral configurations are rarely used nowadays.
Fig. 12.37 Pin–bar–bar external fixators are commonly assembled into
a ‘delta’ (Δ) frame. A triangle is inherently a very stable geometric shape
because it is not possible to distort a triangle without distorting the actual
beams. On the other hand, a square can be distorted to form a parallelogram,
and therefore has less structural strength. This deformation pattern is also
seen in other shapes. However, adding crossbars to these shapes to form
triangles within them significantly increase their rigidity.

External fixation III

A ring external fixator provides three-dimensional stability to bone. It allows
axial loading but inhibits translational and bending deflections.

Biomechanics of ring external fixator

The basic components of a ring external fixator are as follows:

Wires: Thin wires of typically 1.8 mm diameter are passed across the
bone to emerge on both sides of the bone. There are two types of wires.
 Plain wires act as simple supports for the frame.
‘Stopper wires’ have an embedded small bead (olive). These
wires have two main functions. The olive in the wire can be used
to help in fracture reduction by pulling a bone fragment in the
required direction. The stopper wires also provide greater fixation
stability by preventing undue motion.
Pins: Half pins are also utilised in the circular frame construct.
Rings: There are three types of rings: full (closed), partial (open) and
arches. Ring properties have a significant effect on frame stability.
Full ring constructs are stiffer than partial ring constructs. Partial
rings are applied around the joints, where full rings would prevent
normal limb function or positioning.
Wider and thicker rings are stiffer as components, but smaller
diameter rings provide a stiffer overall construct than larger
diameter rings. Different diameter rings may be required within the
same frame to conform to the contour of the limb.
Rods: Threaded rods connect different rings.
Nuts and bolts: These are used to secure the other components together.
Hinges and motors: These are incorporated if controlled motion is
required between different segments of the frame.

All these components are usually made of stainless steel.

Biomechanics of bone–frame construct

A number of factors determine the stiffness of bone–frame construct. The key
factors are discussed below:

Wires are inserted in pairs in bone segments. The optimal angle

between wires for equal stability in all planes is 60°–90°. Stopper
wires provide greater stability than plain wires when the angle between
the wires is very small.
Wires are typically tensioned to 90 kg. Paired wires must be tensioned
simultaneously, as sequential tightening affects the tension in the first
wire as the second wire is tensioned.
A block of multiple rings connected together is much stiffer than a single
ring. Stiffness of a multiple-ring block increases with the number of
 connecting rods. Therefore, a bone segment should ideally be stabilised
with a ring block consisting of two rings, each with two points of
fixation to bone and connected to each other with four rods.
The construct stability is not significantly affected if the bone is not in
the centre of the ring.

Fig. 12.38 The key factors that affect the stiffness of a bone–frame
construct.
Fig. 12.39 The most essential element of a ring fixator is the tension in
the wires. Equally tensioned wires support the ring of the frame in the same
way as equally tensioned spokes of the wheel of a bicycle support the wheel
ring. In both cases, the tension in the wires constantly pulls the ring together,
which generates stiffness in the construct and maintains the overall shape.
This allows the construct to be light and rigid.
13 Trauma meeting: case-based discussions

Mechanics of trauma meeting

The trauma meeting is an integral part of orthopaedic practice. It focuses on
the clinical application of biomechanics in the context of the trauma patient.
There is no ‘one size fits all’ approach to fracture management, but
biomechanical principles apply whichever treatment modality is used.
The management of a trauma patient depends on a number of factors:

patient factors, e.g. co-morbidities, lifestyle and functional demands

the nature of the injury, e.g. the fracture pattern, soft tissue involvement,
neurovascular status and other associated injuries
surgical factors, e.g. local expertise and surgeon's preference of
treatment.

It is therefore often said in orthopaedics that the management of a fracture

depends on the ‘personality’ of the patient, ‘personality’ of the fracture and
‘personality’ of the surgeon.
The aim of this chapter is to highlight biomechanical aspects of different
treatment modalities in relation to specific fractures. It is not intended to
focus on fracture classifications, on surgical technique or to provide an
extensive clinical review of the outcome of treatments. Instead, the chapter
focuses on biomechanical interactions between specific fractures and
fixation techniques. It therefore illustrates the approach to comparing and
contrasting different techniques used in the management of a family of
fractures. In order to keep the discussions simple and manageable, key
characteristics of implants and general trends in management are presented,
and the finer points and controversies have been left out intentionally.
The aphorism ‘a fracture is a soft tissue injury that happens to involve a
broken bone’ emphasises the importance of attention to soft tissues in any
type of fracture treatment. Therefore, although the following sections
evaluate different techniques mainly in relation to the fracture, the impact of
the techniques on the soft tissues is also of paramount importance, and can be
the determining factor in treatment selection (Table 13.1).

Fig. 13.1 It is usually a race between fracture healing and failure of

fracture fixation. This is because the fixation device withstands all or some
of the applied load until the fracture is fully united. Therefore, the fixation
device could eventually experience fatigue failure if the fracture does not
unite. On the other hand, once the fracture is fully united, the fixation device
is superfluous.

Table 13.1. Trauma meeting: clinical perspective

A trauma meeting is typically held at the beginning of each day to

discuss all new acute patient admissions, review the status of all
trauma patients awaiting a surgical procedure, and discuss the
management of any other specific patient. The preliminary order of
‘trauma list’ (i.e. operating list for patients with an acute injury) is
usually also confirmed in the trauma meeting. It is a multi-
disciplinary meeting that is typically attended by:
members of the following teams: orthopaedics,
orthogeriatrics and anaesthetics and/or staff from theatre
department, e.g. scrub nurse or operating department
practitioner
trauma coordinator/ bed manager
members of the ward staff and physiotherapy and
occupational therapy teams looking after trauma patients
 staff involved in research.
doctors from other specialties, e.g. emergency medicine, and
students of the medical, nursing and allied specialties may
also attend the trauma meeting.
representatives from implant companies may also attend the
meeting to learn about the upcoming operations and surgical
equipment requirements.
The meeting is usually chaired by the on-call orthopaedic
consultant. The junior members of the orthopaedic team present
each patient and the case is then discussed. The meeting serves a
number of functions, including patient handover, education and
training, and research and audit.

Fractures of the middle third of clavicle

The clavicle is an S-shaped long bone that acts as a strut between the sternum
and shoulder. In mechanics, a strut keeps two other components apart and
resists compressive forces transmitted between them. Similarly, the clavicle
lateralises the shoulder and transmits compressive forces from the upper
limb to the trunk. In addition to these forces, the clavicle is also subjected to
significant forces from the attached muscles.
The middle third of the clavicle has the smallest cross-sectional area, and
is therefore least stiff. In addition, the ends of the clavicle are relatively
stabilised by their respective articulations. As a result, the middle third of
the clavicle is most prone to failure, and approximately 80% of clavicular
fractures occur in this section. In the past, these fractures were considered
best managed non-operatively, as almost all of these united with conservative
measures. However, although the overall union rate of clavicular fractures is
about 95%, the union rate in displaced fractures is approximately 85%; with
the non-united fractures being quite troublesome. In addition, clavicular
fractures healed in displaced position (mal-union) are also more likely to
lead to problems such as residual pain and poor shoulder posture and
function. Therefore, more recently there has been a shift towards surgically
stabilising displaced clavicular fractures.
 The mainstay treatment for most clavicular fractures is still conservative
management. This usually entails application of a supportive sling or a
figure-of-eight bandage; both forms of support are equally effective. The
support is discarded when there is no significant motion between fracture
fragments and the fracture is pain free.
Surgical treatment is considered when the fracture is shortened by >2 cm
or when the fragments are displaced >100%. Other indications for surgical
intervention include open fractures/ compromise of the overlying skin,
neurovascular injury and floating shoulder. Displaced fractures are usually
caused by high-energy injuries, and are more likely to have the associated
indications for fixations.
The two primary techniques for fixing the clavicle are plate and
intramedullary pin fixation. These techniques were associated previously
with relatively high rates of implant-related complications, which was
another reason for placing more emphasis on conservative management.
However, improvements in surgical technique and design of implants have
reduced these issues significantly.
Plate fixation is the more popular method for surgically stabilising a
clavicular fracture. The modern plates have a low profile and may be used in
locking mode to achieve high construct stability. However, a meticulous
surgical technique is required to avoid injury to the vital structures in close
association with the under-surface of the clavicle, e.g. the neurovascular
structures and the lung. Intramedullary pin fixation allows the fragments to
collapse and come together more naturally. The technique involves minimum
soft tissue dissection, and the hardware is easily removed after the fracture is
healed. For these reasons, intramedullary pin fixation is considered to be
particularly useful for fixing clavicular fractures in children.
Fig. 13.2 The middle third of the clavicle has the smallest cross-
sectional area, and is relatively unsupported compared with the ends of
the clavicle. It is therefore most prone to failure, and approximately 80% of
clavicular fractures occur in this section.

Fig. 13.3 Plate fixation allows for accurate reduction and absolute
stability of clavicular fracture.
Fig. 13.4 Intramedullary pin fixation provides relative stability to
clavicular fracture. The pin may be inserted antegrade or retrograde. It
provides limited rotational stability. (Image courtesy of Sonoma.)

Fractures of the proximal humerus

Fractures of the proximal humerus are classified according to the number of
parts to the fracture. The proximal humerus is considered to have four main
parts: humeral head, shaft and greater and lesser tuberosities. A part is
considered to be displaced if it is separated >1 cm or angulated >45°.
Therefore, the proximal humerus can sustain a fracture of between one to four
parts; the one-part fracture comprising ‘undisplaced’ fragments. The severity
of the fracture, and the risk of avascular necrosis of humeral head, increases
with the number of parts. Other factors associated with increased risk of
avascular necrosis of humeral head are head-split fractures; loss of medial
hinge, i.e. medial support to the humeral head; and fracture dislocations.
The goals of treatment are to maximise shoulder function and minimise
pain. A number of additional factors need to be considered in deciding the
appropriate treatment, e.g. bone quality and patient's age, functional level and
co-morbidities. Approximately 85% of proximal humerus fractures are
undisplaced or minimally displaced, i.e. consist of one part, and can be
managed conservatively. Surgical intervention is generally required only if
there is significant displacement between fragments or if there is an
increased risk of avascular necrosis of humeral head. The two common
surgical treatments for proximal humeral fractures are open reduction and
internal fixation, and shoulder arthroplasty.
Open reduction and internal fixation
The principles of fixing displaced fractures are as follows:

The head and shaft must be aligned correctly. There should be sufficient
stability to allow early mobilisation.
The tuberosities must be repositioned anatomically to balance the forces
produced by the rotator cuff muscles.

The tuberosities can be re-attached using trans-osseus sutures, or anchors in

a tension-band configuration. A locking compression plate is commonly used
to stabilise multi-fragmented fractures. The locking plate transfers the entire
load from humeral head to shaft whilst the fracture is healing, and if the
fracture develops non-union (pseudoarthrosis), the plate eventually breaks
due to fatigue failure.

Shoulder arthroplasty
Shoulder arthroplasty is generally considered when the fracture is not
amenable to fixation, or for a three- or four-part fracture in the older
osteoporotic patient. In general, there is a greater emphasis for fixing any
type of fracture in the young patient and a lower threshold for shoulder
replacement in the older patient. Other considerations that influence the
management are the degree of underlying arthritis and integrity of the rotator
cuff. The standard form of shoulder arthroplasty for proximal humerus
fractures is hemiarthroplasty, and the reversed shoulder replacement is
considered when there is rotator cuff deficiency or if there is a concern about
realigning the tuberosities, e.g. due to severe comminution (also see pages
134–135) (Table 13.2).
Fig. 13.5 The two common surgical treatments for proximal humeral
fractures are open reduction and internal fixation, and shoulder
arthroplasty. A locking compression plate is commonly used to stabilise
multi-fragmented fractures. However, it is associated with high rates of
complications, such as intra-articular penetration of screws, screw
loosening, mal-union, non-union and avascular necrosis of humeral head.
Most of these complications are considered to be related to surgical
technique. The risk of fixation failure is higher in three- and four-part
fractures. Therefore, there is controversy over the best management option, in
terms of fixation or shoulder replacement, for these fractures in the older
person. (Radiograph (b) courtesy of Arthrex.)

Table 13.2. General trends in management of the four types of

proximal humerus fractures in patients of different age groups

Number of parts Age group

to the proximal
humerus fracture

Under 50 50–70 years Over 70 years

years
 1 Conservative Conservative Conservative

2 ORIF ORIF Conservative/

ORIF**

3 ORIF ORIF/ Conservative/

Shoulder ORIF/
replacement* Shoulder
replacement**

4 ORIF ORIF/ Conservative/

Shoulder ORIF/
replacement* Shoulder
replacement**

* Shoulder replacement is usually considered for head-split fractures and for fracture
dislocations, i.e. when there is a higher risk of avascular necrosis.

** The treatment option is matched to the patient's functional requirements and general health.

Fractures of the mid-shaft of humerus

Fractures of the mid-shaft of humerus can be managed conservatively with a
functional brace in up to 90% of cases. A functional brace is a splint that
fully encompasses the injured segment of the limb, but does not impede the
motion of the adjacent joints. Fracture stability is achieved through hydraulic
compression forces of the muscles surrounding the fracture. The humerus can
tolerate up to 3 cm of shortening, 20° of anteroposterior angulation and 30°
varus–valgus angulation, due to the mobility afforded by the adjacent
shoulder and elbow joints.
Indications for surgical treatment include: polytrauma; open fractures;
segmental or other ipsilateral fractures; pathological fractures; and failed
conservative treatment. Fractures with associated neurovascular injury are
generally also best managed with surgery. The two main methods of
operative fixation of humeral shaft fractures are intramedullary (IM) nailing
and plate fixation.

Intramedullary nailing
IM nails can be divided into two groups of design: interlocking and ‘bio’
nails. An interlocking nail is stabilised proximally and distally with
interlocking bolts. A ‘bio’ nail is stabilised at the near end with interlocking
bolts and at the far end with an alternative locking mechanism, e.g. manual
expansion or deployment of divergent rods. Both types of nails can be
inserted antegrade or retrograde. When the interlocking nail is inserted
antegrade, there is a risk of iatrogenic injury to neurovascular structures
during distal locking. The bio nail reduces the risk of this complication;
however, fixation stability is reduced when compared with the interlocking
nail. Therefore, the interlocking nail usually provides adequate stability,
regardless of direction of insertion in relation to fracture location. However,
the stability of fixation with ‘bio’ nail is better if the interlocking bolts at the
near end of the nail are closer to the fracture site. Therefore, a ‘bio’ nail
should ideally be inserted antegrade if the fracture is in the proximal half of
the humerus and retrograde if the fracture is in the distal half of the humerus.

Plate fixation
This is the gold standard for surgical fixation of humeral shaft fractures.
Fracture configuration dictates the mode in which the plate is applied. A
transverse fracture can be fixed with direct compression plating; a short
oblique or a spiral fracture may be stabilised with lag screws with the plate
applied in neutralisation mode; and, in comminuted fractures, the plate is
applied in bridging mode. The use of locking plates is advantageous in the
setting of poor bone quality but not in fracture comminution per se. There
should ideally be a minimum of six cortices' fixation on either side of the
fracture (Table 13.3).
Fig. 13.6 The functional brace works by creating a cylinder of set
volume around the limb. When the muscles contract, their attempted
increase in size produces a hydraulic compressive force that stabilises the
fracture. Therefore, a functional brace stabilises, but does not immobilise,
the fracture. Although the fracture ends are relatively mobile, they return to
the original position because the compressive force is uniformly distributed
within the cylinder. The principle of soft tissue containment does not depend
on the stiffness of the bracing material, but instead on the set volume of the
cylinder formed. (Image courtesy of Patterson Medical Ltd.)

Table 13.3. Clinical and biomechanical considerations in

intramedullary and plate fixation of humoral shaft fractures

Intramedullary nailing Plate fixation

 Intramedullary nailing Plate fixation

Advantages The IM nail is a load-sharing Plate fixation can

implant. Its placement in the be performed
medullary canal aligns it through a variety of
with mechanical axis of the surgical
bone, and therefore it is approaches.
subjected to lower bending The neurovascular
stresses. structures,
It requires smaller surgical especially the
incisions, and involves less radial nerve, are
soft tissue stripping. directly visualised
and their integrity
confirmed.
It is associated with
much lower rates of
fracture non-union
than IM nailing.
 Intramedullary nailing Plate fixation

Limitations There is a high incidence of This is a load-

‘insertion site’ morbidity. bearing implant,
The antegrade technique is and can
associated with shoulder theoretically lead to
complications, such as stress-shielding
rotator cuff dysfunction, osteopenia.
frozen shoulder and Complication rate
subacromial impingement ranges from 3% to
syndrome. The retrograde 13%. The range of
technique is associated with complications is
iatrogenic fractures. similar to those
‘Bio’ nails have a high occurring in plate
secondary complication rate fixation of other
due to insufficient rotational fractures, e.g.
stability. infection, mal-
IM nailing is contraindicated union, non-union
in open fractures and in and neurovascular
fractures associated with injury.
neurological deficit. It is important to
document where the
radial nerve
crosses the plate –
this is useful in
future surgery.

Intra-articular fractures of distal humerus

The standard management for intra-articular fractures of distal humerus is
open reduction and internal fixation with plates and screws. The principles
of fixation are: anatomic reduction of articular fragments; restoration of
normal alignment between humeral shaft and distal fragments; and provision
of a stable construct to allow for early, active elbow mobilisation. The distal
humerus consists of distinct medial and lateral columns. The fixation is more
stable when both columns are plated than when only one column is plated. In
the double-plating technique, the plates can be applied perpendicular or
parallel to each other. In both methods, the fracture is anatomically reduced,
and is often temporarily stabilised with k-wires, before the application of
plates.
The perpendicular plate configuration is the traditional approach to two-
column plate fixation. It involves the application of one plate to the medial
supracondylar ridge of the humerus and the other plate to the posterolateral
aspect of the lateral side of the humerus. The parallel plate configuration is a
comparatively new concept in which the plates are applied along the medial
and lateral supracondylar ridges. The plates used in both configurations are
usually the locking type and pre-contoured. The locking plates provide a
more stable fixation than non-locking plates in poor-quality bone or when the
fracture is highly comminuted.
From a biomechanical perspective, the parallel plate configuration
provides a significantly stiffer and stronger fixation than the perpendicular
plate configuration. This is due to a combination of factors. Firstly, there are
generally more screws fixing the fracture in the parallel plate configuration,
and the screws are usually also longer, as they are all applied across the
medio-lateral width of the humerus. Secondly, according to the principle of
area moment of inertia, parallel plating provides a higher resistance to
bending in the antero-posterior direction, which is the main type of load
acting at the fracture due to flexion–extension motion of the adjacent elbow
joint. Thirdly, the overall construct of parallel plate fixation locks the medial
and lateral columns together in the form of an arch, which is an inherently
stable arrangement.
However, from a clinical perspective, both methods stabilise the fracture
adequately to achieve fracture union. As perpendicular plate configuration
has been used for longer, a number of clinical studies are available to show
the long-term results of this method of fixation. These studies indicate that
this technique provides satisfactory outcome with long-term follow-up. On
the other hand, parallel plate configuration does not have such a long clinical
track record, but the available data indicate that this technique also provides
comparable clinical results.
 Finally, from a surgical perspective, there are some suggestions that plate
application to the lateral condylar ridge can be technically difficult, and that
parallel plate configuration is associated with more extensive subperiosteal
elevation, which can increase the risk of delayed union and non-union.

Fig. 13.7 The distal humerus relies on a triangular construct for

stability. (a) The walls of the triangle are formed by medial and lateral
columns and the trochlea. Intra-articular fractures of distal humerus affect all
three walls. The standard management for these fractures is open reduction
and internal fixation with plates and screws. In the double-plating technique,
the plates can be applied perpendicular or parallel to each other. (b) Parallel
plate configuration provides significantly stiffer and stronger fixation than (c)
perpendicular plate configuration.
Fig. 13.8 According to the principle of area moment of inertia (I),
parallel plate configuration (a) provides a higher resistance to bending in
the antero-posterior direction than (b) in the perpendicular plate
configuration (see pages 56–57 footnotes).
Figures (a) and (b) show models of the axial views of the distal humerus
with plates applied.

Distal radius fractures

These are amongst the commonest bone fractures. The mode of treatment
depends on the fracture configuration and on the functional demands of the
patient. The majority of these fractures are managed satisfactorily with
closed reduction and cast treatment. However, a small proportion may
require percutaneous k-wire fixation or open reduction and internal fixation.

Percutaneous k-wire fixation

This is undertaken in cases where the fracture has enough inherent stability to
be managed in a cast, but either it cannot be reduced fully with manipulation
or it is considered to require further support to maintain satisfactory
alignment until union. This judgement is usually made at the time of closed
reduction of the fracture, although the fracture pattern may provide an
indication of fracture stability beforehand. Different types of fractures
require different numbers of stabilising k-wires, which are applied in
various configurations based on the fracture pattern. The most common
configuration consists of a ‘dorsal’ wire and a ‘radial styloid’ wire. The
patient is managed in a cast and the wires are usually removed at 6 weeks.
The results of k-wire fixation are generally better in younger patients than in
older patients, as k-wires maintain a better hold in normal bone than in
osteoporotic bone.

Kapandji k-wiring technique

This technique is typically used to reduce and stabilise an extra-articular
distal radius fracture. It involves percutaneously inserting a k-wire through
the fracture site. Then, the near end of the k-wire is arced towards the distal
fragment. This manoeuvre levers the distal fragment into the reduced
position. Once an adequate position is achieved, the k-wire is driven through
the far cortex to achieve stability. The k-wire does not fix the distal fragment,
but instead provides buttress support.

Open reduction and internal fixation

The threshold in terms of fracture characteristics for consideration of open
reduction and internal fixation includes radial shortening > 3 cm, dorsal tilt
>10° or an intra-articular step >2 mm after attempted closed reduction. This
is also indicated in fractures that may initially be satisfactorily reduced, but
have a high risk of displacement due to the basic fracture pattern. Fractures
with associated neurovascular injury are generally also best managed with
surgery.
There has been a transformation in the internal fixation technique over
time. In the past, dorsal plating was the most popular method. As most distal
radius fractures are displaced dorsally, this worked on the principle of
buttressing the fragments to prevent fracture displacement. However, the site
of plate application – under the extensor compartments of the wrist – was a
compromise, and dorsal plating was associated with tendonitis and rupture
of extensor tendons. Still, this was the most mechanically sound method of
fracture fixation. The emergence of fixed-angle volar locking plate stimulated
interest in volar plating, and this has now become the most commonly used
internal fixation technique.
Fig. 13.9 An example of the use of k-wires to stabilise a distal radius
fracture. Here, one dorsal and one radial styloid wire are adequate to
provide stable, three-dimensional fixation. The dorsal k-wire has been
applied with the Kapandji tehnique. The principles of k-wire fixation
discussed on pages 166–167 have also been applied.

Fig. 13.10 Kapandji k-wiring technique is typically used to reduce and

stabilise an extra-articular distal radius fracture.
Fig. 13.11 Open reduction and internal fixation enables more accurate
fracture reduction and provides greater stability. The most commonly
used internal fixation technique is volar plating. This technique is effective
because the distal screws holding the bone fragments (which are often very
comminuted) are fastened into the locking plate and therefore are prevented
from backing out. The construct has very high stability, despite relatively
poor bone quality and low intrinsic fracture stability.

Intertrochanteric fractures of the proximal femur

An intertrochanteric fracture divides the proximal femur into two main parts:
one consisting of femoral head and neck and the other consisting of femoral
shaft. These main parts may be fragmented even further, e.g. a typical four-
part fracture consists of femoral head and neck, femoral shaft, greater
trochanter and lesser trochanter fragments. The main fracture line usually
runs obliquely from the greater trochanter to the lesser trochanter. However,
in a ‘reverse oblique’ fracture, the main fracture line runs from
intertrochanteric region distally to lateral femoral cortex. In addition, in a
‘subtrochanteric’ fracture, the main fracture line extends distal to the lesser
trochanter. The intertrochanteric fracture does not affect the blood supply to
the femoral head and is therefore suitable for fixation. However, different
types of intertrochanteric fractures have different intrinsic fracture stability.
The two implants commonly used for treatment of intertrochanteric
fractures are dynamic hip screw (DHS) and IM nail. The goal of fixation is
to achieve a construct sufficiently stable to allow immediate full weight-
bearing and rehabilitation. Both implants utilise a cancellous lag screw to
stabilise the femoral head. The lag screw has a purchase in the femoral head
but not in the shaft; therefore, weight-bearing leads to a directed sliding of
the femoral head, which produces compression at the fracture site. The two
devices differ in how the lag screw is connected to the femoral shaft.

Dynamic hip screw

This consists of a plate with an extension (barrel) that slides over the lag
screw. The plate is fixed to the femoral shaft with a number of screws. The
plate is not secured mechanically to the lag screw; therefore, during weight-
bearing the lag screw slides down within the barrel of the plate, allowing the
femoral head to slide down as well, thereby compressing the fracture.
Weight-bearing also produces a bending moment, which acts to displace the
fracture. The plate–lag screw connection prevents the lag screw from
bending, and the screws fixing the plate to the femoral shaft prevent the plate
from bending.

Intramedullary nail
In this setting, an IM nail is also referred to as a ‘cephalomedullary’ implant.
The lag screw is inserted after the placement of IM nail. The lag screw may
be free to slide through the nail, or be fixed in its position; weight-bearing
allows the femoral head to slide down and compress the fracture in both
cases. The nail is fixed to the femoral shaft with the distal screw(s). Again,
the bending moment is resisted by the lag screw–IM nail construct. There is a
choice of short or long IM nail, depending on the fracture configuration
and/or whether there is a need to support the whole bone.
In both methods, a further ‘derotational’ screw may be used to provide
additional rotational stability to the fracture.
Fig. 13.12 Mechanics of intertrochanteric fracture fixation with
dynamic hip screw. The DHS is a load-bearing implant. Bending stiffness of
DHS plate is proportional to its thickness to the third power. The bending
moment produced is greater than in IM nail fixation, as there is a longer lever
arm between the plate and applied force. (a) The DHS is ideal for use in
intertrochanteric fractures with stable configurations. The fracture becomes
progressively less stable with increasing number of parts, requiring more
support from the implant. (b) The reverse oblique and subtrochanteric
fractures are inherently unstable as there is a loss of ‘lateral buttress’
support; so with DHS fixation, the proximal fragments continue to slide down
unstopped, leading to medial relative displacement of the femoral shaft.
Fig. 13.13 Mechanics of intertrochanteric fracture fixation with an
intramedullary nail. The IM nail is a load-sharing implant. It provides a
more stable fixation than the DHS for two main reasons: the IM nail is a
stiffer implant than DHS because its bending stiffness is proportional to its
radius to the fourth power; and the bending moment produced is less than in
DHS fixation, as the IM nail takes up a more medial position and therefore
the lever arm between the implant and applied force is shorter. However,
clinically, IM nail has a clear advantage over the DHS in (a) reverse oblique
and (b) subtrochanteric fracture configurations. In these fractures, the shaft of
IM nail forms the lateral buttress and limits the displacement of the proximal
fragments. A long IM nail may be required for subtrochanteric fractures or if
there is a need to support the whole bone prophylactically.

Fractures of distal third of tibia

Fractures of distal third of tibia vary in pattern from the simple extra-
articular type to those with intra-articular tibial plafond extension and more
severe pilon fractures. These fractures can be managed with a number of
different methods. Each method has advantages and limitations, and currently
there is no consensus on the most appropriate method for treating these
fractures. This section considers the biomechanical issues related to fixation
of an isolated, simple extra-articluar fracture of distal third of tibia with two
common techniques: IM nailing and plate fixation.

Intramedullary nailing
IM nailing is considered to be the treatment of choice for most fractures of
the mid-diaphysis of tibia. It is also used for fixing fractures of the distal
third of tibia. However, the biomechanical issues in IM nail fixation of the
two types of fractures are quite different. In the mid-diaphyseal fracture, the
bone–implant construct is inherently stable, as the tubular profile of bone in
the diaphyseal region means that fracture generally needs to be quite well
aligned for the nail to cross the fracture site, and so there is extensive contact
between the bone and the nail. However, in the fractures of the distal third of
tibia, the bone–implant construct is inherently less stable, because in the
wider distal third section, there is limited direct contact between the bone
and the nail. This can cause difficulty in achieving the correct alignment of
fracture. The limited bone–implant contact means that most of the stability is
provided by distal interlocking screws. It further means that there is
increased stress on the interlocking screws (see pages 176–177), which
therefore are more prone to breakage.

Plate fixation
A low profile, contoured, locking compression plate is another popular
choice for fixing this fracture. Other types of plate, e.g. a dynamic
compression plate or a T-plate, can also be used. The usual approach for
plating is anteromedial, although an anterolateral approach is also used. The
anterolateral approach has the advantage that it can also be used to fix the
fibula fracture if required.
The plate can be applied with either the standard open reduction and
internal fixation technique, in which the fracture is fully exposed and
anatomically reduced, and the plate is applied with the aim of achieving
interfragmentary compression; or with a minimally invasive plate
osteosynthesis (MIPO) technique, in which the fracture is reduced closed and
the plate is applied through a small incision and the screws are inserted
through ‘stab’ incisions. The MIPO technique places emphasis on restoring
the tibial mechanical axis (rather than anatomically reducing the fragments)
and on protecting the soft tissues around the fracture site.
Fig. 13.14 Fractures of the distal tibia may be surgically managed with
intramedullary nailing or plate fixation. It is usually not possible to use the
IM nailing technique if the fracture line extends to within 5 cm of the ankle
joint. Advocates of IM nailing emphasise that the reaming required before
nail insertion helps to proliferate the fracture repair process and that the zone
of injury is not disturbed by dissection. The plate may be applied in a number
of different modes, e.g. in neutralisation, compression or bridging mode.
Meta-analyses of studies comparing the outcomes of the two techniques show
that IM nailing is associated with a higher risk of knee pain (because it
requires surgical approach around the knee joint for nail insertion) and mal-
union (as it can be difficult to achieve and maintain correct rotation of
fragments with the IM nail), whereas plate fixation is associated with a
higher risk of superficial wound infections. There is no difference between
the techniques in time to fracture union and fracture non-union rates and other
functional results.
Pilon fractures
A pilon fracture is a comminuted fracture of the distal tibia that extends into
the tibial plafond. The term pilon is French for pestle, which reflects the
shape of the distal tibia.
The pilon fracture typically is caused by an axial compression injury that
drives the talus into the tibia. The talus is one and a half times stronger than
the distal tibia, and a severe impact causes the tibial articular surface to
fracture into fragments. The injury may result in damage to the articular
surface of the talus. In addition, there is often also an associated distal fibula
fracture.
This is a severe injury that consists of three elements: fracture
comminution, disruption of the articular surface and associated soft tissue
swelling. The standard treatment for this fracture is surgical fixation. It is
common practice to stabilise the fracture temporarily with a pin-to-bar
external fixator. The definitive surgery can then be delayed until the soft
tissue swelling has settled. In the meantime, it is usually also necessary to
have further imaging of the fracture, in the form of a CT scan, to plan surgery.
The principles of initial management are aptly summarised as ‘span, scan and
plan’.
The surgical techniques for definitive fixation of pilon fracture can be
divided into two groups: open reduction and internal fixation, and external
fixation. The principles of both treatments are management of soft tissues,
reconstruction of joint line and realignment of the mechanical axis of tibia.
Open reduction and internal fixation involves direct reconstruction of tibial
articular surface, bone grafting to support the reconstructed joint line, and
application of a bridging plate and in certain cases buttress plate(s) as well.
Depending on fracture configuration, free-standing screws may also be
applied to fix specific fragments, e.g. if there is a significant fracture line in
the coronal plane, a screw may be applied from anterior to posterior with the
aim of ‘capturing’ and stabilising a large posterior fragment. The procedure
is performed through as small an incision as possible to minimise injury to
the soft tissue envelope.
External fixation involves closed or minimally invasive reduction of
articular surface, which may be supplemented with limited internal fixation,
e.g. free-standing screws, and application of an external fixator to maintain
mechanical alignment until fracture heals. The external fixator utilises the
principle of ligamentotaxis to provide stability. Both pin-to-bar and circular
frames can be utilised, or a hybrid frame consisting of straight bars and rings
may also be used. The pin-to-bar frame is usually monolateral and bridging-
type, i.e. it spans across the ankle joint.
There is no consensus on the management of the associated fibula fracture.
Fixing the fibula fracture can help to maintain the length and alignment of the
pilon fracture. However, this involves an additional incision, which can
increase the risk of wound complications.

Fig. 13.15 A pilon fracture is a comminuted fracture of the distal tibia

that extends into the tibial plafond. It is estimated that up to a quarter of
these fractures are also open. A temporary external fixator commonly is
applied to stabilise the fracture and allow soft tissues management before
further surgery. The temporary external fixator spans the ankle joint and is
usually applied in triangular configuration, i.e. as a ‘delta’ (Δ) frame.
Depending on the reduction and stability achieved, a temporary external
fixator could even be used for long-term treatment. (Image (c) is reproduced
with permission of Tony Meehan.)
Fig. 13.16 These radiographs show an example of a pilon fracture
managed with open reduction and internal fixation. Here, a locking plate
has been applied in bridging mode.

Fig. 13.17 Classification of different external fixation constructs that

can be used to manage pilon fractures. The term ‘bridging’ means the same
as ‘spanning’.

Ankle fractures
The stability of the ankle joint is derived from congruency of the articulating
surfaces and from the soft tissues, i.e. the ligaments. The treatment of ankle
fractures is based on not only the fracture, but also on how it affects the
stability of the joint. In ankle fractures, the term ‘stability’ is more often used
in the context of the joint than the fracture. The two main groups of treatments
for ankle fractures are conservative, and open reduction and internal fixation.
The algorithm for management of ankle fractures is extensive, due to the
variety in ankle fracture patterns and in treatment options in each group. The
following is intended to be a simple guide to the basic principles of the two
types of management.

Conservative treatment
Conservative treatment is ideal for stable ankle fractures, i.e. when the joint
is stable and the talus is undisplaced. The deltoid (medial) ligament is
central to maintaining joint stability in ankle fractures. It originates from the
medial malleolus and inserts onto the medial aspect of the talus, and
therefore prevents the talus from displacing laterally. An ankle fracture can
be managed conservatively if the deltoid ligament is intact. An example of a
stable ankle fracture is the isolated fracture of the lateral malleolus.
However, a lateral malleolar fracture with tenderness over the medial
malleolus, which suggests deltoid ligament rupture, is potentially unstable. It
may still be managed conservatively (at least initially) if there is no
associated lateral talar shift.

Open reduction and internal fixation

This is indicated for unstable ankle fractures, i.e. for when the ankle joint is
not congruent. The examples of unstable ankle fractures include a lateral
malleolar fracture with lateral talar shift and bimalleolar and trimalleolar
fractures. In all these cases, the deltoid ligament is non-functional. In a
trimalleolar fracture, the posterior malleolus usually only requires fixing if it
is more than a third of the articular surface, and therefore is likely to affect
joint stability.
These fractures may also be associated with diastasis of distal tibiofibular
syndesmosis; which appears as widening of the normal space between the
tibia and fibula. If the syndesmosis requires fixing, then a long screw is used
to hold the fibula and tibia in the anatomical position whilst the syndesmosis
heals. According to the biomechanical studies, one 3.5 mm cortical screw
passed across three cortices, i.e. lateral and medial cortices of fibula and
lateral cortex of tibia, about 2.5 cm above the ankle joint line is usually
adequate for stability. It is important that the ankle is in dorsiflexion when the
syndesmosis screw is inserted to ensure that the wider, anterior aspect of
talus is engaged in the mortise, which prevents over-tightening of the screw.
Isolated fractures of the medial malleolus are also fixed routinely to
restore the ankle mortise, although undisplaced medial malleolar fractures
may be managed conservatively.

Fig. 13.18 The ligaments are the most important soft tissues for
maintaining joint stability, as there are no muscles directly acting on the
talus. Disruption of the deltoid ligament allows the talus to displace
laterally. Similarly, injury to the syndesmosis leads to diastasis between the
tibia and the fibula. The diastasis itself does not predispose to lateral talar
shift, but it is associated invariably with deltoid ligament injury and therefore
with an unstable ankle.
Fig. 13.19 This is an example of a stable ankle fracture. The deltoid
ligament is intact and the talus is undisplaced.

Fig. 13.20 This is an example of an unstable ankle fracture. The deltoid

ligament is disrupted and there is lateral talar shift.
Fig. 13.21 This is an example of a severe ankle fracture that has been
internally fixed. The lateral malleolus has been fixed with a plate applied in
neutralisation mode. The medial malleolus has been fixed with two partially
threaded cancellous screws, which by design are acting as lag screws. The
ankle syndesmosis has also been fixed.

Further reading
Banaszkiewicz PA, Kader DF (eds) (2012). Postgraduate Orthopaedics:
The Candidate's Guide to the FRCS (Tr and Orth) Examination. 2nd edn.
Cambridge: Cambridge University Press.
Bucholz RW, Court-Brown CM, Heckman JD, Tornetta P (eds) (2009).
Rockwood and Green's Fractures in Adults. 7th edn. Vol. 1 and 2. New
York: Lippincott, Williams and Wilkins.
Burstein AH, Wright TM (1994). Fundamentals of Orthopaedic
Biomechanics. New York: Lippincott, Williams and Wilkins.
Charnley J (1999). The Closed Treatment of Common Fractures.
Cambridge: Cambridge University Press.
Dandy DJ, Edwards DJ (2009). Essential Orthopaedics and Trauma. 5th
edn. London: Churchill Livingstone, Elsevier.
Denis F (1984). Spinal instability as defined by the three-column spine
concept in acute spinal trauma. Clin Orthop Relat Res. 189: 65–76.
Gougoulias NE, Khanna A, Maffuli N (2009). History and evolution in total
ankle arthroplasty. Br Med Bull. 89: 111–151.
Holdsworth F (1970). Fractures, dislocations and fracture–dislocations of
the spine. JBJS-A. 52: 1534–1551.
Knahr K (ed) (2011). Tribology in Total Hip Arthroplasty. London:
Springer.
Lucas GL, Cooke FW, Friis EA (1999). A Primer of Biomechanics. New
York: Springer-Verlag.
Miller MD, Thompson SR, Hart J (2012). Review of Orthopaedics. 6th edn.
Philadelphia: Saunders.
Mow VC, Huiskes R (2005). Basic Orthopaedic Biomechanics and
Mechano-biology. 3rd edn. New York: Lippincott, Williams and Wilkins.
Nordin M, Frankel VH (2001). Basic Biomechanics of the Musculoskeletal
System. 3rd edn. London: Lippincott, Williams and Wilkins.
Ruedi TP, Buckley RE, Moran CG (2007). AO Principles of Fracture
Management. 2nd edn. Vol. 1 and 2. New York: Thieme.
Solomon L, Warwick D, Nayagam S (eds) (2010). Apley's System of
Orthopaedics and Fractures. 9th edn. New York: CRC Press.
Vickerstaff JA, Miles AW, Cunningham JL (2007). A brief history of total
ankle replacement and a review of the current status. Med Eng Phys. 29:
1056–1064.
White AP, Panjabi MM (eds) (1990). Clinical Biomechanics of the Spine.
Philadelphia: Lippincott, Williams & Wilkins.
 Index
allometry 61
alloys 28
mechanical properties 28
molecular structure 28
strengthening 28
see also metals
alumina 35
anatomy
ankle 150–151
elbow 136–137
hip 98–99
knee 112–115
shoulder 126–129
ankle 150–153
axis of motion 151
dorsiflexion–plantarflexion arc 150
functional anatomy 150–151
gait rockers 154–155
ligaments 205
loads acting on 150
ankle arthrodesis 153
ankle fractures 204
conservative treatment 204
lateral displacement of talus 17
Lauge–Hansen classification 70
open reduction and internal fixation 204–205
stable 205
unstable 205
ankle replacement 152–153
biomechanics 152
component design/alignment 152–153
annealing 26
area moment of inertia 56–57
articular cartilage 88–89
axial force 52

balanced (Hamilton–Russell) traction 162–163

Bankart lesion 126
bony 126
beams 54–57
geometric properties 56
mechanics 54–55
stiffness 57
theory 54–55
types of 55
bending 50–51, 54–57
three-point 54
bending load fractures 66
biceps 11
biceps brachii 136
bioactive ceramics 34–35
bioinert ceramics 34–35
biomaterials 24–48
ceramics 24–25, 32–34
composites 24–25, 40–42
definition 24
metals 24–27
molecular structure 25
polymers 24–25, 36–39
structure and properties 24
biomechanics 2–3
orthopaedic 3
biotribology 80–94
friction 9, 81–83
lubrication 81, 86–93
wear 81, 84–85
blister formation 81
bone 10, 44–47
ageing 46–47, 61
 anisotropy 46
arrangement of 70
brittleness 46
cancellous (trabecular) 44
characteristics 45
cortical (compact) 44
fatigue properties 65
mass 47
material properties 60, 70
mechanical properties 46–47
molecular structure 44–45
remodelling 45
stiffness 46
strength 16
structural properties 70
viscoelasticity 46
see under specific bones and types
bone cement 38–39
time–temperature curve 39
bony Bankart lesion 126
boosted lubrication 90–91
boundary lubrication 86–87
brachialis 11, 136, 140
brittle fractures 62
brittleness 21, 32
bone 46
buckle fracture 68
butterfly fragment fractures 66–67

calcium carbonate 44
calcium hydroxyapatite 34–35
calcium phosphate 44
cancellous screws 169
cancellous (trabecular) bone 44
cannulated screws 169
cantilever beams 55
carbon fibre composites 42
carbon fibre reinforced polymer (CFRP) 43
carrying (cubitus) angle 136–137
case-based discussions 186–206
casts 164–165
biomechanics 164
bone-cast construct 164
fracture reduction 165
full 164
spica 164
ceramic on ceramic bearing surface 110
ceramics 24–25, 32–34
bioactive 34–35
bioinert 34–35
chemistry 33
classification 33
mechanical properties 32
molecular structure 32
orthopaedic uses 34
cerclage wires 167
Charnley stem 102–103
children
bone structure 46, 60
clubfoot deformity 23
cubitus angle 136
femoral fractures 178
fracture patterns 68–69
long bones 60
traction 160–163
clavicle fractures 188
intramedullary nails 189
plate fixation 188–189
cobalt–chrome 30–31
collagen 44
composite beam stem 102–103
composites 24–25, 40–42
laminated 41
mechanical properties 40
 molecular structure 40
orthopaedic uses 42
compression 50–53
stresses 52
compressive load fractures 66
concrete 41
copolymers 36
corrosion 74–75
chemistry 75
protection against 74
types of 74
cortical (compact) bone 44
cortical screws 169
couple 6–7
creep 22–23
crevice corrosion 74
cruciate ligaments 114–115
cubitus (carrying) angle 136–137

deep flexion arc of knee 116

deltoid 129
double-leg stance 98
dough time 39
ductile fractures 62–63
ductility 20
dynamic hip screw 198–199
dynamics 2

elastic deformation 20
elastic stability 179
elastohydrodynamic lubrication 86, 91
elbow 136–141
dislocation 139
flexion forces 11
functional anatomy 136–137
cubitus (carrying) angle 136–137
range of motion 136
 supination–pronation 137
reaction forces 140–141
extension 140
flexion 140
stabilisers 138–139
dynamic 138
static 138
equilibrium conditions 6
erosive corrosion 74
external fixation 180–185
bone–frame construct 182
bone–pin interface 182
ring fixator 184
spatial configuration 182
stiffness 183, 185
humerus fractures 190–191
monobody 181
pin diameters 181
pin-to-bar 180–181, 183
ring 184–185

facet joints of vertebrae 142

fatigue fractures 64
fatigue life 65, 72
fatigue limit 64–65
fatigue strength 65
femoral stem implants
cemented fixation 102–103
cementless fixation 104–105
femur
age-related changes 61
fractures
dynamic hip screw 198–199
external fixation 181
intramedullary nails 178–179, 198–199
fibre reinforced polymer composites 43
filiform corrosion 74
flexible nails 178–179
fluid film lubrication 86–87, 93
fluids, friction in 82
foot 156–157
external fixation of fractures 181
gait rockers 154–155
as rigid lever 156
as shock absorber 156
subtalar joint 150–151, 156–157
force 4–5, 12, 16, 53
alignment 53
axial 52
ground reaction 9
joint reaction 10
moment of 6–7
and musculoskeletal system 5
normal 52
shear 52
stress–strain relationships 17
transverse 52
forearm fractures
external fixation 181
humerus
distal 194–195
mid-shaft 192–193
proximal 161, 190
intramedullary nails 179
radius 196
forearm mass 11
fractures (general) 62–79
acute 64
brittle 62
ductile 62–63
fatigue 64
fundamentals 62
healing 187
mechanics 70
 mechanism 64–65
patterns 66–71
bending load fractures 66
children 68–69
compressive load fractures 66
tensile load fractures 66
torsional load fractures 66
S-N curve 64–65
fractures (specific)
ankle 204
lateral displacement of talus 17
Lauge–Hansen classification 70
clavicle 188–189
intramedullary nails 189
plate fixation 188–189
femur 178–179, 198–199
humerus
distal 194–195
mid-shaft 192–193
proximal 161, 190
radius 196
tibia
distal third 200
pilon 202–203
ulna 181
fracture fixation 187, 158–185
casts 164–165
external 180–185
intramedullary nails 176–179
plates 172–175
principles 158
surgical screws 168–171
traction 160–163
wires 166–167
see under individual methods for detail
fracture gap length 17
fracture healing 76–77
 biomechanical process 78–79
interfragmentary strain 78, 80–94
primary (cortical) bone 76–78
secondary (callus) bone 76–79
fracture point 18
fracture stiffness 78
free-body force diagrams 8
inclined surface 53
interactions between objects 9
lower limb 9
hip 99, 101
knee 119
upper limb 11
elbow 141
shoulder 131
fretting corrosion 74
friction 9, 81–83
in fluids 82
in joints 82
friction equation 82
frictional torque 82
fully threaded screw 170
functional arc of knee 116
functional brace 164–167
functional spinal unit 145

gait cycle 154–155

Gallow's traction 162–163
galvanic corrosion 74
geometric properties 50–51
beams 56
columns and ties 52
long bones 60
shafts 58
glass fibre reinforced polymer (GFRP) 43
glenohumeral joint 126–127
glenohumeral ligaments 128
glenoid labrum 126–127
grain boundary 26
grain size reduction 28–29
greenstick fracture 68
ground reaction force 9

Hamilton–Russell (balanced) traction 162–163

hardness 20
Haversian canals 44
hip 96–110
functional anatomy 98–99
prosthesis see hip replacement
reaction forces 98–101
clinical implications 100
double-leg stance 98
single-leg stance 98–99
with 100N load in opposite hand 100–101
with cane support in opposite hand 100–101
hip replacement 83, 100
bearing surfaces 110
advantages/disadvantages 111
ceramic on ceramic 110
metal on metal 110
metal on polyethylene 110
Oxinium™ 110
cemented fixation 102–103, 105
composite beam 102–103
femoral stem implant design 102–103
load transfer 103
surface finish 102
taper slip 102
vs cementless fixation 104
cementless fixation 104–105
femoral stem implant design 104–105
vs cemented fixation 104
components 106–107
alignment 106
 design 106
femoral head
head–neck ratio 109
jump distance to dislocation 109
size 108–109
wear patterns 91, 108
linear 108
volumetric 108
hybrid fixation 105
Hooke's law 18, 23
humeroulnar articulation 138
humeroulnar joint 136
humerus
distal 136–137, 195
fractures see humerus fractures
head 126
humerus fractures
distal humerus 194–195
external fixation 181
mid-shaft 192–193
functional brace 193
intramedullary nails 192–193
plate fixation 192–193
proximal humerus 161, 190
open reduction/internal fixation 190–191
shoulder arthroplasty 190
hydrodynamic lubrication 86, 91
hydroxyapatite 44
coating of pin threads 182
hysteresis 22–23

incline plane-based simple machines 12–13

inertia
area moment 56–57
moment 79
polar moment 58
infraspinatus 128
interfragmentary strain 78–79
intervertebral disc 144–145
intramedullary nails 176–179
biomechanics 176
bone–nail construct 176–177
flexible nails 178
clavicle fractures 189
elastic stability 179
femoral fractures 178–179, 198–199
flexible 178–179
forearm fractures 178–179
humerus 192–193
stiffness 177
tibial fractures 200–201
working length 177

joint reaction force 10

joints 10
artificial see prosthetic joints
friction in 82
synovial see synovial joints
wear 84
see under individual joints

k-wires 166–167
Kapandji technique 196–197
radius fractures 196–197
Kapandji k-wiring technique 196–197
kinematics 2
kinetics 2
knee 112–125
axes 113
flexion–extension arc 116–117
deep flexion arc 116
functional arc 116
patellofemoral contact zones 116–117
screw home arc 116
 functional anatomy 112–115
geometry and alignment 112–113
patellofemoral joint 80
tibiofemoral joint 112
prosthesis see knee replacement
Q-angle 113
reaction forces 118–119
patellofemoral joint 118
tibiofemoral joint 118–119
soft tissues 114–115
cruciate ligaments 114–115
menisci 114–115
knee replacement
classification of prostheses 121
component alignment 124–125
femoral component 124
patellar component 124
tibial component 124
component design 120–123
fixed bearing vs mobile bearing 122–123
patellar component 122
single-radius vs multi-radius 120–121
stability vs mobility 120
range of motion 125

lag screw fixation 170–171

lateral collateral ligament 139
Lauge–Hansen classification 70
lever-based simple machines 12
levers 6, 7, 13
first-class 6
musculoskeletal system 6, 14
second-class 6
third-class 6
ligamentotaxis 161
ligaments
ankle 205
 elbow 139
shoulder 128
spine 144
see under individual ligaments
load bearing 159
loading 3, 50–51
geometric properties 50–51
material properties 50–51
modes of 50–51
loading curve 23
locking plates 172
locking screws 169
long bones 60–61
geometric properties 60–61
material properties 60
lower limb
ankle 150–153
axes 96–97
anatomical 96
mechanical 96
relationships 96
weight-bearing 96
foot 156–157
fractures see fractures (specific)
hip 96–110
knee 112–125
relationships 97
lubrication 81, 86–93
modes of 86
boundary 86
fluid film 86
mixed 86
prosthetic joints 92
Stribeck curve 88–89, 93
synovial fluid 88
synovial joints 90
boosted lubrication 90–91
 squeeze film lubrication 90–91
weeping lubrication 90–91

machines, simple 12
concept of 12
incline plane-based 12–13
lever-based 12
mechanism of action 13
materials
classification 19
properties 50–51
bone 46–47, 60, 70
mechanical 20
viscoelastic 22–23
see also biomaterials
McKellop's classification 85
mean 3
mechanical properties 20
bone 46–47
ceramics 32
metals 26
polymers 36
viscoelasticity 36
strain-rate dependent 22
medial collateral ligament 139
median 3
menisci of knee 114–115
metal on metal bearing surface 110
metal on polyethylene bearing surface 110
metals 24–27
ductile to brittle transition 63
mechanical properties 26
elasticity 26
plasticity 26–27
molecular structure 26–27
orthopaedic uses 30
see also alloys; individual metals
mixed lubrication 86–87
mode 3
moment of force 6–7
moment of inertia 79
monobody external fixator 181
motion 12
Newton's laws of 4
multimode lubrication 91
muscles 10
shoulder 128
spine 144
see under individual muscles
musculoskeletal system 2
forces acting on 5
levers 6
simple machines 14
static analysis 10

nailing see intramedullary nails

Newton's laws of motion 4
ninety degree–ninety-degree traction 162–163
normal force 52

orthopaedic biomechanics 3
orthopaedic surgery 3
see also prosthetic joints
osteons 44
overhanging beams 55
Oxinium™ bearing surface 110

paper-tearing experiment 70–71

partially threaded screw 170–171
patellofemoral contact zones 116–117
patellofemoral joint 80
reaction forces 118
physes (growth plates) 68
pilon fractures 202–203
pin-to-bar external fixator 180–181, 183
pitting corrosion 74
plaster of Paris 42–43
plastic deformation 18
plate fixation 172–175
biomechanics 172
bone–plate construct 172, 174–175
conventional vs locking plate 172
stiffness 172–173
clavicle fractures 188
humerus fractures 192–193
modes of application
bridging 174–175
buttress/antiglide 174–175
compression 174–175
neutralisation 174–175
tension band 174–175
tibial fractures 200–201
point loads 10
polar moment of inertia 58
polo mint concept 70
polydioxanone (PDS) 38
polyethylene 37
UHMWPE 38–39
polyglycolide (PGA) 38
polylactide (PLA) 38
polymers 24–25, 36–39, 46
biodegradable 38
long-term implantable 38
mechanical properties 36
molecular structure 36
orthopaedic uses 38
viscoelasticity 36–37
polymethylmethacrylate (PMMA) bone cement 38
prosthetic joints
lubrication 92
Stribeck curve 93
 wear 83, 85
see under specific joints
proximal radioulnar joint 136
pulleys 13
musculoskeletal system 14–15

Q-angle 113

radius fractures 196

external fixation 181
Kapandji k-wiring technique 196–197
open reduction and internal fixation 196–197
percutaneous k-wire fixation 196–197
reaction forces
elbow 140–141
extension 140
flexion 140
hip 98–101
clinical implications 100
double-leg stance 98
single-leg stance 98–99
with 100N load in opposite hand 100–101
with cane support in opposite hand 100–101
knee 118–119
patellofemoral joint 118
tibiofemoral joint 118–119
shoulder 130–131
abduction with elbow in extension 130–131
abduction with elbow in flexion 130–131
clinical implications 130
reversed shoulder arthroplasty 135
reversed shoulder replacement 134–135
component design 134
mechanics 134
rotator cuff tear arthropathy 134
ring external fixator 184–185
bone–frame construct 184
rolling 12
rotator cuff muscles 128–129
rotator cuff tear arthropathy 134–135

S-N curve 64–65

scapula 126
screw 13
see also surgical screws
screw home arc of knee 116
self-tapping screws 169
shafts 58–59
geometric properties 58
shear force 52
shoulder 126–135
functional anatomy 126–129
geometry and alignment 126–127
glenoid labrum 126
joint capsule 128
ligaments 128
rotator cuff muscles 128–129
range of motion 129
reaction forces 130–131
abduction with elbow in extension 130–131
abduction with elbow in flexion 130–131
clinical implications 130
shoulder arthroplasty 190
shoulder replacement 132–133
component alignment 133
glenoid component 132
humeral component 132
implant types 133
resurfacing 133
reversed 134–135
component design 134
mechanics 134
rotator cuff tear arthropathy 134
simple beams 55
simple machines, musculoskeletal system 14
single-leg stance 98–99
with 100N load in opposite hand 100–101
with cane support in opposite hand 100–101
skin structure 81
SLAP tear 126
sliding 12
solid–solution strengthening 28–29
spica cast 164
spinal nerves 145
spine 142–149
biomechanics 142–145
curvatures 143
functional spinal unit 145
intervertebral disc 144–145
ligaments and muscles 144
loads acting on 148–149
stability 146–147
three-column classification 146–147
two-column classification 146–147
vertebrae 142–143
spinous process of vertebra 142
spiral fractures 66–67
spring-ball model 53
squeeze film lubrication 90–91
stability 158–159
stainless steel 30–31
standard screws 169
static analysis 8
assumptions 11
musculoskeletal system 10
see also free-body force diagrams
statics 2
stiffness 18, 20, 22, 158
beams 57
bone 46
fracture 78
straight traction 162–163
strain 16
force–stress relationships 17
strain-rate-dependent mechanical properties 22
strain–time curves 23
strength 20
stress raisers 72–73
stress relaxation 22–23
stress–strain curves 18
fractures 63
properties of 19
viscoelastic materials 23
stress–time curves 23
stresses 16
compression 52
force–strain relationships 17
tension 52
torsion 58
Stribeck curve 88–89
prosthetic joints 93
subscapularis 128
subtalar joint 150–151, 156–157
supraspinatus 128–129
surface finish, hip prosthesis 102
surgical screws 168–171
biomechanics 168–169
bone–screw construct 170
characteristics 168–169
core diameter 168
fully threaded 170
functional anatomy 169
lag screw fixation 170–171
partially threaded 170–171
pitch 168
types of 169
sutures, synthetic 42
synovial fluid as a lubricant 88
synovial joints 80
lubrication 90
boosted 90–91
squeeze film 90–91
weeping 90–91
wear 84

talocalcaneal joint 150–151

talocrural joint 151
talus 150–151
taper slip stem 102
tensile load fractures 66
tensile strength 18
tension 50–53
geometric properties 50–52
stresses 52
teres minor 128
Thomas splint 162–163
three-point bending 54
tibial fractures
distal third 200
intramedullary nails 200–201
plate fixation 200–201
external fixation 181
pilon 202–203
external fixation 203
tibiofemoral joint 112
reaction forces 118–119
titanium 30–31
titanium alloys 30–31
torque 6
frictional 82
torsion 50–51, 58–59
stresses 58
torsional load fractures 66
torus fracture 68
toughness 18, 20–21
traction 160–163
balanced (Hamilton–Russell) 162–163
clinical perspective 161
Gallow's 162–163
ninety degree–ninety-degree 162–163
straight 162–163
Thomas splint 162–163
transverse force 52
transverse process of vertebra 142
trauma meeting 186–206
clinical perspective 187
mechanics 186
see also case-based discussions
tribology 80
β-tricalcium phosphate 34
triceps brachii 140

UHMWPE see ultra-high molecular weight polyethylene

ulnar fractures, external fixation 181
ultra-high molecular weight polyethylene (UHMWPE) 38–39
unloading curve 23
upper limb
elbow 136–141
fractures see fractures (specific)
free-body force diagram 11
shoulder 126–135

van der Waal's bonds 36

vertebrae 142–143
facet joints 142
spinous and transverse processes 142
vertebral body 142–143
viscoelastic materials 22–23
bone 46
polymers 36–37
stress–strain curves 23
wear 81, 84–85
joints 84
rate of 85
wear equation 84
wedges 13
weeping lubrication 90–91
wheel and axle 13
musculoskeletal system 14–15
windlass effect 156
wires 166–167
biomechanics 166
bone–wire construct 166
cerclage wires 167
k-wires 166–167
radius fractures 196–197
tension band principle 167
Wolff's law 44
work 12
work hardening 28–29

yield strength 18, 20–21

Young's modulus 18

zirconia 35