Embed Size (px)
Citation preview
"The role of the abdomen in breathing in normal subjects
and tetraplegic patients"
A thesis submitted to the University of London for the degree
of Doctor of Medicine
by
Jonathan Mervyn Goldman
Lung Function Unit, Brompton Hospital
and
National Spinal Injuries Centre, Stoke Mandeville
February 1986
1
ABSTRACT
The abdomen is an integral part of the chest wall which, from a' respiratory
viewpoint, may not have been studied in as much detail as it deserves. This
thesis examines some aspects of its role in breathing, in normal subjects and
patients with tetraplegia. Electromyographic techniques have been modified to
study the actions of the four principal abdominal muscles during respiration in
normal subjects. These experiments suggest that individual muscles act
independently in trunk motion, but in concert during breathing. Tetraplegic
patients have been examined to illustrate the effects of denervation of the
abdominal muscles on chest wall mechanics. An optical contour mapping system has
been used in conjunction with gastric pressure measurements, to define the
pressure volume characteristics of the abdominal wall in supine normal subjects
and patients with tetraplegia. A similar non-invasive method has been used to
partition total respiratory system compliance into its rib cage and abdominal
components. The abdominal wall of cervical cord injured patients is twice as
compliant as that of normal subjects. This limits expansion of the rib cage and
may have a detrimental effect on ventilation. Applying binders to stiffen the
abdominal wall of patients with tetraplegia improved their breathing both whenOthey were seated, and tilted at 70 to the horizontal, suggesting that this widely
used intervention benefits respiration as well as postural hypotension.
2
ACKNOWLEDGEMENTS
A project of this nature requires a great deal of specialist advice and assistance
and I am grateful to the following for their contributions.
Professor David Dension, for his ideas, supervision and encouragement.
Dr. Mike Morgan, who developed and built the optical contour mapping system, for
his continued advice and support.
Lone Rose, who was an able assistant in all my experiments and taught me the
basics of working with tetraplegic patients.
Dr. Steve Williams, Consultant Chest Physician, at Stoke Mandeville, for his
guidance throughout the project.
Dr. John Silver, for the opportunity to use the excellent laboratory facilities
at National Spinal Injuries Centre, and his advice on making EMG recordings in
tetraplegic patients.
Dr. Bob Lehr, for teaching me to manufacture and use fine wire electrodes.
Roy Robinson and the Medical Electronics Department, Stoke Mandeville Hospital,
for technical assistance.
Dr. David Hughes, Nigel Cox and the Computer Department, Brorapton Hospital, for
helping me to use the computer system effectively.
Cathy Evans, for typing the manuscript.
3
The Medical Art department at the Royal Marsden Hospital, for the anatomical
drawings.
Drs. Andy Bush and Ann Millar, my colleagues in the Lung Function Department at
Brompton Hospital, for their support and humour in adversity.
Most importantly to the patients of the National Spinal Injuries Centre, from whom
I have learnt a great deal, some of which may be found in this volume.
4
PUBLICATIONS
Some of the work in this thesis has been submitted for publication.
1. The measurement of abdominal wall compliance in normal subjects and
tetraplegic patients.
Goldman J.M., Rose L.S. , Morgan M.D.L., Denison D.M.
(Thorax in press).
2. An electromyographic study of the abdominal muscles of tetraplegic
patients.
Goldman J.M., Silver J.R., Lehr R.P.
(Submitted to Paraplegia).
3. The effect of abdominal binders on breathing in tetraplegic patients.
Goldman J.M., Rose L.S., Williams S.J., Silver J.R., Denison D.M.
(Submitted to Thorax).
Two further publications are intended.
1. An electromyographic study of the abdominal muscles in man during
postural and respiratory manoeuvres.
Goldman J.M., Silver J.R. , Lehr R.P.
2. The effect of loading expiration on chest wall motion in tetraplegic
patients.
Goldman J.M., Williams S.J., Denison D.M.
5
CONTENTS
Page
TITLE 1
ABSTRACT 2
ACKNOWLEDGEMENTS 3
PUBLICATIONS 5
CONTENTS 6
LIST OF FIGURES 14
LIST OF TABLES 18
ABBREVIATIONS 20
PREFACE 22
CHAPTER 1 - THE ANATOMY AND PHYSIOLOGY OF THE ABDOMEN IN NORMAL MAN
1.1 Introduction 23
1.2 A model of the chest wall 23
1.3 Anatomy of the abdomen 29
1.3.1 Diaphragm 29
1.3.2 Posterior abdominal wall 31
1.3.3 Anterior abdominal wall 32
1.3.4 Pelvic floor 34
1.4 Static properties of the abdomen 35
1.4.1 A model of the abdomen 35
1.4.2 Effects of posture and gravity 35
6
1.4.3 Pressure-volume relationships of the relaxed chest wall 37
1.4.4 Static pressure-volume characteristics of the abdominal 40
compartment
1.4.5 Summary 42
1.5 Dynamics of the abdomen 43
1.5.1 Mechanical actions of the abdominal muscles 43
1.5.1.1 Movements of the trunk 43
1.5.1.2 Expiratory actions of the abdominal muscles 44
1.5.1.3 Abdominal muscle activity in the erect posture 45
1.5.1.4 Inspiratory actions of the abdominal muscles 45
1.5.1.5 Summary 47
1.5.2 The mechanical actions of the diaphragm 48
1.5.2.1 Actions on inspiration 48
1.5.2.2 Interaction between rib cage and diaphragm 48
1.5.2.3 How does intra-abdominal pressure displace the rib cage? 49
1.5.2.4 The actions of the costal and crural parts of the diaphragm 51
1.5.2.5 Summary 51
1.6 Abdominal wall compliance 52
1.7
1.7.1
1.7.2
Effects of posture on chest wall motion
Rib cage and abdominal motion
Length tension relationship of the diaphragm
52
52
53
1.8 Actions of the intercostal and accessory muscles 54
7
CHAPTER 2 - CHEST WALL MECHANICS IN TETRAPLEGIA
2.1 Introduction 55
2.2 Pathophysiology 55
2.3 Lung function after cervical cord injury 58
2.3.1 Chronic stable tetraplegia 58
2.3.2 Effects of posture on lung function 58
2.3.3 Lung function immediately after acute injury 60
2.4 Chest wall motion in tetraplegia 60
2.4.1 Paradoxical motion of the rib cage 61
2.4.2 Effects of posture on chest wall motion 61
2.4.3 Optical contour mapping 63
2.4.4 Summary 65
2.5 Chest wall compliance in tetraplegia 66
2.5.1 Measurement of chest wall compliance 66
2.5.2 Factors influencing chest wall compliance 67
2.5.3 Factors influencing rib cage compliance 67
2.5.4 Factors influencing abdominal wall compliance 68
2.6 Mechanics of posture change 69
2.7 Abdominal binders 72
2.7.1 The effect of binders on chest wall motion 72
2.7.2 The effect of binders on trans-diaphragmatic pressure 73
2.7.3 Present clinical practice and implications 73
8
2.8 Acute changes in chest wall mechanics after cervical
cord transection
74
2.9 Cough 74
2.10 Sensation of breathing 75
2.11 Summary 76
CHAPTER 3 - METHODS AND MATERIALS
3.1 Selection of patients 77
3.2 Selection of normal subjects 78
3.3 Measurement of abdominal wall displacement 78
3.3.1 Introduction 78
3.3.2 Radiological techniques 78
3.3.3 Linear measurements 79
3.3.4 Measurements of circumference 81
3.3.5 Measurements of cross-sectional area 81
3.3.6 Three dimensional measurements 82
3.3.6.1 Partial plethysmography 82
3.3.6.2 Optical contouring 82
3.4 Methods used 85
3.4.1 The optical contour mapping system 85
3.4.1.1 Apparatus 85
9
3.A.1.2 Contour projection 85
3.4.1.3 Image capture 85
3.4.1.4 Reference planes 88
3.4.1.5 Calibration 88
3.4.1.6 Image processing 91
3.4.1.6.1 Projection 91
3.4.1.6.2 Analogue to digital conversion 91
3.4.1.6.3 Data processing 93
3.4.1.6.4 Partitioning chest wall volume 93
3.4.1.7 Assessment of the method 94
3.4.2 Application of optical contour mapping 95
3.4.2.1 The measurement of abdominal wall compliance 95
3.4.2.1.1 Introduction 95
3.4.2.1.2 Protocol 95
3.4.2.1.3 The "Relaxation Manoeuvre" 96
3.4.2.1.4 Camera trigger mechanism 101
3.4.2.1.5 Digitisation procedure 102
3.4.2.1.6 Partitioning chest wall volume 102
3.4.2.1.7 Data handling 102
3.4.2.1.8 Validation of the method 103
3.4.2.2 The effect of loading expiration on chest wall motion 106
in tetraplegic patients
3.4.2.2.1 Introduction 106
3.4.2.2.2 Apparatus 106
3.4.2.2.3 Protocol 107
10
3.4.3 Pressure monitoring techniques 109
3.4.3.1 Introduction 109
3.4.3.2 Airway pressure 109
3.4.3.3 Measurement of airflow 110
3.4.3.4 Gastric pressure 111
3.4.3.5 Oesophageal pressure 112
3.4.3.6 Trans-diaphragmatic pressure on maximal sniff 112
3.4.3.7 Maximum static inspiratory mouth pressure 113
3.4.4 Experimental methods for Chapter 7 115
(The effect of abdominal binders on breathing in
tetraplegic patients)
3.4.4.1 Study positions 115
3.4.4.2 Abdominal binders 115
3.4.4.3 Abdominal girth 117
3.4.4.4 Protocol 117
3.4.5 Electromyographic Methods 119
3.4.5.1. Introduction 119
3.4.5.2 Apparatus 119
3.4.5.2.1 Intra-muscular electrodes 119
3.4.5.2.2 Oesophageal electrode 122
3.4.5.2.3 EMG module 122
3.4.5.3 Method for recording EMG from individual abdominal muscles 123
3.4.5.3.1 Location of muscle layers 123
3.4.5.3.2 Insertion of electrodes 125
3.4.5.3.3 Validation of technique 125
3.4.5.3.4 Respiratory manoeuvres 126
3.4.5.4 Validation of relaxation manoeuvres 127
11
3.A.6 Statistical methods 127
3.5 Summary 128
CHAPTER 4 - ELECTROMYOGRAPHIC STUDIES OF THE ABDOMINAL MUSCLES
4.1 Study 1, Normal Subjects 129
4.1.1 Introduction 129
4.1.2 Results 130
4.1.2.1 Measurements of skin to muscle distance by CT scan 130
4.1.2.2 Validation of method 130
4.1.2.3 Respiratory manoeuvres 139
4.1.2.4 Effects of posture 139
4.1.3 Discussion 139
4.1.4 Summary 146
4.2 Study 2, Tetraplegic Patients 147
4.2.1 Introduction 147
4.2.2 Results 147
4.2.3 Discussion 151
4.2.4 Summary 157
CHAPTER 5 - MEASUREMENT OF ABDOMINAL WALL COMPLIANCE
5.1 Introduction 158
5.2 Results 158
5.3 Discussion 167
5.4 Summary 169
12
CHAPTER 6 ■- THE EFFECT OF LOADING EXPIRATION ON CHEST WALL MOTION
IN TETRAPLEGIC PATIENTS
6.1 Introduction 170
6.2 Results 171
6.3 Discussion 175
6.A Summary 179
CHAPTER 7 ■- THE EFFECT OF ABDOMINAL BINDERS ON BREATHING IN
TETRAPLEGIC PATIENTS
7.1 Introduction 180
7.2 Results 181
7.2.1 Abdominal girth 181
7.2.2 Sniff Pdi 181
7.2.3 Vital capacity 187
7.2.4 Plmax 193
7.2.5 Gastric pressure 193
7.3 Discussion 196
7.4 Summary 200
CHAPTER 8 •- SUMMARY AND DISCUSSION 201
APPENDICES
1. Data for chapter 5 208
2. Data for chapter 6 214
3. Data for chapter 7 217
BIBLIOGRAPHY 220
13
LIST OF FIGURES
CHAPTER ONE
Page
CHAPTER ONE
1.1 A simple mechanical model 24
1.2-1.5 A simple mechanical model of the chest wall 25
1.6 Anatomy of the diaphragm 30
1.7 Anatomy of the anterior abdominal wall 33
1.8 A model of the abdomen 36
1.9 The effect of gravity on abdominal configuration 38
1.10 Abdominal and thoracic pressures at different lung volumes 39
1.11 Static compliance curves of the rib cage and abdomen 41
1.12 Interaction between the diaphragm and rib cage 50
CHAPTER TWO
2.1 Segmental innervation of the respiratory muscles 57
2.2 Rib cage mechanics in C6 tetraplegia 62
2.3 Rib cage mechanics in Cl tetraplegia 64
CHAPTER THREE
3.1 Optical contour mapping system for use in standing subjects 84
3.2 The optical contour mapping system used to study supine 86
patients
3.3 The optical contour mapping system in use 87
3.4 Focusing planes 89
3.5 Chest wall quadrants for digitisation 92
14
3.6 Gastric pressure trace during a "relaxation manoeuvre" 97
3.7 Gastric pressure trace during a "relaxation manoeuvre" 98
with superimposed abdominal muscle activity
3.8 The chest wall of a normal subject at FRC and TLC 99
3.9 The chest wall of a tetraplegic patient at FRC and TLC 100
3.10 A sequence of photographs taken during a passive expiration 102a
3.11 Reproducibility of digitisation of a "relaxation manoeuvre" 104
3.12 Reproducibility of the "relaxation manoeuvre" 105
3.13 Apparatus for loading expiration 107a
3.14 Trace of ventilation and airway pressure during loaded 108a
breathing
3.15 The abdominal binders 116
3.16 Bipolar fine wire electrode and EMG module 120
3.17 Detail of the fine wire electrode ,121
3.18 CT scan of the abdominal muscles 124
CHAPTER POUR
4.1 EM? trace from external oblique 134
4.2 EMG trace from internal oblique 135
4.3 EMG trace from transversus abdominis 136
4.4 EMG trace from rectus abdominis 137
4.5 Ultrasound scan of the abdominal muscles 138
4.6 Normal abdominal nuscle expiratory EMG activity 140
4.7 Expiratory EMG activity in the 4 seperate abdominal muscles 141
4.8 Normal abdominal muscle late inspiratory EMG activity 142
4.9 Abdominal muscle EMG activity during cough 143
4.10 Tonic abdominal muscle EMG activity in a tetraplegic patient 149
15
4.11 Phasic inspiratory abdominal muscle EMG activity in a 150
tetraplegic patient
4.12 Appearance of phasic inspiratory abdominal muscle EMG 152
activity as minute volume increases
4.13 Phasic inspiratory abdominal muscle EMG activity developing 153
into tonic activity
4.14 Tonic EMG activity in the abdominal muscles, normal phasic 154
activity in the diaphragm
4.15 Tonic EMG activity in the diaphragm, no EMG activity in the 155
abdominal muscles
CHAPTER FIVE
5.1 Gastric pressure/volume-displacement characteristic of the 161
normal abdominal wall
5.2 Gastric pressure/volume-displacement characteristic of the 162
abdominal wall
a) 6 normal subjects
b) 6 tetraplegic patients
5.3 Gastric pressure/volume-displacement characteristic of the 163
abdominal wall. Displacement expressed as % volume expired
a) 6 normal subjects
b) 6 tetraplegic patients
5.4 Comparison of mean data from gastric pressure/volume- 165
displacement curves of normal subjects and tetraplegic patients
16
CHAPTER SIX
6.1 Specific compliance curves of the respiratory system of 8 173
tetraplegic patients
CHAPTER SEVEN
7.1 The effect of posture on Sniff Pdi 186O
7.2 The effect of abdominal binders on Sniff Pdi at 70 tilt 188
7.3 The effect of posture on VC 190
7.4 The effect of abdominal binders on VC in the seated posture 191O
7.5 The effect of abdominal binders on VC at 70 tilt 192
17
LIST OF TABLES
CHAPTER 2 Page
2.1 Lung volume measurements made in tetraplegic patients 59
2.2 Trans-diaphragmatic pressure measurements made in 70
tetraplegic patients
2.3 Maximum static inspiratory mouth pressure measurements 71
made in tetraplegic patients
CHAPTER 3
3.1 Manoeuvres performed to activate individual abdominal 126
muscles
CHAPTER 4
4.1 Measurements of the distance from skin surface to 131
individual abdominal muscles in 20 normal subjects
4.2 Mean data for skin to muscle measurements 132
4.3 Results of validation procedure for positioning electrodes 133
within individual abdominal muscles
4.4 Minute volume at which abdominal muscle EMG activity first 144
became apparent in different postures
4.5 Tetraplegic patients in the EMG study 148
18
CHAPTER 5
5.1 Normal subjects in the abdominal wall compliance study 159
5.2 Tetraplegic patients in the abdominal wall compliance study 159
5.3 Analysis of pressure displaced-volume curves over the range 166
of tidal breathing
CHAPTER 6
6.1 Tetraplegic patients in the loaded breathing study 172
6.2 Specific compliance data for tetraplegic patients 174
6.3 Specific compliance data for normal subjects 178
CHAPTER 7
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
Tetraplegic patients in the abdominal binder study
Girth measurements in tetraplegic patients with and without
abdominal binders
Girth measurements in normal subjects
Mean trans-diaphragmatic pressure measurements
Mean vital capacity measurements
Mean maximum static inspiratory mouth pressure measurements
Mean gastric pressure measurements
Comparison of lung function data with previously published
results
182
183
184
185
189
194
195
197
19
ABBREVIATIONS USED
AP = Anterior to Posterior
EO = External Oblique Muscle
10 = Internal Oblique Muscle
TA = Transversus Abdominis Muscle
RA = Rectus Abdominis Muscle
DIA = Diaphragm
OCM = Optical Contour Mapping System
LED = Light Emitting Diode
AVabd = Abdominal Wall Displacement above that at FRC
A V a b d %VE = Abdominal Wall Displacement Expressed as a %
of the Total Volume Expired
Cabd = Abdominal Wall Compliance
Crc = Rib Cage Compliance
SCrs = Specific Compliance of the Total Respiratory System
SCabd = Specific Compliance of the Abdominal Compartment
SCrc = Specific Compliance of the Rib Cage Compartment
TLC = Total Lung Capacity
FRC = Functional Residual Capacity
RV = Residual Volume
VC = Vital Capacity
VE = Volume Expired During A Stated Manoeuvre
ETR = Expiratory Threshold Resistance
Pa = Airway Pressure
Pb = Barometric Pressure
Ppl = Pleural Pressure
Pab = Intra-Abdominal Pressure
Pg = Gastric Pressure
20
R Pg = Resting Gastric Pressure
I Pg = End Tidal Inspiratory Gastric Pressure
S Pg = Gastric Pressure at Maximal Sniff
Poes = Oesophageal Pressure
Pdi = Trans-diaphragmatic Pressure
Sniff Pdi = Trans-diaphragmatic Pressure on Maximal Sniff
Plmax = Maximum Static Inspiratory Mouth Pressure
PImax Pdi = Trans-Diaphragmatic Pressure on Maximum Static Inspiratory
Effort
EMG = Electromyography
V = Ventilation
CT = Computerised Axial Tomography
X = Mean
SD = Standard Deviation
SEM = Standard Error of the Mean
SED = Standard Error of the Difference
SEE - Standard Error of the Estimate
All predicted values are taken from Cotes (1979)
21
PREFACE
At the beginning of the work carried out for this thesis, the objective was to
gain a greater understanding of the role of the abdomen in breathing. To
illustrate the importance of a normally innervated abdominal wall the plan was to
make measurements of abdominal wall compliance in normal subjects and compare and
contrast them with similar measurements from tetraplegic patients. At the same
time studies were begun to examine the EMG activity of the individual abdominal
muscles of normal subjects, and it was hoped that this method might be applied to
cervical cord injured patients.
The results of these experiments, and the experience of working with patients on a
spinal injuries unit, stimulated further work in this field. Initially this
involved looking at a non-invasive method to assess chest wall compliance and its
components in patients with tetraplegia. The natural progression was to examine a
therapeutic intervention designed to alter abdominal wall compliance in
tetraplegic patients, as described in the abdominal binder study.
This thesis is the result of original research carried out by the author at the
National Spinal Injuries Centre, Stoke Mandeville, and the Lung Function Unit,
Brompton Hospital, London, SW3.
22
CHAPTER 1
THE ANATOMY AND PHYSIOLOGY OF THE ABDOMEN IN NORMAL MAN
1.1, Introduction
This chapter will review present knowledge of the anatomy and physiology of the
abdomen in relation to breathing. It will first present a simple model of the
abdomen within the context of the chest wall, before defining its anatomy in
detail. It will then describe the static properties of the abdominal compartment,
followed by its dynamic properties with emphasis on the mechanical actions of the
abdominal muscles and diaphragm. The second chapter will describe the role of the
abdomen in breathing in patients with tetraplegia.
1.2, A model of the chest wall
Mechanical systems like that shown in Figure 1.1, can be analysed by asking such
questions as, what are the fulcra, what are the levers, what are the prime movers,
and what are the objects to be moved? When this simple approach is applied to
breathing, it becomes clear that the abdomen has an important role, which is not
always given the attention it deserves.
The real objects to be moved during breathing are the molecules of gas in the air,
since they are invisible and fluid their movement is not easily analysed. In
practice the objects "visibly" moved by the chest wall are the visceral pleural
surfaces of the lung (Figure 1.2). They are linked via pleural fluid and the
parietal pleura to the diaphragm and rib cage (Figure 1.3). This linkage acts as
a complex lever, allowing the lung to slide on the diaphragm and rib cage,
23
fulcrum
Figure 1.1A simple mechanical model, to illustrate the concept of a mover, a lever, a fulcrum and an object to be moved.
24
Area of apposition
Figure 1.2
Figure 1.3
Figure 1.4
Figure 1.5
The visceral pleura are the ’’objects to be moved" by the chest wall during breathing.The pleural fluid and the parietal pleura act as a complex lever linking the "objects to be moved" to the ribs and diaphragm.When the diaphragm descends, the spine and pelvis act as fulcra, and the anterior abdominal wall as a partial fulcrum.Resistance to descent of the diaphragm causes a rise in intra-abdominal pressure.When intra-abdominal pressure rises, it acts on the "area of apposition" to expand the lower rib cage.
25
accommodating regional differences in expansion, and applying the expanding force
evenly over a broad area. These features reduce stress concentration, which is
important, since easily expanded structures have a low "works of fracture" and
tear easily (Gordon 1978).
The medial surfaces of the lung slide over the mediastinum, as the heart and
oesophagus move and the spine flexes. Though anchored to the mediastinum by their
hila, they may expand in their own plane on breathing. However, in practical
terms, they may be considered as stationary, and in that sense the mediastinum
acts as a fulcrum for lung expansion.
The phrenic surface of the lung moves caudally on inspiration and its excursion
accounts for 75% of tidal volume, and 60% of vital capacity in the supine posture
(Vellody et al 1975, Morgan et al 1985). In the upright posture, its movement
accounts for 25% of tidal breathing but still 60% of vital capacity (Vellody et al
1975, Wade et al 1954). Underneath the diaphragm lie the incompressible but fluid
abdominal contents which act as a lever, passively transmitting pressure. The
abdominal cavity is bounded by the rigid pelvic bowl, and almost rigid spine and
posterior abdominal wall, which act as fulcra, and the flexible antero-lateral
abdominal wall, which acts as a partial fulcrum (Figure 1.4). For practical
purposes any caudal movement of the diaphragm must be accompanied by an excursion
of equal volume of the abdominal wall. It is here that the partial fulcrum
activity of the wall is important, as it will resist the descent of the diaphragm.
The degree of displacement of the abdominal wall depends not only on the mover
activity of the diaphragm, but also on its own compliance. Mover activity of the
diaphragm is only valuable in breathing, if it is resisted by the rib cage, which
in that sense also acts as a fulcrum.
26
Movement of the costal surfaces of the lung accounts for some 25% of quiet
breathing and 35% of the vital capacity in supine people, and 75% of quiet
breathing and 35% of vital capacity in upright people. This movement is only
valuable if it is opposed by the diaphragmatic surface. This allows the
generation of a negative pressure. The interaction between the diaphragm and
abdomen in assisting expansion of the rib cage is complex, and needs to be
mentioned in some detail.
At the end of a normal expiration, a substantial area of diaphragm, around the
costal margins, lies in direct contact with the rib cage. There is no intervening
lung. This area of apposition is exposed to the.radially directed hydrostatic
pressure of the abdomen (Figure 1.5). When abdominal pressure rises, from
whatever cause, it will tend to displace the rib cage upwards and outwards. Since
the ribs are shaped and anchored like bucket handles, there can be no purely
outward motion, i.e. there will always be some upward component. If a simple
mechanical model is applied to this system, it becomes clearer. The movers are,
the abdominal muscles on expiration, and the diaphragm on inspiration. The
fulcrum is the pelvis and posterior abdominal wall, and in addition on
inspiration, the abdominal wall acts as a partial fulcrum. There are a series of
levers, namely the abdominal contents, the area of apposition, the rib cage (which
also acts as a pivot), and the pleura, which transmit the force of the mover to
expand the lung. On expiration the abdominal muscles may contract and thus have
mover activity; in this circumstance their insertions into the lower ribs will
oppose rib cage expansion.
When the diaphragm contracts on inspiration, it may follow any course between two
extremes. At one extreme, its caudal movement is not resisted by the abdomen at
all, the rise in abdominal pressure is negligible, and the action of the diaphragm
expanding the rib cage is slight. At the other extreme, caudal movement of the
27
diaphragm is opposed completely by the abdomen, there is a large rise in abdominal
pressure, pushing the ribs upwards and outwards. This motion is assisted by a
direct cephalad pull on the costal margin, by the diaphragm using abdominal
pressure as a fulcrum and the liver and spleen as pulleys. Here, the fibres of
the costal diaphragm, which run cephalad, and are opposed to the rib cage, are the
movers which develop this so called "insertional" force.
These inital observations suggest the following:-
1) If the antero-lateral abdominal wall was absolutely rigid, no useful
motion of the diaphragm or costal margin could occur, and breathing
would be confined to the upper rib cage, moved by the intercostal and
accessory muscles..
2) At the end of a breath in, when the inspiratory muscles relax, the
diaphragm is restored to its resting end-expiratory position by elastic
recoil of the lung assisted in the supine posture, by the passive
hydrostatic pressure of the abdominal contents.
3) Any movement of the diaphragm cephalad of its resting end-expiratory
position can only occur by raising abdominal pressure. The only means
of doing this without contracting the diaphragm is by contracting
the abdominal muscles, flexing the lumbar spine, or applying external
pressure to the antero-lateral abdominal wall.
Since the abdomen is a major agent of both expiration and inspiration, its
properties form an important part of the mechanics of breathing. This thesis
concerns some aspects of this topic, in particular the way breathing is altered by
paralysis of the abdominal wall in tetraplegia.
28
1.3, Anatomy of the Abdomen
The abdomen is that part of the trunk between the diaphragm and the pelvis. Its
morphology will be divided into diaphragm, posterior abdominal wall, anterior
abdominal wall, and pelvic floor.
1.3.1, Diaphragm (Figure 1.6)
The diaphragm forms the superior boundary of the abdominal
cavity. It consists of a helmet shaped sheet of muscle with a central
tendon. Its peripheral attachments can be divided into 3 parts.
a) Sternal Attachments
These consist of 2 strips of muscle that arise from the back of
the xiphoid process.
b) Costal Attachments
These arise from the inner surfaces of the lower 6 costal
cartilages and interdigitate with the transversus abdominis muscle.
The costal fibres run cranially and at functional residual
capacity are apposed to the rib cage over 30% of its surface area
(Mead 1979).
c) Vertebral Attachments
Two crura arise from the medial and lateral arcuate ligaments
on each side. The right crus attaches to the bodies of the
first 3 lumbar vertebrae, and the left to the first two. These
fibres do not insert directly into the rib cage.
29
Sternal origin
Lumbocostalarch
Costal origin
Lateralarcuateligament
Vena caval -foramen
-Oesophagealhiatus
-Aortic hiatus
Right crus Left crus
Figure 1.6The anatomy of the diaphragm.
30
Muscle fibres running from all 3 peripheral attachments insert
into a trilobed central tendon.
There are openings in the diaphragm for the inferior vena cava,
oesophagus and aorta. The right and left halves of the
diaphragm are innervated by the phrenic nerves which arise from
the 3rd, 4th and 5th cervical segments of the spinal cord.
1.3.2, Posterior abdominal wall
This consists of 5 lumbar vertebrae, the psoas major and quadratus
lumborura muscles, and the thoraco-lumbar fascia.
Psoas Major arises from the lumbar vertebrae, and inserts into the
lesser trochanter of the femur. Quadratus Lumborura arises from the
inferior border of the 12th rib and inserts medially into the •
transverse processes of the lumbar vertebrae and inferiorly into the
ilio-lumbar ligament and the iliac crest. These muscles are enclosed
by the thoraco-lumbar fascia which consists of 3 layers, and is
continuous with the muscles of the anterior abdominal wall. The
posterior abdominal wall is a rigid structure, which alters little with
breathing.
31
1.3.3, Anterior abdominal wall (Figure 1.7)
The lateral aspect of the anterior abdominal wall is formed by 3 sheet
like muscles in separate layers. They are the external oblique outermost,
the internal oblique in the centre, and the transversus abdominis
innermost. Medially, these muscles become aponeurotic and fuse to
fora a sheath around the strap like rectus abdominis muscle.
a) Rectus abdominis
These muscles lie either side of the midline, and consist of
vertically running fibres; they insert superiorly into the
anterior surface of the 5th, 6th and 7th costal cartilages, and
the xiphoid process. Inferiorly, they attach to the pubic crest
and the front of the pubic symphisis. They are innervated by the
lower 5 intercostal nerves (T7-T11).
b) External Oblique
This muscle originates superiorly from the external surface of the
lower 8 ribs, superficial to the intercostal muscles. Its fibres
run anteriorly and inferiorly to insert into the iliac crest.
Anteriorly, the muscle becomes aponeurotic and runs in front of
rectus abdominis attaching to the xiphoid process, linea alba and
body of the pubis. Its nerve supply is from the lower 5
intercostal nerves (T7-T11).
c) Internal oblique
Arises from the thoraco-lumbar fascia, the anterior two thirds
of the iliac crest and the lateral two thirds of the inguinal
ligament. Fibres run upwards and medially inserting into the
32
External oblique Internal oblique
Transversus abdominis Rectus abdominis
Figure 1.7The anatomy of the muscles of the anterior abdominal wall.
33
antero-lateral surface of the lower 3 costal cartilages
Medially an aponeurosis is formed which diverges to enclose
the rectus abdominis muscle and inserts into the linea alba.
It is innervated by the lower 5 intercostal nerves (T7-T11),
the subcostal (T12), the iliohypogastric (LI) and ilioinguinal
(LI) nerves.
d) Transversus abdominis
1 Arises from the inferior surface of the lower six costal cartilages
the thoraco-lumbar fascia, the iliac crest and the inguinal ligament.
Fibres run transversely, and the muscle becomes aponeurotic lateral
to rectus abdominis, and with the internal oblique aponeurosis
reinforces the posterior wall of the rectus sheath. Transversus
abdominis shares innervation with internal oblique.
Deep to transversus abdominis lies the transversalis fascia and
the peritoneum.
1.3.4, Pelvic Floor
The inferior border of the abdomen consists of the boney pelvis.
The pelvic floor separates the pelvic cavity from the perineum and the
ischio-rectal fossae. It is formed by the levator ani muscles
anteriorly and the coccygei muscles posteriorly. For practical
purposes, this border is fixed during breathing.
34
1.4, Static properties of the abdomen
This section describes the pressure volume relationships of the abdomen. A model
of the abdomen will be described and the distribution of pressure within it in
relation to posture. This information will be used to interpret the pressure
volume relationships of the respiratory system in different postures and the
pressure volume relationship of the abdominal compartment.
1.4.1, A model of the abdomen (Figure 1.8)
The abdomen may be likened to a container full of fluid (Duomarco and Rimini
1947), with the abdominal viscera floating within that fluid. The difference
between the pressure at any 2 levels is given by the difference in their height
times the specific gravity of the fluid. The system is closed and its contents
are assumed to be incompressible. (There is approximately 200 ml of gas within
the abdomen but this is a small amount in relation to the total volume of the
system. The pressures generated during breathing are insufficient to appreciably
alter the volume of this gas.)
There are 2 mobile boundaries to the system, the diaphragm and the antero-lateral
abdominal wall. Since the volume of the system is constant when one part moves in
the other must move out. For example, when the diaphragm descends the abdominal
wall is displaced outwards.
1.4.2, Effects of posture and gravity
In the upright posture with the abdominal muscles relaxed, the hydrostatic
pressure in the lower part of the abdominal container is positive. This pushes
the abdominal wall into a convex shape. By contrast, the pressure in the upper
35
Figure 1.8A simple mechanical model of the abdomen, to illustrate the mobile abdominal wall and diaphragm, and the immobile spine and pelvis.
As the abdominal contents are incompressible if the abdominal wall moves inwards the diaphragm must move upwards.
36
part of the abdomen will be negative, sucking the abdominal wall into a concave
shape (Figure 1.9). Between these two areas is the zero pressure point which
coincides with the point of inflection of the abdominal wall. Duomarco and
Rimini"s data show this point to be 3-4 cm below the diaphragm, in upright man at
resting lung volume. The position of the zero point depends on the elastic forces
of the abdominal wall, rib cage, diaphragm, and lung, as well as the effects of
gravity on the abdominal contents. In the supine position, the zero pressure
point is at the anterior abdominal wall, prone it is at the posterior abdominal
wall and in the lateral decubitus position it is between these two locations.
As a consequence of this variability in intra-abdominal pressure with posture and
also with lung volume, the volume pressure relationships of the respiratory system
also vary with position.
1.4.3, Pressure-volume relationships of the relaxed chest wall
In 1960, Agostoni and Rahn measured oesophageal and gastric pressures in normal
subjects relaxing against a closed shutter at different lung volumes. They
assumed that with the diaphragm relaxed, pressure above and below it would be the
same. They found in the seated erect posture, that gastric pressure was 11 cm H20
greater than intra-thoracic pressure. They suggested that 8.5 cm H20 of this was
due to hydrostatic pressure and 2.5 cm H2O was due to gastric tonus. They
subsequently applied a correction factor of 11 cm H20 to gastric pressure
recordings to convert to intra-abdominal pressure.
In the erect posture, if lung volume is plotted against gastric and thoracic
pressure, with the subject relaxing against a shutter (Figure 1.10), the two
curves run parallel until 20% below resting lung volume. In the supine posture
such curves are close at high lung volume, but diverge considerably at low lung
37
0000
p». ...Pa
Figure 1.9 -In liquid filled containers with a flexible wall, the atmospheric pressure level or zero pressure point (Pg) varies in position. The point of inflection indicates the level of Pg, and the flexible wall will be concave above it and convex below it.The abdomen behaves in a similar manner (Agostoni and Mead 1964).
E R E C T SU PINE
% V C % V C
Figure 1.10Lung volume plotted against gastric and pleural pressure during airway , in the erect and supine posture r vol resting volume, similar in both postures but the gastric pressure curve alters
cm
relaxation against a closed The pleural pressure curve is (Agostoni and Rahn 1960).
volume. Agostoni and Mead (1964) stated that the compliance of the chest wall was
increased in the supine position, and since the intra-thoracic pressure volume
curve illustrated alters little with posture, a reasonable assumption was that it
was the compliance of the abdominal compartment which altered.
1.4.4, Static pressure-volume characteristics of the abdominal compartment
In order to study the abdominal compartment separately from the remainder of the
chest wall, it is necessary to measure rib cage and abdominal wall displacements
at different lung volumes. In 1954, Wade, using a radiographic technique, stated
that in the upright position the rib cage contributed 31% to vital capacity and
the diaphragm/abdomen 69%. This method could not however, be applied to more
detailed studies. In 1968(a) Konno and Mead described a method for the separate
estimation of rib cage and abdominal volume displacement, by extrapolating from
the motion of single points on the chest wall (See 3.3). This allowed them to
construct the separate pressure volume curves of each compartment for the first
time (Konno and Mead 1968b). They did this in normal subjects relaxing against a
closed shutter at different lung volumes, in the supine and standing positions
(Figure 1.11). The gradients of these plots represent the compliance of the rib
cage and the abdominal compartments. The rib cage curves are very similar in the
two positions whilst the abdominal plots are different. In the supine position,
abdominal compliance is greater than in the standing position. In addition, it is
notable that the gradients of the abdominal plots in both postures decrease at
high lung volumes, when the abdominal wall is stretched. The question raised is,
what is the effect of change in abdominal compliance during breathing? If one
considers the the simple mechanical model of the chest wall (See 1.2), it is clear
that the degree of partial fulcrum activity of the abdominal wall, i.e. its
compliance, changes during the respiratory cycle. This in turn will effect the
interaction between the diaphragm and the rib cage. This interaction will be
40
E R E C TSUPINE
V rc Vrc
P (cm H O) P (cm H^O)Figure 1.11Separate static pressure volume characteristics of the rib cage and abdomen, in the erect and supine postures. The open circles represent the abdomen, and the closed circles the rib cage. Vab = Abdominal volume expressed as a percentage or the vital capacity. Vrc = Rib cage volume expressed as a percentage of vital capacity . Open circles plot Vab against trans-abdominal pressure. Closed circles plot Vrc against trans-thoracic pressure. The slope of the curve represents the compliance of the compartment.
discussed in the next section of this chapter and further examined throughout this
thesis.
1.4,5, Summary
1) The abdomen is a closed container with mobile and immobile parts.
2) The abdominal contents are incompressible.
3) The pressure volume characteristics of the abdominal compartment vary
with posture and breathing.
4) Abdominal compliance is greater supine than erect.
42
1.5, Dynamics of the abdomen
In this section, the mechanical actions of the abdominal muscles and the diaphragm
will be considered. This information will be used to discuss chest wall motion in
the supine and the erect posture.
1.5.1, Mechanical actions of the abdominal muscles
These will be divided into movement of the trunk, expiratory actions, activity
related to posture, and inspiratory actions.
1.5.1.1, Movements of the trunk
The actions of the individual abdominal muscles can be implied from their origins
and insertions. In 1950, Floyd and Silver noted electromyographic (EMG) activity
predominantly from rectus abdominis on raising the head in the supine posture, and
predominantly from external oblique on straining. The fibres of rectus abdominis
run from xiphisternum to pubis and its actions are to pull the rib cage towards
the pelvis, and to a lesser extent compress the abdominal contents (Campbell et al
1970). Compression of the abdominal contents is achieved to a greater extent by
the three muscles of the antero-lateral group whose fibres run obliquely or
transversely around the abdominal cavity. In addition to this action, they may
also flex the trunk. External oblique was noted by Campbell in 1952, to display
EMG activity on lateral trunk flexion. More detailed analysis by Carman et al
(1972), confirmed the role of external oblique on ipsi-lateral trunk flexion and
internal oblique on contra-lateral trunk flexion. Implicit in these movements is
a depressing action on the rib cage. The fibres of transversus abdominis run
circumferentially around the abdomen. It is difficult to study as it is the
innermost of the 3 muscle layers. Strohl et al (1981) placed 3 fine wire
43
electrodes into the antero-lateral abdominal muscles, at depths corresponding to
ultrasound measurements of the distance from the skin to each individual muscle.
They demonstrated specific EMG activity from the deepest electrode on voluntarily
"pulling the belly in". This suggests that this movement was carried out by
transversus abdominis.
The 4 muscles of the anterior abdominal wall have separate identifiable actions on
movement of the rib cage in relation to the pelvis. The following paragraphs will
describe the actions of the abdominal muscles in relation to breathing. There is
little information on the actions of the individual abdominal muscles during
respiration. This is an area which merits further study and will be examined in
chapters 3 and 4 of this thesis.
1.5.1.2, Expiratory actions of the abdominal muscles
During quiet breathing, there is no detectable EMG activity from the abdominal
muscles (Floyd and Silver 1950). When ventilation increases, the abdominal
muscles become active late in expiration (Campbell and Green 1953a). This implies
that passive recoil of the rib cage is adequate for expiration until the system is
stressed. In most subjects this is at a level of ventilation of 40L/minute
(Campbell and Green 1955). The mode of action of the abdominal muscles was
demonstrated by Campbell and Green (1953b), who remarked on the progressive
increase in gastric pressure with increasing ventilation and abdominal muscle EMG
activity. The abdominal muscles compress the abdominal contents and increase
intra-abdominal pressure, they pull the rib cage downwards and push the diaphragm
into the thorax, forcing expiration. The abdominal muscles are active during any
forced expulsive act such as coughing, sneezing, vomiting or defaecating, and also
on static expiratory effort (Campbell and Green 1953b) and expiration against a
resistance (Campbell 1957). In the latter case they become active at lower levels
44
of ventilation, if a subject is breathing against an expiratory threshold
resistance of 15cm H20 or more, than if they are breathing at rest.
1.5.1.3, Abdominal muscle activity in the erect posture
When people stand up, EMG activity in their abdominal muscles is increased
(Campbell and Green 1955). This activity is more pronounced in the lower
abdominal muscles (Strohl et al 1981). De Troyer (1983a) demonstrated that in
resting subjects at 45 tilt, there was EMG activity over the lower but not the
upper abdominal muscles. In the erect posture, EMG activity could be detected
over the upper abdominal muscles but was of smaller amplitude compared to the
lower muscles. In this study De Troyer (1983a) used magnetometers to demonstrate
that in erect subjects, when the abdominal muscles were active, abdominal AP
diameter was decreased. This opposes any decrease in intra-abdominal pressure due
to gravity, and acts to maintain the configuration of the diaphragm (See 1.7.).
During increased ventilation in the erect posture, expiratory activity of the
abdominal muscles begins at lower rates of ventilation than in the supine posture
(Campbell and Green 1955). The expiratory actions of the abdominal muscles thus
oppose the inspiratory actions of gravity.
1.5.1.4, Inspiratory actions of the abdominal muscles
The expiratory actions of the abdominal muscles have already been described (See
1.5.1.2) It would seem to be contradictory to suggest that these muscles also have
inspiratory actions. However, within a complex system it is possible for a muscle
to have more than one action, and its overall effect will depend on the relative
force of each action in a given situation. An indirect way in which the abdominal
muscles can facilitate inspiration, is by providing an effective expiration such
that end expiratory volume is decreased, and inspiration is then aided by the
45
passive recoil of the chest wall. A more direct inspiratory action is mediated by
increased intra-abdominal pressure. Intra-abdominal pressure inflates the lower
rib cage by a direct effect on the area of apposition of the rib cage and
diaphragm, and by acting as a fulcrum for the insertional forces of the diaphragm
(See 1.5.2). It also displaces the diaphragm cephalad and improves its length and
tension relationship (Braun et al 1982). Grimby et al (1976) suggested that the
abdominal muscles may help inflate the rib cage, by increasing intra-abdominal
pressure as the diaphragm descends. In fact there is EMG evidence of abdominal
muscle activity during late inspiration, at increased levels of ventilation
(Campbell 1952). Grimby's group interpret this finding as the abdominal muscles
working to improve conditions for diaphragmatic contraction, when the system is
stressed.
De Troyer et al (1983) have studied the balance between the inspiratory and
expiratory actions of individual abdominal muscles in the dog. They found that
stimulating rectus abdominis decreased the AP and lateral diameters of the rib
cage, and increased intra-abdominal presssure; stimulating external oblique
increased both AP and lateral rib cage diameters, and intra-abdominal pressure.
On eviscerating the animals, both muscles deflated the rib cage. This suggested
that the intra-abdominal pressure produced by abdominal muscle contraction had
inflated the rib cage. In man, Mier et al (1985) demonstrated that stimulation of
rectus abdominis decreased the AP diameter of the rib cage, whilst stimulation of
external oblique decreased the lateral diameter. Both muscles increased gastric
and trans-diaphragmatic pressure. Thus the abdominal muscles may have both
inspiratory and expiratory actions, and it is the timing of their contraction in
the respiratory cycle, and the actions of other chest wall muscles, that will
determine the overall effect on the system.
46
1.5.1.5, Summary
The abdominal muscles
1) Move the trunk
2) Act in forced expiration
3) Maintain the diaphragm's position in the erect posture.
4) Interact with other inspiratory muscles when ventilation is increased.
47
1,5.2, The mechanical actions of the diaphragm
1.5.2.1, Actions on inspiration
The diaphragm may be likened to an eliptical cylindroid capped by a dome (De
Troyer 1983b). The cylindrical part is muscular and is directly apposed to the
rib cage. The dome consists of the central tendon. When the muscular part
contracts the central tendon descends in relation to its costal insertions. This
piston like action creates a negative intra-thoracic pressure, which sucks air
into the lungs. It also sucks in the walls of the rib cage and thus has an
expiratory effect. As the diaphragm descends it pushes the abdominal contents
downwards, raising intra-abdominal pressure, which in turn expands and lifts the
lower rib cage.
1.5.2.2, Interaction between rib cage and diaphragm
Galen studied the effects of isolated contraction of the diaphragm on the rib
cage, and observed that the rib cage was expanded (Singer 1976). Magendie
redescribed this finding and noted perceptively that "in general the extent of
elevation (rib cage) will be in direct proportion to the resistance of the
abdominal viscera and the mobility of the ribs" (Duchenne 1867). Duchenne
himself, noted that the rib cage expanding effect of the diaphragm was abolished
in eviscerated animals, and that under these circumstances the diaphragm flattened
and pulled the rib cage inwards. He concluded that it was the resistance to
diaphragmatic descent that was important in the outward displacement of the rib
cage, and that the shape of the abdominal viscera maintained the curvature of the
diaphragm and allowed it to act on the ribs in a vertical direction. From this
work, the concept of intra-abdominal pressure driving the rib cage was developed.
Goldman and Mead (1973) took this idea further. They suggested that if the
48
intercostal and accessory muscles were relaxed, the diaphragm drove the rib cage
along its relaxation characteristic. They assumed that the sum of the inspiratory
and expiratory actions of the diaphragm on the rib cage would give the pressure
across it (Prc). The inspiratory actions were defined as trans-diaphragmatic
pressure; that is abdominal pressure minus pleural pressure (Pab - Ppl), and the
expiratory actions as pleural pressure minus barometric pressure (Ppl - Pb).
Thus Prc = (Pab - Ppl) + (Ppl - Pb)
i.e. Prc = Pab - Pb.
They concluded that in the upright posture, Pab is the effective pressure
displacing the rib cage. They strengthened their argument by demonstrating that
on compression of the abdomen, rib cage volume was driven along the same curve by
Pab as without compression, whilst the plot of Ppl against rib cage volume was
substantially displaced.
1.5.2.3, How does intra-abdominal pressure displace the rib cage?
The first mechanism is through the area of apposition of diaphragm to rib cage
(Figure 1.12). Mead (1979) noted that at residual volume, the area of apposition
of the diaphragm covered half of the surface area of the rib cage, whilst at total
lung capacity, this decreases to zero. Intra-abdominal pressure acts directly
through this area to expand the rib cage. It follows that this is most effective
at low lung volumes. At high lung volume when the diaphragm is flat, further
contraction has an expiratory effect, pulling the rib cage inwards.
Mead (1979) also stated that diaphragmatic tension operates on the rib cage
through its insertions in a direction parallel to the internal surface of the rib
cage. This is the basis of the second mechanism by which the diaphragm expands
the rib cage, the so called "insertional" force (Figure 1.13) This force depends
on resistance to the descent of the diaphragm, which is provided by the abdominal
49
CnO
Figure 1.12The forces acting on the rib cage when the diaphragm contracts on inspiration (De Troyer 1983b)
contents and intra-abdominal pressure, allowing the costal insertions of the
diaphragm to lift the rib cage.
1.5.2.4, The actions of the costal and crural parts of the diaphragm
De Troyer et al (1981) described the effect on the rib cage of stimulating the
costal and crural parts of the diaphragm separately in dogs. Direct comparison
between man and animal species is not valid in respiratory mechanics, because of
differences in anatomy and physiology necessary to maintain the upright posture.
However, animal work frequently gives rise to theories which may subsequently be
proven in man. De Troyer noted that the costal part of the diaphragm had an
inflating effect on the lower rib cage even in the absence of a significant rise
in intra-abdominal pressure. By contrast, the crural diaphragm had a deflating
effect under the same circumstances. He concluded that the diaphragm consisted of
2 muscles with different actions, there is additional evidence of separate
embryological origins, fibre composition and nerve root innervation.
1.5.2.5, Summary
When the diaphragm descends it develops three forces which act on the rib cage.
1) An expiratory force via negative pleural pressure.
2) An "appositional" force which is inspiratory, mediated by the effect
of intra-abdominal pressure on the area of apposition of rib cage to
diaphragm.
3) An "insertional" force which is inspiratory, mediated by the effect of
intra-abdominal presssure acting as a fulcrum upon which the
diaphragm can contract to lift the rib cage.
51
1.6, Abdominal wall compliance
The importance of intra-abdominal pressure has been stressed. What determines the
pressure generated when the diaphragm descends? The answer is the compliance of
the only other mobile part of the abdominal cavity, in other words the anterior
abdominal wall. If the abdominal wall is very compliant, the diaphragm will
descend easily without resistance, and the pressure generated to expand the rib
cage will be small. If abdominal wall compliance is decreased, for example when
the abdominal muscles contract, descent of the diaphragm will be resisted, a
larger intra-abdominal pressure will be generated, and the rib cage will be lifted
effectively (De Troyer 1983a). Because this issue is central to chest wall
mechanics, a major part of this thesis is devoted to the examination of abdominal
wall compliance under different circumstances.
1.7, Effects of posture on chest wall motion
1.7.1, Rib cage and abdominal motion
Sharp et al (1975) used magnetometers to partition respired volume between rib
cage and abdominal compartments in normal subjects, in supine and erect postures.
In the supine resting state, abdominal displacement accounted for 75% of chest
wall motion, whilst in the erect position the rib cage accounted for 75% of
respired volume. Vellody et al (1978) measured the effect of gravity on the shape
of the relaxed rib cage and abdomen. They found that abdominal diameter and
cross-sectional area decreased considerably on moving from the erect seated
posture to the supine posture, whilst there was little effect on rib cage
measurements. They attributed this to the greater compliance of the abdomen.
52
1.7.2, Length tension relationship of the diaphragm
In the erect posture the abdominal contents fall forwards under the influence of
gravity, the diaphragm descends and functional residual capacity is increased.
Grassino et al (1978a) have shown that the relationship between the intergrated
EMG signal of the diaphragm (EDI) (which represents the activation of the
diaphragm) and the pressure it can generate (Pdi), depends on thoraco abdominal
configuration and not lung volume. Pdi at constant Edi was 4 times more sensitive
to abdominal shape than rib cage shape. Diaphragm shortening is therefore
reflected most directly by abdominal displacement, as a consequence the diaphragm
may have different lengths at the same lung volume. In their paper on dynamics
Grassino et al (1978b) have shown that diaphragm shortening increases directly as
the rate of abdominal displacement.
Braun et al (1982) used X-ray measurements to show that the pressure generated by
the diaphragm is proportional to its lengths and,not its radius of curvature, as
suggested by La Place's law. The variables that govern the pressure generated by
the diaphragm are thus, its length and the abdominal configuration.
On standing, the length of the diaphragm is decreased by gravity. Druz and Sharp
(1981) have shown that in this situation the abdominal muscles become active,
abdominal diameter decreases and intra-abdominal pressure rises pushing the
diaphragm upwards into a more effective position. In the supine position, the
weight of the abdominal contents mainstain the length of the diaphragm and the
abdominal muscles do not contract during quiet breathing.
53
1.8, Actions of the intercostal and accessory muscles
The observation that normal subjects are rib cage breathers when erect and
abdominal breathers when supine (Sharp et al 1975), cannot be completely explained
by efficient coupling of the diaphragm and rib cage in the erect posture. There
is good EMG evidence that there is more phasic activity in the parasternal
intercostals and scalene muscles in the erect than the supine posture (De Troyer
and Estenne 1984). This activity is thought to stabilise the rib cage and allow
the diaphragm to move it more effectively. Further evidence for the importance of
the role of the intercostal muscles comes from studies of tetraplegic patients, in
whom these muscles are paralysed (Danon et al 1979). Such patients may have
paradoxical movements of the rib cage on inspiration, which is thus displaced from
its relaxation curve. This suggests that in the absence of intercostal activity,
the rib cage is not effectively coupled to the diaphragm. On the basis of present
evidence, it seems likely that in normal standing subjects intercostal muscle
activity is necessary to stabilise the rib cage, and allow the diaphragm to drive
it effectively along its relaxation characteristic.
Tetraplegic patients provide a unique opportunity to study the effects of the
diaphragm on the chest wall isolated from other respiratory muscles. The next
chapter will review the present knowledge of chest wall mechanics in patients with
tetraplegia.
54
CHAPTER 2
CHEST WALL MECHANICS IN TETRAPLEGIA
2.1, Introduction
The previous chapter described the role of the abdomen in breathing in normal
subjects. It emphasised the importance of the anterior abdominal wall, and in
particular, the effects of changes in abdominal wall compliance on the interaction
between the diaphragm and the rib cage. Contraction of the abdominal wall muscles
reduces its compliance, and the situations in which this occurs have been
described. Tetraplegic patients breath predominantly with their diaphragms, and
in this situation it is possible to examine its effects on the respiratory system
when the abdominal muscles are paralysed. The purpose of this chapter is to
describe chest wall mechanics in tetraplegia. This information will then be
applied to our understanding of the abdomen as an organ of respiration.
2.2, Pathophysiology
Acute transection of the cervical spinal cord usually occurs in previously normal
people, due to trauma. At a stroke it interrupts the nerve supply to most of the
respiratory muscles, and disrupts chest wall mechanics to such an extent that many
patients die soon after injury (Silver and Gibbons 1968). To understand the
respiratory deficit in an individual patient requires a knowledge of the segmental
innervation of the respiratory muscles and the spinal level of the injury.
55
The segmental innervation of the respiratory muscles is displayed diagramatically
in Figure 2.1. When the spinal cord is transected, muscles below the level of the
injury will be paralysed and no longer under supra-spinal control.
A critical factor in the patient's prognosis is whether the nerve supply to the
diaphragm remains intact. The diaphragm is innervated by the phrenic nerve, which
originates from the 3rd, 4th, and 3th cervical segments. The majority of cervical
cord transections occur between C5 and C8, and so the diaphragm frequently
functions normally in tetraplegic patients. However as seen from Figure 2.1, a
lesion at this level will paralyse the majority of the remaining respiratory
muscles, which take their nerve supply from T1 downwards. From a knowledge of
respiratory muscle function, one can predict the consequences to breathing of such
an inj ury.
1) Paralysis of the intercostal muscles will decrease inspiratory
capacity and destabilise the rib cage.
2) Paralysis of the abdominal muscles will make forced expiration
impossible, and by increasing abdominal wall compliance disrupt the
coupling of the diaphragm and rib cage .
If the spinal cord is transected at C4 or above, the diaphragm will be paralysed
but the trapezius and stemo-mastoid muscles will remain under voluntary control.
They may inflate the rib cage alone, with a vital capacity of up to 800mls (Danon
et al 1979), but usually such patients will require mechanical ventilatory
support.
There is recent evidence to suggest that the scalene muscles are more important in
breathing than previously thought. In normal subjects they may act as a primary
muscle of inspiration (De Troyer and Estenne 1984). In tetraplegic patients, if
56
C 1__ 2___3___4___5___6
7___8T 1__ 2___3___4___5__ 6___7__ 8___9
10 11 12
L 1__ 2___3___4___5S 1__ 2___3___4
5
DI
A
INTERCOSTALS
Figure 2.1The segmental innervation of the respiratory muscles.(TRAP = Trapezius, SCM = Sterno-cleido-mastoid,SCAL = Scalenes, DIA = Diaphragm, Ex OBL = External oblique INT OBL = Internal oblique, TRANS ABD = Transversus abdominis, RECTUS = Rectus abdominis).
57
scalene function is preserved, it may play a crucial role in stabilising the rib
cage (Estenne and De Troyer 1985).
2,3, Lung function after cervical cord injury
2.3.1, Chronic stable tetraplegia
The effects of cervical cord transection on breathing have been measured over a
number of years mostly in terms of lung volumes. Such measurements present an
overall view of the function of the disabled respiratory system. Loss of lung
function correlates well with the level of spinal cord lesion (Fugl-Meyer 1971).
In this discussion, only data from patients with cervical cord transection will be
displayed. Table 2.1 reviews the published data on lung volumes in patients with
tetraplegia. The mean supine vital capacity falls to 61% of predicted, and the
mean seated vital capacity to 53% predicted. There may be considerable variation
between individuals; for example, the patients in the study of Fugl-Meyer and
Grimby (1971a) had vital capacities varying between 28% and 64% of predicted. In
Table 2.1 the mean seated total lung capacity is reduced to 74% of normal and the
mean residual volume is increased by 47%. The falls in vital capacity are mainly
due to increases in expiratory reserve volume, consequent on paralysis of the
expiratory muscles. In the supine posture FRC and RV should be identical, though
in the seated position there may be some expiratory capacity below FRC. Bergofsky
(1964) has shown this to be due to shoulder girdle movement and forward flexion of
the trunk.
2.3.2, Effects of posture on lung function
Comparison of the vital capacities of supine and seated tetraplegic patients in
Table 2.1 reveals, that the supine posture has advantages for their breathing.
58
Table 2.1
Lung Volume Measurements Made in Tetraplegic Patients
Expressed as a Percentage of Predicted Values
Level No Position VC
% pred.
RV
% pred.
TLC
% pred,
Cameron et al 1955 C5-C8 11 Supine 65
McKinley et al 1969 Cervical 8 Supine 65 - -
McMichan et al 1980 C4-C8 22 Supine 54 141 78
Supine mean 61 141 78
Hemingway et al 1958 C4-C8 29 Seated 65 - -
Stone & Keltz 1963 C4-C7 9 Seated 53 123 74
Bergofsky 1964 C5 10 Seated 65 - -
Fugl-Meyer 1971 C4-C8 17 Seated 42 168 69
Fomer 1980 C5-C8 42 Seated 51 132 74
Estenne et al 1983 C2-T1 10 Seated 42 165 78
Seated mean 53 147 74
59
This contrasts with normal subjects, in whom vital capacity decreases by 7.5% on
lying down (Allen et al 1985). Cameron et al (1955) noted that vital capacityO
decreased by 6% in tetraplegic patients tilted from supine to 15 head up andO
increased by 7% when they were tilted 15 head down . Fugl-Meyer (1971) measured
vital capacity in 10 tetraplegic patients in the supine posture and compared it toOvital capacity in the seated posture and at 70 tilt to the horizontal. When
seated, vital capacity decreased by 28% and when tilted by 45%. Danon et al
(1979) studied 3 Cl tetraplegic patients undergoing phrenic pacing and noted a
decrease in tidal volume from 1100ml to 580ml, when they were tilted from supineO
to 55 to the horizontal. The cause of these alterations in lung volume with
postural change will be discussed later in this chapter (See 2.6).
2.3.3, Lung function immediately after acute injury
Most of the information on lung function in tetraplegia relates to chronic stable
patients. Immediately after injury, vital capacity may be reduced to as little as
300ml, even if the diaphragm is working. If lung function is followed serially
over a period of months, it gradually improves to reach the levels stated in Table
2.1 (McMichan et al 1980, Ledsome and Sharp 1981). This may take place without
evidence of neurological recovery; the mechanisms by which it occurs will be
discussed (See 2.8).
2.4, Chest wall motion in tetraplegia
Duchenne in 1867 first described paradoxical movements of the rib cage on
inspiration, in patients with flaccid paralysis of the intercostal muscles.
Descriptions of this phenomenon have become more detailed as the techniques
available have become more sophisticated.
60
2.4.1, Paradoxical movements of the rib cage
Bergofsky in 1964 used a plethysmograph sealed at the lower costal margin, to
partition respired volume between rib cage and abdominal compartments in
tetraplegic patients, ^e found that in the kupine posture, displacement of the
abdomen accounted for the majority of a given tidal volume, and in some cases was
greater than respired tidal volume. This implied that the rib cage contribution
to breathing was negative, i.e. there was paradoxical rib cage motion. Moulton
and Silver (1970) used calipers to examine rib cage motion in tetraplegia. They
noted paradoxical antero-posterior (AP) motion of the upper rib cage and
paradoxical lateral motion of the lower rib cage. Fugl-Meyer and Grimby (1971b)
applied magnetometers to the rib cages and abdomens of 10 tetraplegic patients
during quiet breathing. They found that at the start of inspiration and
expiration, rib cage and abdominal movements were out of phase, and that this
varied with body posture.
2.4.2, Effects of posture on chest wall motion
Mortola and Sant^Ambroggio (1977) used magnetometers to measure AP and lateral
diameters of the rib cage, in patients with complete transection of the cervical
spinal cord betweem C5 and C7. This information can be conveniently displayed on
a Konno and Mead plot of rib cage diameter against abdominal diameter (Figure
2.2). They found that during tidal breathing in the seated posture, AP rib cage
motion was normal, whilst lateral expansion was reduced. In the supine position,
AP movement was always paradoxical in inspiration, whilst lateral motion was
initially appropriate but became paradoxical mid way through inspiration. Another
important observation was that if the rib cage was distorted during inspiration,
it sprung back to a normal position at the start of expiration. This causes a
delay in expiration whilst volume is redistributed between rib cage and abdomen.
61
SUPINE SITTING
------------------ NORMAL------------------ QUADRIPLEGIA
Figure 2.2The rib cage/abdomen plots of the upper rib cage A-P dimension (upper panel) and the lateral rib cage (lower panel) in supine and sitting positions. A Cg tetraplegic patient is compared to a normal subject (Mortola and Sant'Ambrogio 1978).
62
Danon et al in 1979 described the chest wall motion of a Cl tetraplegic patient
equipped with a phrenic pacemaker. This allowed the effects of the diaphragm on
chest wall movement, to be compared to those of Sterno-mastoid and Trapezius when
the pacemaker was switched off. On the Konno and Mead plots in Figure 2.3,
abdominal movements can be compared to an idealised relaxation characteristic
supplied by mechanical ventilation. When the patient was seated and using his
neck muscles to breathe, the rib cage is expanded but the diaphragm and abdominal
wall are sucked in. When the phrenic pacemaker is activated, the abdominal motion
is appropriate and there is some expansion of the rib cage, though it is not
completely normal since it has to spring back to the relaxation curve to deflate
in expiration. In the supine position, when the diaphragm is paced, there is
obvious rib cage paradox. The reasons for these alterations of chest wall motion
in posture will be discussed later in this chapter (See 2.6).
2.4.3, Optical contour mapping
Recently an optical contour mapping technique (See 3.4.1) has been used to look at
chest wall movements in tetraplegia (Morgan et al 1984, Morgan and De Troyer
1984). This allows a more detailed analysis of motion, as information is
available from the whole of the visible chest wall surface, as opposed to selected
diameters when magnetometers are used. This method has confirmed the presence of
paradoxical movements of the rib cage in some, but not all tetraplegic patients.
AP paradox of the upper rib cage has been shown to be predominant during tidal
breathing, whilst lower lateral rib cage paradox is predominant during increased
ventilation.
63
ABD ABD
SITTING SUPINE
Figure 2.3The rib cage/abdomen plots of A tetraplegic patient sitting and supine. AM illustrates the use of the accessory muscles alone, and PP the phrenic pacemaker. The relaxation curve is obtained from IPPV (Danon et al 1979).
64
2.4.4, Summary
1. The chest wall is driven by the diaphragm in tetraplegic patients, and
abdominal motion is therefore predominant.
2. The rib cage moves paradoxically in some but not all patients. This is
an inefficient way to breath.
3. There is more paradoxical rib cage movement in the supine than the erect
posture. This effects predominantly the AP diameter of the upper rib
cage, and the lateral diameter of the lower rib cage.
65
2.5, Chest wall compliance in tetraplegia
The partition of ventilation between the rib cage and abdominal compartments,
depends on the relative compliances of these compartments at a given lung volume.
This is a dynamic system and as lung volume alters so may the ratio of rib cage to
abdominal compliance.
2.5.1, Measurement of chest wall compliance
Bergofsky (1964) measured abdominal compliance (Cabd), over the range of tidal
breathing. He found that the abdomen was more compliant in tetraplegic patients
than in normal subjects. That is, there was a greater increase in abdominal
displacement for a given gastric pressure change. Abdominal compliance also
varied with posture, being greatest supine. Fugl-Meyer and Grimby (1971b) used
magnetometers to measure abdominal displacement during a relaxed expiration, in
patients with tetraplegia. Their findings were similar to Fugl-Meyer"s, except in
one subject who had spasm of the abdominal muscles, in whom abdominal compliance
was diminished. Estenne et al (1983) applied a weighted spirometer technique
(Heaf and Prime 1964) to the measurement of chest wall compliance in normal
subjects and tetraplegic patients. They measured total respiratory system
compliance (Crs), and by measuring lung compliance (Cl) with an oesophageal
pressure monitoring balloon, derive chest wall compliance by the formula 1/Crs =
1/Cw - 1/C1. Chest wall compliance was found to be significantly reduced in
tetraplegic patients, and it was speculated that this was mainly due to a decrease
in rib cage compliance (Crc). Danon et al (1979) measured Crs and attempted to
partition it into Crc and Cabd, on the basis of the relative motion of rib cage
and abdomen measured with magnetometers. The problem as in the work of Fugl-Meyer
and Grimby (1971b), was that the chest wall in tetraplegic patients moves with
more than 2 degrees of freedom. It is therefore inaccurate to extrapolate from
66
change in diameter assessed with magnotometers, to change in volume (See 3.3.3).
Optical contour mapping would be an ideal way to partition Crs between rib cage
and abdomen (See 3.4.2.2).
2.5.2, Factors influencing chest wall compliance
Immediately after transection of the cervical spinal cord, there is a period
during which there is flaccid paralysis of the respiratory muscles. This is known
as spinal shock and during this time the compliance of the rib cage and the
abdomen will be increased. As recovery begins, there may be a return of tone and
reflex activity to these muscles. This may amount to frank spasticity, and
commonly this occurs in transient episodes known as mass spasms (Michaelis 1976).
These may occur spontaneously or as a reflex for example due to irritation within
the urinary tract, or contact with the skin of the abdomen or thigh. During
spasm, contraction of the abdominal or intercostal muscles will alter chest wall
compliance. These spasms may even spread to the diaphragm and alter ventilatory
pattern (Silver and Lehr 1981a).
2.5.3, Factors influencing rib cage compliance
In some patients the rib cage is stiff and barely moves with breathing (Estenne et
al 1983), in others it is floppy and moves paradoxically on inspiration (Morgan
and De Troyer 1984). These differences may be caused by a variety of factors.
There is evidence to suggest that in time the joints of the rib cage and spine
become ankylosed and heterotopic ossification may develop (Wharton 1952). Another
factor may be the return of reflex activity to the intercostal muscles. Phasic
inspiratory EMG activity has been detected in the intercostal muscles of
tetraplegic patients (Guttmann and Silver 1965, Silver and Lehr 1981b), and it has
been suggested that this is associated with improved respiratory performance (De
67
Troyer and Heilporn 1980). More recently, 2 studies on a total of 30 seated
tetraplegic patients breathing at rest have failed to find this phasic inspiratory
EMG activity, but have detected tonic EMG activity in the intercostal muscles of
some tetraplegic patients (Morgan and De Troyer 1984, Estenne and De Troyer 1985).
This activity was not associated with any detectable respiratory benefit to the
patient. Estenne and De Troyer (1985) studied 20 seated patients, and detected
paradoxical rib cage motion in 8. All of these had either absent or tonic EMG
activity in their scalene muscle, whilst of the remaining 12, 8 had phasic
inspiratory EMG activity of the scalenes. It seems reasonable to conclude that
the compliance of the rib cage depends on the elastic properties of its
infra-structure, and on the presence of normal phasic inspiratory activity in the
scalene muscles. Phasic inspiratory activity of the intercostal muscles may be
important at increased levels of ventilation.
2.5.4, Factors influencing abdominal wall compliance
In the majority of tetraplegic patients, the abdominal wall is displaced further
during breathing than in normal subjects. It has been presumed that abdominal
wall compliance is increased, though this is difficult to measure directly.
Changes in tone within the abdominal muscles will alter compliance, and such
increases in tone may be associated with mass spasms. In this situation, a
decrease in abdominal wall compliance may provide such resistance to descent of
the diaphragm, that it cannot be overcome and apnoea will occur. In addition to
tonic EMG activity of the abdominal muscles during spasm, phasic inspiratory EMG
activity has been reported (Guttmann and Silver 1965). Subsequent studies have
not reproduced this finding in patients during tidal breathing (Silver and Lehr
1981b, Estenne and De Troyer 1985). Phasic inspiratory EMG activity which caused
an increase in abdominal muscle tone could, by resisting the descent of the
diaphragm, improve its coupling to the rib cage (See Chapter 4).
68
2.6, Mechanics of posture change
When tetraplegic patients are tilted head up, they note difficulty in breathing,
and their lung function deteriorates. The decrease in vital capacity on moving
from the supine to the upright seated posture has been detailed (Table 2.1).
There is some information available in the literature, on the effect of change in
posture on trans-diaphragmatic pressure (Table 2.2) and maximum static inspiratory
mouth pressure (Table 2.3).
When normal subjects stand up, their abdominal muscles become active and oppose
the hydrostatic drop in intra-abdominal pressure. This maintains the high arched
configuration of the diaphragm in the thorax. The abdominal wall provides
resistance to the descent of the diaphragm on inspiration, and this promotes lower
rib cage expansion (See 1.5.2.2). In the supine posture, the abdominal wall is
more compliant than when erect, and is easily displaced by the descending
diaphragm causing less lower rib cage expansion.
When tetraplegic patients are tilted up, since the abdominal muscles are
paralysed, the abdominal wall is displaced forwards by the effect of gravity on
the abdominal contents. It becomes stretched and is therefore less compliant.
The diaphragm is pulled by gravity to a lower position with a flatter
configuration, from which it can contract less effectively (Braun et al 1984). As
the diaphragm descends, it is resisted by the stretched abdominal wall, and the
lower rib cage is expanded. In the supine posture, gravity acting on the
abdominal contents maintains the length and tension of the diaphragm, so that it
may contract effectively. The abdominal wall is very compliant, such that there
is little resistance to diaphragmatic descent. As the diaphragm descends easily,
it flattens and further contraction will pull the rib cage inwards. This accounts
for the paradoxical inward motion of the rib cage in supine tetraplegic patients.
69
Table 2.2
Trans-Diaphragmatic Pressure Measurements Made In Tetraplegic Patients
Level No. Position Pdi(cm H20)
Pichurko et al 1985 Cervical 5 Supine 73 (Plmax Pdi)
De Troyer et al 1980 C5-C7 10 Seated 52 (Plmax Pdi)
Normal Range = 82 - 204 cm H20 (Miller et al 1985).
70
Table 2.3
PImax Measurements Made In Tetraplegic Patients
Level No. Position PImax (cm H
McMichan et al 1980 C4-C8 22 Supine 77
Pichurko et al 1985 Cervical 5 Supine 62
Supine ' mean 70
Fugl-Meyer 1971 C4-C8 12 Seated 75
Estenne et al 1983 C2-T 10 Seated 36
Morgan and De Troyer 1984 C5-C6 8 Seated 52
Seated mean 54
Normal Range = 80 - 168 cm H 2O (Black and Hyatt 1969)
Estenne and De Troyer (1985) describe this motion predominantly in the lateral
diameter of the rib cage. This may be explained by the greater surface area of
apposition of the diaphragm to the lateral rather than the anterior chest wall.
Paradoxical motion of the upper rib cage relates to the level of negative pleural
pressure acting on the destabilised rib cage (See 2.5.3). In the supine posture,
the shape of the diaphragm allows it to generate more negative pleural pressure,
than in the upright posture. As a consequence, upper rib cage paradox is greater
in the supine position than in the erect position.
2.7, Abdominal binders
2.7.1, The effect of abdominal binders on chest wall motion
If the abdominal wall of tetraplegic patients could be made less compliant, how
would this effect their breathing? Abdominal binders have been used in
tetraplegic patients for some years, particularly when they are first raised from
the supine posture. The rationale for their use has been to decrease postural
hypotension by increasing venous return to the heart, and to improve breathing.
There is very little scientific literature published on the effect of abdominal
binders on lung function in tetraplegia. One would expect that if a binder was
fitted in the upright posture, that it would increase intra-abdominal pressure by
compressing the abdominal wall. This would improve the length and tension of the
diaphragm, and make the abdominal wall less compliant. The diaphragm would
generate a greater force on contraction, and its descent would be resisted and
lower rib cage expansion increased. Danon et al (1979) bound the abdominal walls
of 2 Cl tetraplegic patients undergoing diaphragmatic pacing. One patient was
studied seated and the other at 55° tilt. The seated patient's tidal volume
improved by lOOmls with the binder and there was a 16% increase in rib cage
expansion. The tilted patient also had an improvement in tidal volume, and rib
72
cage expansion increased by 20%. Strohl et al (1984) reported the case of aOsimilar patient with tetraplegia. When she was tilted to 85 tidal volume fell
from 425ml - 120ml, on binding the abdomen tidal volume increased to 330ml. In
the supine posture, her tidal volume fell when the abdomen was bound, though the
rib cage expansion was improved. This illustrates the point that if the abdominal
wall is bound too tightly, the diaphragm may have difficulty in overcoming the
resistance to its descent, and this will have a detrimental effect on breathing.
Clinically this is important in patients with severe abdominal muscle spasms.
2.7.2, The effect of abdominal binders on trans-diaphragmatic pressure (Pdi)
Celli et al (1985) studied a patient with chronic airflow obstruction, and a very
compliant abdominal wall due to an abdominal hernia. As there was no effective
tone in the abdominal muscles, this situation is analogous to tetraplegia.
Reduction of the hernia with a binder led to an increase in Pdi from 14cm H20 to
27cm H20 in the standing position. There is no published data on the effect of
abdominal binders on trans-diaphragmatic pressure in tetraplegic patients.
2.7.3, Present clinical practice and implications
The abdominal binders currently available in spinal injury units are shaped like a
corset. This means that as well as enclosing the abdominal wall, they also
encroach on the lower rib cage. They may therefore be opposing their own rib cage
inflating effect. The application of abdominal binders to tetraplegic patients is
a clinically important area of chest wall mechanics which has not been fully
investigated. Chapter 7 of this thesis investigates the effect of abdominal
binders on breathing in cervical cord injured patients.
73
2.8, Acute changes in chest wall mechanics after cervical cord transection
In the stage of spinal shock, flaccid paralysis of the respiratory muscles renders
the rib cage and abdomen very compliant. For reasons stated, the former is more
important in the supine posture, and the latter in erect postures (See 2.6). In
the United Kingdom acutely injured tetraplegic patients are nursed in the supine
position for some weeks. Morgan et al 1985 have shown serial improvement in the
lung function of a patient over a period of months after injury. This was due to
a decrease in paradoxical rib cage motion and an increase in its positive
contribution to breathing. This may in part have been due to an increase in the
mechanical stiffness of the rib cage due to boney ankylosis of the joints, or
return of tone to the intercostal muscles. Estenne and De Troyer (1985) state
that duration of injury does not influence the presence of rib cage paradox, and
that phasic inspiratory scalene muscle activity is more important. Their study
only included one patient injured for less than a month and therefore may not
apply directly to the acute situation. It may be that oedema of the acutely
transected spinal cord will initially impair the function of the diaphragm
(C3,4,5) and Scalenes (C4,5,6,7,8), but that this resolves in time and allows
improvement in their function. Alternatively, the return of tone to the
intercostal muscles soon after cord transection may be of some as yet unproven
importance.
2.9, Cough
Tetraplegic patients do not cough effectively because their expiratory muscles are
paralysed (Figure 2.1). The expiratory pressures generated in this situation are
dependent on the elastic recoil generated by the chest wall subsequent to the
previous inspiration. This causes a reduction in the peak expiratory flow rate to
65% of normal (Fugl-Meyer and Grimby 1971a). Pleural pressure during coughing is
74
also considerably reduced and in some subjects never becomes positive (Fomer
1980, De Troyer and Heilpom 1980). As a consequence, the dynamic compression of
the airways is reduced, as is the linear velocity of gas, and cough is
ineffective. It is common practice to manually compress the abdominal wall of
tetraplegic patients at peak inspiration to help them clear their airways.
2.10, Sensation of breathing
The control of breathing depends on sensory input from the lungs, diaphragm and
intercostal muscles. Spinal cord transection or anaesthesia below Tl, does not
impair breath holding or sensation of load (Eisele 1968). The response to added
inspiratory load is also intact in this situation, and immediate adaptation can
take place (Frankel et al 1971, Axen 1982). Addition of an expiratory threshold
resistance (see Chapter 6) elicits a normal response in dogs after Tl cordotomy
(D'Angelo and Agostoni 1975), which can be abolished by vagotomy. These findings
suggest that if the diaphragm is normally innervated, respiratory sensation from
diaphragm receptors can be transmitted to the respiratory centre via an intact
phrenic nerve.
75
2.11, Summary
This chapter has reviewed the present knowledge of chest wall mechanics in
tetraplegic patients. It has highlighted the effect that paralysis of the
abdominal wall has on their breathing, in particular with relation to posture.
Several important questions have been raised.
1) Can direct measurements of abdominal wall compliance be made
in tetraplegic patients?
2) Can the partition of total respiratory system compliance between
rib cage and abdomen help us to understand the dynamics of breathing
in tetraplegic patients?
3) Does reflex EMG activity occur in the abdominal muscles of tetraplegic
patients?
4) Can lung function be significantly improved by binding the abdomen
of tetraplegic patients, and under what circumstances?
The experimental work described in the remainder of this thesis was designed to
answer these questions.
76
CHAPTER 3
METHODS AND MATERIALS
3.1, Selection of Patients
Patients were selected from current in-patients at the National Spinal Injuries
Centre, Stoke Mandeville. All had clinical evidence of complete transection of
the cervical spinal cord due to trauma. This was defined as an absence of
neurological function below the cervical segment at the level of the lesion. When
examined, no patient included in the studies had evidence of any acute respiratory
problem. All patients studied were examined at least three months after their
injury. All patients gave informed consent, and permission to perform all
procedures had been granted by the Stoke Mandeville and Brompton Hospital Ethics
Committees. Four separate studies are reported. The clinical data, physiological
results and discussion of results for each group of patients, are presented in the
following chapters.
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Electromyographic study of the abdominal muscles.
Measurement of abdominal wall compliance.
The effect of loading expiration on chest wall motion.
The effects of abdominal binders on breathing.
These studies were performed over an 18 month period. Since patients remain in
hospital for up to 9 months after their injury, some patients took part in more
than one study.
77
3.2, Selection of normal subjects
In the work described in Chapters 4 and 5, normal controls were studied. These
were volunteers drawn from the staff of Stoke Mandeville Hospital. They were of a
similar age to the patients studied. None were suffering from any acute or
chronic, neurological or respiratory illness. On examination, they were
neurologically intact. They gave informed consent for all procedures undertaken.
Details of normal subjects will be given in the relevant chapters.
3.3, MEASUREMENT OF ABDOMINAL WALL DISPLACEMENT
3.3.1, Introduction
Change in volume of the thoracic cavity during breathing is reflected by
displacement of the rib cage and diaphragm which form its boundaries. Since the
abdominal contents are incompressible, caudal movement of the diaphragm causes
displacement of the abdominal wall. Chest wall motion can therefore be assessed
by measurement of the motion of the rib cage and the abdominal wall, which may be
independent of each other. A variety of techniques have been used to measure the
separate displacements of the rib cage and abdomen during breathing and these will
be briefly reviewed.
3.3.2, Radiological techniques
Wade in 1954 used a radiological method to measure change in AP diameter of the
rib cage, and linear excursion of the diaphragm during breathing. From these
measurements, ventilation could be partitioned into rib cage and diaphragm/abdomen
components, by using 2 simultaneous equations to calculate volume motion
coefficients. More recently, Hedenstierna et al ,(1985) used computerised axial
78
tomography to measure rib cage and abdominal dimensions in normal subjects under
anaesthetic. Such methods are unsuitable for monitoring chest wall motion for
long periods because of radiation dosage.
3.3.3, Linear Measurements
The motion of a point on the surface of the chest wall can be described, by its
separation from an equivalent point on the opposite surface of the body, or from
another fixed point. Konno and Mead (1967) described a method for assessing chest
wall motion, by measuring the displacement of points on its surface from their
resting positions. Originally, they attached a long cord to a point on the skin
surface, and measured its radial displacement during breathing, relative to an
arbitary reference point. In order to extrapolate from the motion of single
points on the surface of the rib cage and abdomen, to the partition of respired
volume between these two compartments, they made certain observations. They
demonstrated that in trained normal subjects, the open respiratory system behaved
as a simple physical system with two independently moving parts, the rib cage and
the diaphragm/abdomen. The system is said to have 2 degrees of freedom of motion,
since the movement of one part does not influence the freedom of motion of the
other. Konno and Mead showed that the volume change of the two compartments was
linearly related to the change in their AP diameters. It was therefore valid to
measure the motion of a single point, and presuming it moved in a similar way to
all other points on the surface of the rib cage or abdomen, to calculate the
volume change of the compartment. Finally, they showed that when the glottis was
closed the system had only a single degree of freedom. Thus motion of one part
could only occur when there was reciprocal motion of the other part. In simple
terms, if the volume of one part of the system decreased, since the volume of the
whole system remained unchanged, the volume of the other part of the system must
increase. This fact was employed to calibrate the system using the "Isovolume
79
Manoeuvre", in which volume was voluntarily shifted between rib cage and abdomen.
During this manoeuvre, since the volume of the system remains constant, the motion
of points on the surface of the rib cage and abdominal compartments must represent
equivalent changes in volume. It is therefore possible to calculate the volume
motion coefficients fot each compartment.
In 1967 Mead et al described respiratory magnetometers, which provided a
convenient way to measure changes in chest wall diameter with breathing. This
method is commonly used to partition ventilation between rib cage and abdominal
compartments (Goldman 1984), but there are problems with its application.
Magnetometers will measure all motion of the rib cage and abdomen, including that
caused by alterations in posture. In addition, outside the range of tidal
breathing the respiratory system ceases to behave with 2 degrees of freedom, so
increase in diameter as measured by magnetometers may not relate directly to
change in volume of the compartment. Goldman et al (1981) have shown that in some
subjects, lateral rib cage diameter increases significantly early in inspiration
during spontaneous breathing, which increases the number of degrees of freedom of
the system. In tetraplegic patients paradoxical rib cage movement increases the
number of degrees of freedom of chest wall motion, as Fugl-Meyer and Grimby
(1971b) found when they tried to partition ventilation between rib cage and
abdomen in such patients. With these reservations, it should be stated that by
using change in linear dimensions it is possible to measure tidal volume and
pulmonary ventilation to within approximately 10% of spirometric values. The
studies described in this thesis require the measurement of abdominal displacement
outside the tidal breathing range in tetraplegic patients. Magnetometers are not
suitable for making these measurements.
80
3.3.4, Measurements of circumference
Measurements of change in chest wall circumference with breathing can be made with
a tape measure or a mercury in rubber transducer. Agostoni et al (1965) used such
a method to observe the relationship between lung volume and rib cage
circumference. They assumed that the rib cage was elipticaL and derived a formula
to measure cross-sectional area, and hence lung volumes. Outside the range of
tidal breathing, this assumption may not be correct.
3.3.5, Measurements of cross-sectional area
Measurement of cross-sectional area removes the necessity to make assumptions
about rib cage configuration. The respiratory inductance plethysmograph consists
of 2 bands containing coils of uninsulated wire, which encircle the rib cage and
abdomen. An oscillating signal passes through the coils, and the
frequency-modulated signal changes in proportion to the alterations in self
inductance of the coils, resulting from change of volume in the enclosed part.
This method as described by Sackner et al (1980) can measure tidal volume at rest
and during exercise, to within an accuracy of 20%. To calibrate this apparatus
requires an isovolume manoeuvre, or the solution of simultaneous equations derived
from rib cage and abdominal motion in different postures. These calibration
procedures again depend on the chest wall having 2 degrees of freedom of motion,
and on the volume motion coefficients being constant and independent of posture.
If the relationship between rib cage and abdominal motion differs at any time from
that during the calibration procedure, then the calculated changes in volume are
not valid. This method is not suitable for use in the assessment of tetraplegic
patients.
81
3.3.6, Three dimensional measurements
3.3.6.1, Partial plethysmography
In order to measure the volume displacement of the abdomen in absolute terms,
Bergofsky (1964) constructed a partial body plethysmograph. This enclosed the
abdomen in a tank whilst the rib cage remained outside. A rubber seal was fitted
along the lower costal margin. This seal had to be air tight, but not interfere
with chest wall motion. The method was criticised because it failed to take into
account cranio-caudal motion of the lower costal margin with breathing.
With the exception of plethysmography, the above techniques are limited, as they
try to describe a 3 dimensional structure by making measurements in 1 or 2
dimensions. A variety of methods have been used to make measurements in 3
dimensions. These include stereophotogrametry, light slit techniques, Moire
fringe interferometry, and more recently optical contouring.
3.3.6.2, Optical contouring
This method involves the projection of a grid pattern onto the object to be
assessed, which is then observed from an angle other than that of the projection.
From the observation point, a series of contours are formed on the surface of the
object. Lovesey (1966) and Cobb (1972) used this method to record face shapes, to
design oxygen masks for aircrew. Williams (1977) applied such a method to
studying the chest, using a large fine grid and a parabolic mirror. This
technique was not quantitative. Kovats (1970,1974,1978) developed a system using
divergent optics to study thoraco-abdominal motion in the upright subject. A grid
pattern was projected onto the front of the subject, who was photographed from one
side. Under conditions of known geometry, he was able to calculate the x,y and z
82
co-ordinates of points on the front of the body. From these, horizontal profiles
of the body were constructed during a breath.
Denison et al (1982), Peacock et al (1984) and Morgan et al (1984) have used
similar principles in conjunction with computer technology, to apply optical
contouring to the study of thoraco-abdominal motion during breathing in the erect
and supine postures. In the optical contour mapping technique described by
Peacock et al (1984), the subject stands in a reference frame, and a line pattern
is projected onto his body from either side. Two 35mm cameras positioned opposite
each other and at 90 to the axis of the projectors, capture images from the front
and back (Figure 3.1). Each pair of photographs is then converted into digital
form, and the spatial co-ordinates of points on the body surface calculated by
computer. Further computer programmes can be used to measure volume and surface
area, or display shape. The linear accuracy of the technique is within 0.5mm.
Peacock et al (1985) tried to apply this method to partition chest wall volume
between rib cage and abdomen, using a horizontal line drawn through the
xiphistemum as a boundary. This method was inaccurate since the boundary chosen
was only a rough approximation to the position of the diaphragm, which in reality
divides the two cavities. Further development of the computer programmes
available has allowed more accurate partitioning of chest wall volume using the
lower costal margin as the dividing line (Morgan et al 1985).
Morgan et al (1984) developed a similar optical contour mapping system for
examining supine patients. The apparatus which he built is situated in a
dedicated light laboratory at the National Spinal Injuries Centre, Stoke
Mandeville. This system was used to make the measurements of thoraco abdominal
motion described in Chapters 5 and 6 of this thesis. The system and the way it
was applied will be described in detail.
83
Projector B
Figure 3.1The optical contour mapping system for use in standing subjects. Contours are projected onto the front and back of the subject and recorded by two cameras.
84
3.4, METHODS USED
3.4.1, THE OPTICAL CONTOUR MAPPING SYSTEM
3.4.1.1, Apparatus
The system consists of a camera and 2 projectors bolted to a rigid framework
(Figure 3.2). The gantry is constructed from lightweight steel tubing (Alta
Structura Ltd). The vertical pillars are 2.4m high and the lintel 2.8m across.
One projector (Kodak carousel 2020 with 35nmi lens) is mounted on each pillar and a
single 35mm still camera (Olympus 0M2n with 35mm Zuiko f 2.8 lens) is mounted on
the lintel.
3.4.1.2, Contour Projection
The contours are produced by projection of a black and white line pattern onto the
subject. The line pattern, printed on transparencies, was designed to produce
about 40 contour lines on the visible surface of the chest wall (Figure 3.3).
3.4.1.3, Image capture
The still camera was mounted exactly between the two projectors in the same
vertical plane. It was set at a shutter speed of l/125th second at f 4. Fast
black and white film (Ilford HP5) was used, and was wound on by a motor drive
which could take up to 4 frames/second. It was fired remotely by a camera control
box (Olympus MAC), which had the facility for external triggering.
85
Figure 3.2The optical contour mapping system.
86
IIIIBM
Figure 3.3The optical contour mapping system in use, note the display of contours.
87
3.4.1.4, Reference planes (Figure 3.4)
The projectors are positioned so that they face each other in the same horizontal
plane, which is designated the optical reference plane. The vertical camera axis
bisects the projector akis at a point which is the optical centre of the system.
The distanced between tiie optical centre and the camera (DZ), and the optical<
centre and each projector (DX), must be known exactly. These measurements are
taken from the front of each lens.
3.4.1.5, Calibration
The camera and projectors are exactly aligned by a series of check procedures.
The vertical camera axis is defined by a plumb line dropped from the centre of the
camera lens cap. The projectors are positioned equidistant from the camera axis
in the same horizontal plane, confirmed by a spirit level placed between them.
The projector axes are aligned by projecting a vertical and horizontal cross hair
pattern onto the centre of each others lens. A grid pattern is then projected
onto a vertical white screen in the centre of the system, and confirmed to be
horizontal. The black white interface which lies on the horizontal reference
plane is designated the reference contour. The reference contour should be
projected to exactly the centre of the opposite projector's lens. If this is not
the case, the whole procedure is repeated. Once correctly aligned, the projectors
are fixed in place by cementing plastic block in juxtaposition to the projector
feet, on its supporting tray.
Two light emitting diodes (L.E.D.s) attached by rods and clamps to the framework
of the system, are positioned equidistant from the optical centre. They are at a
known height from the projector axis in the same vertical plane. The optical
centre of the system is implicit on each photograph, from the position of the
88
Figure 3.4The focussing planes of the camera (f-c) and two projectors (f-pl) and (f-p2).
89
LEDs. This also allows the magnification factor of the system to be calculated.
The camera is focused on the optical reference plane, but because the projectors
do not have sufficient depth of field they are focused 10cm short of the optical
centre (Figure 3.4). Each grid is focused onto a board 10cm short of the optical
centre. The position of 42 black/white interfaces (30 above and 12 below the
reference contour) are marked. Their distance above or below the reference plane
(Z') is then measured in millimetres, and corrected to the position they would
have had at the optical centre (Z) by similar triangle geometry.
i.e. Z = Z' Dz
Dz-100
This procedure is performed for both projectors, and only accepted when the
positions of the contours from each side are exactly matched in the centre of the
system.
The calibration procedure provides a series of measurements, which are then
entered into the computer programmes used for data processing.
Dz Height of camera lens from optical centre.
Dx Distance from projector lens to optical centre.
Z Measurements of the contour positions at the
centre of the system, 40 per projector.
Magnification Factor. This is requested each time the
programme is run; it is calculated from the known distance
between the LEDs, and the distance between the LEDs on the
projected image.
It is easy to run a quick check to ensure that the system is correctly aligned. A
90
plumb line is suspended from the centre of the camera lens cap. The contours
formed on it by the line pattern from each projector should match exactly. The
reference contour should be identified at its known distance (Dz) from the camera
lens. This check was applied each time the system was used. When there was
evidence that the camera or projectors had been displaced, the full calibration
procedure was repeated.
3.4.1.6, Image processing
3.4.1.6.1, Projection
The film is developed in the normal way and the negatives are projected onto a
digital plotting tablet (Tektronix 4954). The projector is positioned above the
tablet, onto which the negative is projected through a wide angle lens (Leitz
Universal + 35mm) via a front silvered mirror. The projector,, mirror and tablet
were aligned by triangulation.
3.4.1.6.2, Analogue to digital conversion
A digitising programme was written (Gourlay et al 1984), to enter the spatial
co-ordinates of points on the chest wall into a main frame computer (Prime 750).
This programme, DIGITX, divides the chest wall into four quadrants (Figure 3.5).
These comprise the segments to the right and left of the optical centre, above and
below the reference contour. These quadrants are known as chest right, chest
left, back right, and back left. The programme calls for 3 calibration points.
The first is the optical centre (x=0,y=0) which is identified as mid-way between
the LEDs. The second and third calibration points (x=10,y=0 and x=0,y=10) are
located in relation to the first calibration point, using graph paper placed on
the digital tablet.
91
Figure 3.5For analogue to digital conversion, the chest wall is divide into 4 quadrants. BR = Back Right, CR = Chest right,CL = Chest left, BL = Back left, OC = Optical centre,RC = Reference contour. The quadrants are digitised in the order CL, CR, BR, BL.
92
The digitising procedure now begins, starting at the chest left quadrant. The
reference contour is identified (Line 0), and the cursor placed on the next
visible black white interface (Line 1). This is identified to the computer and
the cursor is then run down the contour within the area of interest (See
3.4.2.1.2) starting at the head end. The programme records the x value at 1cm y
levels, while the z value is implicit in the contour number. Each contour is then
analysed in sequence within the quadrant, followed by the contours in the
remaining 3 quadrants. The recorded information is then stored as a new data file
(DIGIDATA). The digitisation procedure is labour intensive, taking about 30
minutes for each negative.
3.4.1.6.3, Data processing
The data file contains the spatial co-ordinates of about a thousand points on the
surface of the projected image. All further analysis begins with calculation of
the true co-ordinates of the chest wall surface, by correcting for the optics of
the system. The surface of the chest wall is then reconstructed from this data by
linear interpolation. Programmes are available which measure the whole or partial
volumes of this reconstruction . JGLUNG calculates the total volume of the
reconstruction and partitions it into left and right, chest and back quadrants.
In order to partition volume into rib cage and abdominal compartments, 2 further
programmes are used.
3.4.1.6.4, Partitioning chest wall volume
The CUTTER programme allows the operator to identify a dividing line across the
projected image. After calling for calibration points as in DIGITX, the point at
which each contour is divided is identified to the computer with the cursor. This
is done in sequence of contours and quadrants, similarly to DIGITX. This creates
93
a further temporary data file (CUTTING-LINE), which is then applied to the
original data file using the JGLUNG2 programme. This is a modification of JGLUNG,
which calculates the volume of the two components of the original reconstruction,
using data from CUTTING-LINE as the dividing boundary. The way in which these
programmes are applied to analyse chest wall motion during breathing will be
described later in this chapter.
3.4.1.7, Assessment of the method
Morgan et al (1984) have assessed the accuracy of this method in detail. Their
findings will be summarised.
1) Volume measurements of the visible surface of a tailor^s dummy, to assess
reproducibility.
Mean optical estimate = 7964ml (SD = 26ml)
2) Accuracy of absolute volume measurements, determined on upturned
kitchen bowls, + 2%.
3) Calclulation of percentage of respired volume visible as chest wall
motion to a single camera in supine subjects. Visible portion - 97%.
4) Optical measurement of expired volume compared to spirometric
measurement. Mean error of 250ml in a total volume of 6L. However,
within subject measurements show a deviation of observations about
the regression line of 131ml (SD = 92ml).
The optical contour mapping system is a non-invasive and accurate method of
measuring respired volume. This can then be partitioned into its rib cage and
abdominal components. The next section of this chapter describes how the method
was applied to make the measurements reported in Chapter 5 and 6 of this thesis.
94
3.A.2, APPLICATION OF OPTICAL CONTOUR MAPPING
3.4.2.1, The measurement of abdominal wall compliance
Method for Chapter 5
3.A.2.1.1, Introduction
The aim of this study was to construct a pressure volume curve for the abdomen
during a relaxed expiration. The slope of this curve represents abdominal wall
compliance. Intra-abdominal pressure was measured as gastric pressure (See
3.A.3.A) and "abdominal volume" was measured using the optical contour mapping
system. Since the absolute volume of the abdomen does not change, what was in
fact measured was the visible volume-displacement of1the abdominal wall. The term
"abdominal wall compliance" will be used in preference to "abdominal compliance".
3.A.2.1.2, Protocol
6 normal subjects and 6 tetraplegic patients were examined in the supine posture
They were positioned on a Kings Fund bed, naked to the waist, with a gastric
pressure monitoring balloon in situ.
Each subject was then "marked up", to define the area of interest on the
photographs for the subsequent digitisation. Red adhesive paper spots were placed
at the sternal notch and on the anterior superior iliac spines, to mark the top
and bottom boundaries of the chest wall. Velcro fapes were passed around each
axilla, to define the lateral limits of the rib cage. In most subjects the lower
costal margin was clearly visible, but in a few it was obscured by fat. A marker
line was drawn along the lower costal margin at total lung capacity, and at
95
resting end expiration, in order to clarify its position for the partitioning
procedure.
The bed was wheeled under the optical contour mapping system, and its height
adjusted so that on full inspiration, the reference contour was visible throughout
the area of interest. The pressure monitoring balloon was then connected to
provide a continuous trace of gastric pressure, on a thermal pen recorder (See
3.4.3.4). Each subject was then trained to perform a "relaxation manoeuvre".
3.4.2.1.3, The ^Relaxation Manoeuvre**
This consisted of a passive expiration from TLC to FRC. Normal subjects were
instructed to take a deep breath in to TLC, and then to relax their respiratory
muscles. They allowed air to escape passively through pursed lips over a period
of 8-10 seconds, reaching a natural halt at FRC. During practice manoeuvres
gastric pressure was monitored; it was maximum at TLC, and declined exponentially
to a minimum at FRC during a passive expiration (Figure 3.6). Any expiratory
activity of the abdominal muscles caused a positive pressure pulse (Figure 3.7),
and with the help of the investigator subjects could be taught to avoid this. EMG
recordings from the abdominal muscles of these 6 subjects during the manoeuvre
showed no electrical activity (See 3.4.5.4). Tetraplegic patients performed the
same manoeuvre; their expiratory muscles are paralysed, so to avoid active
expiration was not a problem.
Displayed in Figure 3.8 are a pair of optical contour mapping photographs taken at
TLC and at FRC in a normal subject. In Figure 3.9 are photographs of a
tetraplegic patient taken at the same points in the "Relaxation Manouevre".
Comparison of the two sets of photographs demonstates that there are a similar
number of contours visible at FRC in both cases, however at TLC there are more
96
1 O-i
G A S T R I C P R E S S U R E
( c m H 2 0 )
CO■<1
C A M E R A T R I G G E Rrrmmrmrir- r m inim ii----------- 15 s e c s
Figure 3.6A typical gastric pressure recording of a normal supine subject during a passive expiration from TLC to FRC, with camera trigger marked.
G A S T R IC P R E S S U R E
(cm H O)
C------ -5 s e c s
Figure 3.7Example of a gastric pressure recording from a supine normal subject during a "passive expiration" which was rejected from analysis. During expiration from TLC to FRC there is a positive pressure pulse which represents abdominal muscle contraction.
FRC TLC
Figure 3.8The chest wall of a normal subject photographed by the optical contour mapping system camera.
10
0
FRC TLC
Figure 3.9The chest wall of a tetraplegic patient photographed by the optical contour mapping system camera.
contours visble on the abdomen of the tetraplegic patient. Thus during the
"Relaxtion Manoeuvre" there has been a greater visible abdominal volume
displacement in the tetraplegic patient.
3.4.2.1.4, Camera trigger mechanism
During relaxed expiration the camera was triggered at one second intervals, and
the gastric pressure trace marked as each photograph was taken. It was therefore
possible to match individual photographs to the gastric pressure recorded as they
were taken.
A 1/second signal was generated by the stimulator of a "Medelec MS6" EMG module.
The voltage of the signal was amplified by a battery powered amplifier (Stoke
Mandeville Hospital, Medical Electronics). This connected to the camera control
box (Olympus MAC) and to a channel of the paper recorder. Each signal triggered
the camera and synchronously made a vertical mark on the gastric pressure
recording (Figure 3.6). At the beginning of each passive expiration the EMG
module was switched on manually, and then fired automatically at one second
intervals throughout the manoeuvre.
Each subject performed 12 relaxation manoeuvres. Any that were technically
unsatisfactory were discarded. From the remainder, a representative wave form was
identified and the corresponding single set of photographs (Figure 3.10) analysed
to determine abdominal wall displacement.
101
3.4.2.1.5, Digitisation Procedure
In this study the area of interest was defined between horizontal lines drawn
through the sternal notch, and through the anterior superior iliac spines. The
lateral limits were identified by the velcro bands passed around the axillae. The
digitisation procedure was performed as described (See 3.4.1.6.2).
3.4.2.1.6, Partitioning chest wall volume
The cutter and JGLUNG2 programmes were used. The dividing line on each photograph
was defined as the visible lower costal margin. If this was not clearly seen, the
marker lines (See 3.4.2.1.2) were used as a guide to "best guess" its position at
a given stage in expiration. The costal margin was digitised separately using the
CUTTER programme for each photograph analysed.
3.4.2.1,7, Data Handling
Chest wall volume, and its rib cage and abdominal components were now available at
8-12 points on the gastric pressure curve. The total volume expired during the
relaxation manoeuvre was calculated by subtraction of chest wall volume at FRC
from chest wall volume at TLC. Abdominal wall displacement was expressed as
visible volume of the abdominal compartment, greater than that at FRC. This was
plotted against pressure, in absolute terms ( Vabd), or as a percentage of the
total volume expired ( Vabd %VE). Total volume expired was derived optically, by
subtracting chest wall volume at FRC, from that at TLC.
102
T L C ------------------------------------------------------------- -------------^
► F R C
Figure 3.10A sequence of photographs of the chest wall taken at 1 second intervals, during a passive expiration from TLC to FRC.
102 a
3.4.2.1.8, Validation of the method
In order to test the reproducibility of the digitisation procedure, a single
photograph was digitised on 6 separate occasions (See Appendix 1.4). The mean
volume of the abdominal compartment calculated was 2330.3ml, with 95% confidence
limits of 13.4ml (0.6%). 5 replicate analyses were then performed on a set of
photographs from a single manoeuvre, in one patient (Figure 3.11). The least
squares quadratic curve best fitting the data (See Appendix 1.5) was calculated,
as was the standard error of the estimate. This standard error of the estimate
was +27ml.
In view of the time consuming nature of the digitisation procedure, only one set
of photographs corresponding to a single relaxation manoeuvre was digitised for
each subject. An assessment was made of the reproducibility of the manoeuvre in
one subject. These data support the validity of analysing a single manoeuvre (See
Appendix 1.6). Figure 3.12 shows 5 attempts by a normal subject to reproduce the
passive expiration manoeuvre. The least squares quadratic curve fitting these
data had a standard error of the estimate of + 37 mis.
103
500
Figure 3.11Reproducibility of the digitisation procedure. 1 manoeuvre digitised 5 times with the quadratic curve of best fit.SEE = Standard error of the estimate, Pg = Gastric pressure, iXvabd = Volume-Displacement of the abdominal wall above that at FRC.
A va b d
(m l)
P g (cm H O)
Figure 3 . 12Reproducibility of the "Relaxation Manoeuvre". The points are from 5 manoeuvres performed by one normal subject. The quadratic curve of best fit is shown. SEE = Standard error of the estimate Pg = Gastric pressure, Avabd = Volume-Displacement of the abdominal wall above FRC.
105
3.4.2.2, The effects of loading expiration on chest wall motion in tetraplegic
patients
Method for chapter 6
3.4.2.2.1, Introduction
The aim of this study was to measure the compliance of the total respiratory
system, using a variation of the weighed spirometer technique of Heaf and Prime
(1956), and to partition this into rib cage and abdominal components. Expiration
was loaded with a series of expiratory threshold resistances, and expiratory
airway pressure was measured (See 3.4.3.2). Chest wall volume was measured at end
expiration, using the optical contour mapping system, and partitioned into rib
cage and abdominal compartments (See 3.4.1.6.4, 3.4.2.1.6). The volumes of the
chest wall and its components were plotted against airway pressure in 8
tetraplegic patients.
Each subject was studied supine under the optical contour mapping system. It was
used as described (See 3.4.2.1), with the exception of the timing of triggering of
the camera.
3.4.2.2.2, Apparatus (Figure 3.13)
The optical contour mapping system has been described in detail (See 3.4.1).
Subjects breathed through a system designed to display ventilation and airway
pressure, and to allow the addition of a series of expiratory threshold
resistances (ETR). This allowed the phase of respiration to be matched to
synchronous airway pressure. The system (Figure 3.13) consisted of a mouthpiece
attached to a pneumotachograph (See 3.4.3.3) with a side port to measure airway
pressure (See 3.4.3.2). The pneumotachograph was connected to a 2-way valve, the
dead space from mouthpiece to atmosphere via the inspiratory limb was 130ml. The
expiratory limb was attached via a 2m length of elephant tubing to a 2 way valve.
One limb of this was open to the atmosphere, and the other was adapted to fit a
series of ETR valves. The resistances used were commercially available, spring
loaded positive end expiratory pressure valves (Medic-aid), with manufacturers
specifications to provide a threshold resistance of 2.5 cm H20, 5.0 cm H20, 7.5 cm
H20, 10.0 cm H20 and 12.5 cm H20.
3.4.2.2.3, Protocol
Subjects were positioned under the optical contour mapping system, and breathed
through a mouth piece. The 2-way valve was disconnected from the rest of the
apparatus and subjects were allowed to equilibrate on the remaining system. They
were asked to relax, but were informed that there would be some resistance to
breathing out. If at any time they felt distressed by this, they were told to
come off the mouthpiece.
The operator observed the trace of ventilation and the contours formed on the
subject, and manually activated the camera trigger at FRC. The camera trigger
also connected directly to the paper recorder and as each photograph was taken, a
vertical mark was made on the trace. Each photograph could then be matched
against airway pressure and position within the respiratory cycle (Figure 3.14).
Photographs were taken at FRC over 10 consecutive breaths on each run, and the
first 5 discarded. From the remaining 5, the one taken closest to FRC was
selected for analysis. This selection was based on the proximity of the camera
trigger mark to end expiration on the ventilation trace.
The photograph was then digitised and analysed to give chest wall volume, and its
107
spring loaded valve
pneumotachograph
pressuremonitoring
port
Figure 3.13Diagramatic representation of the apparatus used to apply an expiratory threshold resistance to breathing in tetraplegic patients. There are 3 ports from the pneumotachograph.
107 a
rib cage and abdominal components. These values calculated when the subject was
breathing thorugh a pneumotachograph alone, were defined as volume at resting FRC
and used as a reference zero.
This protocol was initially performed using a circuit incorporating the
pneumotachograph alone. The expiratory limb of the circuit was then added,
followed by the addition of the series of resistances in order of magnitude
starting with the smallest. Subjects were allowed to come off the mouthpiece and
rest between each set of recordings. Before the addition of each resistance to
the circuit, the subject was allowed to take up to five breaths with the
expiratory limb open to the atmosphere.
Chest wall, rib cage and abdominal volumes relative to those at FRC, were
expressed as a fraction of predicted TLC (Cotes 1979) for each patient, and
plotted against expiratory airway pressure. The gradient of these plots represent
the specific compliance of the total respiratory system (SCrs), and its rib cage
and abdominal components. (See Chapter 6 for results and discussion).
108
5cm H O 2airway pressure
ventilation
camera trigger
i r T
'V
-------- i5se c
Figure 3.14A typical trace of airway pressure, ventilation, and the camera trigger, recorded from a tetraplegic patient breathing through an expiratory threshold resistance.
108a
3.4.3, PRESSURE MONITORING TECHNIQUES
3.4.3.1, Introduction
Pressure measurements of some form were made in all the studies reported in this
thesis. Each technique will be described individually and reference made to the
experiments in which it was used. Measurements of trans-diaphragmatic and mouth
pressures, form the major part of the work in Chapter 7, and the experimental
protocol used, will be described at the end of this section.
3.4.3.2, Airway pressure (Pa)
(See 3.4.2.2.2 - Methods for Chapter 6)
Pa was measured at a side port, on a conical pneumotachograph with a diameter of
3cm at the sampling point. The objective was to assess the pressure within the
system, when a series of expiratory threshold resistances were added. In view of
the large cross sectional area of the gas stream, it is reasonable to assume that
the Bemouilli effect was negligible, and that pressure recorded related to flow
resistive pressure change (Milic-Emili 1984). The port was connected by
polyethylene tubing (internal diameter 1mm) to a pressure transducer
(Elema-Schonander EMT 35). The signal was amplified in two stages, both with gain
controls. Firstly, by the amplifier supplied with the transducers (Elema
Schonander EMT 32C), and then by a purpose built differential amplifier (Stoke
Mandeville Hospital, Medical Electronics). The resulting signal was displayed on
a 4 channel thermal pen recorder (Gould 4400). The amplifier settings used to
measure Pa provided a signal linear +_ 0.2cra H2O to 60cm H2O. The system was
calibrated with a water manometer before and after each study.
109
3.4.3.3, Measurement of air flow
(See 3.4.5.2.3 - Methods for Chapter 4)
(See 3.4.2.2.2 - Methods for Chapter 6)
Airflow was measured and the signal integrated to provide a volume trace. The aim
was to display the respiratory cycle, such that events could be timed against it.
Airflow was measured with a conical pneumotachograph. In the EMG study (Chapter
4) during which subjects hyperventilated, a pneumotachograph accurate to flow
rates of 300L/minute was used. In the study on the effect of loading expiraton
(Chapter 6), during which subjects breathed at tidal volumes, a pneumotachograph
accurate to flow rates of 150L/minute was used. The pressure signals generated in
both studies were handled in the same way. The outlet ports of the
pneumotachograph were connected by 0.5cm plastic pressure tubing to the two inlets
of a differential pressure transducer (Elema-Schonander EMT 32c). The signal
generated was processed by a purpose built integrating amplifier (Stoke Mandeville
Hospital, Medical Electronics). The time lag of the integrating amplifier in
response to a square wave was 0.01 seconds. The volume signal was displayed
either on a channel of the thermal pen recorder (Gould 4400 - Chapter 6) or on the
oscilloscope of the EMG module (Medelec MS6 - Chapter 4). The system was
calibrated with a 1L syringe before and after each study.
110
3.4.3.4, Gastric Pressure (Pg)
(See 3.4.2.1.2 - Methods for Chapter 5)
(See 3.4.3.6, 3.4.4 - Methods for Chapter 7)
Pg was measured to reflect intra-abdominal pressure (Agostoni and Rahn 1960).
There is evidence to suggest that there are regional variations in intra
abdominal pressure (Decramer et al 1983) but at present these cannot be quantified
in man without unacceptably invasive procedures.
The naso-pharynx of each subject was anaesthetised with lignocaine spray. A 10cm
latex balloon mounted on a polyethylene catheter (PK Morgan) was passed via the
nose, 65 cm into the gastro-intestinal tract. The balloon containing 1.5ml of air
was connected via polyethylene tubing (internal diameter 1mm) to a pressure
transducer (Elema Schonander EMT 35). After amplification (See 3.4.3.2) the
signal was displayed on a channel of the thermal pen recorder (Gould 4400). The
position of the balloon in the stomach was confirmed by a positive pressure
deflection on sniffing.
For measurements of gastric pressure alone, (See 3.4.2.1.2) the amplification
system was set such that the signal was linear +_0.2cra H20 to 60cm H20. When used
in conjunction with oesophageal pressure to measure trans-diaphragmatic pressure,
(See 3.4.3.6) the amplification system was set such that, the signal was linear +
0.5cm H 20 to 200cm H20. In both studies the system was calibrated against a water
manometer before and after each experiment.
Ill
3.4.3.5, Oesophageal pressure (Poes)
(See 3.4.3.6, 3.4.4 - Methods for Chapter 7)
Poes was measured according to the method of Milic-Emili et al (1964). After
application of local anaesthesia (Lignocaine spray) to the naso- pharynx, a 10cm
latex balloon mounted on a polyethylene catheter (PK Morgan) was passed via the
nose, 45cm into the gastro-intestinal tract. The balloon containing 0.5ml of air
was connected by polyethylene tubing (internal diameter 1mm) to a pressure
transducer (Elema-Schonander EMT 35). After amplification (See 3.4.3.2) the
signal was displayed on a channel of the thermal pen recorder (Gould 4400). The
signal was linear +0.5cm H 20 to 200cm H20. The position of the balloon above the
diaphragm was confirmed by a negative pressure deflection on sniffing. The system
was calibrated before and after each study with a water manometer.
3.4.3.6, Trans-diaphragmatic pressure on maximal sniff (Sniff Pdi)
(See 3.4.4 - Methods for Chapter 7)
Pdi was measured by electrical subtraction of Poes from Pga measured using a
matching pair of pressure transducers and amplifiers (See 3.4.3.4, 3.4.3.5). The
frequency response to a square wave input to the pressure transducers showed that
the signal rose with a half time of 10.1ms with no detectable phase or amplitude
difference between gastric or oesophageal pressure responses.
Sniff Pdi was chosen as a reproducible and accurate measure of diaphragm strength.
Miller et al (1985) demonstrated that Sniff Pdi had a more discerning lower range
of normality (Normal Range 82 - 204 cm H20) than the alternative method of
measuring Pdi during a PImax manoeuvre (Normal Range 16 - 164 cm H 2O). Sniff Pdi
was also more reproducible with a coefficient of variation over 3 days of 7.2%,
compared with 13.0% for PImax Pdi.
Each subject was asked to perform a single short sharp sniff from functional
residual capacity followed by 3 quiet breaths. The manoevure was repeated 10
times and the best sniff used for analysis. Pdi at resting end expiration was
used as a reference zero. Each subject underwent a training period of 2 sets of
10 sniffs, during which they could see the recordings and were encouraged to make
maximal efforts. In the formal study, they were not allowed to see their results.
The Sniff Pdi value measured was the tallest peak of pressure obtained from the
best of 10 sniffs. Tetraplegic subjects were initially studied in their
wheelchairs, and then after transfer to a tilt table, either supine or at 70 tilt
to the horizontal. In order to ensure that the oesophageal balloon provided
accurate recordings of oesophageal pressure in different postures, an "occlusion
manoeuvre" was performed after each change in position (Baydur et al 1982). The
subject made a maximal static inspiratory effort with the glottis open, against an
occluded airway (See 3.4.3.7). If the position of the oesophageal balloon was
satisfactory, then Pa equalled Poes. When this was not the case, the position of
the oesophageal balloon was adjusted accordingly.
3.4.3.7, Maximum static inspiratory mouth pressure (PImax)
PImax is an indication of total inspiratory muscle strength. It was measured
according to the method of Black and Hyatt (1969). Each subject wore a nose clip
and breathed through a mouth piece, to which a shutter system was attached. This
incorporated a port for pressure measurement, and a standard 2mm leak to keep the
glottis open. The patients made maximum inspiratory efforts from FRC, sustained
for l-2s against a closed shutter. Mouth pressure at the port was measured via
polyethelene tubing (diameter 1mm) attached to a pressure transducer/amplifier (SE
113
Labs Sem 425), linked to a thermal pen recorder (Gould 4400). The signal was
linear +_ 0.4cm H2O to 100cm H20. This system was calibrated before and after each
study against a water filled manometer.
Subjects performed a training period of 2 sets of 3 maximum inspirations, while
observing their recordings, and were encouraged to make maximal efforts.
Subsequently they performed sets of 3 maximum inspirations blind, and the best
result from each set was used for analysis. The PIraax value measured was the
maximum inspiratory pressure sustained for 1 second.
114
3.4.4, EXPERIMENTAL METHODS FOR CHAPTER 7
The effect of abdominal binders on breathing in tetraplegic patients
The aim of this study was to examine the effects of 2 different abdominal binders
on breathing, in a group of 7 tetraplegic patients.
3.4.4.1, Study positions
Each subject was studied in 3 different positions. Sitting in their wheelchairs,
lying horizontal and supine on a tilt table, and resting at 70° to the horizontal
on the same tilt table. Patients were held to the table with straps at the knees
and at the pelvis below the binder. Their legs were braced with their feet on a
foot plate, and a safety strap was placed loosely around the chest, which did not
encroach on rib cage motion, unless the patient tipped forward.
3.4.4.2, Abdominal binders
Each subject was provided with a conventional abdominal binder appropriate to his
size (Figure 3.15). This binder was made of 70% viscose, 20% cotton and 10%
elastidiene (Credelast). It was 20cms in width, came in 3 different lengths and
fastened at the front with velcro. The newly designed binder (Fig 3.15) was
tailor made to each subject and was fitted in the supine posture. The front plate
was made of low temperature thermoplastic (Orthoplast), cut and moulded to fit
beneath the lower costal margin and above the boney margin of the pelvis. The
binder was held in place by cotton webbing riveted to the front plate, which was
wound around the body and fastened in front with velcro.
115
OLD NEW
Figure 3.15A conventional elastic binder used in tetraplegic patients, compared to the newly designed binder made of low temperature thermoplastic. Note the old binder encroaches on the lower rib cage.
116
3.4.4.3, Abdominal girth
Abdominal girth was measured at the mid point between xiphistemum and pubis, in
each patient, in all postures. Changes in this measurement with posture were
taken as an index of abdominal wall compliance. Similar measurements were made on
a group of normal volunteers matched for age and physical stature (See Table 7.3).
In order to standardise the degree of abdominal compression provided by each
binder, girth was reduced by a similar amount by both binders in each position.
This reduction was defined as the maximum compression that could be comfortably
applied by the conventional binder.
3.4.4.4, Protocol
Three indices of respiratory ability were measured.
1) Sniff Pdi
2) PImax
3) Vital Capacity (VC)
Sniff Pdi was measured (See 3.4.3.6) initially with the patients in their
wheelchairs. The order in which the conventional binder, the new binder and no
binder were studied was randomised for each posture. The patient was then
transferred to the tilt table, and studied either supine or at 70° tilt to the
horizontal in a random order. The pressure monitoring balloons were then removed.
Gastric pressure was measured separately from Pdi, at resting FRC and at tidal
inspiration. The mean values calculated from measurement of 10 consecutive tidal
breaths were recorded and analysed statistically. Gastric pressure during maximal
sniff, was that measured during the sniff producing the greatest Pdi.
117
Plmax was measured (See 3.4.3.7) with the patient supine or at 70 tilt according
to the randomisation. They were then transferred to their wheelchairs and the
measurement repeated. The order in which the binders were applied was again
subject to randomisation.
Vital capacity was measured after each set of Plmax measurements, using a single
breath wedge spirometer (Vitalograph) accurate to +2% at ATPS. Sets of 3 forced
expiration were performed and the best reading used for analysis.
The protocol was designed so that the patient was lifted on and off the tilt table
only once.
118
3.4.5, ELECTROMYOGRAPHIC METHODS
(Methods for Chapter 4)
3.4.5.1, Introduction
At the simplest level, the presence of EMG activity can be used to indicate
whether or not a muscle is being activated. The aim of the studies reported in
Chapter 4 was to examine the EMG activity of the individual muscles of the
abodminal wall during breathing. The recordings made were qualitative, and the
techniques used were simple. However, a new method had to be developed to make
recordings from the individual layers of muscle of the antero-lateral abdominal
wall. The normal practice is to record from a single needle placed in the muscle
group (Estenne and De Troyer 1985), or from surface electrodes (Campbell and Green
1953a).
3.4.5.2, Apparatus
3.4.5.2.1, Intra-muscular electrodes
Bipolar fine wire electrodes (Figure 3.16) were manufactured in the manner of
Basmajian and Stecko (1962) using 0.06mm stainless steel wire coated with diamel
insulation (Johnson & Mathey). Two 50cm lengths of wire were cut and 0.5mm of
insulation removed from one end with a scalpel. These ends were then fashioned
into hooks of staggered length, such that the bared wires could not short circuit
(Figure 3.17), The unhooked ends of the wires were passed through a 27 gauge
needle, till the hooks rested on the bevel. A 2cm length of the free end was
bared of insulation with a methylated spirit flame. The whole electrode was then
sterilised.
.119
Figure 3.16Upper panel, The "Medelel MS6" EMG Module.Middle panel, A Bipolar five wire electrode.Lower panel, Springs connecting the electrode to the pre-amplifier.
12 0
IN SU LA TED W IRE
Figure 3.17Detail of a bipolar fine wire electrode, to illustrate that the bared metal tips cannot short circuit.
121
Fine wire electrodes can be easily and painlessly placed in muscle with a small
gauge needle, which is then removed. The hooks anchor the electrode in the
muscle, and in these studies they were further secured with tape at the skin. The
wires pick up electrical activity from muscle within 1.5mm of their tips. When in
situ they cause little discomfort, and allow the subject freedom of movement.
They are easily removed with a sharp pull, and do not disrupt muscle because of
their small calibre.
3.4.5.2.2, Oesophageal electrode
These are a form of surface electrode used to make recordings of the electrical
activation of the diaphragm. The bipolar oesophageal electrode used in these
studies consisted of 2 insulated wires contained in a closed and fully insulated
plastic tube. At the tip the 2 wires connect to separate hollow metal rings
moulded onto the surface of the electrode. No anchoring balloon was used, as a
stable baseline was achieved without such a device.
To pass the electrode, the nasopharynx was anaesthetised with lignocaine spray.
The electrode was then passed through the nose, 65cm into the gastro-intestinal
tract. It was gradually removed whilst recordings of electrical activity were
made, until the maximum diaphragmatic EMG signal was obtained. It was then
secured to the nose with tape.
3.4.5.2.3, EMG module
A "Medelec MS6" EMG module was used to make recordings (Figure 3.16). It was
fitted with 4 AA6 amplifiers, an oscilloscope and a light sensitive paper
recorder. The bare ends of the wire electrodes were connected to purpose built
insulated springs, which were plugged into 4 pre-amplifiers (Figure 3.16).
122
Recordings were made from muscle at a gain of 200 mA, the low filter was set at
16HZ and the high filter at 1.6KHZ. Continuous recordings were made at a paper
speed of 5cm/s. Diaphragm EMG was recorded with the oesophageal electrode using
the Same filter settings, but at a gain of 100 mA.
A trace of ventilation was displayed on the oscilloscope, using the method for
measuring airflow described (See 3.4.3.3).
3.4.5,3, Method for recording EMG from individual abdominal muscles
3.4.5.3.1, Location of muscle layers
Computerised tomographic (CT) scans of the abdomens of 20 patients, performed with
an Elscint 2002 whole body scanner at the Brompton Hospital, were analysed. These
patients were not selected, but were the first 20 to be referred for abdominal
scans from the beginning of the study. Their details and results are given in
Table 4.1. The scan taken nearest to the mid-point between xiphistemum and
pubis, was selected from a "scout" view. It was viewed at a window width of 450
HU and a centre setting of 40 HU.
The individual muscle layers of the antero-lateral abdominal wall were clearly
visible on the scan (Figure 3.18). The cursors available on the viewing module
were used as electronic calipers to measure the distance from the surface to each
muscle layer. The measurements were made at the point of maximum thickness of the
abdominal muscles. This was consistently just lateral to the circumferential
mid-point between the coronal and sagital planes. By using a metal marker this
point was located on a subject as 3cm medial to the mid-axillary line. The mean
distances from the surface of external oblique, internal oblique and transversus
abdominis were calculated (Table 4.2). A 27 gauge needle was marked at these
123
ZOOM£ .5 0
* * * BRQMPTQN HQ5 I
BP 365mm~ m
Cl IW 1 4 ;
Figure 3.18CT scan of the abdominal muscles. The three layers of the antero-lateral muscle group are clearly seen, as is rectus abdominis medially. The arrows represent electronic calipers. 3 is positioned at the skin, 2 at the surface of internal oblique and 1 at the surface of transversus abdominis.
124
distances from its tip. In a series of pilot studies the needle was inserted into
the antero-lateral abdominal muscles of 6 normal volunteers, mid-way between
xiphistemum and pubis, and 3cm medial to the anterior axillary line. As each
layer of fascia was penetrated it was possible to feel a decrease in resistance or
"give", and on some occasions there was reflex muscular contraction as the muscle
was penetrated. With experience, the operator became confident as to the position
of the needle inserted within the muscle layers.
3.4.5.3.2, Insertion of electrodes
In the formal studies, a bipolar fine wire electrode was placed into rectus
abdominis, 3cm to the left of the midline, mid-way between pubis and xiphisternum.
3 electrodes were placed within 0.5cm of each other at the surface, into the
antero-lateral muscle group. The entry point was located on the left, 3cm medial
to the mid-axillary line, mid-way between the xiphisternum and pubis. The
objective was to place an electrode into each of external oblique, internal
oblique and transversus abdominis. To do this, the electrodes were mounted in 27
gauge needles with a shaft length corresponding approximately to the expected
depth of the muscle. To insert the wires into external oblique a 1.3cm needle was
used, for internal oblique a 1.9cm needle, and for transversus abdominis a 2.5cm
needle.
3.4.5.3.3, Validation of technique
In order to demonstrate that the electrodes had been place correctly in each named
muscle, a series of manoeuvres were performed. These were designed to produce EMG
activity predominantly in a single muscle layer (See 1.5.1.1).
125
Table 3.1
Muscle Activated Movement
External Oblique Lift left shoulder and point
it towards right hip.
Internal Oblique Lift right shoulder and point
it toward left hip.
Transversus Abdominis Pull "belly" in.
Rectus Abdominis Lift both legs.
(This movement is not specific, but
activates rectus abdominis to ensure
the wire is in place. The electrode
is in a visibly different position
from the other 3 electrodes)
Results of the validation technique are reported in the next chapter (See
4.1.2.2).
3.4.5.3.4, Respiratory manoeuvres
Subjects performed a series of respiratory manoeuvres with the 4 intra-muscular
electrodes and an oesophageal electrode in place. Initially they breathed at
resting tidal volume; they then voluntarily increased the rate and depth of their
breathing up to maximum hyperventilation. They also performed cough, "strain" and
Valsalva manoevures. "Strain" was defined as expiratory effort against a closed
126
glottis. The EMG activity of the abdominal muscles and diaphragm during these
manoeuvres is displayed in Chapter 4.
3.4.5.4, Validation of relaxation manoeuvre (Chapter 5)
A relaxatibn manoeuvre has been described (See 3.4.2\1.3), during which abdominal
wall compliance was measured. In normal subjects compliance would be altered by
activity of the abdominal muscles during the manoeuvre. In order to confirm that
that the abdominal muscles were relaxed, EMG recordings were made as described
above. A single bipolar fine wire electrode was placed in the antero-lateral
abdominal muscle group of the 6 subjects examined. When they performed the
relaxation manoeuvre described, no abdominal muscle EMG activiy was detected.
3.4.6, STATISTICAL METHODS
All statistical analyses were carried out on the Brompton Hospital computer (Prime
750). The arithmetic mean, standard deviation, and standard error of the mean are
used to describe the results of patient and control groups.
The Kolmagorov-Smirnoff test was applied to group data to assess whether it
differed significantly from a normal distribution. Wilcoxon ranked sum tests were
used to compare non-parametric data, and Student's T tests to compare parametric
data.
Linear regression lines were calculated using the method of least squares.
Correlation coefficients were calculated when appropriate. Least squares
quadratic regression curves were plotted for curved data, and the standard error
of the estimate calculated.
127
3.5, SUMMARY
This chapter has described the methods used to study normal subjects and
tetraplegic patients. The data gathered will be presented in the following
chapters.
128
CHAPTER 4
ELECTROMYOGRAPHIC STUDIES OF THE ABDOMINAL MUSCLES
This chapter is divided into 2 studies ih which the same method has been used.
The first study deals with normal subjects and the second with tetraplegic
patients.
4.1, Study 1
Normal Subjects
4.1.1, Introduction
The classical studies of the EMG activity of the abdominal muscles have used
surface electrodes (Floyd and Silver 1950), or needle electrodes placed in rectus
abdominis and external oblique muscles (Campbell 1952). The EMG activity of
external oblique during breathing is assumed to be representative of the
antero-lateral muscle group as a whole. In view of the differences in the origins
and insertions of the 3 layers of muscle which make up this group, it was decided
to test this assumption. The close proximity of the 3 muscle layers has made
separation of their electrical activity difficult, though attempts to do this have
been made (Carman et al 1972, Strohl et al 1981). The aim of this study was to
develop a method (see 3.4.5.3) of sampling EMG activity from the individual
abdominal muscles, and then to apply it during breathing.
129
4.1.2, Results
4.1.2.1, Measurements of skin to muscle distance by CT scan (See 3.4.5.3.1)
20 normal subjects were studied, their details and the measurements made are
displayed in Table 4.1. The mean data for the whole group and for each sex are
displayed in Table 4.2. Using an unpaired Student's T Test, there was no
significant difference in any of the measurements made, between men and women.
The approximate mean muscle depths taken for marking a "pilot" needle were:
External Oblique 10mm
Internal Oblique 15mm
Transversus Abdominis 20mm
4.1.2.2, Validation of Method (See 3.4.5.3.3)
A series of movements were performed by each subject (Table 3.1), to test the
theory that the electrodes were placed in separate muscles. Four electrodes were
placed in the abdominal wall of 6 normal subjects. Their details, and the results
of the validation procedure are displayed in Table 4.3. The manoeuvres designed
were successful in eliciting specific EMG activity from individual muscles. This
is illustrated in Figures 4.1, 4.2, 4.3 and 4.4. A total of 24 electrodes were
inserted, of these the evidence suggested that 20 were positioned in the muscle
for which they were aimed. In subject 4, it was not possible to distinguish
between the EMG activity from the putative external oblique and internal oblique
electrodes. In subject 6, the EMG activity from the electrodes aimed for internal
oblique and tranuversus abdominis could not be distinguished. In the latter
subject, placement of the transversus abdominis electrode had caused some
discomfort: this may be why the procedure failed. An ultrasound scan of the
subjects abdominal muscles showed the anatomy to be normal (Figure 4.5).
130
Table 4.1
Measurements of the distance from skin surface to individual abdominal muscle
layers in 20 normal subjects (EO = External oblique, 10 = Internal Oblique, TA
Transversus abdominis).
Measurements (mm)
Sex Age Surface to Surface to Surface to
(yrs) EO 10 TA
F 75 10 15 21
M 60 13 18 23
F 49 10 15 22
M 58 7 10 22
F 54 14 20 24
F 59 8 13 20
M 71 11 16 22
F 41, 6 12 16
F 50 13 17 23
F 70 8 12 20
M 69 5 11 21
M 64 8 13 19
M . 58 9 16 24
F 58 10 18 23
M 62 7 14 21
F 33 6 10 14
F 54 8 13 18
M 75 13 20 24
M 64 6 12 22
M 32 11 16 25
131
Table 4.2
Males
Females
All
Mean data for measurements of the distance from skin surface
to individual abdominal muscle layers in 20 normal subjects.
Measurements (mm)
Age
(x + SD)
(yrs)
62.4 + 11.7
54.3 + 12.4
58.1 + 12.4
Surface to
External
Oblique
(x + SD)
9.0 + 2.9
9.3 + 2.7
9.15 + 2.6
Surface to
Internal
Oblique
(x + SD)
14.6 + 3.2
14.5 + 3.1
14.55 + 3.1
Surface to
Transversus
Abdominis
(x + SD)
22.3 + 1.7
20.1 + 3.2
21.2 + 2 .8
132
Table 4.3
Results of the validation procedure for placement of fine wire
electrodes into the abdominal muscles of 6 normal subjects.
(+ = satisfactory placement)
Subject Sex Age External Internal Transversus Rectus
Oblique Oblique Abdominis Abdominis
1 M 27 + + + +
2 M 21 + + + +
3 F 31 + + + +
4 M 35 - - + +
5 F 32 + + + +
6 F 25 + — — +
133
»*■ H ■ ■ i»- «
f A
R A" ' ‘ r " - r-k -..... r
Figure 4.1EMG recordings from electrodes placed in the 4 individual abdominal muscles, on lifting the left shoulder and pointing it towards the right hip. There is EMG activity from the electrode placed in the external oblique muscle, which is distinguishable from recording from the other muscles.
134
EO
J A*+mrn W M l 1 .. . ■ » il f 0#»w
R A "rr
Figure 4.2EMG recordings from electrodes placed in the 4 individual abdominal muscles, on lifting the right shoulder and pointing it towards the left hip. There is EMG activity from the electrode placed in the internal oblique muscle, which is distinguishable from recordings from the other muscles.
135
Figure 4.3EMG recordings from electrodes placed in the 4 individual abdominal muscles, on ’’pulling the belly in". There is EMG activity from the electrode placed in the transversus abdominis muscle, which is distinguishable from recordings from the other muscles.
136
Figure 4.4EMG recordings from electrodes placed in the 4 individual abdominal muscles, on raising the legs. This confirms that the electrode placed in rectus abdominis is working. It is positioned distant from the other 3 electrodes, and so must be recording a signal from a separate muscle.
137
Figure 4.5Ultrasound scan of the abdominal muscles of a subject in whom EMG activity from the 4 electrodes supposedly placed in individual muscles could not be distinguished.The figure illustrates normal anatomy. a = external oblique, b = internal oblique, c = transversus abdominis, d = rectus abdominis.
138
4.1.2.3, Respiratory manoeuvres (See 3.4.5.3.4)
In all 6 subjects the abdominal muscles were inactive during quiet breathing in
the supine posture. As ventilation increased, phasic expiratory EMG activity of
the abdominal muscles was detected (Figure 4.6). This activity was similar in
timing and duration in all 4 abdominal muscles examined (Figure 4.7). The
amplitude of the EMG signal increased with minute volume. At peak ventilation,
late inspiratory abdominal muscle EMG activity was recorded (Figure 4.8). During
cough, strain, and Valsalva manoeuvres all 4 abdominal muscles acted together
(Figure 4.9).
4.1.2.4, Effects of posture
Identical respiratory manoeuvres were carried out on a tilt table by each subject,O Oat 10 tilt head down to the horizontal, and 40 tilt head up to the horizontal.
The pattern of abdominal muscle activation was identical but the minute volume at
which EMG activity was first detected varied (Table 4.4). The minute volume was
calculated from the ventilation trace, using the volume and duration of the breath
at which EMG activity was first detected. Wilcoxon ranked sum tests showed that
the minute volume at which the abdominal muscles became active was significantly
greater at 10° head down, than flat (p<0.05), and at 10° head down than at 40Ohead up (p<0.05). There was no significant difference between the flat and 40
head up postures.
4.1.3, Discussion
These results support the hypothesis that it is possible to study the EMG activity
of the individual muscles of the abdominal wall. Ideal ways of validating the
method of electrode insertion described, would have been to image the wires within
139
Y,
Figure 4.6EMG recordings of the abdominal muscles and diaphragm of a normal subject with a display of ventilation (upward deflection represents inspiration). There is normal phasic expiratory EMG activity recorded from the abdominal muscles.
140
141
If H 1 iNfrf 'll 1' 1 11 11 110— H » H * ----#***-----— •#*«--------##■*
| Jufrw.. I lnwf
Figure 4.7EMG recordings from electrodes placed in the 4 individual abdominal muscles of a normal subject. Phasic expiratory EMG activity is of similar onset and duration in all 4 channels.
142
IURA
- ■— ...... ■--- --i
— - »»■■■■— ---ihi--In*-i i n r i f- f* i r r; r t r
J?*— —i— — --- !̂ "~]----------------------- - ‘ f ' f §—:--- Ii____________[
1 s e c
Figure 4.8EMG recordings from the abdominal muscles of a normal subject with a display of ventilation (upward deflection represents inspiration). At increased minute volume early inspiratory EMG activity is visible from the 3 abdominal muscles. It is of similar onset and duration in all 3 channels.
Figure 4.9EMG recordings from electrodes placed in the 4 individual abdominal muscles of a normal subject. On coughing the EMG activity from each muscle is similar, in onset and duration.
143
Table 4.4
Minute volume (L/min at which expiratory abdominal muscle
EMG activity was first detected in different postures.
ibject 10° head down Flat 40° he.
1 123 101 90
2 90 78 69
3 119 100 78
4 86 77 76
5 107 98 94
6 83 75 84
144
the muscle layers, or to dissect the muscles and observe them directly. The
former method was attempted using both CT Scanning and high resolution ultrasound.
The fine caliber of the wires and their insulation, made them invisible to these
imaging systems. The use of a cadaver to practice placement of electrodes and
subsequent dissection to confirm their position was considered. This alternative
was rejected because the "feel” of the technique in post mortem specimens was
thought to be different, and results could not be extrapolated to live volunteers.
It was therefore decided to validate the technique indirectly by using manoeuvres
designed to elicit EMG activity predominantly from a single muscle. Since the
electrodes were placed within 5mm of each other at the surface, the separation of
EMG activity as shown in figures 4.1 to 4.4 was considered good evidence that the
electrodes were in different muscles. Twenty out of 24 electrodes were
successfully inserted into the muscles for which they were aimed, with the other 4
displaced by only one muscle layer. This provided an adequate opportunity to
study the activation of individual abdominal muscles during breathing.
All the subjects studied were able to hyperventilate vigorously without
discomfort, or displacement of the wire electrodes. This is not always the case
with needle electrodes. Campbell and Green (1953b) described the EMG activity of
the abdominal muscles during breathing using surface electrodes. They noted
greater EMG activity during voluntary increases in ventilation, than during
rebreathing carbon dioxide. In order to faciliate the detection of any
differences in activity between individual muscles, recordings were made during
voluntary hyperventilation. No new observations were made on the timing of phasic
expiratory EMG activity of the abdominal muscles, or late inspiratory activity at
peak ventilation.. There was no detectable difference between the timing or
duration of activity of the individual abdominal muscles. This suggests that the
role of the individual abdominal muscle is primarily to alter the position of the
thorax relative to the pelvis, but that they act together during breathing.
145
Posture influenced the level of ventilation at which the abdominal muscles were
recruited for expiration. In this study the levels of ventilation at which they
were activated, were greater than during similar studies during carbon dioxideOrebreathing (Campbell and Green 1953b). At 10 head down, the effect of gravity
on the abdominal contents is expiratory, pushing the diaphragm into the thorax.
In this posture, the added expiratory effect of the abdominal muscles is not
necessary until a higher rate of ventilation, compared to when subjects lie flat.
This type of analysis is also applicable when comparing the 40° head up and flat
postures. At 40° head up, the effect of gravity is inspiratory, thus the
abdominal muscles are recruited to assist expiration at a lower level of
ventilation to oppose the gravitational pull. Comparison of the data from the 40°
head up and flat postures failed to reach statistical significance. In the 40°
head up position the level of ventilation at which the abdominal muscles were
recruited was lower than in the flat position in 5 out of 6 subjects. In the
remaining subject the reverse occurred. This finding may have been due to
variability in the voluntary pattern of hyperventilation.
4.1.4, Summary
1. It is possible to record EMG activity from the individual abdominal
muscles
2. These muscles have distinguishable effects on trunk movement, but act
in concert during breathing.
3. The level of ventilation at which the abdominal muscles become active
in expiration varies with posture.
146
4.2, Study 2
Tetraplegic patients
4.2.1, Introduction
In 1966, Guttmann and Silver detected phasic inspiratory EMG activity in the
abdominal muscles of stable tetraplegic patients. Silver and Lehr (1981b) were
unable to reproduce this finding. More recently, Estenne and De Troyer (1985)
reported no EMG activity from the abdominal muscles of 20 tetraplegic patients
seated and breathing at rest. The aim of this study was to apply the method of
examining abdominal muscle EMG activity described in this thesis, to patients with
tetraplegia.
4.2.2, Results
8 tetraplegic patients were studied (Table 4.5), 5 also consented to the
introduction of an oesophageal electrode to monitor diaphragm EMG.
Tonic EMG activity could be detected from the abdominal muscles when mass spasm
was voluntarily induced. This was achieved by moving a finger firmly across the
skin of the abdomen or inside of the thigh. In each patient, this manoevure
demonstrated that at least 3 electrodes were recording satisfactorily. Figure
4.10 shows tonic EMG activity recorded by 4 wire electrodes. The difference in
the pattern and duration of EMG activity from individual wires may indicate that
they are placed in separate muscles.
In all 8 patients examined, phasic inspiratory EMG activity of the abdominal
muscles was detected (Figure 4.11). This was not present during tidal breathing,
147
Table 4.5
Tetraplegic patients studied electromyographically.
Patient Sex Age Level Time after Oesophageal
(yrs) of injury electrode
Lesion
1 M 21 C5,6 7 -
2 M 25 C6 4 +
3 M 21 C5 9 +
4 F 31 C6 6 -
5 M 24 C5 6 +
6 M 34 C6 3 +
7 M 33 C5,6 10 +
8 F 35 C6 8
148
149
E O
10H M - +-U .-U .— .n . ; AAj uu^A^ur ri(iii:in,M̂ .u u u ---- -
• i l I V i V V f 1 ' w , * ' * ',, . ! ‘Jii’l'i ’ ’ ’
i--------------- 11 s e c
Figure 4.10EMG recordings from the 4 individual abdominal muscles of a tetraplegic patient during muscle spasm. The recordings from each muscle appear different.
150
V
Abd■** > ^ «| <■ ■ t1 ■ i M - H +4-
Dia
t
i----------------11 s e c
Figure 4.11EMG recordings from the diaphragm and abdominal muscles of a tetraplegic patient with ventilation displayed (upward deflection represents inspiration). There is phasic inspiratory EMG activity in the abdominal muscles.
but became apparent as ventilation increased (Figure 4.12). The amplitude of the
EMG activity increased with minute volume and in subjects 2 and 6 developed into
tonic activity, clinically associated with spasm of the abdominal muscles (Figure
4.13).
4.2.3, Discussion
It was not possible to ascertain whether the electrodes were placed in separate
abdominal muscles, since patients could not perform the manoeuvres used in Study 1
(see 3.4.5.3). The objective when placing the electrodes was not to examine
individual abdominal muscle activation, but to make good quality EMG recordings
from the abdominal wall muscles.
There was good evidence that the phasic inspiratory EMG activity detected was
derived from the abdominal muscles themselves, and was not radiating through from
the diaphragm. Firstly, the fine wire electrodes used, only detect electrical
activity within 1.5mm of their tips; secondly, it was possible to demonstrate
tonic EMG activity in the abdominal muscles, whilst normal phasic inspiratory EMG
activity was present in the diaphragm (Figure 4.14); thirdly, on straining,
intense tonic EMG activity was present in the diaphragm due to isometric
contraction, whilst there was no EMG activity in the abdominal muscles (Figure
4.15). It was therefore possible to demonstrate high amplitude tonic EMG activity
separately in the abdominal muscles and diaphragm. This strongly suggests that
the phasic inspiratory EMG activity detected in the abdominal muscles arises
locally.
The cause of this phenomenon is unknown, but may be considered with reference to
chest wall mechanics (see 1.4.2). When the diaphragm descends on inspiration, the
abdominal wall is displaced outwards and the abdominal muscles are stretched. The
151
152
i--------- *
1 s e c
Figure 4.12EMG recording from the diaphragm and abdominal muscles of a tetraplegic patient with ventilation displayed (upward deflection represents inspiration). There is no phasic inspiratory activity from the abdominal muscle during tidal breathing. As minute volume increases this activity becomes apparent.
153
•vm'F m * — ------------ < ■ •****+ ,------------ - ------- ------ - tt++m***
I------------- 11 s e c
Figure 4.13EMG recording from the diaphragm and abdominal muscles of a tetraplegic patient with ventilation displayed (upward deflection represents inspiration). Phasic inspirato.ry EMG activity from the abdominal muscles develops into tonic activity, which was clinically associated with abdominal muscle spasm.
154
i------------- -i1 s e c
Figure 4.14EMG recording from the diaphragm and abdominal muscles of a tetraplegic patient, with ventilation displayed (upward deflection represents inspiration). Tonic EMG activity is present in the abdominal muscles, whilst normal phasic inspiratory activity is present in the diaphragm.
155
Abd
1-------------1 s e c
Figure 4.15EMG recording from the diaphragm and abdominal muscles of a tetraplegic patient, during ’’straining". There is tonic- EMG activity in the diaphragm, but no synchronous activity in the abdominal muscles.
EMG activity detected within these muscles may be due to a strech reflex mediated
at spinal level. The stimulus would be stretching of the abdominal wall, thus
activating muscle spindles and resulting in reflex abdominal muscle contraction.
In practice, it was observed that the presence and amplitude of phasic inspiratory
EMG activity related to the degree and rate of abdominal wall distension.
Patients 2, 5, and 6 had noted prior to this study that rapid deep inspiration
induced spasm in their abdominal muscles. This was shown to be initiated by
inspiratory EMG activity from the abdominal muscles in 2 of the 3 (Figure 4.13).
An increase in tone of the abdominal muscles may decrease the compliance of the
abdominal wall. This in turn increases the resistance to descent of the diaphragm
on inspiration, and leads to increased lower rib cage inflation (See 1.5.2.2).
The compliance of the abdominal wall is important in determining the rise in
intra-abdominal pressure on inspiration and hence the mechanical coupling of the
lower rib cage and diaphragm (See 1.5.2.3). In tetraplegic patients, direct
measurement of abdominal wall compliance has been hampered by technical problems
(See 2.5.1, 3.3). In this context, the optical contour mapping system described
may have advantages over conventional methods of studying chest wall motion. It
was therefore decided to use this system to make direct measurements of abdominal
wall compliance in a group of tetraplegic patients, and compare them with a group
of normal subjects. The results of this study are described in Chapter 5.
156
4.2.4, Summary
1) Phasic inspiratory EMG activity was recorded from the abdominal muscles
of 8 tetraplegic patients studied, during increased ventilation.
2) There is good evidence that the activity recorded, is initiated locally within
the abdominal muscles.
3) This phenomenon may decrease abdominal wall compliance.
157
CHAPTER 5
MEASUREMENT OF ABDOMINAL WALL COMPLIANCE
This chapter describes the measurements made of abdominal wall compliance in 6
normal subjects and 6 tetraplegic patients. (See 3.4.2.1)
5.1, Introduction
Many investigators have noted that the abdominal wall moves more during breathing
in tetraplegic patients than normal subjects. Bergofsky (1964), Mortola and S'Ant
Ambrogio (1978), and Estenne et al (1983) have suggested that it does so because
the rib cage is fixed or moves paradoxically. Danon et al (1979) suggested that
paralysis and disuse atrophy of the respiratory muscles might alter the
partitioning of chest wall compliance between rib cage and abdomen. The work
reported in this chapter was performed to measure the distensibility of the
abdominal wall during a relaxed expiration from TLC to FRC, in normal subjects and
patients with tetraplegia. It tested the hypothesis that the abdominal wall was
more easily displaced, and hence there was less resistance to descent of the
diaphragm, in tetraplegic patients than in healthy subjects. A consequence of
this would be to diminish lower rib cage expansion, which would add to the
respiratory problems of cervical cord injured patients.
5.2, Results
The 6 normal subjects studed had a mean age of 27 years, and the 6 tetraplegic
patients had a mean age of 28 years. Their details are displayed in Tables 5.1
and 5.2. The mean vital capacity of the normal group was 5.13L. The predicted
vital capacity of the tetraplegic group was 4.81L (Cotes 1979), their mean seated
158
Table 5.1 Normal subjects studied
ibj ect Sex Age (yrs) Vital Capa*
1 M 19 5.50
2 M 25 4.69
3 M 35 4.73
4 M 32 5.58
5 M 25 5.37
6 M 28 4.88
Table 5.2 Tetraplegic Patients studied
Patient Sex Age Level Time Predicted VC
(yrs) of after VC(L) % Predicted
Lesion inj ury
1 F 34 C5 8 2.93 29%
2 M 21 C5,6 7 5.03 40%
3 M 36 C7 4 4.97 65%
4 M 24 C6 4 5.23 40%
5 M 19 C5 7 5.24 56%
6 M 33 C5,6 10 5.45 64%
159
Figure 5.1 shows a pressure, volume-displacement curve typical of a normal
abdomen. The change in slope of this curve reflects the compliance of the passive
antero-lateral abdominal wall, at various levels of distension throughout the
manoeuvre. A similar curve was constructed for each normal subject and each
tetraplegic patient: The data used to construct these curves is displayed in
Appendices 1.1 and 1.2. The results are displayed in two forms, either with
abdominal displacement expressed as absolute volume (AVabd), or as a percentage of
the total volume expired during the manoeuvre (&Vabd %VE). Both forms use
abdominal volume at FRC as a reference zero.
All the pressure displaced-volume curves obtained in this study were of a similar
shape. At low lung volumes when intra-abdominal pressure is low, abdominal
displaced-volume rises rapidly, and the abdominal wall is at its most compliant.
At high lung volume, the rate of rise of abdominal displaced-volume decreases, the
pressure volume curve flattens off, and abdominal compliance is diminished.
Figure 5.2 shows single curves typical of each of 6 normal subjects. Gastric
pressure is plotted against absolute volume displacement of the antero-lateral
abdominal wall. The range of gastric pressure measured in supine normal subjects
was -0.5cm H20 to +11.4cm H20. during a relaxed expiration. At TLC gastric
pressure lay between +9.0cm H20 and +11.4 cm H2O, with a mean of + 10.1cm H20 (SD
= 1.1cm H20). On expiring to FRC, the decrease in volume displacement of the
abdominal wall varied between 176ml and 997ml, with a mean of 624ml. This range
was due in part to differences in the total volume expired. Figure 5.3 shows
abdominal wall displacement expressed as a percentage of expired volume. In
normal subjects, the mean abdominal displaced-volume was 31% (SD 7.2%) of the
total expired volume.
vital capacity was 49% of predicted.
160
Figure 5.1A typical gastric pressure/volume-displacement characteristic (PV characteristic) of the abdominal wall of a normal supine subject. A Vabd = Volume-displacement of the abdominal wall, above that at FRC.
161
162
Pg (cm H20 ) Pg (cm H20 )
Figure 5.2Gastric pressure/volume-displacement characteristic of the abdominal wall, in 6 normal subjects and 6 tetrapelgic patients. ^ V a b d = volume-displacement of the abdominal wall, above that at FRC
163
100 100
Figure 5.3Gastric pressure/volume-displacement characteristic of the abdominal wall, in 6 normal subjects and 6 tetraplegic patients.A Vabd(%VE) = Volume-displacement of the abdominal wall, above that at FRC, expressed as a percentage of the total volume expired.
The curves for the 6 tetraplegic patients are displayed in Figure 5.2. The
gastric pressure developed during a relaxed expiration ranged from -1.7cm H20 to
+12.5cm H20. Peak gastric pressure lay between +5.Ocm H20 and +12.5 cm H20, with
a mean of 8.7cm H20 (SD 3.1cm H20). This was less than in normal subjects but the
difference failed to reach statistical significance. The tetraplegic patients had
greater absolute volume displacements of the abdominal wall than normal subjects
(413ml to 2174ml), with a mean of 1219ml. When expressed in terms of percentage
of expired volume (Figure 5.3) the mean percentage volume partitioned to abdominal
wall displacement was 77% (SD 34%).
Figure 5.4 displays the mean data for the 2 groups, with abdominal
displaced-volume expressed as a percentage of the total volume expired. These
data suggest that the antero-lateral abdominal wall is displaced about twice as
easily in tetraplegic patients, ie it is twice as compliant as in normal subjects.
The pressure displaced-volume curves constructed in this study have been analysed
over the range of tidal breathing (Table 5.3). This range was defined by
measuring the mean rise in gastric pressure over 10 consecutive quiet breaths in
each subject. In normal subjects, the mean abdominal volume displaced over a
tidal breath was 314ml, while amongst the tetraplegic patients it was 477ml.
These data can be expressed as a percentage of the total volume displaced by the
chest wall during the relaxation manoeuvre. Normal subjects partition 16.5% (SD
8.6%) of total expired volume to abdominal wall displacement, while tetraplegic
patients partition 34% (SD 27%). The mean gastric pressure rise in the 2 groups
was identical at 2.4cm H20.
164
- 2 0 2 4 6 8 1 0 1 2
P g (cm H2 <0)Figure 5.4Gastric pressure/volume-displacement characteristic of the abdominal wall. Mean data for 6 normal subjects and 6 tetraplegic patients, standard error bars are displayed. Solid lines join data points derived from 6 subjects, broken lines join data points derived from less than 6 subjects. A Vabd(%VE) = Volume-displacement of the abdominal wall above that at FRC, expressed as a percentage of the total volume expired.
165
Table 5.3 Analysis of pressure displaced volume curves over the range
of tidal breathing. Abdominal wall displacement expressed
as absolute volume, and as a percentage of the total volume
expired.
Normal Subjects
Subject Rise in Gastric Abdominal Volume Abdominal Volume
pressure displaced displaced
(cm H2O) (ml) (% volume expired)
1 2.3 153 9
2 2.3 110 14
3 2.8 329 30
4 2.3 573 17
5 2.7 490 21
6 2.0
Tetraplegic Patients
233 8
Patient Rise in Gastric Abdominal Volume Abdominal Volume
pressure displaced displaced
(cm H20) (ML) (% volume expired)
1 1.9 764 90
2 3.3 303 15
3 1.4 754 22
4 2.8 231 22
5 2.3 327 22
6 2.9 481 33
166
5.3, Discussion
The mean vital capacity of the tetraplegic patients was 49% of predicted, which is
similar to a mean calculated from previously published data (Table 2.1) of 53%.
The mean vital capacity of the normal group and predicted vital capacity of the
tetraplegic patients were similar, despite the inclusion of a female patient.
Ideally, the 2 groups should have contained similar proportions of the sexes, but
with this reservation comparison seems valid.
Konno and Mead (1968) measured the static pressure volume characteristics of the
abdomen in 6 normal subjects, in the erect and supine postures. They related the
AP diameter of the abdomen to relaxed gastric pressure at different lung volumes.
In the supine posture, they found that gastric pressure at different lung volumes
tanged from 0-1Ocm H20, which is similar to the findings in this study. They
found "abdominal compliance" to be greatest at low lung volumes and to decrease at
high lung volumes. The shape of the pressure volume curves that they plotted was
thus similar to those described here. The decrease in gradient may be attributed
to stretching of the abdominal muscles at high lung volume, which causes a
decrease in elasticity and hence compliance. At active TLC, Konno and Mead (1967)
noted that 34% of vital capacity was partitioned to the abdominal compartment in
normal subjects, which is in good agreement with the result reported here of 31%
of total volume expired.
Bergofsky (1964) used plethysmography to measure "abdominal compliance" in
absolute terms,in 5 tetraplegic patients at low lung volumes. His figures of
90-120ml/cm H20 are similar to those that can be derived from Figure 5.2 over a
similar range of lung volume, in 4 out of the 6 patients studied. Estenne et al
(1983) used a weighted spirometer technique to measure total respiratory system
compliance in tetraplegic patients. In this paper it was suggested that
167
"abdominal compliance" was greater than in normal subjects, but no direct
measurements were made.
The results reported here suggest that in addition to there being greater
displacement of the abdominal wall in tetraplegic patients (as may be observed
clinically), the abdominal wall itself is more compliant than in normal subjects.
This may seem surprising in patients in whom muscle spasticity can be a problem.
However, muscle spasms tend to be intermittent and disuse atrophy may cause a
decrease in abdominal muscle tone. It is common for the skeletal muscles of the
limbs to atrophy in patients with chronic upper motor neurone lesions. This could
also be the case for the abdominal muscles, which are are less easy to observe and
about which there is no published pathological data.
The percentage of expired volume partitioned to abdominal wall displacement in
tetraplegic patients, showed much more inter-subject variation than in normal
subjects. This is in accord with the findings of Morgan and De Troyer (1984) who
commented on the individuality of chest wall motion in such patients. The
presence of paradoxical rib cage motion on inspiration is particularly important
in this respect (See 2.4.1). Its presence can mean that the rib cage contribution
to ventilation is negative, in which case the abdominal compartment's contribution
has to be greater than 100%, as in patient 1 in this study. The relative
compliances of the two chest wall compartments are thus extremely important, and
have been studied in Chapter 6 of this thesis.
The data collected over the tidal breathing range, suggest that the difference in
abdominal wall compliance noted between tetraplegic patients and normal subjects
during the relaxation manoeuvre, hold for breathing at rest. The question is
raised, as to whether therapeutic intervention to decrease abdominal wall
compliance may be of value to these patients?
168
5.4, Summary
1) The abdominal wall of tetraplegic patients is on average twice as
compliant as that of normal subjects.
2) There is considerable variation between individual patients.
3) This may have a detrimental effect on their breathing.
169
CHAPTER 6
THE EFFECT OF LOADING EXPIRATION ON CHEST WALL MOTION IN TETRAPLEGIC PATIENTS
6.1, Introduction
At the end of a quiet expiration there is no movement of air into or out of the
lungs though the glottis is open. Since there is no flow, alveolar pressure and
barometric pressure are equal. The system is stationary and the various forces
within it must be at equilibrium. If positive end expiratory pressure is added to
the system and quiet breathing continued, it could be expected, and was observed
that the (end expiratory) equilibrium volume of the system increases. The slope
of the pressure volume curve of the chest wall under these circumstances, can be
taken as a measure of the compliance of the relaxed respiratory system at volumes
of resting FRC and above.
In normal subjects, when a positive end expiratory pressure is applied, the
increase in chest wall volume at end expiration is more or less evenly distributed
between rib cage and abdomen, and is in the order of 12ml/cm H 2O/L TLC (Vellody et
al 1978). In seated tetraplegic patients, Estenne et al (1983) have found that
the compliance of the total respiratory system is less than in normal subjects.
They attributed this to rib cage stiffness.
In the study described below, Supine tetraplegic patients were asked to breath
quietly against various spring loaded expiratory resistances, and the changes in
chest wall, rib cage and abdominal volumes were recorded to determine the relative
compliances of the rib cage and abdominal compartments. The technique used was
simple and non-invasive and may have a wider clinical application (see 3.4.2.2).
170
6.2, Results
8 tetraplegic patients with a mean age of 27 years were examined, 6 were male.
Further details are displayed in Table 6.1.
Plots of the pressure volume characteristic of the respiratory system of each
patient are displayed in Figure 6.1. The data from which these plots were
constructed are expressed in terms of increase in volume of the system above FRC,
as a fraction of predicted TLC (See Appendix 2). In this way the data have been
normalised for patient size and the slope of the curve reflects the specific
compliance of the system (SCrs). The relationship between change in FRC and
expiratory airway pressure was linear in 7 of the 8 patients. The regression
lines drawn through these data points had a correlation coefficient of 0.96 or
better (Table 6.2). The correlation coefficient for the remaining patient was
0.64. The gradient of each regression line, forced through the origin, was
calculated and represents the Specific compliance of the total respiratory system
(SCrs) over the range of lung volume shown. SCrs for each patient is listed in
Table 6.2. The mean SCrs for the 8 patients studied was 8.lml/cm H2O /L TLC (SD
4.lml/cm H20/L TLC).
If change in FRC is partitioned into its rib cage and abdominal components, these
can also be plotted against end expiratory airway pressure (Figure 6.1). It is
possible to distinguish 2 different patterns amongst these plots, which illustrate
the relative compliance of the 2 compartments at different lung volumes. In
patients 1 to 6, the abdominal compartment is more compliant than the rib cage.
As the volume of the chest wall at FRC increases with end expiratory pressure, the
majority of this rise is due to abdominal wall displacement. In patients 7 and 8
half or less of the volume rise is partitioned to the abdominal compartment. The
mean percentage of SCrs attributable to the abdomen (SCabd) was calculated for
171
Table 6.1 Tetraplegic patients studied
Patient Sex Age Level Time Predicted
(yr) of After TLC
Lesion Inj ur y (L)
1 M 24 C6 4 yr 6.6
2 F 23 T1 4 mth 5.8
3 M 34 C5,6 7 mth 7.0
4 F 35 C5,6 6 mth • 5.0
5 M 31 C5 6 mth 6.75
6 M 27 C5 4 mth 7.75
7 M 29 C5 9 mth 5.8
8 M 20 C5 19 mth 7.0
172
173
AV(ml/L TLC)
Figure 6.1Specific compliance plots of the total respiratory system and its abdominal compartment in tetraplegic patients. Av = change in volume above resting FRC expressed in ml/L predicted TLC.P = airway pressure in cm/H20. Black data points and solid lines represent data pertaining to the abdominal compartment. White data points and broken lines represent data pertaining to the total respiratory system.
Table 6.2 Specific compliance of the total respiratory system and its
abdominal compartment in tetraplegic patients, with the
correlation coefficients of each plot.
Patient Total Resp. System Abdominal Compartment
Specific Corr. Specific Corr. SCabd/SCrs
Compliance Coeff. Compliance Coeff. (%)
(ml/cm H20/L TLC) (ml/cm H20/L TLC)
1 9.12 0.98 9.56 0.99 105
2 11.02 0.99 10.34 0.99 94
3 14.85 0.99 10.27 0.97 69
4 5.70 0.98 3.06 0.86 54
5 7.87 0.97 4.71 0.89 , 60
6 2.29 0.93 1.86 0.95 81
7 3.59 0.64 4.02 0.77 110
8 11.02 0.99 3.40 0.95 31
174
each patient by plotting the regression lines shown in Figure 6.1, and calculating
their gradient. The mean SCabd expressed as a percentage of SCrs for 8
tetraplegic patients was 75.5% (SD 27.2%).
6.3, Discussion
This study has utilised a simple non-invasive method to examine SCrs and its rib
cage and abdominal components, in a group of tetraplegic patients. The method is
analogous to the weighted spirometer technique of Heaf and Prime (1956). Instead
of weights, expiratory threshold resistances were used to load expiration.
Instead of a spirometer to measure volume, the optical contour mapping system was
used to measure change in chest wall volume. Both the weighted spirometer
technique and the method used in this study measure total respiratory system
compliance. Since airway pressure is measured, it is the compliance of the whole
system including the lungs which is being assessed. To normalise the data for
patient lung volume, this has been expressed as Specific compliance.
Vellody et al (1978) refined the weighted spirometer method by using magnetometers
to measure Crc and Cabd separately in normal subjects. Estenne et al (1983)
applied a weighted spirometer technique to patients with respiratory muscle
weakness, including 10 patients with complete neurological lesions of the cervical
spinal cord (8 due to trauma and 2 due to transverse myelitis). They measured
lung compliance (Cl) with an oesophageal balloon to derive true chest wall
compliance (Cw) from Crs using the formula
1/Cw = 1/Crs - 1/C1
They could not measure Cabd and Crc because of difficulties in using
mmagnetometers to partition respired volume between rib cage and abdomen in
175
tetraplegic patients (see 3.3.3). Using optical contour mapping that problem was
overcome in this study.
Expiratory threshold resistance (ETR) was used to load breathing. Campbell et al
(1961) exposed conscious and anaesthetised normal subjects to ETR, and found that
FRC increased with addition of the load, and decreased to normal after it was
removed. In conscious subjects, this pattern was seen even if the load was
present for only 4-10 breaths. An increase in FRC occurred from the first loaded
breath and was complete after 3 or more breaths. In the present study, ETR was
applied for 10 breaths and FRC measure after at least 5 breaths. D'Angelo and
Agostoni (1975) found that response to ETR was abolished by vagotomy in dogs, and
in addition that it was not altered by T1 cordotomy. This evidence suggests that
response to ETR is mediated via the vagus and should therefore be in tact in
cervical cord injured patients. Axen (1984) has made detailed studies of loading
inspiration in tetraplegic patients and found their response to be similar to
normal subjects.
In the present study, patients behaved similarly to normal subjects when
expiration was loaded. Increase in FRC of the total respiratory system was
related linearly to end expiratory pressure, in 7 out of 8 patients. Patient 7
breathed erratically when ETR was added. This meant that it was more difficult to
trigger the camera at end expiration (see 3.4.2.2.3) and may explain the poor
correlation coefficient of his data.
Though qualitatively the tetraplegic patients response to ETR was similar to
normal subjects, quantitatively it was not. Crs as measured by Vellody et al
(1978) in supine relaxed normal subjects was 0.077L/cm H2O, and in anaesthetised
and paralysed normal subjects 0.078L/cm H20. Using the data from Vellody et al,
SCrs has been calculated for each of the paralysed and anaesthetised normal
176
•
subjects that they studied. This data is displayed in Table 6.3, B.nd ffiE.y be
compared to the data for tetraplegic patients in Table 6.1. The mean SCrs for
paralysed normal subjects was 12.37ml/cffi H20/ L TLC (SO 3.36 ml/cm H20/L TLC).
The tetraplegic patients in the present study had a mean SCrs of 8. 13ml/cm H20/L
TLC (SO 4.1ml/c~ H20/L TLC), which is 65.7% of nonnal. The data fo~ normal l t "-
subjects and t~traplegic batients was compared 'using an uhpaired StJdent's T t~st.
There was a significant difference between the two groups (p(O.OS) •. :
A decreased SCrs in supine tetraplegic patients is in line with the finqings of
Estenne et al (1983) in a sfmilar group in the seated posture. The published data
of Cl and Ow were used to derive Crs for their 10 patients with tetraplegia. Mean
Crs wasO.082L/cm H20, while in normal subjects it was 0.13SL/cm H20. They
suggested that this decrease was due predominantly to a decrease in rib cage
compliance. This has been studied directly using the optical contour mapping
system. The tetraplegic patients in the present study partitioned 24.S% of SCrs
to the rib cage (SCrc), whilst' the normal subjects in the study of Vellody et al
(1978) partitioned a mean of 50% of SCrs to the rib cage. Comparison of the two
groups with a Student's unpaired T test showed that tetraplegic patients
partitioned a smaller proportion of SCrs to the rib cage than no~l subjects
(p(O.OS).
In normal subjects who· have been anaesthetised and paralysed the volume of the rib
cage and abdomen increases in parallel (Vellody et al 1978). In 6 out of the 8
tetraplegic patients studied, the abdominal comparbment contributed most of the
increase in chest wall volume. Since the intercostal and abdominal muscles are
paralysed in both th~se groups, the decrease in sCrc in tetraplegic patients may
be due to the rib cage itself becoming stiffer (for example, due to boney
ankylosis of joints), or to decreased lung compliance.
177
1 I ~
i I'
I I !
I ! i i
percentage partitioned to the abdominal compartment in normal
paralysed anaesthetised subjects from the data of Vellody
et al (1978)
Table 6.3 Specific compliance of the total respiratory system, and the
Subj ect Height TLC SCrs % partitioned to abdominal
(cm) (L) (ml/cm H20/L TLC) compartment
1 175 6.35 10.24 46.2
2 183 6.97 10.19 37.2
3 162 5.34 7.49 60.6
4 188 7.36 14.21 58.05 170 5.96 16.28 46.4
6 183 6.97 13.65 52.6
7 170 5.96 10.40 53.2
8 168 5.80 13.10 42.1
9 173 6.19 13.09 46.9
10 175 6.35 18.74 47.1
11 178 6.58 8.66 62.2
178
There is considerable variation in the relative compliance of the rib cage and
abdominal compartments in patients with tetraplegia. Morgan and De Troyer (1984)
studied chest wall distortion in such patients during forced inspiratory
manoeuvres. They found that some patients developed paradoxical rib cage
movements amounting to a negative contribution to ventilation, whilst others did
not. The data presented in this chapter confirms the variability between
individual patients. There is a range of SCrs between 20% and 120% of predicted
normal. Most patients had stiff rib cages and mobile abdomens, but patients 7 and
8 were comparable to normal subjects, with similar values for SCrc and SCabd. In
assessing the respiratory system of a tetraplegic patient, it is important to
obtain some indication of the relative compliances of the rib cage and abdomen.
6.4, Summary
1) The specific compliance of the total respiratory system of tetraplegic
patients is less than normal subjects due to a decrease in its rib cage
component.
2) To adequately assess the respiratory system of tetraplegic patients
it is necessary to obtain information about the compliance of the abdomen
and rib cage.
The last 2 chapters have emphasised that the abdominal wall of tetraplegic
patients is more compliant than normal. This begs the question, could their
breathing be improved by making it less compliant? The study reported in Chapter
7 was designed to answer this question.
179
CHAPTER 7
THE EFFECT OF ABDOMINAL BINDERS ON BREATHING IN TETRAPLEGIC PATIENTS
7.1, Introduction
When the diaphragm contracts, abdominal pressure rises to an extent determined by
the distensibility of the abdominal wall. If the wall is stiff, the pressure rise
is substantial pushing the lower rib cage outwards, and opposing the descent of
the diaphragm, which is obliged to lift the rib cage instead (See 1.5.2.3). If
the wall is compliant it displaces easily, with slight pressure rise, and little
elevation or expansion of the rib cage. The data in Chapter 5 has shown that in
tetraplegic patients, the abdominal wall is twice as compliant as in normal
subjects. This impairs the rib cage expanding mechanism of the diaphragm,
particularly when such patients are raised to seated or near vertical postures.
Abdominal binders decrease the compliance of the abdominal wall. They have been
used in patients with high spinal injury for many years for postural hypotension
and respiratory problems. The binder in present use has the disadvantage of
binding the lower rib cage as well as the abdomen (Figure 3.1 ). It therefore
increases intra-abdominal pressure, but opposes its own rib cage expanding effect.
This chapter contains the results of a study designed to compare the effects of a
conventional binder and one that did not encroach on the rib cage, on breathing in
a group of patients with complete tetraplegia (See 3.4.4).
.1.8 0
7.2, Results
7 Patients with a mean age of 33 years were studied, their clinical and
anthropomorphic data are displayed in Table 7.1.
7.2.1, Abdominal Girth (See 3.4.4.3)
Measurements of the abdominal girth of each patient, in each posture, with and
without the abdominal binders are displayed in Table 7.2. On being raised from
the supine to the erect position there was a mean increase in girth of 6.7% (SEM
0.36%). Comparing the supine and the seated postures, girth was 11.76% (SEM
0.54%) greater in seated patients. A group of 20 normal subjects mean age 33
years were measured for comparison, data are displayed in Table 7.3. In the erect
posture, their girth was 4.65% (SEM 0.34%) greater than in the supine posture, and
in the seated posture, 6.77% (SEM 0.38%) greater than in the supine posture. An
unpaired Student's T Test was used to compare the data of the two groups. In the
erect posture, the abdominal girth of tetraplegic patients had increased by a
great percentage than that of normal subjects (p<0.01). This was also true in the
seated posture (p<0.001).
7.2.2, Sniff Pdi (See 3.4.3.6, 3.4.4.4)
Six of the 7 patients successfully completed the protocol. Patient 7 could not
tolerate the oesophageal and gastric balloons.
The mean data for 6 patients are displayed in Table 7.4. Individual patient data
is available in Appendix 3. Without binders, there was a significant differenceObetween Sniff Pdi in the supine, seated, and 70 tilt positions (Figure 7.1).
Sniff Pdi was greater supine than seated (p<0.05) by 18.5% (11.0cm H20, SED =
181
Table 7.1
Clinical and anthropomorphic details of the tetraplegic patients
in the abdominal binder study
Patient Age Ht Wt Level Time Predicted VC Actual VC
(yrs) (cm) (Kg) of since (L) (L)
Lesion inj ury seated seated
1 43 184 73 C5 22 yr 5.03 3.10
2 24 178 76 C6 4 yr 5.11 2.32
3 27 193 76 C5 6 mnth 5.84 1.68
4 34 183 64 C5,6 7 mnth 5.16 1.13
5 29 168 64 C5,6 11 mnth 4.48 2.66
6 44 178 73 C6,7 4 mnth 4.68 3.43
7 31 180 73 C5 7 mnth 5.03 1.68
182
Table 7.2
Girth measurements (cm) in tetraplegic patients
with and without abdominal binders.
Compression (cm) is the decrease in girth
caused by binding the abdomen
Supine Seated 70° tilt
Subject Binder Girth compression Girth compression Girth compression
1 81.5 5.5 91.5 7.5 86.0 6.0
+ 76.0 84.0 80.0
2 91.5 4.0 101.5 5.0 96.5 9.0
+ 87.5 96.5 87.5
3 96.5 5.0 110.5 11.5 103.5 8.0
+ 99.0 95.5
4 86.5 6.5 98.0 8.0 92.5 6.5
+ 80.0 90.0 86.0
3 71.0 3.5 78.5 7.5 75.0 6.5
+ 67.5 71.0 68.5
6 73.5 2.5 81.5 5.5 78.5 5.0
+ 71.0 76.0 73.5
7 91.5 6.5 101.5 10.0 99.0 7.5+ 85.0
x = 4.8
91.5
x = 7.9
91.5
x = 6.9
SD = 1.5 SD = 2.3 SD = 1.3
183
Table 7.3
Girth measurements (cm) in normal subjects
Subject Age Height Weight Supine Seated Standing
(yrs) (cm) (Kg)
1 52 176 79 84 96.5 94
2 29 173 73 85 91.5 87.5
3 27 168 51 68 73 76
4 31 187 69 76 81 86
5 20 178 68 72 77 75
6 23 185 73 72 78 76
7 33 167 65 82 89 87
8 55 185 79 79 86 85
9 35 183 80 84 89 87
10 32 175 70 86 89 88
11 30 193 95 88 93 92
12 28 179 65 74 78 77
13 38 182 73 85 92 91
14 28 192 80 80 86 83
15 ' 29 175 69 75 79 78
16 36 174 65 75 80 79
17 39 170 65 74 76 75.5
18 32 164 54 64.5 67.5 67
19 44 169 60 79 85 84
20 33 182 76 76 82 80.5
184
Table 7.4
Mean Sniff Pdi (cm H20) in 6 Tetraplegic Patients.
No Binder Conventional Binder New Binder
Supine x = 70.5 x = 68.5 5T= 74.0
SD = 13.0 SD = 13.0 SD = 10.5
Seated x = 59.5 x = 62.5 x = 63.0
SD = 14.0 SD = 15.0 SD = 14.5
70° Tilt x = 52.0 x = 61.5 x = 65.0
SD = 14.5 SD = 17.5 SD = 13.0
185
S n iff Pdi % Sup ine1 2 0 -
40 -
20-
★ ® p < 0 . 0 5
★ A p < 0.01
★ A p < 0 .02
Supine S ea ted 7 0 °V ilt
Figure 7.1The effect of posture on Sniff Pdi in 6 tetraplegic patients, expressed as a percentage of the supine values. p<0.01 refers to comparison of supine and 70° tilt values.p<0.05 refers to comparison of supine and seated values.p<0.02 refers to comparison of seated and 70° tilt values.
186
3.5cm H20), supine that at 70 tilt (p<0.01), by 36.5% (19.0cm H20, SED = 3.5cm
H 20) and seated than at 70 tilt (p<0.02) by 15% (8cm H20, SED = 2cm H20).
OWhen tilted at 70 , both binders improved Sniff Pdi (Figure 7.2). The
conventional binder (p<0.05), by 19.5% (19cm H20, SED = 3cm H20) and the new
binder (p<0.02) by 26% (13cm H20, SED = 3.4cm H20). Neither binder had a
significant effect on Sniff Pdi when seated or supine, and there was no
significant difference between the effect of the 2 binders in any posture.
7.2.3, Vital Capacity (See 3.4.4.4)
The mean data for 7 patients are displayed in Table 7.5. There was a significant
difference in VC in the three different postures (Figure 7.3). VC without a
binder was greater supine than when seated (p<0.05) by 28% (640ml, SED = 210ml),
and supine than at 70 tilt (p<0.02) by 49% (960ml, SED = 160ml).
When patients were seated both binders impoved VC (p<0.01) (Figure 7.4), the
conventional binder by 11.5% (260ml, SED = 50ml), and the new binder by 8% (180ml,
SED = 40ml).
OAt 70 tilt, both binders improved VC (Fig 7.5). The conventional binder (p<0.01)
by 15% (290ml, SED = 70ml), and the new binder (p<0.001) by 24% (470ml, SED =
70ml). Binders did not alter VC in the supine posture. There was no statistical
difference between the effect of the two binders.
187
Figure 7.2The effect of binding the abdomen on Sniff Pdi in 6 tetraplegia patients tilted at 70° to the horizontal, expressed as a percentage of the unbound value.p<0.02 refers to comparison of the new binder and no binder.p<0.05 refers to comparison of the conventional binder and no binder.
188
Table 7.5
Mean Vital Capacity (L) in 7 Tetraplegic Patients.
No Binder Conventional Binder New Binder
Supine x = 2.91 x = 2.96 x = 2.92
SD = 0.49 SD = 0.53 SD = 0.44
Seated x = 2.27 x = 2.53 x = 2.46
SD = 0.84 SD = 0.85 SD = 0.78
70° Tilt x = 1.95 x = 2.24 x = 2.42
SD = 0.57 SD = 0.57 SD = 0.71
189
VC % Sup ine120 —
4 0 -★ ® p < 0 . 0 5
★ A
20 -
p < 0 .02
Supine S ea ted 7 o° Tilt
Figure 7.3The effect of posture on vital capacity in 7 tetraplegic patients, expressed as a percentage of the supine value. p<0.02 refers to comparison of supine and 70° tilt values. p<0.05 refers to comparison of supine and seated values.
190
% change in VC
+ 4 0 -
- 20-
★ © p < 0.01
★ A p < 0 .0 1- 4 0 -
no b inder conven tiona l b inder new binder
Figure 7.4The effect of binding the abdomen on vital capacity in 7 seated tetraplegic patients, expressed as a percentage change of the unbound value.p<0.01 refers to comparison of the new binder and no binder, and to comparison of the conventional binder and no binder.
191
% change in VC
- 20 -
- 4 0 -
★ • p < 0 . 0 1
★ A. p < 0.001
no b inder conven tiona l b inder new binder
Figure 7.5The effect of binding the.abdomen on vital capacity in 7 tetraplegic patients tilted at 70° to the horizontal, expressed as a percentage of the unbound value.p<0.001 refers to comparison of the new binder and no binder. p<0.01 refers to comparison of the conventional binder and no binder.
192
7.2.4, PImax (See 3.4.3.7, 3.4.4.4)
The mean data for 7 patients are displayed in Table 7.6. The only statistically
significant finding from the PImax data was that in the supine posture, patients
had a greater PImax without a binder than with the conventional binder (p<0.05) by
5% (3cm H20, SED = 1cm H20). A subject performed 10 sets of PImax manoeuvres
seated and without a binder. The mean value was 60cm H20 with a standard
deviation of 3.5cm H20 (See Appendix 3). In view of this, a 5% change in PImax is
probably unimportant.
7.2.5, Gastric Pressure (See 3.4.3.4, 3.4.4.4)
Mean gastric pressure measurements at resting end expiration, tidal inspiration
and during maximal sniff are displayed in Table 7.7. There was no significant
difference between Pg in different postures without a binder, though it tended to
be lower in the supine position. When seated, the conventional binder increased
resting (R) Pg (p<0.05) by 2.3cm H20, end tidal inspiratory (I) Pg (p<0.05) by
2.8cm H20, and maximal sniff (S) Pg (p<0.05) by 4.7cm H20. The new binder when
applied to patients in the seated posture, increased I Pg (p<0.05) by 1.8cm H20,
but did not significantly increase R Pg or S Pg. The conventional binder
increased seated R Pg (p<0.05) by 1.7cm H20 more than the new binder.
With patients in the supine posture, both binders increased R Pg [The new binder
(p<0.02) by 1.3cm H20, and the conventional binder (p<0.01) by 3.0cm H20], I Pg
[The new binder (p<0.02) by 4.0cm H20, and the conventional binder (p<0.01) by
4.3cm H20], and S Pg [The new binder (p<0.05) by 5.3cm H20, and the conventional
binder (p<0.02) by 2.7cm H20].
OAt 70 tilt, both binders increased R Pg [The new binder (p<0.05) by 2.7cm H20,
193
Table 7.6
Mean PI Max (cm H2O) in 7 Tetaplegic Patients.
No Binder Conventional Binder New Binder
Supine x = 54.5 x = 51.5 x = 50
SD = 15.5 SD = 17.0 SD = 17.0
Supine x = 49.0 x = 54.0 x = 53.0
SD = 17.0 SD = 17.0 SD = 18.0
070 Tilt x = 47.0 x = 48.0 x = 48.0
SD = 11.5 SD = 15.0 SD = 13.5
194
Table 7.7
Mean Gastric Pressure measurements (cm H20) in 6 Tetraplegic Patients
No Binder Conventional Binder New Binder
x (SD) x (SD) x (SD)
Supine
RPg 2.5 (1.0) 5.5 (1.5) 4.0 (1.5)
IPg 6.5 (2.5) 11.0 (3.5) 10.0 (4.5)
SPg 11.5 (4.5) 14.0 (4.5) 16.5 (7.5)
Seated
RPg 5.0 (3.0) 7.0 (4.5) 5.5 (4.0)
IPg 10.5 (4.5) 13.0 (5.5) 12.0 (4.5)
SPg 13.5 (3.5) 18.5 (7.0) 16.5 (6.0)
70° Tilt
RPg 4.5 (3.0) 7.5 (3.0) 7.0 (3.0)
IPg 10.0 (3.5) 14.5 (3.5) 16.5 (4.0)
SPg 16.0 (7.5) 24.0 (13.5) 28.0(15.0)
195
and the conventional binder (p<0.01) by 3.2cm H20], I Pg [The new binder (p<0.05)
by 6.5cm H20, and the conventional binder (p<0.01) by 4.3cm H20], and S Pg [The
new binder (p<0.05) by 12.2cm H20, and the conventional binder (p<0.05) by 8.0cm
H20]. There was no significant difference between the effect of the binders on PgOsupine, or at 70 tilt.
7.3, Discussion
This study yielded data on changes in abdominal girth, trans-diaphragmatic
pressure, vital capacity, and maximum inspiratory pressure at the mouth with
posture, and with the application of binders, in a group of stable tetraplegic
patients. The mean data for the group without binders is comparable to previously
published data from similar patients (Table 7.8, See Tables 2.1, 2.2 and 2.3 for
sources of data) suggesting that they were a representative group.
Abdominal girth increased when patients were tilted up from the supine position,
and increased more when they were seated in a chair. These increases were
significantly greater in the patients with tetraplegia than in a group of healthy
men of the same age and stature. When patients are tilted up their abdominal wall
falls outwards due to the effect of gravity on the enclosed viscera. In the
seated posture the situation is complicated by flexion of the spine. The degree
to which the abdominal wall is displaced by the weight of the abdominal contents
will depend on its compliance. Thus changes in abdominal girth on tilting will be
determined in part by abdominal wall compliance, and may act as a measure of it.
The differences between patients and normal subjects agree with the finding of
increased abdominal wall compliance in tetraplegia shown in Chapter 4. The use of
a method as simple as a tape measure to assess abdominal wall compliance is
appealing, and may have wider clinical applications.
•196
Comparison of mean lung function data from this study
and from previously published work on tetraplegic patients
Table 7.8
(See Tables 2.1, 2.2, 2.3)
Parameter This Study Published Data
VC Supine
(% Predicted)
58 61
VC Seated
(% Predicted)
49 53
PImax Supine
(cm H20)
55 70
PImax Seated
(cm H20)
49 54
Pdi Supine
(cm H20)
71 73
Pdi Seated
(cm H20)
60 52
197
When normal subjects stand up from a supine position, VC increases by 7.5% (AllenOet al 1985). When tetraplegic patients are tilted up to 70 to the horizontal, VC
decreases by 45% (Fugl-Meyer 1971). If they are raised to the seated position, in
which the abdominal wall is splinted, VC is decreased by 28% (Fugl—Meyer 1971).O
In the present study, the decreases of VC in the 70 tilt and seated postures were
49% and 28% respectively.
The differences between normal subjects and tetraplegic patients can be attributed
to difference in abdominal muscle tone. When normal subjects stand up abdominal
muscle tone increases (Campbell and Green 1955, Strohl et al 1981, De Troyer 1983)
preventing the abdominal wall from falling outwards, and so maintaining a high
arched diaphragm which can contract effectively. When tetraplegic patients are
tilted up, the abdominal wall falls outwards and the diaphragm descends to a lower
and flatter end expiratory position. This disadvantageous configuration decreases
inspiratory capacity, and the pressure that the diaphragm can generate. This is
borne out by the data in Figures 7.1 and 7.3.
The PImax data in this study were non-contributory to analysis of the effects of
binders and posture on tetraplegic patients. This may have been because the
measurements were not made sufficiently accurately, or because PImax was not an
appropriate measurement to make. The PImax values used for statistical analysis
had 95% confidence limits of 12%. Measurements of PImax are effort dependent and
may be influenced by a training effect. It had been hoped that these factors
would be minimised by carrying out all the measurements on the same day, and using
the patients as their own controls. Measurements were performed in random order
in case the patients tired. Despite these precautions PImax was not as sensitive
a measurement of inspiratory muscle function as Sniff Pdi; this was probably due
to a combination of the factors mentioned above.
198
When patients with tetraplegia were tilted up, binding their abdomens improved VC
and Pdi. This has only been demonstrated previously in one or two atypical cases
(Danon et al 1979, Strohl et al 1984, Celli et al 1985). In the present study
binding the abdomen significantly increased VC when sitting, and Sniff Pdi and VC
at 70° tilt. Abdominal binders had no effect when patients were supine.
Binders act by increasing intra-abdominal pressure and pushing the diaphragm into
a position of greater mechanical advantage. In the supine position, the effects
of gravity on the abdominal contents maintain the diaphragm in an advantageous
position, and binders have no additive effect. It is in the more erect postures
that they can substitute for abdominal muscle tone in paralysed patients, and have
a beneficial result.
Binders also decrease abdominal wall compliance, so that the descent of the
diaphragm on inspiration is opposed. As a consequence, intra-abdominal pressure
is increased to a greater degree, and acts via insertional and appositional forces
to promote expansion of the lower rib cage. It was with this in mind that a new
binder was designed which avoided inhibiting lower rib cage motion.
Application of an abdominal binder should compress the abdomen, and thus cause a
rise in resting gastric pressure. It should also decrease abdominal wall
compliance, as shown by an accentuated increase in gastric pressure on
inspiration. Measurements of gastric pressure in this study demonstrated a
significant increase in R Pg, I Pg, and S Pg in all postures on applying the
conventional binder. However, this was not the case with the new binder which
failed to significantly increase R Pg and S Pg in the seated posture. The reasons
for this are inherent in its design. The front plate was made of inelastic
material, which was moulded to the patient in the supine posture. In the seated
position it did not fit the altered configuration of the abdominal wall, and when
199
it was applied, R Pg did not rise significantly (Table 7.7). This may explain why
the new binder showed no significant advantage over the conventional binder,
despite leaving the lower rib cage free to expand.
7.4, Summary
1) This study suggests that abdominal binders are valuable aids to breathing
when patients with tetraplegia are mobilised.
2) The conventional binder should continue to be used until a better one
is designed.
2 0 0
CHAPTER 8
Summary and Discussion
In this thesis the abdomen has been described as a fluid filled bag bounded by
mobile and immobile parts. The mobile parts are the diaphragm and anterior
abdominal wall, the immobile parts are the pelvic bowel and spine. When the
diaphragm descends on inspiration, the immobile parts act as a fulcrum for rib
cage expansion, whilst the anterior abdominal wall acts as a partial fulcrum. The
descent of the diaphragm and hence its rib cage expanding effect, depend on the
compliance of the abdominal wall, i.e. its partial fulcrum activity. This in
turn is influenced by contraction or relaxation of the muscles that make it up.
The aim of the experimental work reported is to expand on and examine this model.
The anterior abdominal wall is not a simple structure. It consists of 4 main
pairs of abdominal muscles (See 1.3.3). Their fibres run in different directions,
and their individual actions might have different effects on the partial fulcrum
activity of the wall as a whole. For this reason the actions of each separate
muscle have been studied in Chapter 4. The experiments performed suggest that
these muscles act synchronously in respiration, and thus the anterior abdominal
wall may be considered as a single unit in this respect. This being the case, the
next step was to assess the degree of partial fulcrum activity of the abdominal
wall, by measuring its compliance. This work was carried out in normal subjects
and patients with tetraplegia, in the supine posture during a relaxed expiration.
In cervical cord injured patients a common clinical observation is that the
abdominal wall is displaced more than normal during breathing. Does this imply
that the abdominal wall is more compliant? If it is more compliant, is this
important?
2 0 1
Chapter 5 details the measurements made of abdominal wall compliance in a group of
healthy subjects and a group of tetraplegic patients. For a given rise in gastric
pressure the abdominal walls of the patients were displaced further, i.e. they
were more compliant and had less partial fulcrum activity for the diaphragm. To
some extent this was a surprising finding, as cervical cord transection causes an
upper motor neurone paralysis of muscles innervated from below the level of
lesion. The abdominal wall muscles would be expected to be spastic and hence less
compliant. The observations made can be explained if considered in conjunction
with the clinical findings. Though tetraplegic patients develop intermittent
spasms of their muscles, these are not constant, and the method used to measure
abdominal wall compliance excluded such episodes. Between spasms there is no
muscle activity, and it would not be surprising if disuse atrophy developed in the
abdominal muscles, as is indeed the case in paralysed limb muscles.
Unfortunately, there is no simple way to measure any such atrophy of the abdominal
wall muscles. If one now considers the effect of a normal diaphragm working to
displace a wasted anterior abdominal wall, the fact that the wall is displaced
further than in normal subjects, and has been measured as being more compliant,
becomes less surprising.
It is not clear from the work in this thesis, whether increased compliance of the
abdominal wall in tetraplegic patients is an important finding which materially
alters the respiratory capacity of an individual patient. Does it matter if the
rib cage and abdominal wall move more or less during breathing, as long as the
total chest wall motion remains the same? It is easy to imagine a situation
whereby, if the rib cage moves paradoxically on inspiration, the abdominal wall
and hence the diaphragm will have to move more to shift the same volume of gas
into the lungs. The energy cost of this form of breathing will be greater than
normal respiratory motion, which could be detrimental to severely injured
patients, who may be on the brink of respiratory failure. In this situation the
2 0 2
rib cage is more compliant than normal and is sucked in by the drop in pleural
pressure on inspiration. Thus the relative compliances of the two compartments of
the chest wall are important when assessing breathing in tetraplegic patients.
In Chapter 6 a simple non-invasive method of assessing the relative compliances of
the rib cage and abdomen was applied to tetraplegic patients. The results
concurred with those of other authors (Estenne et al 1983), in that the total
respiratory system compliance was reduced. In the majority of cases this was
because the rib cage was stiff. Using this method it is clear that for a given
total compliance, if the rib cage is less compliant, the abdomen must be more
compliant. However, the measurements of abdominal wall compliance in Chapter 5
were made in absolute terms, and reflect the physical properties of the abdominal
wall per se. The conclusion that the abdominal wall of tetraplegic patients is
more compliant than that of normal subjects still holds.
Is this important? Speculation that increased abdominal wall compliance will
impair lower rib cage expansion is theoretically sound, but has not been proven
here. If this was the case, would this have detrimental effects on respiratory
status? To answer such questions, the approach taken was to decrease the
abdominal wall compliance of tetraplegic patients artificially, and see if it
improved their breathing.
In Chapter 7, the effects on breathing of binding the abdomen of patients with
tetraplegia were studied. In the supine posture altering abdominal wall
compliance made no difference to the measured parameters of breathing. However, ,
once the patient was raised to more erect postures, abdominal binders caused a
significant improvement in these measurements.
When normal subjects are tilted head up, the abdominal muscles contract to resist
203
descent of the diaphragm under the influence of gravity. This decreases abdominal
wall compliance, and increases the partial fulcrum effect opposing diaphragmatic
motion. The diaphragm is thus maintained in a mechanically advantageous
configuration, and expands the lower rib cage on inspiration.
When tetraplegic patients are tilted up, their abdominal walls are displaced
forwards by the weight of the abdominal contents, the diaphragm descends to a
position in which it is shorter and flatter, and can contract less effectively.
In the experimental work performed, the Sniff Pdi and vital capacity of patients
tilted to 70° to the horizontal were significantly reduced compared with the
supine posture. Both absence of abdominal muscle contraction and greater
abdominal wall compliance are relevant in this context. The importance of the
latter was demonstrated in tetraplegic patients as compared to normal subjects, by
the significantly greater increase in abdominal wall displacement (measured by
girth) on moving from supine to 70° to the horizontal. The fact that abdominal
wall compliance is increased in tetraplegic patients has a detrimental effect on
their breathing in erect postures.
Abdominal binders have been used for years in spinal injuries units, and the
evidence presented in this thesis has confirmed that they have a beneficial effect
on breathing in patients with tetraplegia. It is recommended that they should be
used routinely when patients are first raised from the supine posture in which
they are originally nursed. Their use should be continued if the patient feels a
benefit, or vital capacity is shown to be improved. Respiratory distress in
seated or tilted positions is an indication for a trial of a binder. Binders are
not useful in the supine posture.
204
This thesis has answered some of the questions originally raised regarding the
role of the abdomen in breathing, both in normal subjects and tetraplegic
patients. It has however raised many more uncertainties, which may form the basis
of future research. Some of these points will be dealt with briefly.
1) In the supine posture, does increased abdominal wall compliance in
tetraplegic patients limit lower rib cage expansion?
This could be investigated using the optical contour mapping system to measure rib
cage excursion during tidal breathing and vital capacity manoeuvres. These
measurements could be compared, with and without a binder which decreased
abdominal wall compliance, but did not limit lower rib cage motion.
2) How compliant is the abdominal wall in more erect postures?
To answer this question it would be necessary to build an optical contour mapping
system attached to a tilt table. A similar method to that described in Chapter 5
could be used to measure abdominal wall compliance. The abdominal wall of normal
subjects can never be said to be relaxed when they are erect, because of tonic EMG
activity in the abdominal muscles (See 1.5.1.3). However, a form of "relaxation
manoeuvre" could be performed by trained subjects, and some assessment of
abdominal wall compliance made. In tetraplegic patients the abdominal wall is
relaxed by definition, however, in erect postures it will be stretched by the
effect of gravity on the abdominal contents. It may be that its pressure
displaced-volume characteristic will merely have been moved further along the
supine compliance curve. Measurement of upright abdominal wall compliance might
be less helpful in predicting the effects of assuming the erect posture on
breathing, than compliance measured in the supine posture.
205
3) What happens to rib cage expansion when the abdomen is bound in the erect
posture?
This might be examined using the tilting optical contour mapping system, in
conjunction with an abdominal binder which did not inhibit lower rib cage
expansion. One would predict that rib cage excursions would be greater with the
binder.
4) Can a better abdominal binder be designed?
In the study reported in Chapter 7 there was no significant difference in the
effect on breathing of a conventional binder which bound the lower rib cage, and a
new binder which did not. This was thought to be because the new binder was
fitted in the supine posture (in which it was likely to have the least effect),
and because of its rigidity did not fit, in the other postures examined. The
simplest way of solving this problem would be to fit a similar new binder, to
patients seated in their wheelchairs, in which they spend most of their time. A
second possibility is to design a new binder of similar shape, but of elastic
material. This would need to be attached to a flexible frame like structure, for
it to maintain its configuration and not "roll up".
Perhaps the most appealing approach would be to use the patients own abdominal
muscles as a binder, by stimulating them electrically with indwelling electrodes.
This type of functional electrical stimulation has been used successfully as a
walking aid in the lower limbs of paraplegics, by using a computer to stimulate
the relevant muscles serially. Such a system applied to the abdominal muscles
could stimulate continuously or phasically, thus imitating a binder throughout the
respiratory cycle, or opposing descent of the diaphragm on inspiration only. In
addition such a system might be triggered by the patient to provide tonic activity
- 206
There is a great deal more work to be done concerning the role of the abdomen in
breathing, particularly in cervical cord injured patients. This may have
important clinical applications which will help improve respiration in these
severely disabled people.
in the abdominal muscles during expiration, to facilitate coughing.
207
APPENDIX 1 - Results from Chapter 5
Results of measurements made to plot the pressure volume displacement curves of the abdominal wall
NORMAL SUBJECTS (1.1)
Pg = Gastric Pressure, A Vabd = Abdominal Wall Displacement above FRC A Vabd (%VE) = Abdominal Wall Displacement above FRC expressed as a percentage of the total volume expired.
SUBJECT 1Pg
(cm H20)A Vabd
(ml)A Vabd (% VE)
10.18.177.36.05.1 4.5 3.73.1 1.9
545525465335245195155550
32.031.027.419.714.411.59.19.1 0
Total Volume Expired = 1700 ml
SUBJECT 2
Pg(cm H 20)
A Vabd (ml)
A Vabd (% VE)
9.35.94.12.92.41.91.5 1.31.2 1.0
17615613311586734535150
22.419.917.0 14.711.0 9.3 5.7 4.5 1.9 0
Total Volume Expired = 784 ml
208
SUBJECT 3
Pg A Vabd A Vabd(cm H20) (ml) (% VE)
9.4 473 43.04.7 441 40.12.7 351 31.91.1 269 24.50.4 166 15.1
-0.1 86 7.8-0.6 0 0
Total Volume Expired = 1099 ml
SUBJECT 4
Pg A Vabd A Vabd(cm H20) (ml) (% VE)
9.0 860 25.55.1 787 23.33.3 685 20.32.4 550 16.31.8 466 13.81.4 384 11.41.1 351 10.41.0 290 8.60.8 197 5.80.6 106 3.10.4 74 2.20.3 0 0
Total Volume Expired = 3,373 ml
SUBJECT 5
Pg A Vabd AVabd(cm H20) (ml) (% VE)
11.4 692 29.76.5 568 24.44.6 437 18.73.6 366 15.72.8 247 10.72.7 98 4.22.6 0 0
Total Volume Expired = 2331 ml
209
SUBJECT 6
Pg A Vabd A Vabd(cm H 20) (ml) (% VE)
11.4 997 34.28.3 837 28.76.7 703 24.15.3 527 18.13.8 245 8.42.2 81 2.81.5 76 2.61.0 0 0
Total Volume Expired = 2,918 ml
TETRAPLEGIC PATIENTS (1.2)
Pg = Gastric Pressure,A Vabd = Abdominal Wall Displacement above FRC, A Vabd (% VE) = Abdominal Wall Displacement above FRC expressed as a percentage of the total volume expired.
PATIENT 1
Pg A Vabd A.Vabd(cm H 20) (nil) (% VE)
5.0 1108 130.52.6 729 85.91.5 678 79.91.0 694 81.70.4 606 71.40 485 57.1
-0.1 182 21.4-0.25 80 9.4
Total Volume Expired = 849 ml
2 1 0
PATIENT 2
Pg AVabd AVabd(cm H20) (ml) (% VE)
12.5 1300 55.910.3 1056 52.36.3 790 39.15.1 590 29.23.9 393 19.53.0 317 15.71.6 331 16.40.5 42 2.10.3 0 0
Total Volume Expired = 2-020 ml
PATIENT 3
Pg AVabd Avabd(cm H20) (ml) (% VE)
8.2 2174 63.44.4 1645 48.03.2 1432 41.82.4 1174 34.21.8 949 27.71.0 695 20.30.7 566 16.50.6 406 11.80.3 317 9.20 215 6.3
-0.1 107 3.1-0.2 0 0
Total Volume Expired = 3,429 ml
PATIENT 4
Pg AVabd AVabd(cm H2O) (ml) (% VE)
8.2 431 41.07.8 412 39.27.1 393 37.45.4 367 34.93.4 245 23.31.8 73 6.90.7 61 5.80.3 0 0
Total Volume Expired = 1,051 ml
2 1 1
PATIENT 5
Pg AVabd AVabd(cm H20) (ml) (% VE)
12.3 1459 98.19.6 1220 82.66.7 802 53.95.3 646 43.44.3 498 33.53.1 328 22.62.1 294 19.81.3 75 5.00.7 37 2.40.4 0 0
Total Volume Expired = 1,487 ml
PATIENT 6
Pg AVabd AVabd(cm H20) (ml) (% VE)
6.3 1054 72.35.1 1032 70.83.4 665 45.63.0 654 44.92.8 613 42.01.9 474 32.50.9 302 20.70 250 17.1-0.6 288 19.8-1.3 199 13.7-1.6 167 11.5-1.7 0 0
Results of 6 Digitisations of a single photograph (1.3)
Digitisation123456
Visible Abdominal Volume (ml)2327.52332.72340.32331.6 2329.92319.8
x = 2330.3 SD = 6.72 (0.29%) SE = 2.74 (0.12%)
2 1 2
Results of reproducibility of the digitisation procedure on a set of photographs representing a single "relaxation manoeuvre" (1.4)
Pg AVabdi H 20 ) (ml)
1 2 3 4 5
7.9 421 391 389 386 4017.4 417 381 357 382 3866.0 354 358 362 337 3514.0 285 257 315 282 3122.6 155 158 161 120 1141.3 112 153 59 60 950.4 0 0 0 0 , 0
Results of reproducibility of the "relaxation manoeuvre" 5 manoeuvres digitised (1.5)
1 2 3Pg AVabd Pg AVabd Pg AVabd8.5 413 8.2 431 7.8 4168.2 385 7.8 412 6.8 3837.5 364 7.1 393 5.3 3755.2 250 5.4 367 3.2 2742.9 64 3.4 245 1.8 1141.3 44 1.8 73 0.8 200 0 0.7 61 0 0
0.3 0 '
4 5Pg AVabd Pg AVabd8.5 397 7.9 3958.0 390 7.4 3747.6 379 6.0 3524.6 333 4.0 2842.8 238 2.6 1491.0 62 0.4 960 0 0 0
213
APPENDIX 2
Results from Chapter 6The effect of loading expiration on chest wall motion in tetraplegic patients
Pa = Airway Pressure (cm H20)AVrs = Change in volume of the total respiratory system above resting FRC
(ml/L TLC).AVabd = Change in volume of the abdominal compartment above resting FRC
(ml/L TLC)
PATIENT 1
Pa AVrs AVabd(cm H20) (ml/L TLC) (ml/L TLC)
0 0 00.5 9.5 8.22.5 7.7 17.44.0 29.6 35.07.0 61.0 80.29.0 64.2 78.616.0 161.1 153.0
PATIENT 2
Pa AVrs AVabd(cm H20) (ml/L TLC) (ml/L TLC)
0 0 01.0 15.5 16.93.0 40.2 36.05.0 48.1 60.07.5 86.2 94.310.0 104.3 102.415.0 168.1 143.2
PATIENT 3
Pa AVrs AVabdQ H20) (ml/L TLC) (ml/L TLC)
0 0 01.0 21.3 9.15.0 76.6 64.16.0 93.0 88.48.0 99.1 89.911.0 168.4 114.115.5 234.1 140.1
214
PATIENT 4
Pa A Vrs A Vabd(cm H20) (ml/L TLC) (ml/L TLC)
0 0 00.5 9.8 6.62.5 9.6 4.05.0 19.4 0.87.0 44.4 12.09.0 49.0 18.613.0 77.6 57.2
PATIENT 5
Pa A Vrs A vabd(era H20) (ml/L TLC) (ml/L TLC)
0 0 00.5 -17.2 -32.42.5 14.1 -10.15.0 29.3 3.66.0 45.2 36.57.0 53.3 41.98.0 75.4 42.2
PATIENT 6
Pa A Vrs A Vabd(cm H2O) (ml/L TLC) (ml/L TLC)
0 0 01.5 -10.6 1.44.5 - 9.8 9.77.5 12.9 13.810.5 23.8 18.111.5 26.6 14.316.0 45.7 35.5
PATIENT 7
Pa A Vrs AVabd(cm H2O) (ml/L TLC) (ml/L TLC)
0 0 02.0 24.1 26.23.5 29.3 29.17.0 10.7 21.09.0 22.8 14.310.5 12.4 42.614.0 76.0 67.3
215
PATIENT 8
Pa A v rs(cm H20) (ml/L TLC)
0 00.5 5.71.5 13.13.5 46.16.5 77.98.0 72.912.5 142.6
21 o
A vabd (ml/L TLC)
00.42.314.121.715.350.0
APPENDIX 3
Results from Chapter 7 VC = Vital Capacity (L)Plmax = Maximum Static Inspiratory Mouth Pressure (cm H20)Sniff Pdi = Trans-Diaphragmatic Pressure During Maximal Sniff (cm H20)
PATIENT 1No Binder Old Binder New Binder
Flat Plmax 68 63 57Sniff Pdi 68 55 73VC 3.78 3.89 3.5
Sitting Plmax 75 78 78Sniff Pdi 57 57 68VC 3.1 3.44 3.26
70 Tilt Plmax 50 59 58Sniff Pdi 44 53 54VC 2.33 2.47 2.98
Reproducibility of single Plmax manoevures, seated no binder.
1 2 3 4 5 6 7 8 9 1054 66 59 58 59 63 59 63 62 60
x = 60.3 SD = 3.53
PATIENT 2
No Binder Old Binder New BinderFlat Plmax 72 69 70
Sniff Pdi 84 76 80VC 3.05 3.10 3.14
Sitting Plmax 59 66 65Sniff Pdi 80 84 80VC 2.3 2.64 2.62
70° Tilt Plmax 61 64 70Sniff Pdi 74 81 80VC 2.04 2.37 3.14
217
PATIENT 3
No Binder Old Binder New BinderFlat Plmax 54 51 48
Sniff Pdi 53 56 58VC 2.85 2.72 2.74
Sitting Plmax 40 44 40Sniff Pdi 50 46 46VC 1.68 1.87 1.83
050 Tilt Plmax 38 44 37
Sniff Pdi 46 48 50VC 1.55 1.87 1.89
PATIENT 4
No Binder Old Binder New BinderFlat Plmax 50 48 46
Sniff Pdi 59 60 66VC 2.2 2.3 2.26
Sitting Plmax 27 30 33Sniff Pdi 40 47 45VC 1.1 1.3 1.4
O70 Tilt Plmax 29 36 36
Sniff Pdi 32 42 58VC 1.35 1.35 1.48
PATIENT 3
No Binder Old Binder New BinderFlat Plmax 66 69 71
Sniff Pdi 73 78 84VC 2.62 2.51 2.53
Sitting Plmax 62 78 74Sniff Pdi 68 68 72VC 2.66 2.68 2.72
O70 Tilt Plmax 60 67 60
Sniff Pdi 54 62 69VC 1.85 2.28 2.40
218
PATIENT 6
No Binder Old Binder New BinderFlat PImax 44 40 39
Sniff Pdi 86 86 84VC 3.15 3.26 3.3
Sitting PImax 45 44 42Sniff Pdi 62 72 66VC 3.4 3.7 3.5
70° Tilt PImax 44 37 39Sniff Pdi 60 84 80VC 3.06 3.2 3.6
PATIENT 7
No Binder Old Binder New BinderFlat PImax 28 22 24
Sniff Pdi - -
VC 2.74 2.97 2.95
Sitting PImax 34 38 44Sniff Pdi - - -VC 1.68 2.09 1.88
70°Tilt PImax 38 30 30Sniff Pdi - - -
VC 1.55 2.12 2.05
219
BIBLIOGRAPHY
Agostoni E., Mead J. Statics of the respiratory system in: Handbook of
Physiology, Section 3: Respiration, Vol. 1. W.O. Fenn and H. Rahn (Eds) - The
American Physiological Society, Washington, DC, 1964; 387 - 409.
Agostoni E., Mognoni P., Torri G., Saracino F. Relation between changes of rib
cage circumference and lung volume. J. Appl. Physiol. 1965; 20: 1179 - 1186.
Agostini E., Rahn H. Abdominal and thoracic pressures at different lung volumes.
J. Appl. Physiol. 1960; 1087 - 1092.
Allen S.M., Hunt B., Green M. Fall in vital capacity with posture. B. J. Dis.
Chest. 1985; 79: 267 - 272.
Axen K. Adaptations of quadriplegic men to consecutively loaded breaths. J.
Appl. Physiol. 1984; 56: 1099 - 1103.
Basmajian J.V., Stecko G. A new bipolar electrode for EMG. J. Appl. Physiol.
1962; 17: 849.
Baydur A, Behrakis P.K., Zin W.A., Jaeger M., Milic-Emili J. A simple method for
assessing the validity of the oesophageal balloon technique. Am. Rev. Resp.
Dis. 1982; 126: 788 - 791.
Bergofsky E.H. Mechanism for respiratory insufficiency after cervical cord
injury. Ann. Intern. Med. 1964; 61: 435 - 447.
2 2 0
Black L.F., Hyatt R.E. Maximal respiratory pressures normal values and
relationships to age and sex. Am. Rev. Respir. Dis. 1964; 99: 696 - 702.
Braun N.M.T., Arora N.S., Rochester D.F. Force length relationship of the normal
human diaphragm. J. Appl. Physiol. 1982; 53: 405 - 412.
Cameron C.S., Scott J.W., Jousse A.T., Botterell E.M. Diaphragmatic respiration
in a quadriplegic patient and effect of position on his vital capacity. Ann.
Surg. 1955; 141: 451 - 456.
Campbell E.J.M. An electromyographic study of the role of the abdominal muscles
in breathing. J. Physiol. 1952; 117: 222 - 233.
Campbell E.J.M. The effects of increased resistance to expiration on the
respiratory behaviour of the abdominal muscles and intra-abdominal pressure. J.
Physiol. 1957; 136: 556 - 562.
Campbell E.J.M., Agostini E., Newsom-Davies J. The respiratory muscles.
Mechanics and neural control (2nd edition). Lloyd Luke London, 1970; 175 - 179.
Campbell E.J.M., Green J.H. The expiratory function of the abdominal muscles in
man. An electromyographic study. J. Physiol. 1953a; 120: 409 - 418.
Campbell E.J.M., Green J.H. The variation in intra-abdominal pressure and the
activity of the abdominal muscles during breathing; A study in man. J. Physiol.
1953b; 122: 282 - 290.
2 2 1
Campbell E.J.M., Green J.H. The behaviour of the abdominal muscles and the
intra-abdominal pressure during quiet breathing and increased pulmonary
ventilation; A study in man. J. Physiol. 1955; 127: 423 - 426.
Campbell E.J.M., Dickinson C.J., Dinnick O.P., Howell J.B.L. The immediate effect
of threshold loads on the breathing of men and dogs. Clin. Sci. 1961; 21: 309
- 320.
Carman D.J., Blanton P.L., Biggs N.L. Electromyographic study of the
antero-lateral abdominal musculature using indwelling electrodes. Am. J. Phys.
Med. 1972; 51: 113 - 129.
Celli B.R., Rassulo J., Berman J.S., Make B. Respiratory consequences of
abdominal hernia in a patient with severe chronic obstructive pulmonary disease.
Am. Rev. Resp. Dis. 1985; 131: 178 - 180.
Cobb J. A projected grid method for recording the shape of the human face.
R.A.E. Technical Report 71184. (MOD London) 1972.
Cotes J.E. Lung function, assessment and application in medicine. 4th edition,
Blackwell, Oxford 1979; 329 - 387.
D'Angelo E., Agostini E. Immediate response to expiratory threshold load.
Respiration Physiology 1975; 25: 209 - 284.
Danon J., Druz W.S., Goldberg N.B., Sharp J.T. Function of the isolated paced
diaphragm and the cervical accesory muscles in Cl quadriplegics. Am. Rev. Resp.
Dis. 1979; 119: 909 - 919.
2.2 2
De Troyer A. Mechanical role of the abdominal muscles in relation to posture.
Resp. Physiol. 1983a; 53: 341 - 353.
De Troyer A. Mechanical action of the abdominal muscles. Bull. Europ.
Physiopath. Resp. 1983b; 19: 575 - 581.
De Troyer A., Borenstein S., Cordier R. Analysis of lung volume restriction in
patients with respiratory muscle weakness. Thorax 1980; 35: 603 - 610.
De Troyer A., Estenne M. Co-ordination between rib cage and diaphragm during
quiet breathing. J. Appl. Physiol. 1984; 57: 899 - 906.
De Troyer A., Heilpom A. Respiratory mechanics in quadriplegia. The respiratory
function of the intercostal muscles. Am. Rev. Resp. Dis. 1980; 122: 591 -
600.
De Troyer A., Sampson M., Sigrist S., Macklera P.T. The diaphragm: Two muscles.
Science 1981; 213: 237 - 238.
De Troyer A., Sampson M., Sigrist S., Kelly S. How the abdominal muscles work on
the rib cage. J. Appl. Physiol. 1983; 54: 465 - 469.
Decrainer M., De Troyer A., Kelly S., Zocchi L., Macklem P.T. Regional differences
in abdominal pressure swings in dogs. J. Appl. Physiol. 1984; 57: 1692 -
1687.
223
Denison D.M., Peacock A.J., Morgan M.D.L., Branthwaite M.A., Gourlay A.R. (Does
the lung work). Shedding light on the subject. B. J. Dis. Chest 1982; 76: 1
- 14.
Druz W.S., Sharp J.T. Activity of respiratory muscles in upright and recumbent
humans. J. Appl. Physiol. 1981; 51: 1552 - 1561,
Duchenne G.B.A. Physiologie Des Movements Demontree A I/aide De I/experimentation
Electrique et De L'observation Clinque, et Applicable a I/etude Des Paralysies Et
Deformations. Bailiere, Paris, 1867; 614 - 620.
Duomarco J.L., Rimini R. La Presion Intra-abdominal En El Hombre. 1947; Buenos
Aires: El Ateneo.
Eisele J., Trenchard D., Burki N., Guz A. The effect of chest wall block on
respiratory sensation and control. Clin. Sci. 1968; 35: 23-33.
Estenne M., De Troyer A. Relationships between respiratory muscle electromyogram
and rib age motion in tetraplegia. Am. Rev. Respir. Dis. 1985; 132: 53 -
59.
Estenne M., Heilporn A., Delhez L. , Yemault J.C., De Troyer A. Chest wall
stiffness in patients with chronic respiratory muscle weakness. Am. Rev. Resp.
Dis. 1983; 128: 1002 - 1007.
Floyd W.F. and Silver P.H.S. EMG study of patterns of activity of the anterior
abdominal wall muscles in man. J. Anat. (Lond) 1950; 84: 132 - 145.
224
Forner J.U. Lung volumes and mechanics of breathing in tetraplegics. Paraplegic
1980; 18: 258 - 266.
Frankel H.L,, Guz A., Noble M. Respiratory sensations in patients with cervical
cord transection. Paraplegia 1971; 9: 132 - 136.
Fugl-Meyer A.R. Effects of respiratory muscle paralysis in tetraplegic and
paraplegic patients. Scand. J. Rehab. Med. 1971; 3: 141 - 150.
Fugl-Meyer A.R., Grimby G. Ventilatory Function in tetraplegic patients. Scand.
J. Rehab. Med. 1971a; 3: 151 -160.
Fugl-Meyer A.R., Grimby G. Rib cage and abdominal volume ventilation partitioning
in tetraplegic patients. Scand. J. Rehab. Med. 1971b; 3: 161 - 167.
Goldman M.D. Measurements of respiratory movements of rib cage and abdomen In:
Techniques in the life sciences, P4/II Respiratory physiology - Elsevier
scientific publishers Ireland Ltd 1984; P416 1 - 20.
Goldman M.D., Gribbin H.R., Bruce E.N., Haacke E.M. Estimation of chest wall
volume-motion coefficients: A third degree of freedom. The Physiologist. 1981;
24: 98.
Goldman M.D., Mead J. Mechanical Interaction between diaphragm and rib cage. J.
Appl. Physiol. 1973; 35: 197 - 204.
Gordon J.E. Structure, or why things dont fall down. Penguin books, London 1978.
225
Gourlay A.P., Kaye G. , Denison D.M., Peacock A.J., Morgan M.D.L. Analysis of an
optical mapping technique for lung function studies. Comput. Biol. Med. 1984;
14: 47 - 58.
Grassino A., Goldman M.D., Mead J., Sears T.A. Mechanics of the human diaphragm
during voluntary contraction: Statics: J. Appl. Physiol. 1978a; 44: 824 -
839.
Grassino A ., Goldman M .D ., Mead J., Sears T.A. Mechanics o f the human diaphragm
during vo lu n ta ry co n tra c tio n : Dynamics. J. Appl. P hys io l. 1978b; 44: 840 -
848.
Grimby G., Goldman M., Mead J. Respiratory muscle action inferred from rib cage
and abdominal V-P partitioning. J. Appl. Physiol. 1976; 41: 739 - 751.
Guttmann L., Silver J.R. Electromyographic studies on reflex activity of the
intercostal and abdominal muscles in cervical cord lesions. Paraplegia 1969; 1:
1 - 22 .
Heaf P .D .J ., Prime F .J . Compliance o f the thorax in normal human sub jec ts . C lin *
S c i. 1956; 15: 319 - 327.
Hedenstiema G., Strandberg A., Brismar B., Lundquist H., Svenson L., Tokics L.
Functional residual capacity, thoraco-abdominal dimensions and central blood
volume during general anaesthesia, muscle paralysis and mechanical ventilation.
Anaesthesiology 1985; 62: 247 - 254.
Hemingway A . , Bors E . , Hubby R.P. An in v e s tig a tio n o f the pulmonary fu n c tio n o f
parap leg ics . J. C lin . In ve s t. 1958; 37: 773 - 782.
226
Konno K., Mead J. Measurement of the separate volume changes in rib cage and
abdomen during breathing. J. Appl. Physiol. 1967; 22: 407 - 422.
Konno K . , Mead J . S ta t ic volume pressure c h a ra c te r is t ic s o f the r ib cage and
abdomen. J . Appl. P hys io l. 1968; 24: 544 - 548.
Kovats F. Piethysmographie Optique Du Tronc. Bull. Physiopath. Resp. 1970;
8: 833 - 845.
Kovats F. Morphometric study of breathing movements using a stereometric method.
Proc. Int. Soc. Photogramm. Symposium on Biostereometrics, 1974; 340 - 377.
Kovats F. Biosteriometric De La Respiration. Applications of Human
Biostereometrics (Nato) Spie. 1978; 166; 279 - 285.
Ledsome, J.R . , Sharp J.M. Pulmonary Function In Acute C e rv ica l Cord In ju r y . Am.
Rev. Resp. D is . 1981; 124: 41 - 44.
Lovesey E .J . A method fo r determ ining fa c ia l contours by shadow p ro fe c tio n .
R.A.E. Technical Report No. 66192. (MOD Lond) 1966.
McKinley A.C., Auchinloss JA., J.M., Gilbert R., Nicholas J.J. Pulmonary function
ventilatory control and respiratory function in quadriplegic subjects. Am. Rev.
Respir. Dis. 1969; 100: 526 - 532.
McMichan J .C . , Luc M ., Westbrook P.R. Pulmonary dys function fo llo w in g traum atic
qua d rip le g ia re co g n itio n , prevention and treatm ent. JAMA, 1980; 243: 528 - 531.
227
Mead, J. The functional significance of the area of apposition of diaphragm to
rib cage. Am. Rev. Resp. Dis. 1979; 119: 31 ~ 32.
Mead J., Peterson N., Grimby G. Pulmonary ventilation as measured from body
surface movement. Science 1967; 156: 1383 - 1384.
Michaelis L.S. Spasticity in spinal cord injuries. In: Vinken P.J., Bruyn G.W.
Injuries of the spine and spinal cord. Part II. New York: American Elsevier Co.
1976; 477 - 87.
Mier A., Brophy C., Estenne M., Moxham J., Green M., De Troyer A. Action of
abdominal muscles on rib cage in humans. J. Appl. Physiol. 1985; 58: 1438 -
1443.
Milic-Emili J. Measurements of pressures in respiratory physiology. In:
Techniques in the lift sciences, P4/II Respiratory Physiology - Elsevier
Scientific Publishers Ireland Ltd 1984; P412: 1 - 22.
Milic-Emili J.,Henderson J.A.M., Turner J.M., Glauser E.M. Improved technique for
estimating pleural pressure from oesophageal balloon^. J. Appl. Physiol. 1964;
19: 207 - 211
Miller J.M., Moxham J., Green M. The maximal sniff in the assessment of diaphragm
function in man. Clin. Sci. 1985; 69: 91 - 96.
Morgan M.D.L., De Troyer A. The individuality of chest wall motion in
tetraplegia. Bull. Eur. Physiopathol. Respir. 1984; 20: 547 - 552.
228
Morgan M.D.L., Gourlay A.R., Denison D.M. An optical method of studying the shape
and movement of the chest wall in recumbent patients. Thorax. 1984; 39: 101 -
106.
Morgan M.D.L., Gourlay A.R., Silver J.R. , Williams S. J., Denison D.M.
Contribution of the rib cage to breathing in tetraplegia. Thorax 1985; 40: 613
- 617.
M ortola J .P . , Sant"Ambrogio G. Motion o f the r ib cage and the abdomen in
te tra p le g ia p a tie n ts . C lin . S c i. 1978; 54: 25 - 32.
Moulton A., Silver J.R. Chest movements in patients with traumatic injuries of
the cervical cord. Clin. Sci. 1970; 39: 407 - 422.
Peacock A., Gourlay A., Denison D.M. Optical measurement of change in trunk
volume with breathing. Bull. Eur. Physiopathol Respir. 1985; 21: 125 - 129.
Peacock A . J . , Morgan M .D .L., Gourlay, Turton C ., Denison D.M. O p tica l mapping o f
the thoraco abdominal w a ll. Thorax 1984; 39: 93-100.
Pichurko B., McCool F.D., Scanlon P., Loring S., Scutsky A., Rossier A., Sakarati
M.; Brown R. Factors related to respiratory function recovery following acute
quadriplegia. Am. Rev. Respir. Dis. 1985; 131: A 337.
Sackner J.D., Nixon A.J., Davis B., Atkins N., Sackner M.A. Non-invasive
measurement of ventilation during exercise using a respiratory inductive
plethysmograph. Am. Rev. Respir. Dis. 1980; 122: 867 - 871.
229
Sharp J.T., Goldberg N.B., Druz W.S., Danon J. Relative contribution of rib cage
and abdomen to breathing in normal subjects. J. Appl. Physiol. 1975; 39: 608
- 618.
Silver J.R. , Gibbon N.O.K. Prognosis in tetraplegia. BMJ. 1968; 4: 79 - 83.
Silver J.R. , Lehr R.P. Dyspnoea during generalised spasms in tetraplegic
patients. J. Neurol. Neurosurg. Psychiat. 1981a; 44: 842 - 845.
Silver J.R., Lehr R.P. Electromyographic investigation of the diaphragm and
intercostal muscles in tetraplegics. J. Neurol. Neurosurg. Psychiat. 1981b;
44: 837 - 842.
Singer C. Galen on anatomical procedures published for the Wellcome Histological
Medical Museum, by G. Cumberlege, Oxford University Press, London, New York,
Toronto, 1976.
Stone D.J., Keltz H. The effect of respiratory muscle dysfunction on pulmonary
function. Studies in patients with spinal cord injuries. Am. Rev. Resp. Dis.
1963; 88: 621 - 629.
Strohl K.P., Mead J., Banzett R.B., Lehr J., Loring S.H., O'Cain C.F. Effect of
posture on upper and lower rib cage motion and tidal volume during diaphragm
pacing. Am. Rev. Resp. Dis. 1984; 130: 320 - 321.
230
Strohl K.P., Mead J., Banzell R.B., Loring S.H., Kosch P.C. Regional differences
in abdominal muscle activity during various manoeuvres in humans. J. Appl.
Physiol. 1981; 51: 1471 - 1476.
Vellody V.P., Nassery M., Druz W.S., Sharp J.T. Effects of body position change
on thoraco-abdominal motion. J. Appl. Physiol. 1978; 45 : 581 - 589.
Wade O.L. Movements of the thoracic cage and diaphragm in respiration. J.
Physiol. (Lond) 1954; 124: 193 - 212.
Wharton C.W., Morgan T.M. Ankylosis in the paralysed patient. J. Bone & Joint
Surg. 1952; Y105 - 112.
Williams A.R. Techniques in applied photography: contour mapping by light
sectioning. B.J. Phot. 1977; 265 - 266.
231
