Basic Science For Anaesthetists

BASIC SCIENCE FOR ANAESTHETISTS
This is a revised edition of a book originally titled Anaesthetic Data Interpretation. The
new title better reflects the contents of the book, which contains additional chapters relevant to the Primary FRCA examination. New topics covered include the ventilatory
response to oxygen and carbon dioxide, which is now a core knowledge requirement,
new concepts in cardiovascular physiology, receptor types and the molecular actions
of anaesthetics. Some of the revisions reflect advances in technology; for example, the
uses of the capnograph and the oxygen analyser have advanced considerably in recent
years. The aim is to provide a concise and understandable review of the physics,
mathematics, statistics, physiology and pharmacology of anaesthesia. Basic Science for
Anaesthetists is a concise and informative text, which will be invaluable for trainee
anaesthetists and an aid to teaching for the trainers.
S Y L V A D O L E N S K A qualified from Charles University, Prague, trained as an anaesthetist in the UK and is currently Consultant Anaesthetist at William Harvey Hospital,
Ashford, Kent. She has also acquired the KSS Deanery Certificate in Teaching. Her
other key professional interests are airway management and obstetric anaesthesia.
BASIC SCIENCE FOR

ANAESTHETISTS
BY
SYLVA DOLENSKA MD LMSSA FRCA
Consultant Anaesthetist, Department of Anaesthetics,
The William Harvey Hospital, Ashford, Kent, UK
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521676021
Cambridge University Press 2006
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2006
eBook (NetLibrary)
ISBN-13 978-0-511-16895-6
ISBN-10 0-511-16895-0
eBook (NetLibrary)
paperback
ISBN-13 978-0-521-67602-1
paperback
ISBN-10 0-521-67602-9
Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Every effort has been made in preparing this publication to provide accurate and up-todate information which is in accord with accepted standards and practice at the time of
publication. Although case histories are drawn from actual cases, every effort has been
made to disguise the identities of the individuals involved. Nevertheless, the authors,
editors and publishers can make no warranties that the information contained herein is
totally free from error, not least because clinical standards are constantly changing through
research and regulation. The authors, editors and publishers therefore disclaim all liability
for direct or consequential damages resulting from the use of material contained in this
publication. Readers are strongly advised to pay careful attention to information provided
by the manufacturer of any drugs or equipment that they plan to use.
To my husband
CONTENTS
List of abbreviations and symbols
List of figure captions
Forewords
Preface
Preface to the second edition
Part 1: Physics, mathematics, statistics, anaesthetic

apparatus
1. Gas compression, relationship of volume, pressure
and temperature
2. Real gas compression
3. Flow and resistance
page xi
xiii
xvii
xxi
xxiii
1
2
4
8
4. Heat, vaporization and humidification
14
5. Simple mechanics 1: mass, force, pressure
18
6. Simple mechanics 2: work and power
22
7. Mathematical concepts
26
8. Exponentials 1: the curves
30
9. Exponentials 2: properties of exponential decay curve
36
10. Descriptive statistics
40
11. Presentation of data
44
12. Receiver operating characteristic curve

13. Gas supply and pressure
48
14. The circle system

15. The Mapleson A (Magill) breathing system
54
16. T-pieces
60
17. Lung filling with automatic lung ventilators
62
Part 2: Clinical measurement

1. Basic measurement concepts
52
58
67
68
2. Electromanometers, frequency response and

damping
72
3. Pulse oximeter principle
76
vii
Contents
4. Oxygen content and oxygen tension measurement
80
5. Capnography
86
6. [H+], pH and its measurement
92
7. Principles of measurement of volume and flow in

gases and liquids
96
8. Cardiac output measurement by thermal dilution

technique
100
9. Measurement of the mechanical properties of the

chest
102
10. Lung volumes and their measurement
104
Part 3a: Physiology: the cardiovascular system

1. The cardiac cycle and the intravascular pressure
waveforms
109
114
3. Control of cardiac output, regulation of cardiac

function
116
4. Cardiac cycle: pressure-volume relationships
118
5. Blood pressure and blood volume relationship
122
6. Cerebral blood flow
124
7. Coronary circulation
128
Part 3b: Physiology: the respiratory system

1. Oxyhaemoglobin dissociation curve
viii
110
2. Cardiovascular effects of intermittent positive

pressure ventilation
133
134
2. Respiratory mechanics 1: Static properties, factors

affecting compliance, closing volume
136
3. Respiratory mechanics 2: Dynamic properties,

factors affecting resistance
142
4. Ventilationperfusion relationship
146
5. Oxygen cascade, oxygen therapy and shunt fraction
152
6. Gas R line, solution of the ventilation/perfusion

model
156
7. Ventilatory response to oxygen
160
8. Ventilatory response to carbon dioxide
162
165
1. Drug elimination
166
2. Uptake and distribution of inhalational anaesthetic

agents
170
3. Pharmacodynamic effects of drugs
174
4. Minimum alveolar concentration and lipid solubility
180
5. Receptor types, molecular action of anaesthetics

6. Context sensitive half-time
182
Index
191
Contents
Part 4: Pharmacology
186
ix
ABBREVIATIONS and SYMBOLS

Units are shown in parentheses
a
acceleration (m s2 )
A
ampere
A
area (m2 )
c
concentration (g l1 )
C
compliance (l Pa1 )
C
coulomb
Cd
candela
C
degrees Celsius
d
distance (m)
d
rate of change (derivation)
D
diameter (m)
e
base of natural logarithms
E
extinction coefficient
F
force (kg m per s = N)
g
gram
I
light or current intensity (Cd or A)
J
joule
K or k constant
K
kelvin
l
length (m)
l
litre
m
metre
mol
amount of substance that contains as many elementary particles as
there are atoms in 0.012 kg carbon-12
n
number
N
newton
P
power (J s-1 = W)
p
pressure (Pa ); see chapter on gas pipeline pressure for other units
and their conversion
Pa
pascal
Q
electric charge (C)
Q
flow (l s-1 ) (also denoted as dV/dt)
r
radius (m)
R
resistance (Pa l-1 per s)
R
universal gas constant
Re
Reynolds number (dimensionless)
STP
standard temperature and pressure (0 C, 1 atmosphere = 273 K,
101.3 kPa)
t
time (s)
xi
List of abbreviations and symblos

xii
T
v
V
W
W
absolute temperature (K)

velocity (m s-1 )
volume (1)
watt
work (kg m2 )
Greek symbols
(eta)
(mu)
(pi)
(ro)
(sigma)
(capital sigma)
(tau)
(zeta)
viscosity
population mean
3.141592653 . . .
density
population standard deviation (SD)
summa = total
time constant
damping
FIGURE CAPTIONS
Part 1
1. Three gas laws: Boyles, Charles and Gay-Lussac.
page 3
2. Avogadros hypothesis.
3
3. Isothermic compression.
5
4. Adiabatic decompression.
7
5. Laminar flow as the product of area and velocity. The influence of
doubling the radius on flow.
9
6. Laminar flowpressure relationship.
11
7. Laminar flowresistance relationship.
11
8. Turbulent flowdriving pressure relationship.
13
9. Turbulent flowresistance relationship.
13
10. Latent heat of vaporization of water.
15
11. Latent heat of vaporization of nitrous oxide.
15
12. Water vapour content (absolute humidity) of air fully saturated
with water, as a function of temperature.
17
13. Force as a product of pressure and area.
19
14. Pressure generated in different size syringes with constant force.
19
15. Pressure-reducing valve.
21
16. Work as the product of force and distance.
23
17. Work (energy) as the product of pressure and volume.
23
18. Power as a derivative of work; cardiac power at two levels of
inotropy.
25
19. Cardiac power as a product of force and velocity.
25
20. Cardiac power as a product of pressure and flow.
25
21. Linear relationship.
27
22. Reciprocal relationship the rectangular hyperbola.
27
23. Square function parabola.
29
24. Sine waves and their addition.
29
25. Exponentials: (a) exponential growth curve. (b) growth of bacteria
with time.
31
26. Exponentials: (a) exponential decay curve. (b) exponential decline
plasma remifentanil concentration.
31
27. Exponentials: (a) saturation exponential curve.
(b) lung filling with a constant pressure generator.
33
28. Logarithmic curve and exponential growth curve.
33
29. Exponentials: rate of decay, time constant.
37
30. Area under the exponential curve.
39
31. Normal distribution of height in adult men.
41
32. Gaussian distribution blood glucose measurements in a large
normal sample.
41
xiii
33. Positive skew distribution of hydrogen ion concentration in

humans.
34. Negative skew distribution of haemoglobin concentration in
women.
35. Bimodal distribution.
36. Pie diagram.
37. Histogram (bar diagram).
38. Frequency histogram.
39. Scatter diagram: correlation coefficient = 1.
40. Scatter diagram: correlation coefficient = 0.
41. Line drawing with error bars.
42. Frequency distribution of fasting blood sugar in normal and
diabetic population.
43. ROC curves.
44. Pressure/volume relationship in medical gas supply.
45. The circle system.
46. The effect of nitrous oxide uptake and fresh gas flow in the circle
system on alveolar oxygen concentration.
47. Uptake of volatile anaesthetics as a function of time.
48. The Mapleson A breathing system.
49. Hypothetical pressure and expiratory flow in the Mapleson A
system.
50. T pieces.
51. Inspiratory flow and fresh gas flow in the Bain system.
52. Lung filling with a constant pressure generator.
53. Lung filling with a constant flow generator.
Part 2
54. Linearity principle: (a) Measurement accurate but imprecise.
(b) measurement precise but inaccurate.
55. Drift: (a) baseline (zero) drift, (b) sensitivity drift.
56. Hysteresis loop.
57. Critical damping.
58. Electromanometer principle.
59. Low damping.
60. High damping.
61. Optimal damping.
62. Extinction coefficients of reduced haemoglobins, in red and
infra-red light.
63. Relationship between saturation of arterial blood with oxygen and
the light absorbances ratio of haemoglobin.
64. Components of the pulse oximeter waveform.
65. Oxygen measurement: van Slyke apparatus.
66. Oxygen measurement: (a) fuel cell, (b) Clarke electrode.
67. Paramagnetic oxygen analyser.
68. Quadrupole mass spectrometer.
xiv
43
43
43
45
45
45
47
47
47
49
51
53
55
55
57
59
59
61
61
63
65
69
69
69
71
73
75
75
75
77
77
79
81
83
83
85
87
87
89
91
91
93
95
69. Principal components of a capnograph.

70. Single breath analysis of expired CO2 .
71. Abnormalities of single breath curve. (a) chronic obstructive
airways disease. (b) cardiogenic oscillations. (c) dip in plateau.
(d) rebreathing.
72. Hypercapnia: (a) carbon dioxide overproduction or absorption
(b) rebreathing.
73. Hypocapnia: (a)overventilation, (b) collapse, (c) disconnection.
74. [H+ ] pH relationship on a linear and logarithmic scale.
75. The pH electrode.
76. Flow measurement; relationship between flow and resistance for
constant pressure.
77. Linear relationship between driving pressure and flow.
78. Flow as a derivative of volume with respect to time.
79. Area under the line of flow.
80. Decreasing flow.
81. Cardiac output measurement.
82. Pressure changes in constant volume plethysmograph.
83. Lung volumes.
84. Measurement of functional residual capacity (FRC) by helium
wash-in method.
85. Measurement of functional residual capacity (FRC) by nitrogen
washout.
86. Single breath nitrogen washout.
97
97
99
99
99
101
103
105
105
107
107
Part 3a
87. Intravascular pressure waveforms on the right side of the heart
during pulmonary artery catheterization.
111
88. Arterial blood pressure and respiratory swing.
113
89. Arterial blood pressure and pulse during the Valsalva manoeuvre. 115
90. FrankStarling curves.
117
91. Left ventricular and diastolic pressurevolume relationship.
119
92. (a) Left ventricular pressurevolume loop at steady state;
(b) The effect of increased inotropy on left ventricular
pressurevolume loop. (c) The effect of left ventricular failure
(reduced inotropy) on left ventricular pressurevolume loop.
121
93. Classes of haemorrhage.
123
94. Cerebrovascular autoregulation.
125
95. Cerebral elastance curve.
125
96. Autoregulation of cerebral blood flow within physiological limits. 127
97. Cerebral blood flow as a function of arterial carbon dioxide tension
127
(paCO2 ) and oxygen tension (paO2 ).
98. Coronary artery flow and arterial blood pressure.
129
99. Coronary autoregulation.
129
100. Effect of inhalational anaesthesia with halothane on coronary blood
flow and myocardial oxygen consumption.
131
xv
Part 3b
101. Oxyhaemoglobin dissociation curve.
102. Respiratory mechanics. (a) resting position, (b) pressure gradients
after a tidal breath.
103. Static compliance.
104. The effect of age on lung and chest wall compliance.
105. Changes in closing volume (CV) and functional residual capacity
(FRC) during adulthood.
106. Size of airway and total cross-sectional area.
107. Airway resistance and age.
108. Dynamic compliance loops during spontaneous respiration.
109. Lung pressurevolume diagram in a young, healthy adult.
110. Alveolar volume in relation to distance from lung apex.
111. Diagram of the West zones of the lung.
112. Schematic drawing of ventilationperfusion relationship in the
three zones of the lung.
113. The oxygen cascade.
114. The effect of shunt function and inspired oxygen fraction
(fio2 ) on arterial oxygen (Pa O2 ).
115. Gas R line.
116. Ventilatory response to oxygen.
117. Ventilatory response to carbon dioxide.
Part 4
118. First-order kinetics of remifentanil and alfentanil.
119. Zero order and first order elimination of blood alcohol.
120. Biexponential decline in plasma concentration of a drug after
intravenous injection in a two-compartment model.
121. Factors influencing alveolar gas or vapour concentration during
uptake.
122. Rate of rise of alveolar concentration for different anaesthetic
agents.
123. Drug dose and effect.
124. Opioid drugs log doseresponse curves.
125. Competitive and non-competitive antagonism of norepinephrine.
126. ED50 , LD50 .
127. MAC and lipid solubility of volatile agents in 100% oxygen and in
66% nitrous oxide.
128. Multisubunit ligand-gated ion channel.
129. G-protein coupled receptor.
130. Receptor as enzyme.
131. Drug levels in three compartments after a short infusion.
132. Drug levels in three compartments after a prolonged infusion.
133. Plasma alfentanil concentration after short and long infusion.
134. Context-sensitive half-time after short infusion.
135. Context-sensitive half-time after prolonged infusion.
xvi
135
137
137
139
141
143
143
145
147
147
149
151
153
155
157
161
163
167
167
169
171
173
175
177
179
179
181
183
183
185
187
187
189
189
189
FOREWORD
The syllabus for the Primary FRCA examination is broad, covering basic
anaesthesia and associated skills together with an in depth knowledge of the
principles of basic science which underlie clinical practice. Added to this,
is the requirement to pass the examination at an early stage of the trainees
career. Often, it is an inadequate understanding or wariness of concepts which
involve physics or simple mathematics that is the impediment to success in
the examination.
The author has written a book which explains the principles of physics,
mathematics and statistics and applies many of them to an understanding of
anaesthetic apparatus, clinical measurement, cardiovascular and respiratory
physiology, and general pharmacology. Each concept is supported by a graph
or diagram which is explained in the text. A graphical display of data or a
good diagram is often the key to interpretation and conveying a thorough
understanding of subject matter to an examiner. This approach applies equally
when responding to a question in an oral examination or when supplementing
a written answer.
This book is undoubtedly aimed at the candidate sitting the Primary examination, however, the Final FRCA candidate should not forget that the theme
of questioning in the second oral examination is basic science applied to anaesthesia, intensive care and pain management. This book should not be regarded
as a substitute for the standard textbooks but will be invaluable as a supplement and also for revision. Senior colleagues will find in this book a concise
refresher course on basic science principles that will be of personal value and
will assist in teaching trainees. A proportion of candidates fail the oral section
of an examination having done well in the written part. This would suggest
they have the knowledge but fail in their verbal presentation. There is a fund
of questions, diagrams and graphs in this book that can form the basis of mock
vivas for candidates to improve their fluency of presentation in preparation
for the examination proper.
Leslie E. Shutt
Bristol
January 2000
xvii
FOREWORD
The successful and safe practice of anaesthesia depends, amongst other things,
upon a good comprehension of the scientific foundations of the subject. It is
for this reason that all examining boards set scientific questions in various parts
of their examinations, whether in conventional multiple choice or single best
answer format, formal essays, short answers, OSCEs (objective structured
clinical examinations) or in the oral examinations. Candidates have much
more difficulty with the basic and applied science sections of the examination
than with any other parts. In particular, the understanding of physics and the
application of physical principles are not easy. Many candidates, quite frankly,
lack basic education in these topics when starting at medical school; moreover,
they are less easy to learn as one grows older, especially when embarking on
a busy clinical career in anaesthesia.
I can remember from my own experiences as a candidate for the Primary
FRCA (then called the FFARCS) a legendary examiner who would push a
sheet of paper over to the unfortunate candidate during a viva and invite him or
her to draw the structure of pethidine (merperidine, Demerol). Thank goodness that does not happen nowadays, but reliance on the production of drawings or graphs to illustrate a point is very common, for indeed a good picture is
worth a thousand (some say ten thousand) words. The interpretation of radiographs and electrocardiograms has stood the test of time. Moreover, many
examiners now rely upon previously produced drawings or photographs of varying clarity and quality - as part of the examination, and I must confess
that I have produced some of my own over the years.
Sylva Dolenska originally intended to use the apt subtitle do you get the
picture? for this book but it was changed to Anaesthetic Data Interpretation and
the Primary FRCA examination became her target. Nevertheless, success
in any examination in anaesthesia, wherever in the world, relies upon the
grasp and understanding of basic scientific facts. Hence her approach of using
illustrations (linked to explanations) that have almost come straight from
the examiners briefcase provides welcome help for candidates. Examples are
drawn from everyday clinical anaesthesia: the use of medical gases, respiratory and circulatory physiology, the behaviour and distribution of drugs, and
concluding with concepts of receptors. Many current and future candidates
for examinations in anaesthesia should be grateful for the help this will give
them.
Anthony P. Adams
Professor of Anaesthetics in the University of London at the Guys, Kings and
St. Thomas School of Medicine, Kings College, London.
January 2000
xix
PREFACE
There are many textbooks to chose from when preparing for the FRCA
examination; the candidate suffers not from lack of information but rather
from being inundated with it. The candidate then has the task of information
sorting and data compression to memorize and utilize all this information.
Graphic representation of data is an excellent form of data compression; figures or drawings are frequently asked about at the viva examination, particularly since the candidates understanding of a problem comes across most
clearly when drawing a figure or a using a picture. For anaesthetists whose first
language is not English, figures are also a good way of approaching a topic
I certainly find it easier to find words when describing a plot.
I constructed parts of this book when revising for the Primary Examination and afterwards when preparing tutorials. The book differs from most
in that the text accompanies the pictures, rather than the pictures complementing the text. In many cases, the text is simply a legend to the figure or diagram, expanded by background information. For this reason, the
figures are described only by the names of the axes and their units along
with identification of any other important lines and symbols. The layout
each page of text opposite the relevant figure(s) conveys the essential link
between picture and text, and I hope it makes orientation and understanding
easier.
Not all knowledge required for the FRCA (Primary or Final) is suitable
for graphical representation. The properties of anaesthetic drugs, for instance,
lend themselves to tabulation rather than to diagrammatic representation,
and they require little in the way of understanding of fundamental concepts. I
therefore recommend the reader to read basic, comprehensive textbooks before
beginning this text because, first, this book is not intended as comprehensive
and, second, because a complete textbook will give a fuller perspective on the
topics represented.
The book was updated according to the latest FRCA syllabus, and it
shows the relevance of basic science to clinical anaesthesia in practical examples throughout. A choice had to be made, however, even among the topics suitable for illustration. Most of the topics I have chosen rank highly
in order of importance to anaesthetists (Jones, Anaesthesia (October 1997),
930). Descriptive statistics and mathematical concepts, although not popular, are included as they appear in the syllabus and because they constitute
the basic knowledge on which the candidate can build an understanding of
other subjects seen as more relevant to anaesthesia (such as the principles of
measurement).
Although the book is intended mainly for the Primary FRCA candidate,
it would also make an excellent aide-memoire for clinical tutors and all
xxi
Preface
xxii
practising anaesthetists who undertake teaching and wish to remain connected

with the basic principles on which anaesthesia is built. I hope the prospective
FRCA candidate will find the book useful.
S. D.
October 1999
London
PREFACE TO THE SECOND EDITION

This is a revised edition of the book originally titled Anaesthetic Data Interpretation, published by Greenwich Medical Media in 2000. The new title better
reflects the contents of the book.
New chapters have been added on the direct or indirect advice of Primary
FRCA examiners. Ventilatory response to oxygen and ventilatory response to
carbon dioxide are now core knowledge requirements, and form the basis of
clinical decision making. New concepts in cardiovascular physiology, such as
the end-systolic pressurevolume relationship are important to our understanding of control of cardiac function. The concept of total intravenous
anaesthesia has evolved around pharmacokinetic research into how drugs
behave when injected at a steady rate. Receptor types and molecular actions
of anaesthetics, although may seem far removed from clinical practice, are
however part of the examination syllabus, providing wider background which
helps to understand how anaesthetics work. Receiver operating characteristic,
an idea recently introduced in medical statistics from aviation, is a concept
that will help to understand scientific articles.
Basic science does not change but technology does and some of the revisions reflect this. Technical advances in monitoring continue apace. The
capnograph and the oxygen analyser are no longer the heavy cumbersome
machines that were difficult to maintain. It is important to know how these
machines work, in order to understand what problems may arise and to troubleshoot.
Updated chapters are based on contemporary texts and new concepts which
are now contained in the syllabus. Many chapters have been improved with
additions of new diagrams.
The text is fairly didactic and describes mostly what is usual, or what the
usual deviation from norm is, since this would be the expectation at the
Primary FRCA examination.
The second edition is still slimline and I hope it will continue as a useful
aid to learning for trainee anaesthetists.
SD, Ashford 2005
xxiii
Part 1
Physics, mathematics,
statistics, anaesthetic
apparatus
1
Part 1
Gas compression, relationship of

volume, pressure and temperature
The universal gas equation describes the equation of state for 1 mole ideal
gas:
p V = R T,
where R = molar (universal) gas constant = NA .k (Avogadro and Boltzman
constants).
The simple equation expresses three different ideal gas laws, depending
on which variable is chosen to be constant (and therefore taken out of the
equation). The assumption for an ideal gas is that the molecules do not occupy
any space; this clearly is not true in practice.
In Figure 1, all three gas laws are depicted. The individual curves (rectangular hyperbolas) relate to Boyles law: they show the relationship between
pressure and volume when temperature is constant pressure and volume
are inversely related. Each curve shows the relationship for a certain temperature, and therefore is called an isotherm. The crosses relate to Charles law:
when pressure is constant, the volume is directly proportional to temperature.
The dots illustrate GayLussacs law: when volume is constant, pressure is
directly proportional to temperature.
Avogadros hypothesis states that the number of molecules per unit volume is independent of the gas concerned, at a given temperature and pressure.
At standard temperature and pressure (STP) it is 6.022 1023 molecules in
22 litres (Avogadros number). This number of molecules is equal to 1 mole
gas. This means that 1 mole of a gas, allowed to expand until it reaches equilibrium with atmospheric pressure, will expand to fill a volume of 22 litres.
Conversely, the pressure exerted by a given number of molecules of ideal
gas in a given volume and temperature is constant (6.022 1023 molecules
in 22 litres will exert a pressure of 1 atmosphere) and is independent of its
molecular weight. For an illustration, see Figure 2, the molecules, heavy or
light, are floating in the given space; the number of molecules and the average
distance between them is the same. Random thermal movement of heavier
molecules will be less at a given temperature than that of lighter molecules;
the resulting kinetic energy will be the same, and so will be the pressure inside
the container.
Figure 2. Avogadros hypothesis.
Gas compression, relationship of volume, pressure and temperature
Figure 1. Three gas laws: Boyles, Charles and GayLussac.
Real gas compression
Part 1
Forces of adhesion in the gas lessen the impact on the container. The result
is that the pressure measured is less than that predicted by the universal gas
equation. The effect is magnified in a smaller volume.
Also, the molecules are not negligible in size; their total volume lessens
the volume of the container, decreasing the distance of travel; a correction for
the volume of the molecules (Vo ) has to be applied.
When gas is compressed at a sufficiently low temperature, the forces of
adhesion eventually cause its liquefaction (i.e. the forces of attraction overcome the random thermal motion).
Isothermic compression decompression
Figure 3 shows slow compression of nitrous oxide under various temperature
conditions. Because compression is slow, there is sufficient time for temperature equilibration with the surroundings. This pressurevolume change is
called isothermic.
The top isotherm for 50 C behaves as an ideal gas isotherm. At 36.5 C the
isotherm just touches the lightly shaded area in the graph, which represents
the gas and liquid phase. This temperature is the critical temperature of
nitrous oxide, above which the gas cannot be liquefied at any pressure. At
lower temperatures, here 20 C, the gas can be liquefied. If compression is
slow to allow temperature to remain constant, the pressure in the container
remains constant until all gas is liquefied: the decrease in volume of the
container is matched by a decrease in volume of the gaseous phase (which is
now called vapour) as it is being liquefied. The space above the liquid phase is
saturated with the vapour, and the pressure inside the container at the given
temperature is the saturated vapour pressure (which is constant at a given
temperature).
Once the total contents are liquefied, and if compression is continued, the
pressure inside the container rises steeply, as liquids are virtually incompressible. Slow decompression would follow the same isotherm in the opposite
direction.
The phenomenon of liquefaction is used in practice to increase the amount
of substance in a container: nitrous oxide can be liquefied at ambient temperature in a moderate climate (but not in the tropics); by contrast, oxygen,
with a critical temperature of 119 C, has to be liquefied in special insulated
vessels to prevent its warming.
1
50
Figure 3. Isothermic compression.
1
Part 1
6
Adiabatic compression decompression

Figure 4 shows a sudden decompression of a real gas. As the gas suddenly
expanded from volume V1 to volume V2 , thermal energy was lost but time
did not allow it to regain heat from the surroundings. The gas therefore
moved from its position on the higher isotherm to the lower one; this abrupt
pressurevolume change is called adiabatic. After reaching temperature
equilibrium, the system returned to the original isotherm (from pressure
p2 to pressure p3 ).
Anaesthetists using nitrous oxide cylinders can observe both phenomena.
Starting with a full cylinder, the pressure at first hardly changes as the vapour
is used. The system is on the flat part of the liquid/vapour area of Figure 3;
vapour being released from the container is immediately substituted by the
formation of more vapour from the liquid phase. Any pressure drop in the
cylinder is due to energy loss from the liquid phase because of the latent heat
of vaporization, and the system moves onto a lower isotherm. The cylinder
starts frosting at its base.
Once all the liquid phase is used up, if using high flow, adiabatic decompression takes place as the pressure inside starts decreasing noticeably and at
a higher rate than expected (Figure 4). When the cylinder is turned off, temperature can equilibrate with the ambient air and the gas inside moves back
onto a higher isotherm, with the result that the pressure gauge now gives a
higher reading because a warmer gas exerts a higher pressure.
1
Figure 4. Adiabatic decompression.
Flow and resistance
Part 1
Flow is defined as the volume of gas or liquid passing a cross sectional area
per unit of time. If the volume in a cylinder (or a large tube) is given by the
product of its area and length, then the flow in this tube can be thought of as
area multiplied by the velocity (see Figure 5):
V = A.
Q = dV/dt = A.d/dt = A.v
(equation 1)
Laminar flow
In laminar flow the fluid moves in a steady manner without eddies or turbulence. A slowly flowing river in a straight stretch is a good approximation.
You may notice that the flow near a river bank is very slow while in the centre
it is the fastest. This is because of frictional forces between the flow and the
side. The speed, or velocity, is therefore related to the distance from the side.
In fact it can be shown that the maximum velocity is directly proportional
to the square of the radius; mean velocity is half the maximum velocity
(
v = v max /2).
The following are factors affecting the flow velocity:
r Square of the radius, as mentioned above.
r Pressure gradient between the beginning and the end point (a
mountain river with a steeper gradient flows much faster than a river near
the sea where the gradient is less).
r Viscosity: we know from experience that more viscous fluids (e.g. oil)
flow more slowly than less viscous fluids (e.g. water). Viscosity in physics
is denoted by the Greek (eta).
r Length of the tube: friction slows down the flow at the sides of the tube.
The longer the tube, the longer acting the frictional force becomes and
the slower the flow.
Mathematically expressed:
v r2
v P
v 1/
v 1/.
From equation 1, this equation has been derived for laminar flow:
= A.
Q
v
where
Pr 2
(the factors above and a numerical factor of 8 are
8
not derived here)
v =
8
A1
V2
V1
r2
r1
Flow and resistance
A2
Figure 5. Laminar flow as the product of area and velocity. The influence of doubling
the radius on flow.
and
A = r 2 .
Part 1
Therefore, substitution for area and average velocity gives:

2
2
4
= r .Pr = Pr .
Q
8
8
(equation 2)
This is the HagenPoiseuille equation that describes laminar flow.

From this we can see that the radius influences the flow by its fourth
power. This is because, as shown above, altering the radius alters both the
area and the velocity, which are the determinants of flow, and each is related
to the radius by its square function.
When looking at a given tube of fixed radius and length, and passing a
flow of a certain fluid of given viscosity, the only variable that remains in
the equation is pressure difference: the greater the pressure, the faster the
flow they are directly proportional (see Figure 6). All the other factors
(radius, length, viscosity plus the numerical factors) are fixed for the given
tube and fluid, independent of flow (see Figure 7), and this combination of
factors is known as the resistance of the tube. We know from experience
that flow and resistance are inversely proportional; resistance then becomes
the denominator in the HagenPoiseuille equation:
=P/R
Q
and therefore, by substitution from equation 2
8
r 4
It can be seen if we wish to alter the resistance to flow, the most influential
factor is the radius of the tube: doubling the radius of a tube will reduce the
resistance (increase the flow) 16 times (24 ). The length is inversely related to
the flow but only by a power of 1. Therefore short and thick does the trick.
The markings on the Rotameter are of unequal length for different flow
rates. At low flows the resistance of the orifice around the bobbin behaves
according to the HagenPoiseuille formula. At high flows it becomes a function of the flow, as the flow becomes turbulent.
R=
10
1
Flow and resistance
Figure 6. Laminar flow linear relationship between pressure and flow.
s per
Figure 7. Laminar flow resistance is independent of flow.
11
Turbulent flow
Part 1
If laminar flow passes through a constriction, it changes to turbulent flow;

the manner of flow is no longer smooth, but it forms eddies. Now there is
no difference in fluid velocity across the tube (i.e. between the flow at the
side and the centre of flow), as there was in laminar flow. The velocity in
the constriction is increased overall (as the same amount of fluid has to
pass through a narrower tube portion). An example of this can be found on
a mountain river reaching a narrow gorge. Because of the lateral movement
in eddies, friction is a lot greater; the resistance to the flow therefore is
no longer constant. It now depends on the flow: the friction is greater,
the greater the speed, i.e. the greater the flow. When measuring resistance
in turbulent flow (e.g. during breathing), the flow rate therefore has to be
specified.
Also, because of the lateral movement in eddies, flow is no longer directly
proportional to the pressure difference; some of the velocity (which, of course,
is increased overall) is wasted on the sideways movement, and it can be shown
that the flow is now related to the square root of the pressure difference as
shown in Figure 8 (to double the flow, pressure difference has to be quadrupled).
Usually, the ordinate and abscissa are reversed, so that the pressure difference becomes the dependent variable, and the familiar parabola is vertically
orientated. Because resistance is now related to flow, it too becomes a squareroot function of the pressure difference (see Figure 9).
Other factors that affect turbulent flow are:
r Radius: flow is proportional to its square (not to the fourth power as
before), i.e. an increase in flow with a greater diameter is less easily
achieved.
r Length of tube an inverse relationship remains.
r fluid density (because it is density that maintains the momentum of
the lateral out of stream movement).
Turbulence (the lateral movement), apart from density, is encouraged by
higher velocity (also higher momentum), tube diameter, D (more lateral
space) and lower viscosity (laminar flow streams less adherent). This is mathematically expressed as the Reynolds number, Re:
Re = D.v . /.
When Re > 2000 (note it is dimensionless as the various units cancel each
other out), turbulence occurs. For a given gas, this depends on flow velocity
and tube diameter. In severe upper respiratory tract obstruction (e.g. tracheal
compression), flow may be improved by providing an inspired mixture of
oxygen and helium: because of the low density of helium, the Reynolds
number may be reduced sufficiently to convert turbulent flow through the
constriction into laminar flow.
12
1
Flow and resistance
--1
Figure 8. Turbulent flow: driving pressure is the square function of target flow.
sper
Figure 9. Turbulent flow: resistance is the square function of flow.
13
1
Part 1
14
Heat, vaporization and

humidification
Heat is one form of energy; temperature is a measure of the random thermal
movement of molecules or atoms (the thermal state of a substance).
Physical changes occur in substances when their temperature is changed
by the addition or removal of heat. These are due to increased or reduced
thermal movement of molecules.
With progressive addition of heat, change of phase happens (from solid
to liquid, from liquid to vapour) as cohesive forces are overcome by random thermal movement. The heat necessary to overcome the cohesive forces
during change of phase is called latent heat. According to the first law of thermodynamics, substances with a higher temperature (higher thermal state)
will pass their heat onto substances with a lower temperature (lower thermal
state). Accordingly, to convert a substance from one phase into another, latent
heat must be supplied or removed.
Specific latent heat refers to 1 kg of a substance converted from one
phase to another at a constant temperature. Temperature and substance must
be stated as specific latent heat is temperature-dependent, and each substance
has different thermal properties. Water is a liquid at 0 C and vaporizes at
100 C at atmospheric pressure (see Figure 10). At higher pressures it would
vaporize at higher temperatures, until it reached its critical temperature (data
not shown); at this point the specific latent heat is zero. Nitrous oxide is a gas
at atmospheric pressure; it must be compressed to liquefy. To turn it back
into gas, latent heat of vaporization must be supplied (see the chapter on real
gas compression). Figure 11 shows that specific latent heat of nitrous oxide
is much higher than that of water at 0 C but it decreases quickly to zero at
36.5 C, the critical temperature of nitrous oxide.
Wa
Figure 10. Latent heat of vaporization of water. Reproduced with permission from
Butterworth-Heinemann.
Figure 11. Latent heat of vaporization of nitrous oxide. Reproduced with permission
from Butterworth-Heinemann.
Heat, vaporization and humidification
15
1
Part 1
Humidification requires addition of latent heat. During inhalation, ambient air is warmed and humidified in the upper airway; latent heat must be
supplied for this. During exhalation, the expired gas is cooled in the upper
airway and some, but not all, of the heat and water is returned to the nasal
mucosa. Because ambient air is usually cooler and less humid than expired
air, water and heat from the body are lost during breathing (about 250 ml
water and 350 calories per day). The biggest part of heat thus lost is the latent
heat of vaporization.
The amount of vapour present in air is limited by temperature. When air
contains the maximum amount of vapour, it is said to be saturated by it.
Figure 12 shows the relationship between temperature and the maximum
amount of water vapour in the air. The shape of the line is a parabola but note
that it does not go through zero on the y-axis. Notice that at 20 C, the air
contains a smaller mass of water than at 37 C.
Vaporization
Anaesthetic vaporizers are designed to produce a concentration of anaesthetic
vapour that corresponds to the vaporizer setting regardless of gas flow passing through them. The gas in the vaporizer chamber is fully saturated with
anaesthetic vapour, but the amount of vapour depends on temperature, as
shown above: as the anaesthetic vapour is removed from the vaporizer chamber by the dry gases, latent heat is removed from the remaining liquid and the
vaporizer walls. Unless temperature compensation is applied, the amount of
vapour and its saturated vapour pressure would drop, as would the vaporizer
output. Modern vaporizers have several sophisticated methods of temperature compensation, based on a temperature-controlled valve which adjusts
the splitting ratio: as temperature inside the vaporizer decreases, more gas is
allowed to pass through the vaporizing chamber to compensate for the loss of
saturated vapour pressure. At 20 C saturated vapour pressure of desflurane is
higher than that of most agents. However, because of its low lipid solubility,
it has a fairly high minimum alveolar concentration. The vaporizer, therefore,
has to be heated to produce sufficient amounts of vapour.
16
1
Heat, vaporization and humidification
Figure 12. Water vapour content (absolute humidity) of air fully saturated with
water, as a function of temperature. Reproduced with permission from ButterworthHeinemann.
17
1
Part 1
18
Simple mechanics 1: mass, force,

pressure
Mass is a fundamental quantity; the other above-mentioned mechanical
quantities are derived from it. Its unit, 1 kilogram, is a basic SI (Syst`eme
Internationale) unit. Mass means quantity of matter. This description is
more or less a tautology; it shows the difficulty in defining the fundamental.
Our perception of mass is that of weight, or of an objects opposition to an
attempt to move it, which means applying force. In physics, weight in fact
is a force, defined as the product of mass and acceleration: F = m.a. Its
unit is 1 newton (N) or 1 kg m per s2 . The gravitational pull is the same in
most places on Earth, and gives all objects an acceleration of 9.81 m per s2 .
Thus weight is directly proportional to mass, and the two quantities are interchangeable. The kilogram is therefore used as a unit of weight, instead of
newton, which is the correct unit. The relationship between mass and weight
can be readily demonstrated on weighing scales: 1 kg of mass placed on one
side of the scale produces a force (weight) of 9.81 newtons. Consequently,
when training operating department assistants, to exert a force of 20 to 40
newtons (for the application of cricoid pressure), the weight they should be
aiming for on the scales is 2 to 4 kg. If the legendary apple, which struck Sir
Isaac Newtons head as it fell off the tree, weighed approximately 100 grams,
it would have struck with a force of about 1 newton.
We know from experience, when pushing an object such as a car or a full
supermarket trolley from a stationary position, the greatest force has to be
applied before the object starts moving, i.e. gains acceleration. Minimal force
is required to keep it moving at a constant velocity (acceleration is zero, force is
just sufficient to overcome friction). When stopping, force from the opposite
direction has to be applied (i.e. braking) to achieve deceleration.
Pressure is defined as a force per unit area p = F/A. Its unit is 1 pascal
(Pa), equal to 1 N m-2 . Because 1 newton is a small force and 1 m2 is a fairly
large area, 1 pascal is a tiny unit of pressure. Force can then be thought of
as a product of pressure and area (F = p.A). A reciprocal relationship exists
between pressure and area for a constant force as shown in Figure 13.
In Figure 14, a constant force is applied to a small and then a large syringe.
The flow from the small syringe (point B, Figure 13) is visibly higher (flow
being related to the pressure exerted), i.e. syringe emptying is easier than with
the large syringe (point A, Figure 13).
During intravenous induction of anaesthesia, constant rate of injection
(flow), and therefore constant pressure on the syringe plunger is required. A
greater force must be applied to a large syringe with a large cross-sectional
area. Anybody who attempted to push fluid out of a 50 ml syringe remembers
the aching hand afterwards.
1
Simple mechanics 1: mass, force, pressure
p
A:F = p2 A2
P2
B:F = p1 A1
P1
A1
A2 A
Figure 13. Force as a product of pressure and area; reciprocal relationship between
pressure and area for constant force.
Q ~ P1
Q ~ P2
A1
A2
st.)
st.)
Figure 14. Pressure generated in different size syringes with constant force.
19
1
Part 1
20
In anaesthetics, the principle of balancing forces is used in pressure reducing valves, where high and low pressure gas push against a large area of a
diaphragm connected to a rod, as shown in Figure 15. The sum of these
forces is opposed by the tension of a spring attached to the other side of the
diaphragm. This is set slightly above the force of the high pressure gas acting
on the rod. When low gas pressure decreases, the tension of the spring overcomes the forces produced by the gas and the rod is pushed out, opening the
high pressure gas inlet. Low pressure then rises sufficiently inside the valve
to overcome the tension of the spring and high pressure gas inlet is shut, and
the cycle is repeated as low pressure gas escapes.
1
Simple mechanics 1: mass, force, pressure
Figure 15. Pressure-reducing valve. Reproduced with kind permission from

21
1
Part 1
Simple mechanics 2: work and

power
Work is the product of force acting along a distance: W = F.l . Its unit is 1
joule (J), equal to 1 N m, which equals 1 kg.m2 per s2 .
Doing work can be visualised as moving an object using a certain force from
one place to another; the force which contributes to the work is in the direction
of the movement. The work involved is the product of the force used and the
distance between the two places. Reciprocal relationship between force and
distance for constant amount of work applies. This reciprocal relationship is
used in practice in simple mechanical devices to save force: a ramp was used
by the ancient Egyptians to move some giant stone blocks up the pyramid.
The longer the ramp, the less force was needed to push a block up to the same
height (pushing an object up a steep slope against gravity needs greater force,
and therefore is more exhausting). In Figure 16, the rectangular hyperbola is
the line of equal work (isoerg). The points chosen on the line illustrate that
for a distance n-times longer, the force used is n-times smaller to make the
same amount of work.
When doing work, we expend energy; work is in fact equivalent to energy,
hence the other non-SI unit of energy kilocalorie (kcal). Thermal energy,
kinetic energy, potential energy are all forms of energy; they can change
from one form to another but the total energy remains the same. All work
can be converted to thermal energy, for instance frictional force in breaking is
dissipated as heat. The conversion in the other direction (from thermal energy
to work) is also possible but not all heat can thus be converted to work. ) By
substitution for force from the pressure formula we obtain:
W = F l = p A l = pV
The product of area and length is called the volume (e.g. the volume of syringe
is the product of its cross-sectional area and the length of the barrel).
In other words, work can also be defined as the product of pressure and
volume. Boyles law, isothermic compression of ideal gas, is an example of
this formula: p V = constant (thermal state, or energy, is the equivalent of
work here). The example in Figure 17 shows how compressed gas from a
cylinder is used to fill a number of balloons. At point A the volume of gas is
low but pressure high. When balloons are filled at point B, the total volume
of gas is high but the pressure is lower.
Pumps generate pressures and displace volumes; their work is best defined
in terms of these parameters.
22
First low (the law of conservation of energy) and second law of thermodynamics, which
deals with entropy, are beyond the scope of this book.
1
Simple mechanics 2: work and power
F (N)
l1
F1
F2
l2
l2
l1
Figure 16. Work as the product of force and distance; reciprocal relationship of force
and distance for constant work.
p
A: E = P1.V1
P1
B: E = P2.V2
P2
V1
V2
Figure 17. Work (energy) as the product of pressure and volume, reciprocal relationship
of pressure and volume for constant energy.
23
1
Part 1
Power is the rate at which work is done: P = dW/dt . The unit is 1 watt
(W), equal to 1 J s1 . Unfortunately, the symbol for work and for the unit of
power is the same letter (W). The symbol d means change, or derivative, that
is what increment in work was done in a given time. Derivatives are dealt with
in more detail in the chapter on flow measurement. Notice that whilst work
has a lot to do with force, time becomes an important factor when dealing
with power.
If the force used in some work is constant, by substituting the product F.l
for work we obtain:
P = F.dl /dt = F.v (in words, power is the product of force
and velocity).
Imagine a pump worked at a constant pressure to shift fluid, then by substituting pressure and volume for work we obtain P = dW/dt = pdV/dt = p.Q
(in words, power is the product of pressure and flow. From the above
formulae, it can be seen that there are three ways of expressing power. Applied
to the heart muscle, these will be:
r power as the slope of the work curve against time (see Figure 18). If
cardiac work is plotted against the systolic ejection time, the slope of the
resulting line is steeper for the heart driven by inotropes. Note that the
plot would resemble the Starling curve but the parameters are different.
r power as the product of force and velocity: this is relevant to the
isolated papillary muscle preparation. The force of its contraction
and the velocity of the contraction, as measured in the experimental
situation, define the power. Initially, the muscle is stretched to a starting
tension Fo , and the velocity is zero. The muscle is then stimulated and as
it contracts, the passive stretch is lost and velocity increases. In Figure 19,
muscle shortening velocity is plotted against the initial passive stretch
(tension). Papillary muscle power, the product of the force and velocity of
contraction is the shaded area under the force-velocity line. The
force-velocity line is an approximation; force and velocity of contraction
are reciprocally related i.e. the line should in theory by a rectangular
hyperbola. The graph shows how the area would increase with increased
preload (greater initial stretch, higher initial tension but the same
maximum velocity), and with increased inotropy (higher initial tension
and higher maximum velocity, much higher power).
r power as the product of pressure and flow: cardiac power is then a
product of mean blood pressure and cardiac output (see the relevant
chapter). Under normal conditions blood pressure and cardiac output are
directionally proportional to each other (assuming the aortic impedance Z
is constant). In Figure 20 the plot of blood pressure as a function of
cardiac output for our purposes is a linear relationship ( p = ZQ). The
slope of this line is the impedance, and the cardiac power is the shaded
area under it. Notice that the x-axis here is not the time axis.
24
Ad
P2
P1
W1
t
W2
W1
Figure 18. Power as a derivative of work; cardiac power at the two levels of inotropy.
Simple mechanics 2: work and power
He
W2
t
P1
P2
V
gh
rm
ig
Hi
No
al
Power
Figure 19. Cardiac power as a product of force and velocity.
z
Q
Figure 20. Cardiac power as a product of pressure and flow.
25
Mathematical concepts
Part 1
Graphs show relationships between the depicted variables. If this relationship

follows a certain rule, or law, then the graph of that relationship will show a
certain pattern, which can be described by a formula. The x-axis is sometimes
described as the abscissa, and the y-axis as the ordinate.
The linear relationship
y = k.x is the simplest example of a linear relationship. When plotting this
on a graph, we could calculate y for several xs, plot each as a point and then
join them by a line. We know that this line will be a straight line passing
through zero, with a slope proportionate to k (see Figure 21). If we wished
to obtain the sum of all ys for their respective xs up to X, this would be given
as the area of triangle XYO, otherwise known as the area under the line k, and
the process of obtaining this area is called integration. More on integrals in
the chapter on flow measurement.
Rectangular hyperbola
An inverse, or reciprocal relationship, y = k/x, is depicted by a rectangular
hyperbola, k being a constant. y = 1/x is a specific form of this relationship
when k = 1. The line representing this function is symmetrical across an
axis passing through zero at 45 . It asymptotically approaches (never quite
reaches) both axes, i.e. neither x nor y can be zero. The reciprocal relationship
of y and x is illustrated by looking at rectangles drawn from any point on the
hyperbola whose sides are the coordinates of that point: they all have the same
area, which is also equal to the area of the square drawn at the intersection of
the hyperbola with the 45 axis. In Figure 22 several hyperbolas are drawn,
each for a different k, which is the product of y and x (area of the rectangles
formed by the coordinates) as shown above. There are numerous examples
in anaesthetic practice of variables that are the product of other variables;
in particular Boyles law (the thermal state of ideal gas as a product of its
pressure and volume), or end-tidal CO2 concentration as a product of minute
ventilation and fresh gas flow during controlled ventilation in a T-piece (see
the appropriate chapters). The relationship of drug dose and effect plotted on
a linear scale also produces a rectangular hyperbola but the hyperbola is rising,
i.e. it is inverted against the x-axis and shifted up to be above it: y = a k/x
(see the chapter on pharmacodynamics).
26
1
y = kx k = 1
y=kx k=2
Figure 21. Linear relationship.
Axis of
symmetry
k
k
k
x
Figure 22. Reciprocal relationship the rectangular hyperbola.
27
Parabola
Part 1
The parabola is the graphical representation of the square function y = k.x 2 ;

in anaesthetic practice we come across this shape when considering pressure
and flow, or resistance and flow under turbulent flow conditions. In laminar
flow, the velocity of the flow is a square function of the tubes radius, while
the cross-sectional area is also a square function of the radius. Hence laminar
flow is related to the radius by its fourth power (see the chapter on flow).
Figure 23 shows two typical examples for two values of K.
Sinusoid waveform
The sine wave is familiar to all who know about alternating current. The graph
is a vertical oscillatory movement plotted against time. The formula is
y = k sin x. The wave repeats itself in cycles with identical waveforms; its
characteristics are amplitude (maximum y) and frequency. The frequency
of repetition is reciprocal to the length of each cycle. A shift of the waveform
along the horizontal axis is called a phase shift; at x = 0, y will not be zero
(the waveform starts mid-cycle).
If two oscillatory movements were added up, the amplitude, frequency or
both characteristics of the resulting wave would change. For example, exactly
reciprocal waves (the sine wave and its mirror image across the horizontal axis)
would cancel each other out, two waves of same frequency and orientation
would increase the amplitude, while waves of different frequencies would
add up in a more complicated manner, altering the resulting frequency and
shape of the waveform (see Figure 24). A sine wave of double the frequency
plotted with the original waveform completes two cycles over the time in
which the original waveform completed just one cycle. The faster frequency
is called the second harmonic (it runs in harmony with the original waveform, i.e. there is no phase shift). Breaking down of complex waveforms to
their fundamental waveforms, the harmonics, is the basis of Fourier analysis.
Any complex waveform can be broken down into its sine wave components;
rounded waveforms contain a smaller number of harmonics while spikes consist of a large number of high-frequency components. For instance, analysis
of aortic pressure waveform produces 10 harmonics, an electrocardiogram
contains up to 50 harmonics of frequency range 1100 Hz and intracardiac
electrical potentials contain 400 high-frequency harmonics.
A measuring apparatus following an oscillatory movement can only be
accurate if its frequency response matches the highest frequency contained
in the waveform (see the chapter on electromanometers).
28
1
k=1
k = 0.5
Figure 23. Square function parabola.
y
y3 = y 1 + y 2
y1
Ti
y2
Figure 24. Sine waves and their addition.
29
Exponentials 1: the curves
Part 1
Exponential growth (e.g. chain reaction, growth of bacteria)

An exponential is a power by which we raise a number. General exponential
growth formula is:
y = b x.
The graph of exponential growth (Figure 25a) represents the formula y = 2x .
The value of y never can be zero, as it is a number (b) to the power of another
number (x). When x = 0, y = 1 because any number to the power of zero =
1 (n0 is the same as nx /nx ). Apart from the variable x, the exponent (power) can
also have a fixed component. For exponential growth, this constant exponent
is called the growth rate constant; it can be any number except zero, i.e. it can
be a fraction. The formula then becomes
y = bk.x
where k is the growth rate constant. (Note: y = x 2 , i.e. the square function
is a different shape parabola, and it goes through 0.)
The bacterial growth graph (Figure 25b) represents the formula y = 2t /20 .
The x-axis became in this example the time axis, and doubling time of the
bacteria is 20 min. The growth rate constant is 0.05, or 1 /20 . The negative part
of the x-axis is not shown. This is because in physiological processes we are
interested in their time-course from time zero; what happened before time
zero usually does not conform to the exponential curve, but after time zero,
the starting point, it does.
Exponential decay (e.g. drug elimination, decrease of
transmitted light intensity with distance)
Decay being the reverse of growth, the sign changes and so does the orientation
of the curve; it is inverted around the y-axis. The mathematical formula is:
y = b x .
The figure of exponential decay (26a) represents the formula y = 2x .
We shall concentrate on the part of the graph to the right of 0 on the
x-axis. Figure 26b illustrates the exponential decline in plasma remifentanil
concentration after a rapid intravenous bolus injection. Mathematically, this
is y = yo .2t /4.3 . The half-life, i.e. the time it takes for the starting value
to decline to 50%, is 3 min (see below). Apart from time, there are other
factors which determine the rate of exponential decline; these can be summed
together as the elimination rate constant k. The elimination rate constant
is an analogy of the growth rate constant in the exponential growth process.
The formula for physiological processes then becomes
y = yo .ekt
30
where yo , is the starting value, e is the base of natural logarithms, k is the

elimination rate constant and t is time. We shall return to this formula later.
1
(a)
(b)
Time
Figure 25. (a) Exponential growth curve. (b) growth of bacteria with time.
(a)
(b)
Time
Figure 26. (a) Exponential decay curve. (b) exponential decline plasma remifentanil
concentration.
31
Saturation exponential (e.g. pumping up a tyre, or intermittent

positive pressure ventilation with a ventilator that generates
constant pressure)
Part 1
The process is reverse of exponential decay, but this time inverted around the
x-axis, and it works against a limiting value. The formula therefore is
y = a b x ,
where a is the limiting value, and the minus sign denotes the change in orientation around the x-axis (Figure 27a). A pure mathematical conversion of
exponential decay would have no limiting value (y = b x ). The curve is
symmetrical with the exponential growth curve across the zero point. Lung
filling during intermittent positive pressure ventilation is depicted in Figure
27b. The exponential coefficient for lung filling is influenced by lung compliance and airway resistance. The larger either of these factors, the slower
the lung filling, and conversely the smaller they are, the faster the lung filling.
Thus the filling rate constant is the inverse value of the product of lung
compliance and airway resistance (1/C.R); their product, the reciprocal of the
filling rate constant (C.R) is the time constant of the process (see below).
Figure 27b represents the formula
Y = 500 (500.et /300 ).
The exponential formula is for convenience multiplied by the limiting value
to obtain 0 for y at the beginning. The negative exponential factor 1/300 in
the formula indicates that lung filling has a time constant of 300 ms.
Logarithm
Logarithm is the exponent to the power of which the base has to be raised
to get the specified number. This complicated definition means that this is
the reverse of exponential growth function (just as square root is the reverse
of square function). Figure 28 resembles superficially saturation exponential, but it is different. The logarithmic curve is symmetrical with the exponential growth curve around the 45 axis, and it asymptotically approaches
(never reaches) the y-axis: a logarithm of zero or negative numbers cannot
be obtained, if the base (10 or e) is a positive number. y has no limit, which
means it reaches infinity in both directions.
By contrast, a saturation exponential is symmetrical with the exponential
decay curve around the x-axis; it asymptotically approaches the horizontal
axis, or a limiting value above it, i.e. y has an upper limit.
32
1
Time
(a)
(b)
Figure 27. (a) Saturation exponential curve, (b) lung filling with a constant pressure
generator.
Figure 28. Logarithmic curve and exponential growth curve.
33
1
Part 1
34
The exponential and logarithmic curves can be transformed into lines

by plotting the values against a logarithmic scale. This scale has different
properties from the usual linear scale: there is no zero (an exponential can
never be zero), and the marks on the axis are equidistant for powers of the
base. For instance, in the graph of [H+ ]pH relationship (see the chapter on
pH measurement), notice that the logarithmic scale does not have zero and
the distances between 1 and 10, and 10 and 100, are the same. The linear
transformation of the curve is shown in that graph. The logarithmic scale is
used in pharmacokinetics for linear transformation of elimination curves (see
the chapter on pharmacokinetics).
1
35
1
Part 1
36
Exponentials 2: Properties of
exponential decay curve
1. Decrease in constant proportion: in practice this means that the rate of
decrease depends on the mass present. The mass present at any
point is graphically represented by the value of the ordinate (y-axis). The
rate of decay is not constant, it is the proportion of the mass that is
lost that is constant: the previously introduced elimination rate
constant in fact determines this fixed proportion eliminated per unit of
time. For example, when emptying a bath across a fixed resistance (pipe),
the flow rate (of emptying) depends directly on the hydrostatic pressure
(the level of water) that changes as the bath empties, while the diameter
of the pipe determines the proportion of contents that is emptied per unit
of time, i.e. the elimination rate constant.
2. Every section of the curve differs from any other section by scale only:
this follows from the first point: because the units of the y-axis are
different from those of the x-axis (the time axis), the scale of the y-axis
can be arbitrarily chosen. By the same token, any point of an exponential
decay curve can be chosen as a starting point, and the vertical scale can
be expanded if values are very small. In practice, this also means that we
can project back and forward if we know a part of the curve.
3. The slope (tangent) at any point is directly proportional to y (the
ordinate): this is also another way of saying the first point (the slope of
the curve is the rate of decrease).
4. The vertical projection of the slope of the curve on the x-axis is constant.
Properties 1, 2 and 4 are shown in Figure 29. Because the x-axis in
natural processes represents time, is called the time constant. The
time constant was already mentioned as the reciprocal value of the
elimination rate constant. Therefore, this point is also indirectly
connected with the first point: if we say a constant proportion per unit of
time is lost, then the time it takes to lose this proportion must be
constant. Think of the time constant as the time to finish had the rate
of decay been maintained. This relationship holds no matter which
starting value of y is chosen. In the bath analogy, the elimination rate
constant is given by the resistance of the effluent pipe. This resistance is
fixed for a given bath and can only let a certain flow of water through for
a particular pressure. It follows that if the flow through the pipe were
constant (i.e. driving pressure was constant), it would take a certain
amount of time to pass a given volume of water: this time would be equal
to the time constant of emptying by hydrostatic pressure. For different
baths, the higher the resistance the smaller the flow for any given
pressure, and the longer the emptying. In other words, the elimination
rate constant (pipe resistance) and the time constant (time to empty if
initial flow rate were maintained) are reciprocally related ( = 1/k), as
said above. The exponential decay equation can then be written as:
y
t
Figure 29. Rate of decay as a function of the ordinate, time constant as a projection
of the tangent to the curve on the x-axis.
Exponentials 2: Properties of exponential decay curve
37
1
Part 1
y = y 0 .ekt or y = y 0 .et / . number e is the base of natural logarithms

(= 2.718) and it has some unique properties, e.g. de t /dt = et (i.e. the rate
of change of et with time has the same value as et ). Returning to the bath
analogy, this mathematical property means that volume or flow (the
derivative of volume) can be charted on the ordinate, and the exponential
decay curve will be identical. Any number can be chosen for the base
constant, with appropriate adjustment of the scale; e was chosen because,
as shown above, differentiation (and therefore integration) are very easy.
Integration, or the area under the curve, is dealt with below.
5. Area under the curve: from any point to the infinitely remote point on
the x-axis where the process may be thought of as having finished is
proportional to the ordinate and it is the product of the two
fundamental determinants, initial mass and time constant (yo and ).
In Figure 30, this is equal to the area of rectangle YOTY. Calculation of
the area under a curve is mathematically known as integration (see the
chapter on flow). Thus, an operation that looks complicated (integration)
is in fact very simple: 0 yo .et / /dt = yo .. Integration (for instance
in the bath analogy) can be thought of as the summation of flow over
time to measure volume. For a constant flow equal to the initial flow rate
Q0 , the entire volume (area under the top horizontal line) would have
taken time to disappear: that is the area of rectangle YOTY. For the
exponentially decreasing flow rate the process takes in theory infinity, but
the total volume, the area under the exponential curve, in the end is the
same.
Half-life, time constant and time to finish
In pharmacology, half-life is used more than the time constant. It is the time
taken for the mass to decrease to 50%. and half-life (t 50 ) are related and
the latter is another type of time constant, but one with a more complicated
relationship to the elimination rate constant than .
The percentage of mass left after a given time can be calculated for any
exponential decay process. At time , the amount is:
y = y o .e/ = y o .e1 = y o .0.37,
i.e. just over one-third of the initial amount. It follows that must be longer
than the half-life because the proportion that is left is smaller. By substituting
yo /2 for y we find that is 1.44 times higher than the half-life.
Although in theory the process takes infinity, in practice we need to know
how long it takes to complete. After a time equal to three time constants, y will
be y o .e3 , or 5% of the initial value. After four time constants, y will be y o .e4 ,
or about 1% of the initial value. In other words, after four time constants the
exponential decay process is for all intents and purposes finished.
38
Exponentials 2: Properties of exponential decay curve
Figure 30. Area under the exponential curve equals the area of rectangle
39
Descriptive statistics
Part 1
The purpose of statistical analysis is to obtain information from a sample that

can then be extrapolated to a whole population; to analyse a whole population
would be very lengthy, expensive and difficult to achieve. Care must be taken
that the sample has the same characteristics as the target population (e.g.
paediatric data cannot be extrapolated to the adult population).
Descriptive statistics categorise and summarise data. This chapter deals
with summarisation of parametric (numerical) data.
To describe a population, or a sample, we should know its mode of distribution, a central point and its variability.
Variability is given by range, or by deviation from average. Only standard
deviation is illustrated (but not derived) in this text.
Mode of distribution
Normal (Gaussian) distribution
This is the most frequent mode of distribution. The measured variable tends
to cluster around the central, most common value, while extreme values either
side of the centre are rare (e.g. adult male height, adult male haemoglobin). It
is a bell-shaped curve only if the measurement was conducted in the whole
population (e.g. the UK). Figure 31 shows the distribution of height in adult
men. Observed values of height are plotted on the x-axis, and the frequency of
observation (f) of each value is plotted against the y-axis. SD and x in the graph
stand for standard deviation (SD) and mean (see below). 68% of population
values lie within 1 SD from the population mean; 95.44% lie within 2 SD;
and 95% lie within 1.96 SD. When studying a sample, the resulting frequency
histogram (see the chapter on data presentation) would be less regular than
the classic bell shape. Nevertheless, if the outline of a frequency histogram
(Figure 32) approximates the bell shape, Gaussian distribution is assumed
with all its characteristics.
40
1
Figure 31. Normal distribution of height in adult men.
Number
t
(m
Figure 32. Gaussian distribution blood glucose measurements in a large normal
sample.
41
1
Part 1
Log-normal or skewed distribution

This finding is fairly common in biological processes; the distribution can be
converted to normal, i.e. Gaussian, by substituting natural logarithms for the
observed values. For example, the distribution of hydrogen ion concentration
in a normal population will be heavily skewed towards higher values, as the
body tends towards acidosis. The most frequent value, 40 nmol/l, is lower
than the arithmetic mean of the measured values (Figure 33).
Values < 20 are extremely rare, while those between 40 and 100 are rather
common. A skew towards higher values is called a positive skew: the most
frequent value is lower than the other measures of central tendency.
By converting hydrogen ion concentration to pH, the skew can be eliminated and the distribution will then be symmetrical around the central and
most common value of 7.4, with the usual range narrowly clustered around
this value.
An example of a negative skew is haemoglobin in women of reproductive
age, where anaemia is fairly common: the most frequent value is higher than
the other measures of central tendency (see Figure 34).
Other types of distribution
1. Bimodal distribution has two most common values, and thus gives the
double hump appearance (Figure 35). Often a more detailed analysis
reveals that, in fact, there are two different populations.
2. J-shaped distribution resembles a rising exponential process.
3. U-shaped distribution (e.g. cardiovascular mortality in relation to
alcohol consumption) could perhaps be an inverse normal distribution:
if the curve was redrawn with survival rate rather than mortality rate as
the y-axis parameter, it would then resemble normal distribution with a
central value between two and three units of alcohol.
Measures of central tendency
Mean (
x ) : x = (xi )/n
The arithmetic mean, or average, is a common measure of central tendency.
In a population normally distributed this is the value which is also the most
frequent (mode), and which stands at midpoint (median) see below. The
process of summation is expressed as .
Median
This is the middle value, above and below which we find the values of
50% each for the rest of the population. In a positively skewed distribution
the median will be higher than the most frequent value, but less than the
arithmetic mean; the reverse applies in a negatively skewed distribution.
42
Mode
The mode is the most frequent value. In a skewed distribution it will be the
x-axis value at the peak of the curve (note: not the peak value!). Bimodal
distribution, as its name suggests, has two modes, i.e. two most frequent
values.
1
Figure 33. Positive skew distribution of hydrogen ion concentration in humans.
Figure 34. Negative skew distribution of haemoglobin concentration in women.
Figure 35. Bimodal distribution.
43
Presentation of data
Part 1
The chapter on mathematical concepts deals with graphs as illustrations of

relationships between variables. The chapter on descriptive statistics introduces statistical terms.
Diagrams
Diagrams are used to illustrate statistical data (see the chapter on descriptive
statistics for explanation of terms).
The pie diagram is the simplest form. It is often used to illustrate data
collected from audit. It is useful if there are several categories (portions)
of a whole, for instance age groups, different surgical specialties, items of
expenditure in budget, etc. The pie diagram gives an immediate idea of how
the whole is divided up. Figure 36 shows all patients who experienced nausea
after surgery according to the type of operation.
The histogram is best used to show counts integer numbers of patients,
objects or days. Bands or columns are used for the illustration; although
each number has only one dimension, plotted against the y-axis, the second
dimension the width of the band makes the visual distinction between
groups easier and distinguishes a histogram from a line graph. Figure 37
shows the comparison of the incidence of postoperative nausea among men
and women. Measurements of continuously variable data are best represented
by scatter diagram (see below) although the histogram is often used in this
situation.
The frequency histogram illustrates the distribution of different values
of a measured variable in a certain sample or population. Figure 38 shows the
distribution of hypothetical laboratory measurements of fasting blood glucose
concentration in all employees of a hospital. The x-axis is divided into bands
of integer numbers of blood glucose measurements; the height of each band
represents the number of people with this value. Notice that the histogram is
used correctly in this situation: integer numbers of people are plotted on the
y-axis. If the number in each group is expressed as a percentage of the total,
this histogram becomes the frequency histogram. For a normal sample the
outline of the frequency histogram will resemble the bell-shape of Gaussian
distribution (see the chapter on descriptive statistics).
44
1
Figure 36. Pie diagram proportion of nausea and vomiting according to the type of
surgery in a large sample of day surgical patients.
No
Na
Women
n
Men
n
Figure 37. Histogram (bar diagram) incidence of postoperative nausea and vomiting
in men and women after day surgery procedures.
Number
Bl
Figure 38. Frequency histogram.
45
1
Part 1
46
Scatter diagrams or line drawings should be used for plotting results of

measurements; for few measurements, each result is plotted as a point defined
by its coordinates. If a trend is observed, this is represented by the line of best
fit, or regression line. Individual measurements should lie close to this line.
The measure of this closeness is called the correlation coefficient. For a perfect fit the correlation coefficient is 1 or 1 (Figure 39); when there is no agreement between the points and regression line, the correlation coefficient is 0
(Figure 40). Care must be taken to include sufficient number of valid data:
an isolated outlier in a small sample of otherwise random values probably
means that the measurement was incorrect or that the individual comes from
a different population.
For a large number of measurements each point in the graph represents the
central value, usually the mean, and error bars should be included to indicate
the dispersion of each group. The points are usually joined by lines to show
any trend line drawing. Figure 41 shows mean pain scores in a group of
20 surgical patients on intramuscular as required analgesia, and 20 surgical
patients on patient-controlled analgesia. Error bars overlap immediately after
the operation but there seems a trend for improved pain relief with patientcontrolled analgesia.
Me
1
Time
Figure 39. Correlation coefficient = 1.
Me
Time
Figure 40. Correlation coefficient = 0.
Pain
Time
Figure 41. Line drawing with error bars average pain scores after total abdominal
hysterectomy in patients on morphine patient controlled analgesia (o) and on intramuscular morphine on demand ().
47
1
Part 1
48
Receiver operating characteristic

curve
Sensitivity (detection of true positives) and specificity (detection of true
negatives) of a test are important properties. In an ideal case each should be
equal to 100% or the value of 1.0: all true positives should be identified by
a test and all true negatives rejected, or fall below a cut-off value. In practice,
this is rarely the case. A trade-off usually has to be sought between a good
enough sensitivity and an acceptable specificity.
The name receiver operating characteristic (ROC) curves was conceived
during the Second World War to name the ability of radar operators to discriminate between friend or enemy blips on the screen, whilst also discounting
the background noise (false positives). When tested, correct or false answers
could be plotted and the result was the ROC curve. This was later taken up
by the medical profession to describe the accuracy of laboratory tests.
To visualize how false positives and false negatives are related to the
true values and to each other, consider Figure 42. It shows the distribution
of fasting blood sugar in the normal (non-diabetic) population and in the
diabetic population. Note that the distribution curve for diabetics is skewed
to the right. Some overlap exists between the two populations, as not all
diabetics will always have a raised fasting blood sugar, whilst a raised fasting
blood sugar may on occasions be seen in a non-diabetic person.
The area where the two curves overlap is shown in grey. It is a roughly
triangular area with a peak at point A (the cut-off blood sugar level of
5.5 mmol l1 ). To the left of the separation point in this grey area lie the
false negative (diabetics classified as normal), and to the right the false
positives (non-diabetics diagnosed as diabetics).
If we artificially move the cut-off point for normal to the left, say
point B (peak of the distribution curve for non-diabetics), we would increase
the number of false positives while at the same time the false negative rate
would decrease. Diabetes would be wrongly diagnosed in a higher proportion
of normal population but very few true diabetics would be missed. The reverse
would happen if the cut-off point was moved to the right, to a higher level of
fasting blood sugar (point C).
B
0
C
10
mmol 11
Figure 42. Frequency distribution of tasting blood sugar in normal and diabetic population.
Receiver operating characteristic curve
Frequency
49
1
Part 1
50
Figure 43 plots ROC curves, as described above, for various tests. Sensitivity (true positive rate) is plotted on the y-axis against (1 - specificity)
true negative rate on the x-axis. Looking at Figure 1, it is obvious that if there
was no overlap between the two populations tested, the distribution curves
would be completely separated. There would be no false positives and no
false negatives if the cut off point was placed correctly. Testing such ideal
populations with this cut-off point would show a sensitivity and specificity of
1.0. By artificially moving the cut-off point to lie within one or the other population, we would allow either false positives (by moving to the left towards
point B) or false negatives (by moving to the right towards point C). The
ROC plot of this ideal situation is shown in Figure 43 as a thick black line and
is formed by the y-axis and its coordinate at the value of 1.0. The area under
this curve is equal to 1.0 (it is a square equal to the area of the graph).
In real life, as shown in Figure 42, overlap exists and therefore false positives
and false negatives co-exist. For most tests, it makes sense to put the cutoff point at point A the mid-point between the peaks of the distribution
curves, i.e. where both specificity and sensitivity are highest. A test which
discriminates well between the two populations tested produces a good ROC
curve. This is shown as a thin black line, and it will approximate the ideal
described above. The area under the curve will approximate 1.0. In this case,
the test discriminates well between the two populations, and such test would
be a gold standard.
In some circumstances, it may be better to increase sensitivity at the
expense of specificity or vice versa. One such case is prediction of the
difficult airway. It is desirable that every difficult airway is identified as
such (sensitivity of 100%). Because of overlap of normal and difficult, and
because the cut-off point is moved into the normal area (analogous to point
B in Figure 42), one has to accept a high number of false positives normal
airway identified as difficult. In practice no single test fulfils even this criterion
and we have to apply a battery of tests to improve the results.
Figure 43 also shows a straight interrupted line, the diagonal of the area
of the plot. This represents an imaginary test which does not discriminate
between the two popluations tested, i.e. a useless or irrelevant test. Identification of true positives and true negatives in this case is purely by chance.
The area under this line is equal to 0.5 (half of the square). The more a ROC
curve approximates this line, and the area under it 0.5, the less useful the test.
1.0
Receiver operating characteristic curve
Sensitivity 1.0
1 specificity
Figure 43. ROC curves.
51
Gas supply and pressure
Part 1
Medical gases supplied in cylinders are compressed to a high pressure in order

to store a big mass in a relatively small volume (it follows from Boyles law
that the higher the pressure inside the cylinder, the bigger the final volume of
decompressed gas). The human respiratory system operates at low pressures;
high pressure can produce barotrauma. Gas supply pressure has to be reduced
in stages and safety features are incorporated. The wide range of pressures in
the medical gas supply and in the breathing system has led to the establishment
of various units suiting each particular purpose: bars for high pressures, lb/psi
for moderate pressures, mmHg for atmospheric and lower pressures, and
cmH2 O for very low pressures measured as a column of water. (See more on
gas pressure in the chapter on real gas compression, and on units in simple
mechanics.)
Figure 44 shows the pressures inside the gas supply system on a logarithmic
scale; in this way, it is possible to include a wide range of pressures in one
graph. In particular, notice that the gas supply operates in terms of hundreds of
kPa (bars) but the breathing system pressures are kPa units or their fractions.
The logarithmic scale of the ordinate also demonstrates that each stage of
decompression reduces the pressure very approximately by a factor of 10.
The volume shown on the abscissa is that which would be obtained by
decompression of 1 litre gas at the pressure indicated to atmospheric pressure.
This volume can be calculated by using the formula
V = pc + pa = pc + 1
for 1 litre gas decompressed to 1 atmosphere, where pc = cylinder pressure.
Pressure conversion
1 atm = 101.3 kPa = 760 mmHg = 1033 cmH2 O = 14.69 lb/in2
100 kPa = 1 bar. Notice that 1000 kPa = 1 MPa (megapascal) = 10 bar,
and, by approximation, 1 kPa = 10 cmH2 O = 7.5 mm Hg.
52
1
Gas supply and pressure
Fir
Pi
Back
Blow
Posi
V
Figure 44. Pressure/volume relationship in medical gas supply.
53
The circle system
Part 1
The circle system is a semiclosed or closed absorption system; with the

absorber out of circuit it becomes a semiclosed rebreathing system. Low
flow can be used because carbon dioxide (CO2 ) is absorbed and this gives
economy of use and produces little pollution. Low flow has been defined
as a total fresh gas flow of 1 l/minute, and minimum flow as 0.5 l/minute.
Most new anaesthetic machines incorporate a circle system with absorber,
unidirectional valves and exhaust valve removed from the patient end
(see Figure 45). Modern highly efficient vaporizers are situated outside the
circuit.
The most common technical problem with the circle system is disconnection of fresh gas flow at the fresh gas flow outlet and shutting off
the expiratory valve when the intended mode of ventilation is spontaneous
breathing (by leaving the control knob turned to ventilator instead of bag).
Performance of the circle system under low fresh gas flow
conditions
Inspired gas composition will be identical to the fresh gas composition at high
fresh gas flows, as the expired gas will be flushed out of the system, and CO2 is
absorbed. When reducing the gas flow to low levels, the expired gas will make
an increasingly bigger contribution to the gas composition inside the circuit.
Oxygen flow must not be reduced below the minimum 250 ml/min, the basic
metabolic requirement being 200 ml/min. The concentration effect will
influence the inspired gas composition at induction, when anaesthetic (nitrous
oxide, or N2 O) uptake is high. Alveolar oxygen concentration will increase,
and N2 O concentration will decrease; if very low gas flows were used during
induction, alveolar N2 O concentration could be below MAC. Therefore, it is
necessary to maintain a high fresh gas flow during induction and until expired
concentration of the anaesthetic is near inspired concentration (about 10 min).
After this, anaesthetic uptake is low, and thus alveolar concentration of the
anaesthetic rises, while alveolar pO2 decreases. Such hypothetical situations
are illustrated in Figure 46: 50% O2 and N2 O are supplied in a circle system,
with assumptions about O2 consumption, CO2 production and absorption.
At fresh gas flows > 2 litres, inspired gas composition closely approximates
the fresh gas composition, regardless of uptake. At lower flows, nitrous oxide
uptake plays an increasingly important role: when nitrous oxide uptake is
high, alveolar oxygen concentration rises steeply (the concentration effect is
more pronounced) but when the uptake decreases to < 200 ml/min, alveolar
oxygen concentration decreases substantially < 50%, as nitrous oxide is then
taken up at a lower rate than oxygen.
The inspired concentration of anaesthetic vapour is affected by its own
uptake; therefore, when uptake is high the inspired concentration of the
vapour will decrease and vice versa.
54
Patient
Ex
The circle system
Fresh
From
Figure 45. The circle system.
FAo2
V N 2o
VF
Figure 46. The effect of nitrous oxide uptake and fresh gas flow in the circle system on alveolar oxygen concentration. Reproduced with permission from ButterworthHeinemann.
55
1
Part 1
56
Another factor affecting the inspired gas composition is the accompanying

steady washout of nitrogen from the body. One litre of nitrogen contained in
the lungs is washed out during the induction with high gas flow. Thereafter,
nitrogen accumulates in the circuit at very low flows, further lowering the
inspired pO2 . The volatile agent concentration is also lowered by the accumulated nitrogen, and this tends to offset the expected rise in inspired vapour
concentration when vapour uptake is low.
Uptake of inhalational agents is an exponential decay process, and accumulation of nitrogen in the circuit a saturation exponential. Both processes
have a time constant , that is directly proportional to the volume of the
system V, and inversely proportional to the difference between fresh gas flow
and the sum of all gas uptake. Thus, in theory if the fresh gas flow exactly
equalled all gas uptake, the time constant would be infinitely high (because
then = V/0). With fresh gas flows being somewhat (e.g. 10%) above the
basal requirement, the time constant would still be several hours, offering a
protection from a dangerous mixture being delivered. At the other end of the
spectrum, with fresh gas flow much in excess of anaesthetic uptake the time
constant is short but at these high flow rates anaesthetic uptake and nitrogen
washout do not affect the inspired gas composition.
At minimal flow, the anaesthetist must also consider if a sufficient amount
(mass) of vapour is being delivered. In practice, this could happen with
enflurane which has a relatively high MAC and a higher blood/gas solubility
than isoflurane. Its uptake is therefore much higher than that of halothane
or isoflurane. Figure 47 illustrates the uptake of the three inhalational agents
during anaesthesia in a patient weighing 90 kg when the inspired concentration of the agent is set so that the expired concentration is around 0.8 MAC.
The uptake of isoflurane after 10 minutes is less than 30 ml/minute but uptake
of enflurane is still nearly 60 ml/minute. Assuming that after 10 minutes fresh
gas flow is reduced to 0.5 l/minute with the highest vaporizer setting (5%), the
system will only deliver 25 ml of vapour every minute (5% of 0.5 l is 25 ml).
The inspired concentration of isoflurane can thus just about keep up with its
continued uptake but if enflurane is used the system is unable to deliver sufficient vapour at this minimal flow; flow reduction can only be effected when
uptake has decreased near the maximum vaporizer output at minimal flow.
In leaner patients, this can be done after some 15 to 20 minutes. Alternatively,
a higher inspired concentration could be used from the start (e.g. to obtain
expired concentration 1.2 MAC, to accelerate the initial high uptake period)
or flow could be reduced in stages. At one litre per minute, vaporizer output
at 5% is 50 ml and would cover enflurane uptake after 15 minutes even in the
oversize patient.
1
The circle system
(ml/min)
(min)
Figure 47. Uptake of volatile anaesthetic as a function of time. Reproduced with permission from Butterworth-Heinemann.
57
1
Part 1
58
The Mapleson A (Magill) breating

system
Conway (1985) described the geometry of the Mapleson classification and
the behaviour of the breathing systems under conditions of spontaneous and
controlled respiration. The Mapleson A (Magill) system (Figure 48) has the
fresh gas inlet remote from the subject while the expiratory valve is near
the subject. Dead space is shown as the shaded area. This makes the system
particularly economic to use during spontaneous respiration.
Figure 49 shows the hypothetical pressure and flow inside the Mapleson A
system during expiration. Flow during expiration, which is usually a passive
process, is maximal at the beginning of expiration, and it falls off exponentially. Pressure inside the system rises exponentially but at a faster rate than the
expiratory flow falls because of accumulation of fresh gas in the system; this
rise in pressure is cut off when the expiratory valve opens at 56 kPa and this
pressure is then maintained until end-expiration. Areas under the flow curve
give expired volumes (see the chapter on flow and volume measurement);
dead space gas (VD ), which is identical in composition to fresh gas is expired
first until point A on the graph. When the expiratory valve opens at point
B, alveolar gas (VA ) beyond that point is vented out. Furthermore, alveolar
gas that was deposited inside the system between points A and B (start of
alveolar gas expiration and opening of expiratory valve) is after the opening
of expiratory valve being pushed out by the fresh gas inflow (VF ). If sufficient time is allowed after the opening of the expiratory valve to allow venting
of the alveolar gas deposited during the valve closure, at end-expiration the
system will only contain fresh gas and dead space gas (identical in composition). Thus, during spontaneous respiration, provided that fresh gas flow is
at least equal to alveolar ventilation, i.e. about 70% of the minute volume, there will be no rebreathing of CO2 -containing gas. During controlled
ventilation the economy of gas flow is lost: positive pressure is applied
during inspiration and the pressure relief valve therefore opens, venting fresh
gas; furthermore, the valve needs to be set to a higher pressure (tightened)
and therefore during expiration it opens later, allowing retention of CO2 containing gas inside the system. Higher flows are, therefore, needed during
controlled ventilation. This disadvantage of the Magill system, and the large
weight of the expiratory valve and scavenging close to the subject (i.e. to
the mask or airway connection), caused a steady decline in the popularity of
the Magill system. It is being increasingly replaced by the T-piece or circle
absorption system.
1
The Mapleson A (Magill) breating system
Figure 48. The Mapleson A breathing system.
VE
Ti
VF
VD
VA
Ti
Figure 49. Hypothetical pressure and expiratory flow in the Mapleson A system.
59
1
Part 1
60
T-pieces
T-pieces (Figure 50) are geometrically opposite of the Magill system: the
fresh gas flow inlet is near the subject while the expiratory valve or port is
the furthest. E type is a valveless system suitable for spontaneously breathing
subjects only. F type is a paediatric valveless system.
Figure 51 shows theoretical inspiratory gas flow during spontaneous
breathing in a T-piece system. The inspiratory waveform is a sine wave.
Integration of flow (area under the curve) yields volume inspired. This is
again divided into gas that enters alveoli (VA ) and dead space gas (VD ). Superimposed on the graph is the fresh gas flow, which is constant. The area under
the straight line of the fresh gas flow is the fresh gas volume delivered into the
system. It can be seen that at point A the inspiratory flow exceeds the fresh
gas flow. Until that point, excess fresh gas was being deposited in the tubing.
After that point, inspiratory flow is supplied in part (the part exceeding fresh
gas flow) by gas previously deposited in the tubing. Provided that the roughly
triangular area between the fresh gas flow line, the inspiratory curve up to
point A, and the y-axis, which is the volume deposited in the large bore
tubing, equals the area under the peak inspiratory flow above the fresh gas
flow line up to point B (which is the volume of gas drawn from the system
in excess of fresh gas flow during that time), no CO2 -containing gas will
be rebreathed. These conditions will apply if fresh gas flow is approximately
twice the minute volume during spontaneous respiration.
During controlled ventilation, fresh gas flow mixes with expired gas
as the peak ventilatory flows are higher; the inspirate will then contain CO2 .
Under these conditions the magnitude of rebreathing will depend on fresh gas
flow and minute ventilation: an increase in either of these will lower arterial
CO2 tension. In fact, arterial partial pressure of CO2 becomes the product of
fresh gas flow and minute ventilation. To maintain the CO2 level, an inverse
relationship must be maintained between fresh gas flow and minute volume.
If minute ventilation is increased, fresh gas flow can be reduced; this
provides economy of flow during controlled ventilation.
1
T-pieces
Figure 50. T-pieces.
VE
VF
VD
VA
Time
Figure 51. Inspiratory flow and fresh gas flow in the Bain system. Reproduced with
permission from Butterworth-Heinemann.
61
1
Part 1
Lung filling with automatic lung

ventilators
Constant pressure generator
This type of ventilator is driven by fresh gas flow (Manley) or by electricity
(East Radcliffe). It is suitable for lungs with relatively normal characteristics
because the generated pressure is relatively low (2530 cmH2 O). When a step
change in pressure is generated, the difference (p) between pressure at the
mouth (pm ) and alveolar pressure (pA ) results in gas flow. As the lungs are
being filled by the gas, this pressure difference falls in an exponential fashion, and so does flow. The lung filling curve (volumetimerelationship) is a
saturation exponential. The time constant of the lung filling process is the
product of lung compliance C and airways resistance R ( = C.R): see
the chapter on exponentials. Figure 52a shows the filling of a normal lung,
the pressures, flow and volume are plotted against the time. PA denotes alveolar pressure; PM pressure at the mouth; V = gas flow and V = volume
achieved. For a high airways resistance (Figure 52b) the time constant is long:
the filling takes longer. This is because the high airways resistance reduces
the flow from the beginning (V = p/R). For a poorly compliant lung (Figure 52c), the time constant is shortened. The flow initially is unaltered but it
decreases quickly as the stiff lung soon opposes further filling. Thus the final
volume achieved is reduced.
62
PM
PA
Normal lung
Time
Time
Time
(b)
PA
COAD ( Resistance)
Time
Time
Time
PM
(c)
PA
PM
Time
Time
Time
Restrictive lung disease

(Compliance)
Lung filling with automatic lung ventilators
(a)
Figure 52. Lung filling with a constant pressure generator.
63
1
Part 1
64
Constant flow generator

This type of ventilator is suitable for lungs with abnormal characteristics.
It can generate constant flow by passing gas driven by high pressure through
a flow restrictor. High-pressure gas can be the sole driving force (Penlon) or
sophisticated electronic circuitry generates the pressures and flows required
(Ohmeda 7900, or most ventilators suitable for intensive therapy). Figure 53
again shows lung filling for a normal lung, a lung with a high airway resistance
and a poorly compliant lung. The sequence of graphs is altered to progress
from the flow generated, through volume delivered to pressures achieved.
Constant flow generated by the ventilator is reflected in a constant volume
delivered in a chosen time, as volume is the integral of flow. Thus the flow
and volume graphs are the same for all three situations. From experience,
it is known that pressures achieved will depend on lung characteristics. As
the normal lungs are filled (first column), alveolar pressure, pA , rises in line
with lung volume. Alveolar pressure can be calculated as the ratio of lung
volume delivered and lung compliance (as lung compliance is the ratio of
lung volume over alveolar pressure: C = V/ p A ). Pressure difference between
alveolar pressure and pressure at the mouth is given as the product of flow
(which is constant) and resistance (p = V.R), and is therefore constant.
Pressure at the mouth is a simple sum of alveolar pressure and the pressure
/C + V
.R).
difference ( p m = V
For a high airway resistance (second column), the pressure difference
between alveolar pressure and mouth pressure will be higher, as is obvious from the above formula. Pressure at the mouth, therefore, will reflect
this higher pressure difference. The set lung volume will be delivered but at
the expense of a higher pressure at the mouth.
For a poorly compliant lung (third column), alveolar pressure will rise
more steeply as it will reach a higher value for a given lung volume (alveolar
pressure and lung compliance are inversely related see the formula above),
and this will be reflected in a higher mouth pressure. A set lung volume will
be delivered again at the expense of a higher pressure at the mouth.
To avoid barotrauma, pressure limits must be set on this type of ventilator;
for very stiff lungs, pressure-controlled ventilation is used that keeps the
driving pressure constant at a chosen setting. The price to pay is a smaller
volume delivered, leading to CO2 retention permissive hypercapnia.
(a)
Normal lung
Time
PM
PA
Time
Time
(b)
COAD (Resistance)
Time
PA
PM
Time
Time
(c)
PM
PA
Time
Time
Time
Restrictive lung disease

(Compliance)
Lung filling with automatic lung ventilators
Figure 53. Lung filling with a constant flow generator.
65
Part 2
Clinical measurement
67
Basic measurement concepts
Part 2
Physical and physiological variables are measured using an instrument that

converts the measured quantity (the input signal) into an electrical (output)
signal. The signal is then amplified, displayed and recorded in analogue (wave)
or digital (numeric) form. Figures 5456 illustrate some comparisons between
the measured quantity (the true value) represented by the black continuous
line, and the output of an instrument, represented by the individual dots
(measurements).
For accurate measurement, the system must fulfil the following criteria:
1. Static accuracy this means that:
r The output signal is directly proportional to the input signal the
linearity principle. For linear relationship, see the chapter on
mathematical concepts. Figure 54 shows test results of two new
glucometers compared with a gold standard. The first apparatus
conforms to the linearity principle but deviates somewhat from the true
value on either side: it is fairly accurate, but not precise. The second
apparatus shows consistently lower results: it is precise but not accurate.
The gold standard apparatus, of course, must be both precise and accurate.
r There is neither baseline nor sensitivity drift, i.e. the output signal does
not deviate from true zero in time, with changes in temperature or other
factors see Figure 53.
r There is absence of hysteresis, i.e. that the systems response is identical
whether the input signal is applied in an incremental or decremental
manner. Systems exhibiting hysteresis thus produce a signal that forms a
loop (see Figure 56); its rate of rise is slow at first and faster at higher
values of upward increments. Conversely, when the input signal is
decreasing, the output signal falls off slowly at first but the rate of decrease
is faster near the lowest point. Such a loop resembles the compliance loop,
and indeed the property of hysteresis can be thought of as the
compliance of the system.
2. Physiological reactance this means that the measuring system does not
affect the process being measured.
3. Dynamic accuracy to fulfil this requirement, the system must faithfully
record rapidly changing events. This means that the system must
accurately reproduce both the amplitude of each harmonic component
(see the chapter on mathematical concepts and sine wave), and their
phase relationship.
68
Re
Re
(b)
Time
Time
Figure 54. Linearity principle: (a) measurement accurate but impercise. (b) measurement precise but inaccurate.
(a)
Re
Re
(a)
(b)
Time
Ti
Figure 55. Drift: (a) baseline (zero) drift. (b) sensitivity drift.
Ou
In
Figure 56. Hysteresis loop.
69
2
Part 2
70
To avoid amplitude distortion the system must have a frequency

response higher than the highest harmonic. For instance, at a heart rate
of 120 beats min1 , the frequency of an associated signal (e.g. blood pressure)
is 2 Hz. Assuming that at least 10 harmonics are needed for an accurate reproduction of the input signal, the systems amplitudefrequency response must
be about 20 Hz.
Phase distortion is frequency-dependent. Critical damping extinguishes natural oscillations of the system and abolishes phase distortion; all
the harmonics will be recorded with the same time delay and the output
signal will be undistorted although delayed in time. See the chapter on electromanometers and Figure 57 for more detail.
Noise, i.e. a false signal, will also reduce dynamic accuracy depending on
its frequency. Static accuracy will not be affected as new machines can filter
out background noise or disregard it. The stronger the input signal and the
weaker the noise, i.e. the higher the signal-to-noise ratio, the better dynamic
accuracy.
To satisfy the above criteria, calibration is required. This is achieved
by the application of static increments and decrements of input signal and
recording the response. Drift is established by leaving the system with the
desired input signal for several hours.
The dynamic response is tested by a stepwise application of a large signal
and its sudden removal, as illustrated opposite.
Time
Figure 57. Critical damping ( = 1).
Time
71
2
Part 2
Electromanometers, frequency
response and damping
An electromanometer is a device for direct (invasive) blood pressure measurement. The sensor is a pressure transducer, a device that converts mechanical
energy to electrical signals: blood pressure oscillations are conducted along a
fluid-filled tube (catheter) to a chamber that has a flexible diaphragm on the
end (shown in Figure 58). Displacement of the diaphragm by these oscillations is sensed electrically and converted into an electrical signal.
Frequency response
After a single change in pressure inside the system, the fluid continues to
move in oscillations as it is reflected from the diaphragm on one side and
blood vessel wall on the other. The frequency of this oscillatory movement
is inherent to the set-up and is called the resonant or natural undamped
frequency, fn .
To follow the measured pressure the system must be able to respond
quickly: its frequency response must be higher than the highest frequency
measured.
The frequency response (natural resonant frequency) is calculated using
the formula:

f n = r /2 S/l
It can be seen that a large-bore catheter (large radius r) of short length (l), a
stiff diaphragm (S), plus fluid of low density ( ), are the requirements for a
high natural resonant frequency of an electromanometer.
Damping
For correct operation, the system has to be able to respond adequately to the
highest frequency contained in the transducer waveform. Therefore, it has to
respond quickly. However, a high natural resonant frequency of the system
would produce natural oscillations at that frequency, which could interfere
with the recorded frequency. Damping is the tendency to extinguish natural oscillations through viscous and frictional forces. In the mechanical
model this is provided by immersing a spring in a viscous fluid, which slows
down and extinguishes the natural oscillations. The same factors that influence natural frequency also influence damping but in an inverse manner; in
addition, the viscous resistance, as known from the HagenPoiseuille formula, is directly proportional to viscosity, . In the clinical situation, the
specific fluid viscous resistance and density are constant and can be ignored.
Dextrose is used because it is a non-ionic solution.
72
Catheter
Diaphragm
Figure 58. Electromanometer principle.
Electromanometers, frequency response and damping
Blood
vessel
73
2
Part 2
It follows from the above that the factors that we can use to manipulate
the response of the system are catheter length, its diameter and diaphragm
stiffness. Long and narrow catheters reduce the frequency response, and
increase damping.
Figures 57 and 5961 illustrate four situations with different degrees of
damping.
1. In a system where damping is virtually absent ( 0), a stepwise
pressure change results in an overshoot followed by oscillations at natural
frequency. Such a situation would arise if the catheter is relatively wide
and short. In the corresponding arterial blood pressure trace, note the
natural oscillations after a flush and on the trace, and overshoot of the
signal.
2. At the other extreme, if a long and very narrow catheter is used, damping
is greatly increased ( 1), and the system is unable to respond quickly
to the pressure change. The graphic recording is that of a slow-rising
saturation exponential curve. The corresponding blood pressure trace has
a more rounded form with loss of detail; oscillations after flush are
absent. In the extreme case, the pressure trace almost converges on the
mean blood pressure.
3. If we manipulate the parameters in such a way that the system responds
more quickly but just avoids overshooting, damping is said to be critical
( = 1) (Figure 57). Natural oscillation after flush is absent in the
hypothetical arterial blood pressure trace but some loss of detail is still
present.
4. The situation above is clearly not the best solution, as the response of the
system may not be fast enough to follow the higher frequencies.
Therefore, damping is further reduced (catheter shorter and wider,
diaphragm stiffer) to allow a small overshoot. Experimentally it was
found that the best compromise between accuracy and speed is achieved
when damping is two-thirds of critical ( = 0.66): this is optimal
damping (Figure 61). On the arterial blood pressure trace, the small
overshoot is found as a single oscillation after flush, and it can be seen
that detail is preserved without distortion by oscillations. A too damped
trace on the pressure recording may be due to the presence of air, or to
partial blockage of the catheter by clot, resolved by removal of air and by
flushing of the line.
Another factor that can be accidentally or deliberately introduced is an air
bubble: this interrupts the fluid column and in effect reduces the stiffness of
the diaphragm, as air is compressible. The presence of air therefore reduces
the frequency response of the system, and increases damping. See also the
chapter on basic measurement concepts.
74
Time
Figure 59. Low damping ( 0).
Time
Electromanometers, frequency response and damping
Time
Time
Figure 60. High damping ( >> 1).
Time
Figure 61. Optimal damping ( = 0.66).
Time
75
Pulse oximeter principle
Part 2
An oximeter is an instrument designed to measure oxygen saturation of

haemoglobin in whole blood. Oxygen saturation (Sa O2 ) is defined as the
ratio of oxygen content of haemoglobin over oxygen carrying capacity.
Oximeters operate on the principle of light absorption; the formula below
expresses the LambertBeer law:
I = Io ec.d .E ,
where I, Io = received and initial light intensity; c = concentration of absorbing medium; e = the base of natural logarithms; d = distance travelled; and
E = the extinction coefficient of a particular medium.
The exponential coefficient has three components: E, which is fixed
for each medium, c (concentration) and d (distance), which can be varied.
Figure 62 shows light absorbances of oxyhaemoglobin, deoxyhaemoglobin,
methaemoglobin and carboxyhaemoglobin, in the red and infra-red spectrum.
It can be seen that the lines of oxy-and deoxyhaemoglobin cross several times,
and deviate most widely at 660 nm wavelength. The points where these two
lines cross indicate the wavelength at which the light absorbances are equal;
such a point is called an isobestic point. The 800 nm point was used as
a reference point to compensate for changes in total haemoglobin concentration (light absorbance at this point does not depend on oxyhaemoglobin
concentration). The light absorbance of a given sample of blood at 660 nm (or
any other point chosen) will lie somewhere between the two curves, depending on the percentage of oxyhaemoglobin concentration (c), which is in fact
oxygen saturation.
Pulse oximeter components and method of measurement
The pulse oximeter consists of two light-emitting diodes, a photocell detector
and a microprocessor with a visual display unit. The light-emitting diodes
are illuminated in turn at a set frequency, and an off phase is incorporated
to compensate for extraneous light. The wavelengths chosen in practice are
actually 660 nm (red) and 940 nm (infrared), since this gives a better separation
of the wavelengths (i.e. the newer machines do not utilize the isobestic point).
At 940 nm the light absorbance of deoxyhaemoglobin is less than that of
oxyhaemoglobin, while at 660 nm the absorbances are reversed. However,
linear relationship of the light absorbances ratio to oxygen saturation still
applies, as shown in Figure 63.
The detector cannot distinguish between different wavelengths, therefore
it assumes that light received is that from the light-emitting diode currently
on. Light absorption varies with cardiac cycle because of the difference in
oxyhaemoglobin percentage between arterial and venous blood. This variation
is detected at the two wavelengths, and pulse rate calculated.
76
2
Me a
Ox
Re
Ca
SpO2 (%)
Figure 62. Pulse oximeter principle. Extinction coefficients of oxyhaemoglobin, reduced

haemoblobin, ethaemoglobin and carboxyheamoblobin in red and infrared light.
A660/A940 (nm)
Figure 63. Relationship between saturation of arterial blood with oxygen and the light
absorbances ratio of haemoglobin.
77
2
Part 2
78
The pulse oximeter detects the cyclical change in light absorption in the
arterial end of the capillaries as well as constant light absorption by venous
blood and tissues. The constant absorption is ignored, and pulse waveform
displayed based on the changing absorption by arterial blood, as shown in
Figure 64.
There are numerous sources of error and interference which include physiological and physical factors (vasoconstriction, hypotension, additional arterial
or venous pulsations, scattering and refraction of light, time delay), intrinsic
and extrinsic factors (other types of haemoglobin, dyes, nail varnish, patient
movement, diathermy, ambient light). Inherent error is less than +/3% at
saturations > 70%.
Pulse oximetry as a form of monitoring has many advantages: it is a continuous non-invasive method of measurement of haemoglobin oxygen saturation,
gives early warning of hypoxic events, and is superior to clinical judgement.
Easy to use, with little interpretation needed, it is one of the minimum monitoring requirements. The main danger is a false sense of security when used
to detect hypoxaemia at high inspired oxygen concentrations.
2
Figure 64. Components of the pulse oximeter waveform.
79
Oxygen content and oxygen

tension measurement
2
Part 2
Definitions
r Oxygen content: oxygen bound by haemoglobin + oxygen dissolved in
r
r
r
r
solution. Measured as volume at STP or molecular mass (the relationship

between volume at STP and mass is constant, see Avogadros law).
The amount of oxygen bound by haemoglobin depends on haemoglobin
concentration, its oxygen-carrying capacity and oxygen saturation. The
first and last factors are variable. Oxygen carrying capacity is fairly
constant.
Oxygen tension: partial pressure of oxygen in solution (pO2 ). Measured
in units of pressure (mmHg or kPa).
Oxygen concentration: strictly speaking applies only to concentration
in a gaseous mixture (e.g. air has 21% oxygen). Sometimes applied to
oxygen content.
Oxygen saturation: oxygen content/oxygen carrying capacity. See pulse
oximetry and oxygen dissociation curve.
Oxygen content measurement

Physical method: van Slyke (Figure 65)
This is a manometric method. Instead of volume the apparatus measures
pressure at constant volume and temperature. Change in pressure then accurately reflects change in the mass of gas (Avogadros and Boyles laws apply).
Oxygen has to be released from the blood sample by haemolysing the red
cells with saponin and treating haemoglobin with ferricyanide to convert it
to methaemoglobin this releases all oxygen bound to haemoglobin. At the
same time, CO2 is released by the application of lactic acid. The released gases
are extracted into vacuum and compressed to a standard volume. Vacuum
extraction and subsequent compression is done by the means of lowering and
raising the mercury levelling bulb. The gases are absorbed in turn (oxygen
with sodium hydrosulphite, CO2 with sodium hydroxide) and the resulting
change in pressure measured. The burette is surrounded by a water jacket for
temperature maintenance; temperature is measured and adjusted for standard
temperature.
The apparatus is glass and is therefore breakable; the method is slow and
cumbersome but very accurate. It is suitable for calibration of other analysers.
80
(b)
Figure 65. Van Slyke apparatus (a) water jacket; (b) mercury levelling bulb. Reproduced
with permission from Butterworth-Heinemann.
Oxygen content and oxygen tension measurement
(a)
81
Electrochemical method: fuel cell (Figure 66a)

This method uses the oxygen reaction with water and an electric current (the
electron is denoted as e ).
Part 2
O2 + 4e + 2H2 O 4(OH) .
The amount of current consumed in the above reaction is directly proportional to the rate of reduction of oxygen at the cathode, and since the
only source of oxygen in the fuel cell was the blood sample introduced, the
ampermeter readout, after electronic integration, will give oxygen content.
Temperature compensation needs to be applied by means of a thermistor.
The fuel cell is a battery and has, therefore, a limited life. External voltage
need not be applied as the battery produces its own. The amount of oxygen
used in the reaction depends on oxygen uptake at the cathode and thus on its
partial pressure. To obtain the oxygen bound in haemoglobin in solution,
oxygen has to be released from haemoglobin. This is done in an apparatus,
called Lex-O2 -Con, by haemolyzing the blood sample with distilled water in
a scrubber and passing oxygen-free carrier gas through it. The carrier gas is
nitrogen with 1% CO. The CO reacts with haemoglobin in a similar manner as ferricyanide in the van Slyke apparatus, i.e. it releases all the oxygen.
Oxygen partial pressure in the carrier gas then directly corresponds to the
concentration in solution as the gaseous and liquid phase are in equilibrium.
Polarographic method: Clark electrode (Figure 66b)
This method uses the same electrochemical reaction as described above. However, no battery is incorporated and therefore an external voltage of 0.6V
must be applied. Note that electrodes are different a platinum cathode and
silver/silver chloride anode. Current flow is then measured which, as in the
case above, depends on oxygen tension at the platinum cathode.
The cathode is protected from deposits by a thin plastic membrane; this
separates it from the blood sample, and blood equilibrates with a small amount
of electrolyte surrounding the cathode. Temperature is again controlled at
37 C.
This method is commonly used in blood gas analysers. The oxygen
electrode must be kept clean and the plastic membrane regularly checked for
deposits or puncture and then replaced.
82
Paramagnetic method (Figure 67)

This technique uses oxygen paramagnetic properties (oxygen is attracted
into a magnetic field). This is because the electrons in the outer shell of an
oxygen molecule are unpaired. Most other gases, e.g. nitrogen, are diamagnetic (repelled from a magnetic field).
The gas analyser contains a gastight chamber with a glass dumb-bell filled
with nitrogen, which is suspended on a filament. A non-uniform magnetic
field is applied across the chamber. When oxygen is introduced into the chamber, it displaces the glass dumb-bell filled with nitrogen (as oxygen is attracted
into the magnetic field, while nitrogen is repelled). This displacement can be
measured because the dumb-bell rotates on the filament. A mirror attached
(b)
Figure 66. Oxygen measurement: (a) fuel cell (b) Clarke electrode. Reproduced with
permission from Butterworth-Heinemann.
Figure 67. Paramagnetic oxygen analyser.
(a)
83
2
Part 2
to the dumb-bell reflects a light beam and in older versions indicated the
position on a calibrated scale. In the newer null-deflection system two photocells receive the light beam their output is compared and used to generate
a current around the dumb-bell which produces an opposing magnetic field
and thus keeps the dumb-bell in its resting position. This apparatus is very
accurate and fast; it is used in most modern on-line anaesthetic gas analysers
for breath-by-breath analysis.
Mass spectrometry (Figure 68)
This method can be used for measurement of any gas. It separates gases
according to their molecular weight.
A gas sample introduced is ionized and the ions are then separated, based
on their respective charge-to-mass (Q/m) ratio. Only a small amount is
analysed, which diffuses into the system via a molecular leak. Ionization is
achieved by bombarding the sample with electrons. Separating can be done
by means of eliminating all ions except those with a specific Q/m ratio. This
happens in the quadrupole system: the electrical potential across the four
rods is varied so that the ions oscillate as they travel. As the oscillations increase,
the ions hit the rods and lose their charge; only the measured gas is allowed
to travel through. In a sector system, the separating is done by means of
a magnetic field applied across the flow of ions; these are then deflected
according to their Q/m ratio and the deflected beams collected in specific
positions in collector cups.
This instrument allows rapid simultaneous breath-by-breath analysis of
respiratory gases and vapours. It is used in the latest gas analysers. Its cost is
prohibitive for day-to-day use but it is useful for research purposes. Size is
no longer a problem; a portable system has been developed.
In summary, paramagnetic and mass spectrometric methods are suitable
for analysis of respiratory gas mixtures; van Slyke, fuel cell and polarographic
methods are suitable for blood gas analysis.
84
2
Figure 68. Quadruple mass spectrometer.
85
Capnography
Part 2
Capnography is the measurement of carbon dioxide in a gas mixture. It uses

the method of infrared absorption spectrophotometry. All gases with
dissmilar atoms in the molecule, such as carbon dioxide and nitrous oxide,
absorb infrared light. Peak absorbance by carbon dioxide is at 4.26 m, fairly
close to, but distinguishable from, the two nitrous oxide peaks at a slightly
higher wavelength. Modern capnographs therefore utilize narrow bandwidth
infrared light of 4.26 m wavelength. Figure 69 shows as schematic representation of a capnograph. After passing through the monochromatic filter and
the sampling chamber, residual light is detected by an infrared light detector
and processed electronically. The windows of the sampling chamber have to
be made from halogen crystals or other materials that do not absorb infrared
light. To avoid drift, most instruments incorporate a reference chamber and
a chopper close to the source of light to provide a discontinuous source of
light (not included in the diagram).
There are two types of infrared CO2 monitors: sidestream and mainstream
capnographs; they differ in the gas sampling technique. Sidestream capnographs sample from the breathing system via a sampling line. The apparatus
is lightweight and is suitable after adaptation for unintubated subjects. Mainstream capnographs have a measuring head placed in close proximity to the
endotracheal tube, and the measuring chamber is heated to prevent condensation. These are heavy and cumbersome, are sensitive to external contamination with dirt and have the potential to cause burns and to break. However,
they have a faster response time, less scope for error and are unaffected by
expiratory flow rate.
A problem common to both types is response time the signal is a saturation exponential function of time with a time constant about 0.1 s. Therefore
a 90% response time is achieved in about 0.2 s (two time constants). Most
capnographs can then cope with respiratory rates up to 60 min1 but if further
delay is introduced by a long sampling line, inaccuracies can occur at rates of
40 min1 .
Waveform analysis (5 mm s1 )
In Figure 70, a single-breath gas washout curve is depicted. Phase I between
points A and B represents dead space gas, in which there is no CO2 , unless
it is being added (dead space gas is the fresh gas). After a rapid rise, points
BC or phase II, the end-tidal CO2 curve reaches the alveolar plateau
points CD (phase III), which is often sloping slightly upwards because of
uneven emptying of different alveolar regions with varying time constants,
and because the lung volume is getting progressively smaller, whereas CO2
excretion continues. A further increase may occur at the end between points
D and E (phase IV), if expiration proceeds below closing capacity.
86
2
CO2
Capnography
CO2
Figure 69. Principal components of a capnograph.

From left to right: infrared light source, filter, sampling chamber, light detector, processor, output.
Figure 70. Single breath analysis.
87
2
Part 2
88
Abnormalities of single breath waveform (see Figure 71)

(a) Steep phase III as shown in Figure 71a. Alveolar emptying is more
uneven. This happens in asthma or chronic obstructive airways disease,
and is due to different time constants in different lung regions with
various degrees of altered compliance and airways resistance.
(b) Cardiogenic oscillations (Figure 71b). This may happen in paediatric
circuits, during low anaesthesia, and at low respiratory rates. In these
cases, gas sampling rate exceeds expiratory flow. The ripples are caused
by cardiac activity and pulmonary blood flow. The oscillations can be
eliminated by application of positive end-expiratory pressure or moving
sampling port most distally.
(c) Dip in alveolar plateau (usually only seen in ventilated patients). The
dip means the patient makes a small inspiratory effort during the course
of expiration. Action by the anaesthetist will depend on the stage of
operation and postoperative plan.
(d) Rebreathing (Figure 71d). The capnographic waveform does not return
to zero on inspiration and gradually shift upwards, with a same or
higher amplitude. Rebreathing is eliminated by removing dead space.
2
Capnography
(a)
(b)
(c)
(d)
Figure 71. Abnormalities of single breath curve, (a) chronic obstructive airways disease,
(b) cardiogenic oscillations, (c) dip in plateau, (d) rebreathing.
89
Trend analysis (display at 25 mm min1 )
Part 2
Hypercapnia (end-tidal only)

As shown in Figure 72 can be caused by:
(a) Increased CO2 production (pyrexia) or a sudden CO2 load (release of
tourniquets or aortic clamp, i.v. bicarbonate, insufflation of CO2 ).
Reduced CO2 elimination depression of respiratory centre, or
reduction of ventilation due to neuromuscular problems or iatrogenic
(IPPV).
(b) Baseline drift upward from zero indicates rebreathing or a calibration
error: compare the two graphs of hypercapnia.
Hypocapnia (Figure 73)
(a) Increased CO2 elimination hyperventilation (common cause): slow
exponential fall.
(b) Decreased CO2 production hypothermia: slow exponential fall.
Major disturbance to pulmonary or systemic circulation (shock, air
embolism, cardiac malfunction): fast exponential fall in end-tidal CO2 .
(c) Endotracheal tube kinking or disconnection: abrupt fall.
In the last two situations, end-tidal CO2 does not accurately reflect arterial
pCO2 . The pa CO2 ETCO2 gradient will then be increased over the usual
23 mmHg (5 mmHg in anaesthetized subjects).
Uses of capnography
r Confirmation of endotracheal intubation (oesophageal intubation
false-positive only in first few breaths).
r During IPPV to determine the correct level of minute ventilation.
r During anaesthesia with spontaneous breathing to monitor respiratory
activity, determine fresh gas flow and anaesthetic agent requirement (in
particular during low flow).
r Detection of disconnection.
r Trend analysis as above.
r Differential diagnosis of hypotension (induced hypotension versus
cardiovascular collapse).
90
2
Capnography
%CO2
5
(a)
%CO2
5
0
(b)
Figure 72. (a) Hypercapnia:(a) carbon dioxide overproduction or absorption, (b)
rebreathing.
%CO2
5
(a)
%CO2
%CO2
(b)
(c)
Figure 73. Hypocapnia (a) overventilation, (b) collapse, (c) disconnection.
91
[H+], pH and its measurement
Part 2
Basics of acidbase balance

Maximum strength of acid = 1 g/l [H+ ] ( = 100 g/l). Maximum strength
of alkali = 1014 g/l [H+ ]. Water is halfway between 100 and 1014 = 107
g/l = 100 nmol/l (1 nmol = 109 mol).
To make these small numbers easier to handle, and to accommodate the
large range, Sorensen proposed a logarithmic scale, so that
pH = log10 [H+ ].
For the definition of a logarithm, see the chapter on exponentials. The logarithmic scale has equal distances between the powers of 10, or e in the case of
natural logarithms, rather than between the units. It converts an exponential
curve into a straight line, as can be seen in Figure 74. For the effect of a logarithmic scale on the population distribution of hydrogen ion concentration
and pH, see the chapter on statistics.
The physiological concentration of H+ is near water but slightly on the
alkaline side: 40 nmol/l. Thus physiological pH = log10 40 109 :
= log10 4 108 = (0.6 8) = (7.4) =
= 7.4.
92
The normal pH range is 7.367.44, i.e. very narrow. pH < 7 is already a severe
acidosis.
When converting [H+ ] to pH, we have to appreciate the properties of the
logarithmic scale. While small changes into the alkaline direction of hydrogen
ion concentration result in significant changes in pH, the reverse happens on
the acid side: between pH 8 and 7, the H ion concentration changes 10-fold,
10100 nmol/l; between pH 7 and 6, the H ion concentration changes 10-fold
again, 1001000 nmol/l, while the pH only lost another unit. Figure 73 shows
this relationship between the pH values of 6 and 8. Bold linear scale on the
x-axis relates to the bold exponential curve. The scale of the logarithmic x-axis
is arbitrarily chosen to fit of [H+ ] 40 and 100.
In practice, pH users know that pH about 6 is an extremely severe metabolic
acidosis, and are not fooled by the seemingly small change in pH. Furthermore, on the alkaline side the physiological consequences of a smaller deviation are equally devastating, i.e. severe metabolic alkalosis, or perhaps more,
since the body is not well equipped to deal with an alkali load (or [H+ ] deficit);
the change on the alkaline side in hydrogen ion concentration from 40 to
10 nmol appears rather small and belies the seriousness of the condition.
A practical way for an approximate conversion of [H+ ] into pH follows
from the properties of the exponential curve: for each 0.3 change in pH
between 7 and 8, halve the hydrogen ion concentration (from 100 at 7 to 12.5
at 7.9); for each 0.3 unit change in pH between 7 and 6, double the hydrogen
ion concentration (from 100 at 7 to 800 at 6.1). The numbers are not exact
because the factor of 0.3 is not exact.
2
[H+ ], pH and its measurement
logarithmic
Figure 74. [H+ ]pH relationship on a linear and logarithmic scale.
93
2
Part 2
It is possible to argue that because physiological processes of elimination

tend to happen exponentially, perhaps using pH makes more sense than the
H ion concentration. It would have been even more logical to use natural
logarithms. These are not, however, easy to handle mathematically. Notice
that the pH electrode reacts in direct proportion to pH, not [H+ ].
pH meter (Figure 75)
The instrument is a potentiometer, i.e. measures change in electrical potential across two electrodes. It acts as a battery, i.e. no electromagnetic force
(voltage) need be applied. The Hg/HgCl electrode is a reference electrode
that maintains a constant electromagnetic potential. The Ag/AgCl electrode
is the pH electrode and incorporates pH-sensitive glass: the potential across
the glass changes in direct proportion to unit pH (not [H+ ] another example
of a natural exponential process), i.e. 60 mV per unit pH. When the sample
chamber is empty the battery is disconnected. When a blood sample is introduced the circuit is completed and the potential can be measured across the
two electrodes.
94
Sample
out
Capillary of pHsensitive glass
Reference
electrode
Hg/Hg Cl
(calomel)
Ag/Ag Cl
electrode
[H+ ], pH and its measurement
pH readout
Buffer
solution
Porous plug
Membrane
Water jacket at 37 C
Sample
In
Figure 75. The pH electrode. Reproduced with permission from ButterworthHeinemann.
95
2
Part 2
Principles of measurement of flow

in gases and liquids
Flow measurement is based on the physical principles underlying the definition of flow.
Q = A.v
In words, flow is measured as a function of velocity (for an illustration,
see the chapter on laminar flow). This principle is used, for instance, in the
electromagnetic flowmeter: electric potential develops across a magnetic
field through which blood is moving; the magnitude of this potential is directly
proportional to the average velocity of blood flow. Ultrasonic flowmeters
use the change in frequency of an audio signal when it is reflected from moving
red blood cells. This again depends on blood velocity. If the diameter of the
vessels is measured at the same time, blood flow is computed and displayed
simultaneously.
Two methods of gas flow measurement are based on this formula:
96
Q = p/R
r Variable orifice (and thus resistance), constant pressure. Flow is then
inversely related to resistance or directly related to the size of the
orifice. Laminar flow increases 16 times if the radius of the orifice doubles
(see the chapter on flow). Conversely, flow is reduced 16 times when the
radius is halved and two times if the radius is only 15% less. In Figure 76,
water (rather than gas) flow from a mains pipe with a tap is plotted against
the resistance offered by the tap. It is shown that closing the tap only
slightly from the maximum at point A, to reduce its diameter by 15% at
point B, halves the flow. The Rotameter and the Wright peak flow meter
are examples of variable orifice flowmeters. Constant pressure in the
former is provided by the pressure of the bobbin; the pressure is the ratio
of the product of the mass and gravity over the cross sectional area of the
bobbin (p = F/A = m a/A). The Rotameter is tapered, and thus the
magnitude of the orifice around the bobbin corresponds to the magnitude
of gas flow. In Wrights peak flow meter the opposing pressure is the
tension from a coiled spring acting on the rotating vane area.
r Constant orifice (and resistance), variable pressure. The flow in this
case is directly proportional to the pressure drop (Q = R1 p,
where R = constant), i.e. a linear relationship (Figure 77). The
pneumotachograph is based on this method: a pressure drop across a
gauze screen or a light, plastic valve of known resistance is measured and
transduced. In Figure 75, the relationship between pressure and flow is
illustrated as flow from a vessel with a fixed outlet resistance. The
magnitude of flow at any given moment is directly proportional to the
hydrostatic pressure in the vessel. Pressure is higher at point B on the
plot, and flow is correspondingly higher. (The sequence of events in
practice would be from point B to point A.)
2
(D = 7 mm)
(D = 6 mm)
s per litre)
Figure 76. Reciprocal relationship between flow and resistance for constant pressure.
Principles of measurement of flow in gases and liquids
Q
1
R
p
Figure 77. Linear relationship between driving pressure and flow.
97
2
Part 2
98
Q = dV/dt
Flow as the rate of change of volume with time. d means change or
derivative. Cardiac output by the thermal dilution method uses this formula
(see the next chapter).
It may be helpful to define the derivative and the reverse process, integration at this point. The derivative is the slope of the tangent to the curve
representing a given function. It is described as the rate of change of the
function.
Figure 78 illustrates constant flow as a derivative. Volume of mains water
that passed through the area of interest (the measurement point) is plotted
against time. The flow of water through a pipe under mains pressure is constant, and the rate of rise of volume collected with time is constant: over 1 s,
we collect, say, 100 ml, over 1 min, 6000 ml. When volume against time is
plotted, the slope of this linear relationship is also the slope of the derivative,
the flow. Its numerical value is given by the fraction Vt /t; this is constant at
any point on the line: the flow is constant.
Exponentially falling flow, such as when emptying a bath, is the subject of
the chapter on exponentials.
If we chart flow and wish to know what the volume collected was over a
specified time, integration is applied. Derivation of volume with time yielded
flow; integration of flow, the reverse operation, yields volume. Calculation of
volume from flow is easily done by multiplication of the flow by the specified
time. Mathematically and graphically this is called integration: a small volume
passed in a unit of time can be depicted as the area of a small rectangle with
sides equal to the two variables (unit of time and increment of volume over
time i.e. flow). In Figure 79 the rectangles under the line of flow are the
individual increments. As the flow is constant, it is obvious that the sum of
these incremental rectangles is equal to the total area under the line. If the
flow were variable e.g. decreasing such as in Figure 80, and the incremental
rectangles infinitely narrow, so that the irregularity of the curve at the top
of the rectangle is minimised, their sum would again be equal to the area
under the curve (AUC). To sum up integration with respect to time is
summation of increments over time in order to obtain total.
Principles of measurement of flow in gases and liquids
V
t
V
t
Figure 78. Flow as a derivative of volume with respect to time.
t
Figure 79. Area under the line of flow = volume passed. Constant flow.
V
Q
l over 2 minutes
)
t
Figure 80. Decreasing flow.
99
2
Part 2
Cardiac output measurement by

thermal dilution technique
Bedside cardiac output measurement (Figure 81) using the thermal dilution
technique is based on the formula Q = dV/dt.
It is the measurement of blood volume that passed the measuring point
during one circulation time.
Since collection of aortic blood flow over any period is not a bedside option,
the blood volume that passed over the measurement period is measured indirectly, as a change in blood temperature, or concentration of cold. A known
quantity of cold saline is injected into the bloodstream and is thus mixed with
circulating blood, reducing its temperature. The volume of the saline (10 ml)
is negligible compared with the diluting volume and does not require correction in the calculation. The quantity of cold, or negative heat, is also small
but it is sufficiently large to produce a measurable temperature drop, which
of course changes with time as the mixing is done in the bloodstream. Blood
temperature is recorded intravascularly with a thermistor at the tip of the
pulmonary artery flotation catheter, and the signal is processed by computer.
Remember there is a reciprocal relationship between volume and
concentration. In other words, what we measure (concentration of cold)
is inversely related to the result (blood flow). The diluting blood volume
calculated as the ratio of mass of cold and concentration of cold (V = m/c)
is then used in the formula above for the calculation of cardiac output:
V = m/c
To know the final hypothetical concentration of cold, the cardiac computer
applies integration (see the previous chapter): integration of the instantaneous
change in concentration (dc/dt) yields total final concentration over specified
time, t. The integral, as shown in the previous chapter, is the area under
the curve (AUC) of the instantaneous recording in the concentration. The
reciprocal relationship between cardiac output and AUC still applies, as
we merely substitute AUC for concentration of coldin the volume formula:
V (diluting volume) = m/c = amount of cold/AUC
Cardiac output is then calculated by substituting the result, diluting volume,
V, which is the volume of one circulation, into the flow formula.
In Figure 81, the instantaneous change in blood temperature is shown
with its mirror image, the concentration of cold. The lower curve is the
familiar shape produced on the cardiac computer screen. Remember that the
curve is not the pulmonary artery flow but the concentration of cold saline,
the blip on it is not the dicrotic notch but the change in temperature with
recirculation, and the area under it is not cardiac output but a hypothetical
final concentration of cold saline. Because of recirculation the first large curve
must be extrapolated to zero and the area under this curve only is calculated.
100
(a)
Te
(b)
Te
Co
t
Figure 81. Cardiac output measurement (a) high output, (b) low output.
Cardiac output measurement by thermal dilution technique
Co
The AUC is identical for the temperature and concentration curves for
a particular cardiac output. Compare the high output graph with the low
output one: the peak of the low output curve is smaller but the total AUC
is larger. This makes sense if we keep in mind that a fast flow will produce
a swiftly rising and falling concentration curve, with a greater instantaneous
change in temperature but a small AUC, while slow flow will do the opposite.
101
2
Part 2
Measurement of the mechanical

properties of the chest
Static compliance (C = V/p)
The change in volume is measured as the patient takes a breath from a
spirometer.
A change in pressure is given as the difference between pleural and
alveolar pressure. Direct pleural pressure measurement would produce a
pneumothorax; oesophageal pressure is a good approximation. With the
glottis open and after equilibration, the mouth pressure is (for measurement purposes) equal to alveolar pressure. The measurement is taken at
different lung volumes, and compliance is calculated from the slope of the
pressure/volume plot. See the chapter on respiratory mechanics for the plot.
Airways resistance (R = p/Q = palv pmouth Q)
The pressure gradient is the difference between alveolar and mouth pressures
during flow.
Mouth pressure can be measured directly by a pressure transducer; alveolar pressure has to be derived, for instance, as a reciprocal value to a change in
pressure inside an airtight box (body plethysmograph) during tidal breathing. In Figure 82, the pressure waveform of alveolar pressure and pressure
inside the box are plotted against each other to show the relationship.
Flow is measured directly by a pneumotachograph, and resistance calculated.
102
P
Figure 82. Pressure changes in constant volume plethysmograph.
Measurement of the mechanical properties of the chest
103
2
Part 2
Lung volumes and their

measurement
Figure 83 shows the lung volumes. The capacities are sums of individual
volumes.
Tidal volume and the volumes above can be measured directly by a
BenedictRoth spirometer, not described here. The volumes below resting
expiratory level are best measured indirectly.
Functional residual capacity (FRC)
The FRC is of more interest than residual volume, particularly the relationship
of FRC to the closing volume. The indirect measurements utilize the principle
of dilution of a known mass of a gas in the unknown volume.
Closed circuit helium
The closed circuit helium wash-in method employs this gas because it is
relatively insoluble in blood: gas uptake is assumed to be zero. The patient
rebreathes from a closed circuit with a known concentration of helium. CO2
is absorbed and sufficient oxygen supplied for adequate oxygenation. After
equilibration the mass, m, of helium in the volume composed of FRC and
the volume of the apparatus, V, must be the same as the mass of helium which
was originally in the apparatus only
m1 = m2
m=Vc
V [He1 ] = (FRC + V ).[He2 ].
Therefore FRC = V [He1 He2 ]/He2 .
This is illustrated in Figure 84: a known volume of a known concentration of
helium is mixed with the unknown volume (FRC). After mixing, the helium
concentration in the total volume is lower (as there was no helium in FRC).
Helium concentration is measured by a katharometer (a measurement apparatus based on the change of thermal conductivity).
104
Figure 83. Lung volumes.
Lung volumes and their measurement
Figure 84. Measurement of functional residual capacity (FRC) by helium wash-in

method.
105
2
Part 2
Open circuit nitrogen washout

Open circuit nitrogen washout similarly employs nitrogen dilution in the
lungs after a period of oxygen breathing. The exhaled volume has to be collected and measured. The mass of nitrogen found in the exhaled collected
gas (calculated as the product of final expired concentration of nitrogen, fE ,
and the sum of exhaled volume, VE , plus the volume of the apparatus, VApp )
must be the same as the mass of gas which left the lung (given by the product
of FRC and the difference between alveolar concentration of nitrogen before
and after the period of oxygen breathing, fA1 fA2 ):
(VE + VApp ).fE = FRC.(fA1 fA2 ).
Therefore FRC = (VE + VApp ).fE /fA1 fA2 .
Figure 82 illustrates this situation. What is in effect done is mixing a known
concentration of nitrogen (80%) from an unknown volume (FRC) into a
larger volume (expired volume plus the volume of the apparatus), and the
resulting concentration is measured. Because not all nitrogen is washed out,
alveolar concentration of nitrogen at the end of oxygen breathing has to be
subtracted from the alveolar concentration at the beginning. Alveolar concentration is assumed to equal end-tidal concentration of nitrogen. During
oxygen breathing, nitrogen is also being washed out from blood and the
tissues; this proceeds more slowly and is not taken into account.
The above methods can also be used to determine the distribution of
inspired air. Poorly aerated lung zones have long time constants to achieve
equilibrium when gas composition is changed. With helium wash-in, the
number of breaths needed to achieve equilibrium is the measure of the time
elapsed (1530 breaths, or <3 min). Control measurement is needed to compensate for the characteristics of the apparatus. Nitrogen washout takes longer;
in normal lungs alveolar concentration of nitrogen after 7 min is < 2.5%. Values greater than this indicate non-uniform distribution of inspired gas.
Dead space volume, closing volume (single breath analysis of
expired nitrogen)
106
During and after a single breath of oxygen, the change in volume is plotted against time, and simultaneously nitrogen concentration in expired air is
monitored.
In Figure 83, the volume plot against time is shown in the upper half,
and expired nitrogen concentration in the lower half. During inspiration of
oxygen, nitrogen concentration will be zero; it stays at zero while dead space
gas is breathed out. When alveolar gas reaches the measuring point, nitrogen
concentration abruptly rises and stays at a plateau for most of the expiration.
Towards the end of expiration, when closing volume is reached, it rises again,
as more concentrated nitrogen is breathed out from a smaller lung volume.
Dead space volume and closing volume can simply be read from the plot
between the points indicated.
In a patient with emphysema and a high physiological dead space, the alveolar plateau of nitrogen is never reached nitrogen concentration continues
to increase from the beginning of expiration until the end.
2
Lung volumes and their measurement
Figure 85. Measurement of functional residual capacity (FRC) by nitrogen washout.
V
V
Sp
Figure 86. Single breath nitrogen washout. Reproduced with permission from
107
Part 3a
Physiology: the
cardiovascular system
109
3a
Part 3a
110
The cardiac cycle and the

intravascular pressure waveforms
With the exception of systemic arterial pressure, intravascular pressure waveforms can be observed during pulmonary artery catheterization.
The balloon-tipped catheter is introduced into a central vein and with
appropriate scale adjustment the central venous pressure trace, similar to
the one opposite, can be seen. The A wave corresponds to atrial contraction,
and is followed by the C wave: tricuspid valve closure simultaneous with
ventricular contraction. X descent follows, as the right atrium and central
vein are now empty and ventricular pressure is now indirectly transmitted
across the valve. The right atrium starts filling until the tricuspid valve opens
again. The V wave corresponds to the valve opening, following which central
venous pressure drops momentarily (the Y descent) as the right atrium empties
passively into the right ventricle until the next atrial systole.
The scale is usually adjusted to monitor right ventricular pressures, and
the central venous pressure trace is then an undulating waveform around the
usual 68 mmHg. A straight line indicates a technical problem such as wrong
connection or a kinked line.
At 15 cm the balloon is inflated and the catheter advanced downstream.
After 20 cm the catheter should enter the right atrium. Right atrial pressure
is related to central venous pressure, their difference being the hydrostatic
pressure difference; in the supine patient this difference is very small.
Right ventricular pressure trace is instantly recognized when the systolic pressure reaches 2025 mmHg, while the diastolic pressure should be
close to zero.
At about 30 cm, the catheter tip should pass the pulmonary valve and the
pulmonary artery pressure trace should be seen. Its shape and amplitude are
similar to the ventricular pressure trace; however, the effect of the pulmonary
valve closure can be seen as the pressure trace does not return to zero, and
the diastolic pressure is about 812 mmHg.
Figure 87 shows a semi-diagrammatic representation of several related
parameters on the right side of the heart.
For better illustration, heart sounds are added to relate the pressure recordings to the mechanical events (valve closures), and an ECG waveform to relate
these events to the electrical cycle.
Note how the pulmonary artery systolic pressure is identical with the right
ventricular systolic pressure; the difference between these pressures is in the
diastolic pressure. Pulmonary artery occlusion pressure is not shown here, as
this is a measurement relating to the left side of the heart.
3a
0
AS
VS
VD
(s)
1.0
ECG
Audio
Pressures
20
(mm Hg)
PAP
PAP
a c
0
a c
RAP
RAP
RVP
RVP
Legend:
PAP pulmonary artery pressure
RAP right atrial pressure
The cardiac cycle and the intravascular pressure waveforms
Time
RVP right ventricular pressure

PVO
PVC
TVC tricuspid valve closure

TVO tricuspid valve opening
PVO pulmonary valve opening
TVC
TVO
PVC pulmonary valve closure
Figure 87. Intravascular pressure waveforms on the right side of the heart during pulmonary artery catheterisation. Adapted from: C. G. Caro, T. J. Pedioy, R. C. Schroter &
W. A. Seed, The Mechanics of Circulation, OUP, 1978.
111
3a
Part 3a
After a further 510 cm, the balloon occludes the pulmonary artery and a
continuous column of blood should exist between the catheter tip and the left
atrium. A pressure trace now recorded is the pulmonary artery occlusion
pressure (wedge pressure). This corresponds to left atrial pressure and left
ventricular filling pressure, provided the following conditions are fulfilled
r The pulmonary intravascular pressure exceeds extravascular pressure, to
obtain a continuous column of blood. This condition is not fulfilled if the
intravascular pressure is low, e.g. when the catheter is lodged in West
zone 1, or if the extravascular pressure is abnormally high, as with the
application of positive end-expiratory pressure.
r The pressure in the left atrium reflects the left ventricular end-diastolic
pressure (LVEDP). In mitral valve disease left atrial pressure can be much
higher than LVEDP. Conversely, in heart failure when sinus rhythm is
preserved, LVEDP may be higher than the mean left atrial pressure: left
atrial pressure increases to the high ventricular level only during the atrial
systole.
Wedging of the pulmonary artery flotation catheter can be seen on the screen as
a sudden decrease in the amplitude of the pressure waveform to < 12 mmHg.
(Left heart filling pressure is higher than on the right side.) On deflation of
the balloon, a normal pulmonary artery trace should be seen again.
Arterial pressure waveform is shown in Figure 88 with airway pressure
during intermittent positive pressure ventilation. Blood pressure trace shows
changes in phase with respiration, called the respiratory swing. (For more
details of the effects of intermittent positive pressure ventilation on blood
pressure see next chapter.) As the intrathoraic pressure is only increased briefly
during inspiration, overall effect is a decrease in arterial systolic and diastolic
pressure which lags behind the increase in intrathroacic pressure. Respiratory
swing is pronounced in hypovolaemia, and is a valuable tool in the estimation
of fluid deficit. Notice also that the dicrotic notch is low and the arterial
pressure trace is narrow in hypovolaemia. Many ITU monitors have different
sweep speeds for arterial pressure and for respiration; association of blood
pressure swing and airway pressure is then not observed on the monitor.
112
3a
Airway
pressure
Figure 88. Arterial blood pressure and respiratory swing.
The cardiac cycle and the intravascular pressure waveforms
Arterial
blood pressure
113
3a
Part 3a
Cardiovascular effects of
intermittent positive pressure
ventilation
Intermittent positive pressure ventilation causes changes in intrathoracic pressure and blood volume, which are reflected in the cardiovascular parameters.
Four phases can be recognized (see Figure 89):
r Phase I: blood is forced into the left atrium by the increased intrathoracic
pressure blood pressure rises and reflex bradycardia follows.
r Phase IIa: blood pressure falls because of decreased venous return the
result of the raised intrathoracic pressure.
r Phase IIb: blood pressure is restored to normal by reflex tachycardia.
r Phase III: venous return is decreased on lowering the intrathoracic
pressure as there is now increased venous capacity in the lungs. Blood
pressure falls initially but normalizes soon as reflex tachycardia persists.
r Phase IV: persisting tachycardia results in an overshoot of blood
pressure; heart rate then drops back to normal (baroreceptor reflex) and
blood pressure then also normalizes.
The magnitude of the effect on blood pressure as a result of effects on right
and left ventricular filling depends on blood volume and the integrity of
the sympathetic nervous system control, i.e. effects will be exaggerated in
the presence of hypovolaemia, sympathetic blockade (this includes general
anaesthesia) or autonomic dysfunction (e.g. diabetic neuropathy), or a combination of factors (ill patients, e.g. sepsis), as compensation is lost (blocked
Valsalva). The ventilator settings also play a role fast ventilation rates with a
long inspiratory phase will allow little time for compensation; high ventilation
pressures will cause greater effects.
In patients with pulmonary hypertension (e.g. severe ARDS), increased
airway pressure during intermittent positive pressure ventilation results in
increased pulmonary vascular resistance. Interventricular septum may shift
to the left and compromise the left ventricular filling, and eventually decrease
the stroke volume. Because the lung is leaky in ARDS, inspiration pressure
has to be controlled rather than fluids administered to improve left ventricular
filling.
114
3a
DP 80
P 40
Figure 89. Arterial blood pressure and pulse during the Valsalva manoeuvre. SP, DP
systolic and diastolic pressures. P pulse.
Cardiovascular effects of intermittent positive pressure ventilation
SP 120
115
3a
Part 3a
116
Control of cardiac output,

regulation of cardiac function
Cardiac output is the product of stroke volume and heart rate. Both are
under the control of the sympathetic nervous system. Stroke volume is also
affected by changes in preload, contractility and afterload, and their interaction. This chapter will deal with preload and afterload only. For the effect
of contractility on cardiac performance, see chapter Cardiac cycle: pressure
volume relationships. Venous return (preload) depends on blood volume,
posture and venous tone. Venous tone is under the control of the sympathetic
nervous system. Sympathetic output changes in response to the peripheral
metabolic need.
Figure 90 shows the FrankStarling relationship between the left ventricular end-diastolic pressure (LVEDP) and stroke volume (SV). The Frank
Starling curve mechanism allows a ventricle to match its output (stroke volume) to the volume of blood that enters it (the preload).
On the x-axis, end diastolic pressure is substituted for fibre length, and on
the y-axis stroke volume is substituted for force of contraction (fibre length
and force of contraction were the variables used in the original research on
isolated muscle fibre). In practice, we use surrogate parameters that can be
directly measured and make assumptions which are valid under normal physiological conditions, the assumption being that LVEDV (left ventricular end
diastolic volume) LVEDP = LAP (left atrial pressure) = PAOP (pulmonary artery occlusion pressure). The relationship between LVEDP and SV
is linear up to a certain unphysiological point. After that, the mechanism fails
(in the original research this happened because of fibre disruption).
Point OP on the middle curve represents the operating point under normalconditions, with a filling pressure of about 8 mmg Hg and a stroke volume
of 70 ml. The curves to the right and left show how the relationship changes
under varying conditions of inotropy and afterload. The slope of the curve
is increased with increasing inotropy or decrease in afterload. Conversely, the
slope is less steep (the curve moves down) with increased afterload or a loss
of inotropy.
In critically ill patients, we optimize the preload first in order to improve
cardiac performance. Patients in circulatory failure due to volume loss will
need generous levels of preload for optimum cardiac function. Patients in
shock because of pump failure (e.g. myocardial infarction, chronic heart failure) will need cautious filling to lower targets and a may need a reduction
in afterload. Septic shock is a mixture of pump failure due to myocardial
depression, and volume loss (absolute due to pyrexia, or relative due to
vasodilatation); optimizing the preload is often a result of a trial and error
process. Volume filling is done in small increments whilst at the same time
manipulating the systemic vascular resistance to counter vasodilatation.
3a
Control of cardiac output, regulation of cardiac function
100
OP
SV
50
(ml)
0
0
10
20
LVEDP (mmHg)
Figure 90. FrankStarling curves. Changes in afterload and inotropy shift the Frank
Starling curve up or down.
117
3a
Part 3a
Cardiac cycle: pressurevolume

relationships
Left ventricular compliance and elastance
The heart fills during the diastole by venous return, augmented, if the heart is
in sinus rhythm, by atrial contribution. The pressurevolume relationship in
an object being filled is called compliance (dV/dp) in this case the left ventricular compliance, left ventricular end-diastolic volume and left ventricular
end-diastolic pressure.
In practice, ventricular pressures are more easily measured than ventricular
volumes, and therefore the inverse of compliance, i.e. elastance or stiffness
(dp/dV), is of interest. In Figure 91, the left ventricular pressure is chosen as
the dependent variable, and the slope of the relationship between LVEDP and
LVEDV corresponds to the stiffness of the left ventricle: the steeper the slope,
the stiffer (less compliant, more thick walled or restricted) is the left ventricle.
The relationship is linear up to a certain point of maximum stretch, beyond
which the LVEDP rises steeply with any further increment of volume. Under
physiological conditions, the heart operates on the linear part of the curve,
and therefore it is possible to assume that LVEDP will correspond to LVEDV.
Note that the left ventricular end-diastolic volume corresponds to the
initial fibre length of the FrankStarling relationship. When measuring the
pulmonary artery occlusion pressure (PAOP), we estimate the left ventricular filling from a pressure measurement inside a small pulmonary artery.
The relationship between the PAOP and the LVEDV is complex (see more
in chapter Control of cardiac output); ventricular compliance (or stiffness)
is only one of the factors which can alter this relationship: a stiffer ventricle will achieve a smaller filling volume before the ventricle reaches the
maximum capacity point, or conversely higher end-diastolic pressures are
needed in order to achieve a comparable end-diastolic volume. This has to
be borne in mind when interpreting the results of cardiac catheterization: the
assumption that left ventricular end-diastolic pressure (estimated from
the pulmonary artery occlusion pressure) corresponds to left ventricular
end-diastolic volume only applies provided that left ventricular compliance remains unaltered.
Volatile anaesthetics decrease left ventricular compliance to various
degrees, and this may be the basis for the extent to which they depress myocardial function.
The left ventricle maintains flow in the systemic circulation. It is a highpressure system and, therefore, the left ventricle is thick walled.
The right ventricle and pulmonary circulation are a low-pressure
system; the filling pressure is about 5 mmHg, and the systolic pressure about
20 mmHg. The ventricle is much more compliant than the left ventricle and
thus can match left ventricular output at low pressures.
118
3a
100
High elastance
Normal
100
200
LVEDV (ml)
Figure 91. The relationship between left ventricular end-diastolic volume and enddiastolic pressure.
Cardiac cycle: pressurevolume relationships
LVEDP
(mm Hg)
119
3a
Part 3a
120
Pressurevolume loops
This is a dynamic representation of the relationship between left ventricular
volume and left ventricular pressure during each phase of the cardiac cycle.
The pressure volume loop can be used to describe cardiac performance
and the effect of various changes in physiological parameters.
Starting from the ED (end-diastole) point in Figure 92(a), the left ventricle
is filled to maximum and the ventricular systole starts. During the isovolumetric phase of the systole, tension develops in the muscle fibre which
will relate to the force of contraction in the next phase. Ventricular volume
during this phase remains unchanged.
During the isotonic phase of the systole, cardiac muscle contracts, i.e.
fibre shortening occurs. The aortic valve opens and blood is ejected out of
the ventricle. Ventricular pressure during this phase changes little until the ES
(end-systolic) point, when it decrases abruptly as the ventricle relaxes during
the isometric phase of the diastole.
The tangential line at the ES point describes the end-systolic pressure
volume relationship (ESPVR). Its slope corresponds to the inotropic state
of the heart (see below).
The bottom line describes the pressurevolume relationship during the
isotonic phase of diastole (ventricular filling). Its slope inversely relates to a
particular state of ventricular compliance.
Figure 92b shows how the pressurevolume loop changes with an
increased state of inotropy. Notice that the slope of the ESPVR is
steeper, allowing the ventricle to contract more forcefully and therefore to
increase the stroke volume (the difference between ES and ED point). The
ES point therefore is lower on the x-axis (LV volume).
Figure 94c demonstrates the effect of heart failure. The slope of the
ESPVR is lower, and the ventricle has to operate at a much higher filling
volume in order to generate pressures comparable to normal conditions. As
emptying is inefficient, stroke volume is smaller and the ES point is much
higher on the x-axis.
In acute heart failure due to myocardical depression such as occurs in
septic shock, the heart has a relative lack of inotropic drive. The remedy in
this case is to use inotropic infusion in order to improve the ESPVR and
enable to ventricle to contract more efficiently. The loop will shift to the left
(normal).
By contrast, the efficiency of the heart in chronic heart failure or in states of
high afterload can be improved by a decrease in afterload. This improves the
slope of the FrankStarling curve (see chapter Control of cardiac function),
which improves the stroke volume for a given ventricular filling pressure.
Thus the ES point and the whole loop moves left, and the ESPVR improves
as a result.
(a)
200
3a
ES
100
ED
100
200
LV volume (ml)
Figure 92(a). Left ventricular pressurevolume loop at steady-state.
(b)
200
Increased
Inotropy
LV
pressure 100
(mmHg)
Cardiac cycle: pressurevolume relationships
LV
pressure
(mm Hg)
Control
Loop
0
0
100
LV volume (ml)
200
Figure 92(b). The effect of increased inotropy on left ventricular pressurevolume loop.
Heart rate and aortic pressure held constant.
(c)
200
Loss of
Inotropy
LV
pressure
100
(mm Hg)
Control
Loop
0
0
100
LV volume (ml)
200
Figure 92(c). The effect of left ventricular failure (reduced inotropy) on left ventricular
pressurevolume loop. Heart rate unchanged.
121
3a
Part 3a
Blood pressure and blood volume

relationship
Following blood loss, the body draws on its reserves to maintain blood flow
and oxygen delivery to the tissues. Initially, water is retained by the kidney
and extracellular fluid is drawn into the intravascular compartment. If blood
loss continues, physiological compensatory mechanisms bring about changes
in other physiological parameters to maintain blood flow to the tissues, and in
a more severe haemorrhage to the vital organs only. Haemorrhage is classified
into four degrees of severity:
r Class I: 15% or 750 ml blood loss. Stroke volume may fall minimally
at lower levels of loss, resulting in a minimal tachycardia to maintain
cardiac output. This is the situation induced by venesection in a blood
donor.
r Class II: 1530% or 7501500 ml blood loss. Tachycardia is noticeable
while systolic blood pressure is still maintained; diastolic pressure,
however, rises due to the higher level of circulating catecholamines. Mean
blood pressure is maintained but flow to the organs without
autoregulation of blood flow is reduced (e.g. muscle, skin). Because of the
reduced skin blood flow, core-to-skin temperature difference starts to
rise. Renal blood flow is minimally affected and urine output is
maintained at a physiological minimum. Cerebral blood flow is
maintained due to autoregulation but anxiety due to the circulating
catecholamines is evident.
r Class III: 3040% or 15002000 ml blood loss. The compensatory
mechanisms are being exhausted and circulatory failure starts to develop.
Tachycardia is marked and there is a measurable fall in systolic blood
pressure. Tachypnoea is present due to reduced O2 delivery to the tissues.
Urine output decreases significantly, core-to-skin temperature difference
increases further and mental changes are pronounced.
r Class IV: 40% or 2000 ml blood loss results in circulatory failure,
with compromised blood flow even to vital organs. Tachycardia is very
high, systolic blood pressure markedly depressed, while diastolic pressure
is still high; thus pulse pressure is very narrow. Urine output is negligible
or zero, and if mean pressure falls below the autoregulatory level of the
brain, the level of consciousness may be depressed. The skin feels cold
because of vasoconstriction. Blood lactate, not shown in Figure 92, is
significantly increased. Loss of 50% or 2500 ml blood results in loss
of consciousness and total circulatory failure.
122
Only some measurable physiological parameters are depicted in Figure 93;

other changes such as skin colour and sweating are important in the evaluation
of blood loss.
The response to haemorrhage described here is a typical one of a healthy
individual. Heart disease, neuropathy or medication may produce a different
response; some individuals respond with a paradoxical bradycardia.
3a
Blood pressure and blood volume relationship
Sy
Me
Di
Pu
He
Ur
Re
Di
Figure 93. Changes in physiological parameters with blood loss (classes of haemorrhage).
123
3a
Cerebral blood flow
Part 3a
Normal cerebral blood flow is 54 ml/100g/min (750 ml/min). The formula for
flow indicates that flow is given by the ratio of pressure gradient (in this case
cerebral perfusion pressure, which is the difference between mean arterial
pressure and intracranial pressure), and resistance
MAP ICP
.
(1)
R
Autoregulation exists to maintain, within physiological limits, a constant
blood flow to the brain regardless of variations in arterial blood pressure and
intracranial pressure; this is done by changes in cerebrovascular resistance,
or by changing the perfusion pressure. As shown below, these factors are
interrelated.
r Mean arterial pressure: Figure 94 shows how autoregulation maintains
Q = p/R =
124
within physiological limits (normally 60160 mmHg) constant cerebral

blood flow. This is affected by changes in cerebral vascular resistance (see
below). In chronic hypertension (or hypotension), the autoregulatory
limits are moved up or down respectively. Outside the autoregulatory
limits cerebral blood flow is pressure-dependent. When cerebral
perfusion pressure is low due to high intracranial pressure (> 2530
mmHg), baroreceptor stimulation produces systemic hypertension and
cardioinhibitory centre stimulation produces reflex bradycardia
(Cushings reflex).
r Intracranial pressure: the relationship between intracranial fluid
volume and pressure is shown in Figure 95 cerebral elastance (or
stiffness), defined as p/V. Since the intracranial pressure is the
dependent variable, it is logical to plot the pressure on the y-axis, and to
use the term elastance in this context. Elastance is the inverse of
cerebral compliance, which is defined as V/p. In the graph elastance
is given by the slope of the curve at any point, as shown. The skull is a
rigid container filled with brain, cerebrospinal fluid and blood. As fluid is
essentially incompressible, any increment in cerebral volume (e.g. in
cerebral oedema) must be matched by a corresponding decrease of one of
the other fluid components; cerebrospinal fluid can be to a certain extent
pushed out of the foramen magnum, and venous blood is propelled out as
the veins are compressed inside the skull. Since venous blood pressure is
one of the components of intracranial pressure, a fall in the former (as
venous blood is squeezed out) is accompanied by a fall in the latter. Thus,
initially, a small increase in cerebral volume can be accommodated
without a significant rise in intracranial pressure. When the compensatory
limits are over stretched or lost, any small increment of cerebral volume
(e.g. cough, drug or hypercarbia-induced vasodilatation) will be reflected
in a steep rise in intracranial pressure, with a detrimental effect on
cerebral blood flow.
3a
Cerebral blood flow
No
Hy
Hy
Figure 94. Cerebrovascular autoregulation.
V
p
V
p
V
Figure 95. Cerebral elastance curve.
125
3a
Part 3a
126
r Cerebrovascular resistance: the matching of cerebrovascular resistance

to cerebral perfusion pressure is the basis of autoregulation of cerebral
blood flow. The factors determining cerebrovascular resistance are:
1. Partial pressure of CO2 : cerebral arteriolar resistance is under the
direct influence of local pCO2 across a wide physiological range.
Vasodilatation under normal intracranial pressure conditions results in a
higher blood flow; this is partly the mechanism of autoregulation: if
cerebral blood flow is reduced or if the cerebral metabolic rate is high,
cerebral pCO2 rises and pH falls, leading to vasodilatation and restoration
of blood flow. In head injury, a mild degree of hypocapnia is preferable
when ventilation is controlled under anaesthesia, provided blood
pressure is maintained: intracranial pressure will fall as a result of
venoconstriction and cerebral blood flow will improve. Very low levels of
pCO2 (< 3.4 kPa) lead to symptomatic cerebral ischaemia. Hypercapnia
abolishes cerebrovascular autoregulation: cerebral blood flow is then
directly proportional to mean arterial blood pressure, (as shown in Figure
96). Conversely, cerebrovascular response to CO2 is blood
pressure-dependent: in severe hypotension cerebral blood flow does not
change with changes in CO2 tension, as shown in Figure 97.
2. Oxygen partial pressure in the major cerebral arteries plays a role
outside the physiological range; global cerebral blood flow only starts to
rise when hypoxia is already significant. At the tissue level, however, O2
partial pressure is probably the mechanism regulating local or regional
blood flow in response to hypotension: reduced local blood flow with
resulting tissue hypoxia produces immediate arteriolar vasodilatation.
High levels of pO2 are associated with a mild degree of vasoconstriction,
i.e. reduced cerebral blood flow (as shown in Figure 97).
3. Hydrogen ion concentration: the effect is similar to, but independent
of, pCO2 ; the hydrogen ion is the mediator of flow metabolism coupling
but it is not involved in the response to hypotension.
4. Blood viscosity influences vascular resistance as shown in the
HagenPoiseuille formula (see above). The higher flow in haemodilution
compensates (in part only) for otherwise reduced O2 delivery due to the
reduced haemoglobin concentration. This mechanism applies in cerebral
circulation as in any other part of the body.
5. Neurogenic and myogenic control: apart from the humoral factors,
the cerebrovascular tone is under sympathetic nervous control, and the
autoregulation curve is shifted to the left or right according to the
sympathetic tone. Myogenic response (increased tension in response to
increased stretch) probably plays a part similar to other parts of the body.
In head injury, autoregulation is lost and the intracranial pressure may be
elevated due to the presence of haematoma, cerebral oedema or both. Ventilation may be depressed with resulting hypercapnia: cerebral blood flow is then
pressure-dependent. To prevent secondary brain injury, ventilation must be
controlled and an adequate O2 delivery to the brain ensured. A cerebral perfusion pressure > 60 mmHg is thought to provide adequate flow. From formula
1 it is then obvious that the desired mean arterial pressure must be kept at a
level calculated as the sum of intracranial pressure + 60 mmHg.
3a
Cerebral blood flow
per min)
Figure 96. Autoregulation of cerebral blood flow within physiological limits; it is abolished by hypercapnia.
(PaCO2)
(PaO2)
(PaCO2)
(PaO2)
Figure 97. Cerebral blood flow as a function of arterial carbon dioxide tension (Pa CO2 )
and oxygen tension (Pa O2 ).
127
3a
Part 3a
128
Coronary circulation
Myocardial blood flow is 200 ml/min, or 4% of cardiac output, for an organ
weighing only 0.4% of body weight. Oxygen consumption of the heart is also
high, 23 ml/min, or 9% of total. This is for a good reason: the heart is a pump
that perfuses the rest of the body; its work is hard and it needs a constant
energy supply from aerobic metabolism.
The coronary arteries are the first to receive oxygenated blood from the
aorta; their perfusion depends on the pressure gradient generated by
the heart. It is important to remember that coronary arteries run on the
epicardial surface. The coronary arterial pressure gradient is thus from epicardium to endocardium, while the intramural pressure gradient during systole is in the opposite direction. Therefore, there is practically no endocardial
flow during systole (see Figure 98 where coronary artery flow is plotted against
the arterial pressure waveform) while the flow in epicardium is maintained.
To compensate for the lack of perfusion in systole the subendocardial arteries
are thought to be in a chronic state of dilation during diastole. At times of
increased demand for perfusion, e.g. tachycardia, hypertension, this region
is then unable to increase flow further and thus it is more susceptible to
ischaemia.
Anaesthetic agents depress myocardial performance and oxygen consumption falls in line with reduced myocardial work (see Figure 100). Coronary blood flow therefore is reduced.
Myocardial ischaemia occurs when myocardial oxygen demand exceeds
supply, i.e. when coronary blood flow and oxygen flow fall below the minimum required. In a diseased myocardium symptoms of ischaemia occur at a
higher perfusion pressure, inside the lower limit of autoregulation. The prediction of cardiac events during anaesthesia is difficult but prevention should
be practised: cardiac performance (power) is the product of mean arterial
blood pressure and cardiac output. The output is the product of stroke volume
and heart rate. Stroke volume is not easily assessed by bedside (or operating
tableside) measurements but heart rate and blood pressure are monitored.
Thus the ratepressure product remains a useful clinical tool when estimating myocardial oxygen demand, and therapeutic manoeuvres can be directed
at optimizing myocardial performance to maintain oxygen flow to the systemic circulation (which requires adequate cardiac output and haemoglobin
concentration) while not overloading the heart by excessive pressures and
rates.
3a
Figure 98. Coronary artery flow and arterial blood pressure.
Figure 99. Coronary autoregulation.
129
3a
Part 3a
130
Autoregulation
Unlike the brain, which at times of need receives a higher perfusion pressure
via baroreceptor stimulation, the heart cannot effectively increase its oxygen
flow by increasing its perfusion pressure since the heart generates the pressure:
myocardial oxygen consumption rises proportionately with myocardial work.
The heart, therefore, regulates its perfusion only via changes in coronary artery
resistance. The oxygen tension in the myocardium or a related parameter is
the governing factor. Autoregulation maintains a constant blood flow in the
coronary circulation within a wide range of pressures, 60140 mmHg (see
Figure 99). It must be kept in mind that the aortic diastolic pressure (not the
mean arterial pressure) is the coronary perfusion pressure.
The sympathetic and parasympathetic nervous system affects the
coronary vascular resistance, but this is modified by autoregulation: adrenergic stimulation produces coronary vasoconstriction but if coronary
blood flow is compromised autoregulation results in vasodilatation. Parasympathetic stimulation, if unopposed, results in bradycardia. The accompanying
fall in oxygen demand produces vasoconstriction; however, if a fast heart rate
is maintained, vasodilatation prevails.
3a
f
Figure 100. Effect of inhalational anaesthesia with halothane on coronary blood flow
and myocardial oxygen consumption.
131
Part 3b
Physiology: the respiratory

system
133
Oxyhaemoglobin dissociation
curve
3b
Part 3b
The oxyhaemoglobin dissociation curve (see Figure 101) describes oxygen

binding to haemoglobin and its release in terms of haemoglobin saturation as a
function of partial pressure oxygen. It is a sigmoid curve and it rather resembles
the drug doseeffect relationship curve (see the chapter on pharmacology).
However, in the case of the oxyhaemoglobin dissociation curve, the dose,
i.e. oxygen partial pressure, is on a linear scale. This means that small, linear
increments in oxygen partial pressure produce large changes in haemoglobin
saturation on the straight part of the curve; at the extreme ends of the curve
the opposite applies.
The central value on the straight part of the curve, which corresponds to
haemoglobin saturation of 50%, is called the p50 ; it is normally 3.6 kPa (26.6
mmHg). This is an unphysiological but convenient value to describe the
position of the oxyhaemoglobin dissociation curve with respect to the pO2 .
Points on the normal curve that correspond to the arterial and venous point
are denoted as A and V. A shift of the oxyhaemoglobin dissociation curve to
the right results in a higher p50 , and conversely a left shift lowers the p50 . The
higher the p50 , the lower the affinity of haemoglobin for O2 .
Oxygen affinity (and therefore P50 in a reciprocal manner) is affected by
the following interrelated factors:
r 2,3-Diphosphoglycerate (2,3-DPG) concentration in the red cell: this is
r
r
r
the primary control mechanism. 2,3-DPG is the allosteric effector that

binds selectively to the deoxygenated haemoglobin, producing changes in
haemoglobin molecule conformation. Oxygen affinity of haemoglobin is
inversely related to 2,3-DPG concentration, therefore the oxyhaemoglobin dissociation curve is shifted to the right with a rise in 2,3-DPG.
[H+ ]: in this particular situation, [H+ ] is preferable to pH since hydrogen
ion concentration affects O2 affinity in the same way as the other factors.
A negative feedback mechanism through the effect of the hydrogen ion
concentration on the 2,3-DPG concentration restores p50 back to normal.
pCO2 : increasing the partial pressure of CO2 increases p50 . This and the
effect of pH was described by Bohr (Bohr effect). pH and pCO2 are
interrelated, but each is also a separate component.
Temperature: There is also a negative feedback mechanism through the
2,3-DPG concentration.
Mean corpuscular haemoglobin concentration: Mean corpuscular
haemoglobin concentration changes rapidly with shifts in pH, to balance
the slower changes in 2,3-DPG concentration.
For ease of O2 delivery to the tissues, oxygen should bind easily to haemoglobin in the lung, and it should be easily released in the periphery.
134
3b
A
t, [H+], pCO2, 2,3DPG
75
50
25
pso
10
Oxyhaemoglobin dissociation curve
t, [H+],p CO2, 2,3DPG
SO2(%)
100
pO2(kPa)
A arterial point at 96% saturation (10.7 kPa)

V venous point at 74% saturation (5.3 kPa)
Figure 101. Oxyhaemoglobin dissociation curve.
The body normally tends towards acidosis, constantly produces CO2 and
produces heat. These conditions result in a lower affinity of haemoglobin for
O2 , to facilitate O2 release in the tissues. The price for this is the necessary
higher O2 partial pressure that must be supplied. Conversely, to achieve adequate oxygen delivery, fetal haemoglobin must have a higher affinity than
maternal haemoglobin for O2 . The fetal pH range is lower than maternal pH
and this allows O2 to be released in the fetal tissues.
135
3b
Part 3b
Respiratory mechanics 1: Static

properties, factors affecting
compliance, closing volume
Elastic properties of the lungs and chest wall
From the mechanical point of view, the lungs are elastic sacs filled with
air through tubes of varying resistance, attached on their outer surface to a
container that is also elastic but which contains a rigid structure the chest
wall. Lung expansion has its limits; as the lungs are filled, pressure inside rises.
The stiffer the lung, the higher the pressure rise will be for any given volume.
Compliance is the reciprocal entity to stiffness, or elastance, and is defined
as the change in volume over the change in pressure. It is the measure of the
elastic properties of the lungs, chest wall or the total:
C = V/p(1 kPa1 ).
Pulmonary pressures are usually measured in cmH2 O. Normal lung compliance would then be, for instance, 600 ml over 3 cmH2 O, or 200 ml.cmH2 O1 .
Depending on where the pressure change is measured, compliance can be
calculated for the lung, CL (pressure difference measured from alveolus to
pleural space), chest wall, CCW (pressure difference measured from pleural
space to atmosphere); the total, thoracic compliance CT , is the sum of its
components. When adding up compliance, it must be kept in mind that it is
reciprocal to stiffness. Total stiffness is the sum of the individual components,
while reciprocal values are added up for compliance.
1/CT = 1/CCW + 1/CL
ST = SCW + SL .
Under normal resting conditions, the retractive forces (elastic recoil) of the
lung are balanced by the expansile forces of the chest wall; this balance governs
the resting position. The elastic recoil of the lung draws it inwards; the resting
position for the lung is at the end of expiration the resting expiratory position.
On the other hand, the elastic forces of the chest tend to expand it; the resting
position for the chest wall is at the tidal volume level (500 ml higher than
the resting position for the lung). In Figure 102a, the resting position of
the isolated lung and isolated chest is shown for the right lung, and for the
expanded lung on the left. The pressure gradients are shown for the resting
expiratory position, and in the next drawing (Figure 102b) for the tidal breath
position.
To expand the lung by 200 ml, a pressure difference of 1 cmH2 O has to
be applied across the lung and 1 cmH2 O across the chest wall. Therefore the
sum for total compliance would be:
136
CT = V/p = 200/1 + 1 = 100(ml cmH2 O1 ),
3b
(b)
p A
p A
p A
Figure 102. Respiratory mechanisms. (a) resting position of the isolated lung and chest
(right lung) and resting expiratory position of the normal lung at FRC (left lung). Atmospheric pressure p = A. Pressure gradients are shown as vectors, (b) pressure gradients
after a tidal breath.
C
C
Respiratory mechanics 1: Static properties, factors affecting compliance, closing volume
(a)
p
Figure 103. Static compliance of the lung (CL ), chest wall (CCW ) and total compliance
(CT ).
137
3b
Part 3b
which means that total thoracic compliance is smaller than that each component. It is the stiffness of each that adds up directly. Therefore, to expand the
thorax by 500 ml, a pressure difference of 5 cm has to be applied.
The volumepressure relationship of the lungs and chest wall are not
linear at the extremes of lung volume: at a high lung volume the lung opposes
further filling, i.e. its compliance is smaller (compliance slope less steep). In
Figure 103, the compliance is the slope of the line depicting the pressurevolume relationship. At a low lung volume below the resting expiratory level,
the expansile force of the chest opposes further expiration, so that chest wall
compliance is smaller. This is reflected in the shape of each compliance curve.
Notice that the total compliance then becomes a sigmoid curve. The nonlinearity should be kept in mind when comparing different compliances: the
pressure difference for each value should be stated.
Factors affecting compliance are:
r Disease: in atelectasis, when the lung is relatively stiff, the point of
138
balance (resting expiratory volume) will be reached at a lower lung

volume, as there will be greater pull inwards; this will predispose to
further atelectasis. The excessively compliant emphysematous lung has
less elastic recoil; resting expiratory volume will then be greater, since the
natural tendency for the chest wall to expand is maintained.
Age: the immature lung in the infant is less elastic than in adulthood.
Elasticity is highest in young adults and it decreases slowly with advancing
age: lung compliance is therefore lowest in young adults. Chest wall
compliance is highest at birth and slowly declines with age (see Figure
104).
Posture: thoracic compliance is lower in the supine position, as the
gravitational pull of the abdomen, which existed in the upright position, is
reversed and the diaphragm is pushed into the chest by the abdominal
contents.
Anaesthesia: several factors (supine position, airway closure, changes in
intrathoracic blood volume, accumulation of fluid, direct effect of drugs,
altered muscle tone, external pressure) influence compliance under
anaesthesia; generally compliance is decreased.
Obesity: the effect is compounded by supine or lithotomy position.
3b
C
C
Figure 104. The effect of age on lung compliance (CL ) and chest wall compliance (CCW ).
139
3b
Part 3b
140
Airways closure
Pleural pressure is influenced by the pressure of thoracic contents: it is less
subatmospheric in the dependent lung zones then in the apex. When a normal
subject exhales below the resting expiratory level, and towards the residual
volume, pleural pressure eventually approaches or exceeds atmospheric pressure. The pressure gradient distending the small airways is thus lost, and the
small airways will close; they will re-open only when a sufficient gradient is
reached for their distension. The lung volume at which the airways close is
called the closing volume. It increases more rapidly with age than the functional residual capacity. At around 65 years of age closing volume exceeds
the FRC and airway closure occurs then during tidal breathing. The resulting mismatching of ventilation and perfusion is the explanation for the
higher shunt fraction in the elderly. The relationship of closing volume and
FRC is shown in Figure 105 for the upright and supine positions, and for the
supine-anaesthetized patient. The further the closing volume line from the
FRC line at a given age, the higher the shunt fraction. (See also the chapter
on ventilationperfusion relationship.)
3b
Age
Figure 105. Changes in closing volume (CV) and functional residual capacity (FRC)
during adulthood.
141
Respiratory mechanics 2: Dynamic

properties, factors affecting
resistance
3b
Part 3b
Airway (non-elastic) resistance

Airway (non-elastic) resistance depends on the pattern of airflow (laminar or
turbulent see the relevant chapters), the rate of breathing and the radius and
length of the airway, as described in the HagenPoiseuille formula. Remember that resistance is independent of flow in laminar flow but it rises linearly
with flow in turbulent flow. Much higher pressures have to be achieved in turbulent flow to pass the same flow. (See also the chapter on flow for Reynolds
number and the influence of viscosity and density.)
The radius decreases as the airways branch; the nasopharynx and larynx
account for half of the total airway resistance, and the trachea and smaller
airways constitute the other half. Airways smaller than 2 mm internal diameter contribute very little (see Figure 106). This is because their total crosssectional area is large, and airflows in the small airways are lower, therefore
laminar.
Factors affecting airway resistance are those which alter the diameter of
the airway:
r Body size: the diameters of the larynx and trachea account for the large
r
r
r
r
142
part of total resistance. Airway resistance is highest in neonates

(30 cm l 1 per s) and declines with increasing body size until a functional
residual capacity of 2500 is reached. Thereafter, the diameter of the
airways and airways resistance does not change substantially as shown in
Figure 107.
Thickness of bronchial mucosa, e.g. swelling or secretions.
Bronchial muscle tone: constriction due to parasympathetic stimuli,
histamine release, irritant gases, dilatation produced by sympathetic
stimuli or parasympathetic block.
Pressure gradients: high-pressure gradient in atelectasis, etc. will
distend airways, low-pressure gradient in emphysema results in smaller
diameter of the airways.
Anaesthesia: airway obstruction, effects of drugs.
3b
Respiratory mechanics 2: Dynamic properties, factors affecting resistance
Diameter
Order
Figure 106. Size of airway and total cross-sectional area.
Age
Figure 107. Airway resistance (R) and age.
143
3b
Part 3b
144
Dynamic compliance
This is measured during normal tidal breathing. Figure 108 shows the plot of
lung volumes and pressures for two sizes of breath in the normal lung, and
a large breath in chronic obstructive airways disease. The slope of the axis
of each loop is identical with the static compliance. It can be seen that for
a normal (tidal) breath in the normal lung, the compliance slope is steeper
than for a large breath, i.e. the lung is more compliant during normal tidal
breathing than during a vital capacity breath. During a vital capacity breath
in chronic obstructive airways disease, compliance slope is identical to the
normal lung (or it could be higher if the disease progresses to emphysema)
but the loop encompasses a greater area, indicating a greater work of breathing
(see chapter on simple mechanics, work and power). This is understandable
as the work is done against a higher airways resistance.
3b
Small
and
e.
Large
and
e.
Large
airways
e.
Figure 108. Dynamic compliance loops during spontaneous respiration.
Respiratory mechanics 2: Dynamic properties, factors affecting resistance
145
3b
Part 3b
146
Ventilationperfusion relationship
Alveolar ventilation and its distribution
Approximately two-thirds of total ventilation reach the alveoli; one-third
remains in the bronchi and is not available for gas exchange dead space gas.
Fresh gas that enters the alveoli per minute is known as alveolar ventilation;
the normal value is about 5 l min1 .
Alveolar ventilation is not distributed evenly in the lung: dependent regions
are better ventilated during spontaneous respiration than the upper zones.
This is because of the effects of gravity: pleural pressure at the base is less
negative, and the alveoli at the base are less expanded than apical alveoli. In
Figure 109, apical alveolus, represented by point A, is already nearly 80%
expanded, while the basal alveolus at point B is only about 30% expanded. At
maximum inspiration the change of size is higher for the basal alveoli than
the apical ones, which had less scope for expansion. The above only applies
for the normal lung at normal lung volumes. If basal alveoli are collapsed they
receive no ventilation if the tidal breath, and the pleural pressure change is
small. This is illustrated by point C (closed alveolus) in Figure 109. Such a situation occurs in the normal lung at residual volume, in old age when closing
volume exceeds functional residual capacity or in chronic obstructive airways
disease when functional residual capacity is high and the tidal breathing range
is near the closing volume. Pleural pressure at the lung base exceeds atmospheric pressure in these situations, and a greater change in pressure gradient
is required to re-expand basal alveoli, as shown.
Figure 110 illustrates how alveolar size changes across the lung from apex
to base.
3b
p
Figure 109. Lung pressurevolume diagram in a young, healthy adult.
Figure 110. Alveolar volume in relation to distance from lung apex.
147
3b
Pulmonary perfusion and its distribution
Part 3b
Blood flow distribution in the lung is also uneven, as the arteriovenous

pressure gradient, which is a function of the hydrostatic pressure in
the pulmonary arterial tree, is affected by alveolar pressure, which itself
is a function of intrapleural pressure. The reduction in the arteriovenous
pressure gradient (and therefore of flow) will be the greatest in the apex, where
alveolar pressure is the highest; at the base, where alveolar pressure is low,
the arteriovenous pressure gradient will be hardly affected by it. The result
is that pulmonary blood flow is highest in the region where the ventilation is
also highest the base: this is the mechanism of physiological ventilation
perfusion matching.
The lung can be divided into three zones the West zones according to
relative magnitude of the pulmonary arterial, venous and alveolar pressure.
This is illustrated in Figure 111.
r In zone 1 at the apex or non-dependent region, alveolar pressure may
exceed the capillary pressure. The capillary then collapses and receives no
flow. The ventilation that such an alveolus receives is dead space
ventilation. In the normal lung, apical capillaries are just expanded; in
hypotension or if alveolar pressure is artificially increased (e.g. in positive
pressure ventilation), the capillaries collapse and give rise to dead
space effect.
r In zone 2 alveolar pressure is smaller than pulmonary arterial pressure
but higher than the venous pressure. Here the pressure gradient for
blood flow will be the difference between arterial and alveolar
pressure (Starling resistor effect). Since the arterial pressure increases
downward because of the hydrostatic effect, the pressure gradient and
flow increase downwards too.
r In zone 3 venous pressure exceeds alveolar pressure and all capillaries are
held open. Blood flow is then the function of the arteriovenous
pressure gradient as in other tissues.
At low lung volumes, blood flow to the base is reduced in the normal lung
(thus matching reduced ventilation at the low volume) presumably because
the basal blood vessels would be more compressed, rather than being held
open wide as is the case at a normal lung volume. Another factor may be folds
in pulmonary capillaries.
148
3b
Pressure gradients
West zone
Figure 111. Diagram of the West zones of the lung.
149
3b
Part 3b
150
Ventilationperfusion ratio
Blood flow increases more rapidly with the distance from apex in the upright
lung than ventilation. Blood flow at the base is 20 times higher than blood
flow at apex, while basal ventilation is only three times higher than apical ventilation. As a consequence the ventilation/perfusion ratio (V/Q) diminishes
from apex to base, to about 15% of the apical value at base. In other words, the
apex is relatively over ventilated while the base is relatively over perfused.
This is the basis of the physiological ventilation/perfusion inequality.
Figure 112 illustrates this schematically. Ventilation of unperfused alveoli
at the apex contributes to the physiological dead space effect, as mentioned above: inhaled fresh gas is wasted and is unavailable for gas exchange.
Perfusion of unventilated alveoli at the base means that blood flowing through
this region will receive no oxygen, and will contribute to the physiological
shunt. In the normal lung the base receives most of the pulmonary blood
flow; as the blood there is less well oxygenated, the result is a venous admixture, normally up to 5%. Well ventilated but under perfused apical regions
cannot compensate for the shunt by increasing oxygenation of the blood
that they receive, as the blood is already fully saturated with oxygen; this is due
to the shape of oxygen dissociation curve. Carbon dioxide exchange, however, is a linear function of its partial pressure; apical alveoli can compensate
to a large extent for the excess CO2 arising from the non-ventilated basal
alveoli. Thus end-tidal (alveolar) to arterial gradient for CO2 is < 1 mmHg,
or 0.13 kPa, while the end-tidal to arterial gradient for O2 is several mmHg.
Increased total ventilation of a normal lung can compensate for both, as the
basal well-perfused alveoli will then receive more ventilation.
During anaesthesia changes in dead space and shunt fraction occur: as
shown previously, closing volume is above functional residual capacity in most
anaesthetized patents. Endobronchial intubation increases the shunt fraction
substantially (one lung perfused but not ventilated). A higher inspired oxygen
concentration can only partly compensate for this. Dead space changes follow
changes in cardiac output, e.g. hypotension.
3b
e.
D
Air
e.
De
d.
e.
Blood
Ox
d.
Figure 112. Schematic drawing of ventilationperfusion relationship in the three zones

of the lung.
Air
151
3b
Part 3b
Oxygen cascade, oxygen therapy

and shunt fraction
Oxygen cascade
The oxygen cascade depicted in Figure 113 shows the decrease in oxygen
tension at each stage of oxygen transport from the ambient air into the tissue.
Notice that alveolar oxygen tension, PA O2 = 13.3 kPa, is substantially
lower than oxygen tension in ambient air. This is because alveolar gas has
a different composition from ambient air: inhaled air is fully saturated with
water vapour on inspiration and it then mixes in the alveoli with gas that
already underwent gas exchange.
Alveolar oxygen tension can be calculated from the Nunn equation (see
the chapter on Gas R line for more details).
The arterialalveolar tension gradient will depend on the diffusing
capacity of the alveolar membrane, cardiac output, rate of binding of oxygen
with haemoglobin (these are the factors affecting gas transfer) and on shunt
fraction. In the normal lung, and with the other conditions being normal, the
most important factor is the physiological shunt.
Oxygen tension in the capillary is further reduced as a result of the uptake
by the tissues. If oxygen extraction is poor (for instance, in low cardiac output
states when oxygen delivery does not match oxygen demand, or in mitochondrial poisoning), capillary and venous oxygen tension is relatively high while
the tissues suffer hypoxia.
Each step of oxygen transport down the cascade can be altered by abnormal conditions or pathological changes. To achieve the target mitochondrial
oxygen tension of 0.131.3 kPa, inspired oxygen tension must be raised to
increase the pressure gradient if such abnormal conditions exist. In the case
of mitochondrial poisoning, this is very difficult as only the unaffected mitochondria utilize the oxygen supplied. The oxygen pressure gradient has to
be increased to 100%; even so, the extent of tissue hypoxia will depend on
the degree of poisoning. At the other end of oxygen cascade it is possible to
lower inspired oxygen concentration in the ambient air without significantly
affecting oxygen supply (this is the basis of expired air mouth-to-mouth resuscitation). It can be seen that the further down the oxygen cascade, the higher
the inspired oxygen tension must be to compensate for a lesion that makes
the gradient steeper.
152
3b
20
15
E
A
10
c
5
m
I = inspired (warmed humidified) air.
E = end-expiratory gas.
A = ideal alveolar gas.
a = arterial blood.
c = mean capilary blood.
m = mitochondria.
Oxygen cascade, oxygen therapy and shunt fraction
pO2 (kPa)
Figure 113. The oxygen cascade.
153
3b
Part 3b
154
Iso shunt lines

Iso shunt lines shown in Figure 114 illustrate the effect of increasing the
inspired oxygen tension on arterial oxygen tension, for various shunt fractions.
Assumptions are made about the other variables affecting oxygen delivery
(normal haemoglobin, normal oxygen extraction, normal arterial pH and CO2
tension, normal gas transfer). Inspired oxygen tension (fI O2 ) is the controlled
parameter, and arterial oxygen tension (pa O2 ) is its function. Zero shunt
is included although it only exists in theory: there is a linear relationship
between fI O2 and pa O2 . Physiological shunt of up to 5% and pathological
shunt of up to 10% are highly responsive to oxygen therapy: the slope of
linear relationship is steep with the exception of the lower end of the line, and
the line is parallel to the zero shunt line. Between an fI O2 of 0.21 and about
0.35 for 5% shunt, or 0.45 for 10% shunt, the slope is less steep (less increment
of arterial pO2 with increased fI O2 ). With increasing shunt fractions the slope
is reduced up to progressively higher inspired oxygen concentrations. A 50%
shunt is hardly affected by increasing the fI O2 at all; arterial pO2 stays low
even when breathing pure oxygen. Therapeutic efforts in this case must be
directed towards removing the cause of the shunt.
3b
Oxygen cascade, oxygen therapy and shunt fraction
paO2
Pa
v
fi O2
Figure 114. The effect of shunt function and inspired oxygen fraction (fi O2 ) on arterial
oxygen (pa O2 ). Assumptions as above are made.
155
3b
Part 3b
Gas R line, solution of the

ventilation/perfusion model
The gas R line is the O2 /CO2 diagram shown in Figure 115. It depicts the
relationship between lung volumes and the partial pressures of the respiratory gases. It shows how the anatomical and physiological dead space
influence the partial pressure of CO2 , and how the magnitude of alveolar
ventilation and partial pressure of inspired O2 affect alveolar partial pressure
of O2 .
The volumes in the diagram are shown on the R line:
VDalv = alveolar dead space (points AE),
E),
VDanat = anatomical dead space (points E
`
VDphys = physiological dead space (points AE),
VA = ideal alveolar ventilation (points EI),

VE = ventilatory volume measured during expiration (points AI).
Point A denotes alveolar; point I denotes inspired.
The partial pressures are in the following gases:
pa = arterial gas
pE = end-tidal gas
pE = mixed expired gas
pi = inspired gas.
An assumption is made that
p aCO2 = p ACO2 (arterial p CO2 = alveolar p CO2 ).
Partial pressures of CO2 are plotted on the vertical axis, and partial pressures
of O2 on the horizontal axis. The distance between various points on the R
line corresponds to the lung volumes shown.
The basis of the graph is the fact that gas exchange for O2 and CO2 occurs
in opposite directions: inspired gas contains a higher concentration of O2
and no CO2 ; after gas exchange, the concentration of O2 in the expired gas
is lower than that in inspired gas while the concentration of CO2 increased
from zero to the end-tidal value.
The gas equations have been derived from this plot. The coordinates of
the various points on the R line, the axes, and the R line itself form rightangled triangles; according to the laws of geometry the ratios of any pair of
similar sides are equal. Therefore
VDphys /VE = AE/AI

= p aCO2 p ECO2 / p a CO2 (the Bohr equation),
156
or, in other words, the ratio of dead space ventilation to total ventilation is the
same as the ratio of the difference between partial pressures of CO2 in arterial
3b
Figure 115. Gas R line.
V
p
V
p
Gas R line, solution of the ventilation/perfusion model
157
3b
blood and the expired gas to arterial p CO2 . Another variant of this equation
is:
Part 3b
VDalv /VE = AE/AI

= p aCO2 p ECO
/ p aCO2 ,
2
or, in other words, the ratio of alveolar dead space to total ventilation is the
same as the ratio of the arterial end-tidal pressure gradient for CO2 to arterial
p CO2 . The nominator in the equation, the arterial to end-tidal pressure
gradient for CO2 , is easily measured and serves to assess the magnitude of
dead space.
The ideal alveolar oxygen tension is calculated from the Nunn equation.
Notice that point X is not zero as alveolar pO2 is not zero; therefore pAO2 has
to be subtracted from the values on the horizontal axis:
IX/IX = AI/EI,
therefore
p iO2 p AO2 / p iO2 p EO
2 = p aCO2 / p EO
2.
All gas tensions, except pAO2 , can be measured. Then
),
p AO2 = p iO2 paCO2 ( p iO2 = p EO
2 / p ECO
2
where p iO2 and p EO
2 are measured directly by the paramagnetic oxygen analyser (or any of the other oxygen analysis methods), p aCO2 is measured directly
from arterial blood sample and p ECO
is measured in expired gas collected into
2
a Douglas bag, with an infrared CO2 analyser.
The Nunn equation is used for the calculation of the alveolar p O2 , which
is then used in the calculation of the shunt fraction, not derived here.
Qs /Qt = c c c a /c c c v ,
where c is oxygen content, and the subscripts are c for end-pulmonary capillary, a for arterial and v more mixed venous blood. Arterial and mixed venous
oxygen content is measured directly by the Van Slyke method or any of the
other oxygen analysis methods; mixed venous blood is sampled from the pulmonary artery catheter. End-pulmonary capillary blood cannot be sampled
but oxygen tension in the pulmonary capillary is assumed to be in equilibrium with alveolar oxygen tension, derived from the Nunn equation. Oxygen content in the pulmonary capillary is then calculated from the saturation,
assuming a normal oxyhaemoglobin dissociation curve.
158
3b
Gas R line, solution of the ventilation/perfusion model
159
3b
Part 3b
160
Ventilatory response to oxygen

Hypoxia is a powerful respiratory stimulus. In figure 116, pulmonary ventilation in litres per minute is plotted against arterial oxygen tension in kilopascals
(thick line). The horizontal axis is truncated above 17 kPa and shows extrapolation at abnormally high levels of hyperoxia.
It can be seen that the curve has a shape similar to a rectangular hyperbola,
whereby the minute ventilation stays almost the same above 12 kPa of arterial
oxygen tension but starts rapidly increasing when arterial pO2 is reduced
below this level. From the chapter on mathematical concepts in Part 1 we
know that a rectangular hyperbola describes a reciprocal relationship, in
this case between arterial pO2 and minute ventilation. Hypoxia below
7 kPa produces a steep rise in minute ventilation, theoretically ad infinitum.
Notice that the vertical asymptote is at 4 kPa, as oxygen tension below this
level becomes incompatible with life. The horizontal asymptote is at a level
slightly below normal minute ventilation (although in this particular case the
normal level is somewhat high at almost 10 litres per minute).
The thin line in the graph represents the relationship between arterial
oxygen saturation and minute ventilation. This is an approximately linear relationship with a negative gradient. Notice that the horizontal axis for
arterial oxygen saturation is only shown for values between 70% and 90%,
as the relationship between these two parameters is roughly linear within
these limits (see oxyhaemoglobin dissociation curve, page 131). Ventilatory
response to oxygen below arterial oxygen saturation of 70% has not been
tested in humans.
There is a high individual variability in the hypoxic response, which
explains poor adaptation of some individuals to high altitude.
The hypoxic response is augmented by hypercapnia: small increases of
arterial carbon dioxide move the threshold of hypoxic response (curve shifted
to the right) and increase the slope of the curve.
Anaesthetic drugs reduce the hypoxic ventilatory drive, even at sub anaesthetic doses. This is only one of the reasons why anaesthetised patients and
patients recovering from an anaesthetic are given an oxygen enriched gas mixture, usually above 30%. The hypoxic ventilatory drive is also reduced during
natural sleep, explaining reduced oxygen saturations during sleep. In patients
with obstructive sleep apnoea, this response is so diminished that the patients
intermittently become apnoeic, and only profound hypoxia as a result of the
apnoea provides a sufficient stimulus for the respiration to resume.
3b
Ventilatory response to oxygen
M V(l min1)
30
20
10
8
70
12
80
16 paO2 (kPa)
90 SaO2 (%)
Figure 116. Ventilatory response to oxygen.
161
3b
Part 3b
162
Ventilatory response to carbon

dioxide
Carbon dioxide, up to an arterial tension of about 15 kPa in normal subjects, is a respiratory stimulant. In figure 117, pulmonary minute volume as
percentage of normal is plotted against arterial pCO2 . Pulmonary minute volume increases in an approximately linear fashion up to 15 kPa. The response
decreases sharply above 15 kPa and the slope has a reverse gradient after
20 kPa.
Beyond 20 kPa, hypercapnia becomes irreversible because of the vicious
spiral of reduced minute ventilation leading to a further rise in Pa CO2 , leading
to a further reduction in minute volume. Simply removing the original cause
of hypercapnia, for instance rebreathing from an anaesthetic circuit, will not
reverse this course it is necessary to start artificial ventilation to remove the
excess carbon dioxide.
Only the straight portion of the ascending limb has been determined in
humans. Minute ventilation below normocapnia is reduced and the line is
extrapolated to zero at both ends.
Normal responsiveness of the respiratory centre to carbon dioxide is on
average 15 litre. min1 kPa1 (2 litre min1 per mmHg).
In patients with chronic obstructive airways disease, the curve is shifted
to the right, so that minute ventilation is low at carbon dioxide levels which
would be stimulant in normal subjects. It is useful to know the normal
level for a patient with chronic lung disease before commencing artificial
ventilation, in order not to hyperventilate when discontinuing the artificial
ventilation. Because of their decreased sensitivity to carbon dioxide, hypoxia
becomes the predominant stimulus in patients with type II respiratory failure,
and oxygen therapy needs to be controlled in this situation.
Conversely, ventilatory response to carbon dioxide is increased in pregnant
women and the curve is shifted to the left.
Carbon dioxide was used as a respiratory stimulant during induction of
anaesthesia when the intended mode of ventilation was spontaneous respiration. It was added in a concentration of 5% to the inspired gas mixture; the
advantage of this method was that manual hyperventilation could be used
to ensure fast uptake of the inhalational agent, without causing hypocapnia and respiratory depression. Unfortunately, accidents with carbon dioxide
flowmeter being left open at full flow resulted in carbon dioxide narcosis
and therefore this practice is no longer encouraged. Carbon dioxide cylinders
must not be left on an anaesthetic machine, and modem machines do not
have the CO2 yoke. Circle systems allow taking the absorber out of circuit,
and this also produces hypercapnia and respiratory stimulation.
3b
Ventilatory response to carbon dioxide
MV
(% of normal)
500
400
300
200
100
10
20
30
40
paCO2 (kPa)
Figure 117. Approximate shape of carbon dioxide response curve.
163
Part 4
Pharmacology
165
Drug elimination
Part 4
Drug elimination, like other natural processes of elimination, is in many

instances an exponential decay process (see the chapter on exponentials for
more details and graphs). The elimination constant, k, of this process is
directly proportional to clearance of the drugs and indirectly proportional to
the volume of distribution. The time constant is inverted elimination rate
constant. Therefore
k = Cl/VD
= 1/k = VD /Cl.
The concentration of the drug in the body at any time can then be calculated
from the basic exponential decay equation.
C = Co et / = Co ekt = Co .et
Cl/V
First- and zero-order kinetics

From the properties of the exponential decay curve it is known that the rate
of decay at any time is proportional to the ordinate at that point; applied to
drug elimination this means that rate of elimination is dependent on the
drug concentration at the chosen time point. Mathematically denoted this
would be:
dc /dt = kc 1 ,
where k = elimination rate constant, which is characteristic for each drug.
Figure 118 of the drug illustrates the first-order kinetics formula: the higher
the concentration, the higher the rate of change in concentration. The slope
of the relationship between drug concentration and its rate of elimination is
the value of the elimination rate constant. Remifentanil, which is cleared fast,
has a steep slope (high elimination rate constant). Alfentanil, which is cleared
more slowly, has a gentle slope.
By contrast, the relationship of alcohol elimination to alcohol concentration has two phases, the first phase being zero-order kinetics. Note that in
the equation above, concentration is raised (mathematically quite unnecessarily) to the power of 1; it would have sufficed just to write c. The power
of 1 relates to the term first-order kinetics. In zero-order kinetics, when the
physiological processes of drug elimination are overwhelmed (enzymes saturated), the rate of drug elimination is independent of drug concentration
and depends on the elimination rate constant only. In Figure 119 the
rate of alcohol elimination is constant during the initial phase, conforming to
the formula
dc /dt = k c o = k.
166
When blood alcohol is < 0.05 mg ml1 , first-order kinetics apply.
Al
Drug elimination
dc dt
Re
c
Figure 118. First-order kinetics of remifentanil and alfentanil.
Zero
First
Time
dc dt
c
Figure 119. Zero-order and first-order elimination of blood alcohol from 1 mg1 to
complete elimination.
167
4
Part 4
First-order kinetics are also called linear pharmacokinetics, because the

rate of drug elimination has a linear relationship to the drug concentration.
However, drug concentration as a function of time falls exponentially! The
exponential decay curve can be transformed into a straight line when drug
concentration is plotted on a logarithmic scale (see the chapter Exponentials
1). Drug elimination half-life can then be easily read off this plot.
Compartmental models
So far, the pharmacokinetic model assumed that the drug, after intravenous
injection, was rapidly and uniformly distributed throughout the body, and
then rapidly excreted. This is called the one-compartment model, i.e. the
body behaves as one pharmacokinetic compartment. An example that can be
applied to this situation is an injection of suxamethonium.
For most drugs, however, their tissue concentration depends not only on
the plasma concentration but also on their lipid solubility and protein binding.
Tissue concentration in richly perfused organs approaches that in plasma;
these form a small central compartment. The rest of the body then form a
larger peripheral compartment, which slowly achieves equilibrium with
the central compartment, depending on drug transfer constants between the
compartments. This two-compartment model approximates reality better
than the single-compartment model.
Drug concentration in the central compartment (plasma) will then follow
a more complicated pattern than that of a single exponential decay; the drug
can leave the central compartment by either transfer (distribution) to the
peripheral compartment or by elimination from the body. Initially, when drug
concentration is high in the central compartment but low in the peripheral
compartment, the drug leaves the central compartment chiefly by distribution
to the peripheral compartment. This is the distribution or phase, with
its own transfer constant, k , and distribution half-life, t1/2.
When the peripheral compartment concentration equilibrated with that
in the central compartment, the drug then leaves the compartment by the
process of elimination from the body. This is the final elimination phase ,
with its elimination rate constant, k , and elimination (or terminal) half-life,
t1/2 . Drug elimination can then be described by the biexponential equation:
C = C1 et + C2 et ,
where C1 = initial drug concentration at time t = 0, and C2 = theoretical
drug concentration at time t = 0 if the elimination phase had applied to the
whole process. and denote the elimination rate constants in this formula.
In Figure 120 C2 is the intercept of the extrapolated slope of the elimination
phase with the ordinate. Notice that the plot is semi-logarithmic; during the
transitional phase the line is curved as the two processes (distribution and
elimination) happen simultaneously.
168
4
Drug elimination
Time
Figure 120. Biexponential decline in plasma concentration of a drug after intravenous

injection in a two-compartment model.
169
4
Part 4
Uptake and distribution of

inhalational anaesthetic agents
This chapter deals only with uptake in high gas flow. The following formula describes the factors involved in the uptake of inhalational anaesthetic
agents:
an /V
,
fA an = fi an V
A
where f A an , f i an = alveolar and fractional inspired concentration of
the agent, respectively, Van = uptake of the agent and VA = alveolar ventilation.
In other words, alveolar concentration depends on how much is put in
( f i an ) and taken out (Van /VA ).
The rate of rise of f Aan as a function of time can also be plotted as the rate
of rise of the fraction f A / f i . This merely describes alveolar concentration as
a fraction of the inspired concentration. From the equation above, it follows
that
an /V
A f i ,
f A/ f i = 1 V
It can be seen from this equation that alveolar concentration as a fraction of
inspired concentration ( f A / f i ) will be approaching 1 (the desired aim of swift
A f i is small or negligible. The bigger the
an /V
induction) when the fraction V
A , f i ), the smaller the fraction. In other words,
factors in the denominator (V
high alveolar ventilation and high fractional-inspired concentration of the
anaesthetic increase the rate of rise of alveolar concentration. The reverse is
true of the factors that enhance the uptake of the inhalational agent namely
cardiac output, shunt fraction and blood solubility of the agent.
The rate of rise of alveolar concentration can then be plotted as shown
in Figure 121; the factors mentioned above influence it in the predicted
manner: an upward shift of the curve occurs with increased alveolar ventilation, increased inspired concentration, reduced cardiac output, reduced
shunt fraction, lower blood solubility and vice versa. The time axis on the
top diagram has no markings as this is an idealised representation; the markings would be different for each inhalational agent, depending on their blood
solubility.
170
4
Uptake and distribution of inhalational anaesthetic agents
Figure 121. Factors influencing alveolar gas or vapour concentration during uptake.
171
4
Part 4
172
The effect of blood solubility of an agent on the rate of uptake is shown

in Figure 122. Sevoflurane has the lowest blood solubility and is taken up
almost as rapidly as nitrous oxide, with the fA to fi ratio greater than 50% in
about 2 minutes. The rate of washout on discontinuation of the agent would
be similarly rapid.
Interaction occurs among the factors above: changes in alveolar ventilation
and/or cardiac output affect most those anaesthetic agents with high blood
solubility. With agents such as nitrous oxide, where the rate of rise of alveolar
concentration is fast, the effect of changes in alveolar ventilation and/or cardiac
output is negligible.
The premise that the partial pressure of the anaesthetic in the brain (and
therefore the speed of induction) closely follows the anaesthetic partial pressure in the alveoli only holds for normal cerebral perfusion. Thus the effects
of altered minute ventilation or cardiac output on the speed of inhalational
induction may be modified by the effect of these factors on cerebral vasculature. Cerebral vasoconstriction caused by hyperventilation could in theory
slow down the rate of induction (instead of increasing it). Similarly, in decompensated shock there may be a reduction in cerebral blood flow, which may
offset a theoretical rapid induction.
In practice, however, this is rarely observed: during inhalational induction,
patients usually breathe from a semi-closed rebreathing circuit. The rise in
arterial pCO2 as a result of rebreathing would mostly offset the effects on
the cerebral perfusion mentioned above during hyperventilation. As regards
decompensated shock, in this situation inhalational agents are used with great
caution, in doses that would be subanaesthetic in haemodynamically stable
patients, because of myocardial depression and the already altered state of
consciousness. Swift inhalational induction aided by high inspired concentration of the inhalation agent is not the aim here maintenance of mean
arterial pressure and cerebral perfusion pressure take precedence. The theoretical principles considered here are useful to keep in mind, but do not forget
the wider picture.
4
N2O
Sevoflurane
1.0
Isoflurane
Halothane
0.5
10
20
30
40
50
t (min)
Figure 122. Rate of rise of alveolar concentration for different anaesthetic agents.
Uptake and distribution of inhalational anaesthetic agents
fA /fi
173
Pharmacodynamic effects of drugs
Part 4
Most but not all drugs produce their effects by binding to receptor sites. When
an active drug binds to its receptor site, the receptor changes its configuration
to its active state, and this initiates an intracellular response in the effector
tissue, and eventually results in the drugs effect.
This binding is reversible, with an equilibrium that depends on the association and dissociation constants of the process.
Drug + receptor
k1
k2
drugreceptor complex effect.
Doseresponse curve
The doseresponse curve is the plot of the dose of a drug against its effect,
usually expressed as the percentage of maximum effect. Initially, the effect
rises steeply up to 50%; thereafter, progress to the maximum effect is slower,
with much higher doses required to produce the ceiling effect. The graphical
result is a rectangular hyperbola (Figure 123), conforming to the formula
y = a 1/x.
If the graph is re-plotted with log dose on the ordinate, the result is a sigmoid
shape. This has advantages: it allows a greater range of doses to be represented,
and it allows easy comparison of drug dosages on the straight part of the curve
by the means of comparing their efficacy and potency. To produce an effect,
a drug has to bind to the receptor site; its affinity for the receptor therefore
affects both its efficacy and its potency. These terms are explained below.
Affinity
Affinity is a measure of the ability of a drug to bind to the receptor site and
form a stable complex. Pure agonist drugs have a high affinity for the active
state of the receptor, whereas pure antagonists have a high affinity for its
inactive state. Thus, drugs with a high affinity for the same receptor can
have very different effects, depending on their intrinsic activity, or efficacy
(see below). Affinity may be expressed by an IC50 , which is the concentration
producing 50% inhibition of the binding of a highly selective ligand. Among
agonist drugs, there is a high correlation between their IC50 and their potency.
The relationship between affinity and efficacy is not only complicated by the
degree of agonism, as mentioned above, but also by other limiting factors that
may be involved in the process of the biological response, e.g. the presence of
other ligands, second messengers, etc. A well-known example is the degree
of acetylcholine receptor occupancy by non-depolarizing relaxants versus the
degree of neuromuscular block.
174
4
Figure 123. Drug dose and effect.
175
Efficacy (intrinsic activity)
Part 4
The maximum effect that a drug is capable of producing is its efficacy, i.e. it is
the plateau of the doseresponse curve; in other words, efficacy indicates
the limit of the doseresponse relationship on the y (effect)-axis.
Pure agonists have a high affinity for the active state of the receptor site
(e.g. morphine and the receptor), and are 100% efficacious. Antagonist
drugs (e.g. naloxone and the receptor) have no intrinsic activity (they have
a high affinity for the inactive state of the receptor site), and are not efficacious. Partial agonists (e.g. nalorphine and the receptor, -adrenergic
blockers with intrinsic sympathomimetic activity and the -adrenergic receptor) have only low affinity for the active receptor site, so that even at maximum
receptor occupancy not all receptor sites will be activated. Therefore, partial
agonists have a lower intrinsic activity or efficacy < 100%. Full and partial
agonist opioid log doseresponse curves are shown in Figure 124; the plot for
naloxone is identical with the x-axis. Buprenorphine and nalorphine have not
only lower affinity than fentanyl but also a lower efficacy than all three full
agonists they do not achieve 100% response.
Potency
The potency of a drug is a measure of the mass of the drug required to
produce a certain level of effect. Clearly it is not useful to chose 100%
effect as the doseresponse relationship is lost at that level. A reference
point on the doseresponse curve is therefore chosen to allow a comparison,
usually 50% of the maximum effect. A drug that produces the same (50%)
effect with only 1/10th of the dose (mass) of another drug is 10 times more
potent (e.g. pethidine and morphine). Relative potency of morphine versus
pethidine is therefore 10, calculated as the ratio of equipotent doses. Potency
can be illustrated as the position of the log doseresponse curve along the
x (dose)-axis; the less potent a drug, the further its doseresponse curve lies
along the x-axis (Figure 124).
Antagonism
Competitive antagonists
Competitive antagonists occupy reversibly some of the available receptor
sites; if a higher dose of active drug is given, it can still achieve its maximum
effect. Competitive agonists, therefore, in effect reduce the potencyof a
drug, but not its efficacy (the doseresponse curve shifts along the x-axis,
the shape is unchanged). Naloxone reversibly antagonizes opioid drugs; in
its presence the log doseresponse curves are shifted to the right, as shown
opposite on the example of pethidine.
176
Bu
Mo
Al
Pe
Pe
Na
Na
100
Fe
Figure 124. Opioid drugs log doseresponse curves.
177
4
Part 4
Non-competitive antagonists
Non-competitive antagonists bind irreversibly to the receptor sites; thus they
make some (or all) of the receptor sites unavailable to the active drug. Increasing the dose of the active drug does not achieve a full effect. Non-competitive
antagonists thus reduce the efficacy of a drug, while its potency, at least
in the lower part of the curve, is relatively unchanged. The picture produced is the same as that of partial agonist drugs. In Figure 125, compare the
log doseresponse curve of norepinephrine under normal conditions, in the
presence of a competitive blocker (phentolamine) and in non-competitive
antagonism by increasing doses of phenoxybenzamine.
To quantify the degree of partial agonism by various agonists, pA2 has been
devised. This is the negative logarithm of the concentration of the antagonist
necessary to produce a particular level of response. Thus, pure agonists at
one type of opioid receptor, such as (morphine, -endorphine) have the
same pA2 for naloxone (7), while pA2 for leu-enkephalin is about 6. In other
words, ten times more naloxone is needed to antagonise the effects of leuenkephalin than to antagonize -endorphine. (Remember that a difference of
1 unit on the logarithmic scale is ten times the difference on the linear scale.)
This suggests that these two agents act on different receptor populations,
an observation which was postulated even before the receptor (at which
leu-enkephalin is an agonist) was isolated.
Therapeutic index
The therapeutic index is a measure of the safety of a drug in clinical use.
Reference points on two integrated probability curves (otherwise known as
centile curves) need be compared; one for a desired clinical effect, the other
for a toxic effect. The centile curve resembles the log doseresponse curve but
is different: the y-axis represents the percentage of population that responds at
the given dose; the x-axis scale is linear, as shown in Figure 126. The reference
point is a dose that produces the chosen effect in 50% of the individuals. ED50
is the median effective dose (which produces the clinical effect in 50% of the
population, whether human or experimental animal), and LD50 is the median
lethal dose (which produces death in 50% of experimental animals). The ratio
LD50 /ED50 is the therapeutic index. Drugs that are relatively non-toxic have
a high therapeutic index (LD50 much higher than ED50 ), while the reverse
is true of the more toxic drugs, such as digoxin (LD50 closer to ED50 ). The
slope of the doseresponse curve is usually an indicator of the drugs toxicity
the lower the therapeutic index, the more steep the rise of the slope. In the
example opposite, LD50 is about ten times higher than ED50 ; the therapeutic
index is sufficiently large.
178
4
Ph
No
Ph
Log dose
of norepinephrine
Figure 125. Competitive and non-competitive antagonism of norepinephrine effect on
log doseresponse curve.
po
De
De
Do
Figure 126. ED50 , LD50 .
179
4
Part 4
180
Minimum alveolar concentration

and lipid solubility
Minimum alveolar concentration (MAC) is defined as such a concentration
of a volatile anaesthetic agent that produces immobility in 50% of experimental animals subjected to a standardized noxious stimulus. It resembles
ED50 for intravenous drugs, and serves the same purpose: comparison of the
potencies of various drugs. Potency of inhalational anaesthetic agents is usually directly proportional to their lipid solubility, although many exceptions
exist. MAC is indirectly related to lipid solubility (the more potent a drug, the
lower its MAC). Figure 127 illustrates this relationship. Both axes have logarithmic scales; the oil/gas partition coefficient is plotted on the x-axis against
MAC in atmospheres on the y-axis. It can be seen that the points representing the position of individual agents lie close to a straight line. The product
of MAC and solubility on this graph is about 2. The straight line seen on
the graph is the result of double logarithmic transformation of a rectangular
hyperbola, the shape that would depict the reciprocal relationship between
MAC and lipid solubility on a linear scale. The logarithmic scales have the
advantage of accommodating a wide range of numbers in a relatively small
space.
MAC of nitrous oxide is above atmospheric, and its lipid solubility is very
low. MACs for the other agents are given in oxygen for the upper line, and 66%
nitrous oxide on the lower line. MAC is reduced by 60% by this concentration
of nitrous oxide, as the effects are additive. This reduction is represented by
the vertical arrows joining the corresponding points for each agent.
Notice that isoflurane and enflurane are one of the exceptions to the reciprocal relationship rule: isoflurane has a lower MAC (is more potent) than
enflurane (1.15 v.s 1.68%), yet it has a slightly lower lipid solubility (91 v.s 96).
The ascending order of the names from halothane at the lower end to nitrous
oxide at the upper end applies to the minimum alveolar concentration but
not to lipid solubility in the case of isoflurane and enflurane.
MAC is usually converted to percentage of atmospheric pressure, i.e. partial
pressure. For instance, the MAC of halothane is 0.76% in 100% oxygen,
or < 1%. The corresponding point for halothane is indeed slightly below
0.01 atm., or 1%.
4
Minimum alveolar concentration and lipid solubility
Figure 127. MAC and lipid solubility of volatile agents in 100% oxygen and in 66%
nitrous oxide.
181
4
Part 4
Receptor types, molecular action

of anaesthetics
Receptors are molecules in the cell membrane whose function is to cause
a response in target tissue after the binding of a specific ligand (a molecule
for which the receptor has high affinity). The binding usually results in conformational change, and this triggers the response. Three types of receptors,
the ligand gated ion channel, the G-protein coupled receptor system, and
the transmembrane enzymatic receptor, will be described here. It should be
noted that the voltage sensitive ion channel is not strictly speaking a receptor
as it does not bind any molecules.
Multisubunit ligand gated ion channel
Examples of this type are the nicotinic, GABA1 and glycine receptors. The
macromolecule is a conical structure with the thick end protruding into the
extracellular space. It is composed of four or more usually five subunits (it is a
tetra- or pentamer). These subunits are called by the Greek letters , , , ,
and ; a variant called the type exists in the GABA1 receptor. Their structure
is very similar to each other. Each subunit has four transmembrane domains
called M1 to M4, where the protein crosses from extra- to intracellular or vice
versa. The domains are shown opposite in Figure 128 as the grey rectangles.
The M2 domain is involved in ligand (anaesthetic) binding. It can be seen
that both the amino- and the carboxyl residue are extracellular; the amino
residue is long, which forms the extracellular bulk mentioned above. The
subunits are arranged in the tetra or pentamer as shown in Figure 128, a
cross section of the whole receptor structure at the level of the cell membrane.
The receptor molecule is like a channel going through the membrane, with
the subunits on the periphery of the channel. The transmembrane domains
in each subunit are shown and it can be seen that they, too, are organised
symmetrically inside the subunit.
G-protein coupled receptor
Examples of this type are the adrenergic receptor (adenyl cyclase), the phospholipase, and some muscarinic receptors. This receptor has seven transmembrane domains, with short amino and carboxyl residues on opposite sides of
the cell membrane, as shown in Figure 129. In the vicinity of this peptide is the
G-protein, the effector which converts guanosine triphosphate into guanosine diphosphate after receptor stimulation. Further aspects of this receptor
are beyond the scope of this book.
182
Extracellular space
Cell
Intracellular space
(b)
Figure 128. (a) Multisubunit ligand-gated ion channel.

(b) Cross-section of a tetramer (subunits ).
Receptor types, molecular action of anaesthetics
(a)
Extracellular space
Cell
Intracellular space
Figure 129. G-protein coupled receptor.
183
4
Part 4
184
The transmembrane enzyme receptor

Examples of this type are the antinuclear factor, growth factor, neurotropic
factor and others. These receptors are protein kinases mostly. As shown in
Figure 130, the ligand binds to the extracellular receptor site; this produces a
conformational change at the intracellular site with the result that the intracellular process is catalysed at a much faster rate.
4
Extracellular space
Cell
Catalysis
Figure 130. Receptor as enzyme.
Intracellular space
Receptor types, molecular action of anaesthetics
Ligand
185
4
Part 4
186
Context-sensitive half-time
This concept was developed to describe drug elimination after an intravenous
infusion, designed to produce a steady plasma concentration of a drug, is discontinued. Drugs which depend on the liver or kidneys for their elimination
demonstrate this phenomenon. In practice, this encompasses all hypnotic
drugs and opioids used in total intravenous anaesthesia, with the exception of
remifentanil, which is eliminated from the body by tissue and plasma esterases.
Half-time is the time required for the drug concentration to halve. Recovery
cannot be expected before the plasma concentration decreases below 50% or
the clinically effective concentration.
The half-time is context sensitive because drug elimination depends on
the duration of infusion. To understand it, consider the three-compartment
pharmacological model, consisting of the central compartment-C1 -(the
intravascular space) and two peripheral compartments, one which equilibrates with the central compartment fairly rapidly-C2 -(well perfused tissues
such as the brain) and the other which equilibrates more slowly because of
poorer perfusion-C3 -(fat). The rich or poor perfusion is illustrated in the
figures opposite by a wide or narrow band connecting the compartments.
After a short duration infusion, the peripheral compartments have not
yet reached equilibrium with the central compartment (see Figure 131).
Therefore, after the infusion is discontinued, plasma concentration declines
fairly rapidly as the drug continues to redistribute into the peripheral compartments, until equilibrium is reached at a lower plasma concentration. Because
redistribution is faster that elimination, context-sensitive half-time in this
situation is closer to the redistribution half-life.
After a prolonged infusion, the peripheral compartments reached equilibrium with the central compartment and drug concentration is the same in
all three (steady state), as shown in Figure 132. On discontinuing the infusion,
plasma concentration can only decline by drug elimination from the central
compartment via liver or kidneys. As the drug is being eliminated, more drug
is moved from the peripheral compartments into the central compartment,
adding to the drug load to be eliminated. Hence the context-sensitive halftime in this situation is close to elimination half-life.
4
C3
C1
K10
Elimination
C2
C1 central compartment
C2,C3 peripheral compartments
Figure 131. Drug levels in three compartments after a short infusion.
C2
C3
C1
K10
Elimination
Figure 132. Drug levels in three compartments after a prolonged infusion.
187
4
Part 4
188
Figure 133 shows the observed decline of plasma alfentanil concentration

after an infusion of 10-minute and 3-hour duration. Although the initial
plasma concentration is the same (100 ng/ml), the decline clearly varies for
different duration of the infusion. The half-time after the 3-hour infusion is
four times greater.
Context-sensitive half-time for various anaesthetic drugs is plotted in
Figure 134 (short infusion) and 135 (prolonged infusion). Remifentanil,
which has a half-time of 3 minutes, independent of duration of infusion,
is included for comparison. For other drugs, context-sensitive half-time
increases more or less in the form of a sigmoid curve, until after several
hours a steady state is reached. The point at which the context-sensitive halftime reaches a plateau depends on the volume of distribution of the drug in all
three compartments: the larger the VD , the longer it takes to reach steady state.
Notice that the line for fentanyl intersects the alfentanil and thiopentone line.
This, in practice, means that after a short infusion of up to 2 hours, fentanyl
will be cleared more rapidly than alfentanil, but the reverse is true when the
duration of infusion is longer than 2 hours. This is because of the difference between the redistribution and elimination half-life of the two drugs.
Because of its high lipid solubility, context-sensitive half-time for fentanyl
has not reached a plateau even after 12 hours, whereas the other drugs have
reached, or are near, a plateau after 3 hours. Therefore, after a prolonged infusion, fentanyl ends up with the slowest elimination of the drugs illustrated in
the drawing.
In practice, context-sensitive half-time may be expressed in the form of
a fraction, with the redistribution half-time (concerning a short infusion
of 10 minutes) in the nominator and elimination half-time (concerning
a prolonged infusion of 3 hours) in the denominator. Thus remifentanil
has a context sensitive half-time of 3/3, alfentanil 10/40 and fentanyl 3/70.
Pharmacokinetic data for fentanyl are inconsistent and other values may be
found in different sources.
Devices for target-controlled infusion use the model of context-sensitive
half-time to predict the decline in plasma concentration after the infusion is
stopped. Therefore, it is possible to calculate the expected time to waking up
effectively to a subhypnotic plasma concentration. Propofol, with a context
sensitive half-time of 5/9, is a suitable drug for total intravenous anaesthesia.
Even after a 12-hour infusion the half-time is 18 minutes. Thiopentone with
a context sensitive half-time of 4/85 is not suitable for continuous infusion
where rapid wakefulness is desired after stopping the sedation. Thiopentone
may still be used in specific indications, for instance status epilepticus, for its
pharmacodynamic properties.
100
10
10
20
30
40
t (min)
Figure 133. Plasma alfentanil concentration after short and long infusion.
CSHT
(min) 50
3-hour infusion
10-min infusion
C
(mg/ml)
50
Thiopentone
Alfentanil
Fentanyl
Propofol
Remifentanil
10
10
20 30 40 50
Duration of infusion (min)

Figure 134. Context-sensitive half-time after short infusion.
Fentanyl
CSHT
(min)
200
Thiopentone
100
Alfentanil
Propofol
Remifentanil
1
2
3
4
5
6
Duration of infusion (hrs)
Figure 135. Context-sensitive half-time after prolonged infusion.
189
4
Part 4
190
Further Reading
Kenny, G., Davis, P. D. Basic Physics and Measurement in Anaesthesia. Butterworth Heinemann,
2003.
Miller, R. D. Anaesthesia. Edinburgh: Churchill Livingstone, 2000.
Peck, T. E., Williams, M. Pharmacology for Anaesthesia and Intensive Care. Greenwich Medical
Media, 2000.
Mepleson, W. W. Fifty years after reflections on The elimination of rebreathing in various
semi-closed anaesthetic systems. Br. J. Anaesth. 2004; 93: 31921.
Pinnock, C., Lim, T., Smith, T. Fundamentals of Anaesthesia. Greenwich Medical Media, 2002.
Prys-Roberts, C. The Circulation in Anaesthesia. Oxford: Blackwell, 1980.
Scurr, C., Feldman, S. Scientific Foundations of Anaesthesia. Oxford: Butterworth-Heinemann.
www.frca.co.uk.
INDEX
Page references in italic indicate figures.
abscissa (x-axis) 26
absorption system see circle system
acid, hydrogen ion concentration 92
acidbase balance 92, 93, 94, 95
acidosis 92
adiabatic compression and decompression 6,
7
adrenergic receptor (adenyl cyclase) 184
age effects
airway resistance 144, 145
airways closing volume 142, 143
thoracic compliance 140, 141
agonist drugs 176, 178, 179, 180
air, saturation with water vapour 16, 17
airway diameter, and airway resistance 144,
145
airway resistance (non-elastic), factors
affecting 144, 145
airways closure (closing volume) 142, 143
alcohol, rate of elimination 168, 169
alfentanil 179
context-sensitive half-time 190, 191
effects of duration of infusion 190, 191
elimination rate constant 168, 169
alkali, hydrogen ion concentration 92
alkalosis 92
alveolar oxygen tension 154, 155
alveolar partial pressure of O2 158, 159, 160
alveolar pressure, calculation 64
alveolar size, changes from lung apex to base
148, 149
alveolar ventilation and its distribution 148,
149
amplitude (sine waves) 28
amplitude distortion 68, 70
anaesthesia
changes in lung dead space 152
effects of IPPV 114
effects on airway resistance 144
effects on airways closing volume 142, 143
effects on hypoxic ventilatory response
162
effects on left ventricular compliance 118
effects on myocardial performance 129,
131
effects on thoracic compliance 140
estimation of myocardial oxygen demand

129, 131
use of capnography 90
see also breathing systems
anaesthetic agents
lipid solubility 182, 183
minimum alveolar concentration (MAC)
182, 183
potency 182, 183
rate of rise of alveolar concentration 172,
173
speed of induction 174
uptake 54, 55, 56, 57, 172, 173
use with decompensated shock 174
anaesthetic machine
circle system 54, 55, 56, 57
pressure reducing valves 20, 21
anaesthetic receptor types
G-protein coupled receptor 184, 185
multisubunit ligand-gated ion channel 184,
185
transmembrane enzyme receptor 186, 187
anaesthetic vaporizers 16, 54, 56, 57
antagonist drugs 176, 178, 179, 180, 181
antinuclear factor 186
aortic pressure waveform, harmonics 28
ARDS, effects of IPPV 114
area under the curve see AUC
arterialalveolar tension gradient 154
arterial pressure waveform 112, 113
arteriovenous pressure gradient in the lung
150, 151
atelectasis
airway resistance 144
effects on lung compliance 140
AUC (area under the curve) 98, 99, 1001,
101
autoregulation of blood flow
cerebral blood flow, 124, 125, 126, 127
perfusion of the heart 129, 130
responses to blood loss 122, 123
Avogadros hypothesis 2, 3
Avogadros number 2
bacteria, exponential growth 30, 31
Bain system 60, 61
191
Index
192
barotrauma 52, 64
baseline drift 68, 69, 70
bimodal distribution 42, 43
blood donation, compensatory responses
122
blood flow
in the lung 150, 151
measurement 96, 97, 98, 99
blood flow autoregulation
cerebral blood flow, 124, 125, 126, 127
perfusion of the heart 129, 130
responses to blood loss 122, 123
blood gas analysers 80, 81, 82, 83, 84
blood loss 122, 123
blood oxygen saturation measurement 76, 77,
78, 79
blood pressure
and blood volume 122, 123
respiratory swing 112, 113
blood pressure measurement,
electromanometer 72, 73, 74, 75
blood viscosity, and cerebral blood flow 126
blood volume
and blood pressure 122, 123
and effects of IPPV 114
body size, and airway resistance 144
Bohr equation 158, 159, 160
Boyles law 2, 3, 26
breathing
loss of water and heat 16, 17
turbulent flow 12
breathing systems
circle system 54, 55, 56, 57
Mapleson A (Magill) system 58, 59
T-pieces 60, 61
bronchial mucosa thickness, and airway
resistance 144
bronchial muscle tone, and airway resistance
144
buprenorphine 178, 179
calibration of instruments 70
capnography 86, 87, 88, 89, 90, 91
response time 86
trend analysis 90, 91
types of capnograph 86, 87
uses 90
waveform analysis 86, 87, 88, 89
carbon dioxide (CO2 )
end-tidal concentration 26
former use in anaesthesia 164
infrared light absorbance 86
measurement (capnography) 86, 87, 88, 89,
90, 91
ventilatory response 164, 165
see also CO2 partial pressure
cardiac cycle
end-systolic pressurevolume relationship
(ESPVR) 120, 121
intravascular pressure waveforms 110, 111,
112, 113
left ventricle and systemic circulation 118
left ventricular compliance and elastance
118, 119
left ventricular volume and pressure
relationship 118, 119, 120, 121
pressurevolume loops 120, 121
pressurevolume relationships 118, 119,
120, 121
right ventricle and pulmonary circulation
118
cardiac output
heart rate 116
improving in critically ill patients 116, 117
measurement by thermal dilution method
98, 99, 1001, 101
stroke volume 116, 117
sympathetic nervous system control 116
treatment for pump failure 116
treatment for volume loss 116
cardiac performance, pressurevolume loops
120, 121
cardiac power, ways of expressing 24, 25
cardiac stroke volume
afterload 116, 117
contractility 116
FrankStarling relationship 116, 117
preload (venous return) 116, 117
relationship with LVEDP 116, 117
cardiovascular effects of IPPV 112, 113, 114,
115
central venous pressure trace 110, 111
cerebral blood flow
and partial pressure of CO2 126, 127
autoregulation 124, 125, 126, 127
cerebrovascular resistance 124, 126, 127
Cushings reflex (reflex bradycardia) 124
intracranial pressure and cerebral elastance
124, 125
mean arterial pressure 124, 125
normal range 124, 125
perfusion pressure 124, 125
responses to blood loss 122
cerebral elastance 124, 125
Charles law 2, 3
chest
airways resistance 102, 103
mechanical properties 102, 103
static compliance 102
chest wall
compliance 138, 139, 140, 141
elastic properties 138, 139, 140, 141
damping 72, 74, 75

data presentation
correlation coefficient 46, 47
frequency histogram 44, 45
histogram 44, 45
indications of trends 46, 47
line drawing 46, 47
pie diagram 44, 45
regression line (line of best fit) 46, 47
scatter diagram 46, 47
dead space
influence on CO2 partial pressure 158,
159, 160
measurement of 158, 159, 160
dead space gas 148

decompensated shock, use of inhalational
anaesthetic agents 174
derivative (rate of change), definition 98, 99
descriptive statistics see statistics
desflurane 16
2,3-diphosphoglycerate (2,3-DPG) 134,
135
doseresponse curves 176, 177, 178, 179, 180,
181
drift (measurement) 68, 69, 70
drug elimination
and duration of infusion 188, 189, 190,
191
compartmental models 170, 171
context-sensitive half-time 188, 189, 190,
191
elimination rate constant (k) 168
exponential decay curve 30, 31
first-order kinetics 168, 169, 170
half-life 30, 31, 38, 188, 189, 190, 191
three-compartment model 188, 189, 190,
191
zero-order kinetics 168, 169
see also pharmacokinetics
drugs
binding to receptor sites 176
dose and effect 26
redistribution half-life 188, 189, 190, 191
safety 180, 181
therapeutic index 180, 181
toxicity 180, 181
see also pharmacodynamics;
pharmacokinetics
dynamic accuracy of measurement 68, 70, 71
ED50 180, 181, 182
elastance 118, 119
electrocardiogram, harmonics 28
electromagnetic flowmeter 96
electromanometer 72, 73, 74, 75
damping 72, 74, 75
frequency response (natural resonant
frequency) 72
elimination rate constant (k) 30, 36, 168
emphysema
airway resistance 144
effects on lung compliance 140
lung volumes measurement 106, 107
endobronchial intubation, effects on shunt
fraction 152
-endorphine 180
endotracheal intubation, use of capnography
90
energy, forms of 22
Index
chronic lung disease, ventilatory response to

carbon dioxide 164, 165
chronic obstructive airways disease
closing volume 148
dynamic compliance 146, 147
ventilatory response to carbon dioxide 164,
165
circle breathing system 54, 55, 56, 57
Clark electrode method (oxygen content
measurement) 82, 83, 84
closing volume (airways closure) 142, 143
functional residual capacity 148
CO2 partial pressure (pCO2 )
and cerebrovascular resistance 126, 127
and O2 affinity for haemoglobin 134, 135,
136
influence of dead space 158, 159, 160
see also carbon dioxide
compartmental pharmacokinetic models 170,
171
compliance 118, 119
chest wall 138, 139, 140, 141
factors affecting 140, 141
lungs 138, 139, 140, 141
total thoracic 138, 139, 140, 141
concentration effect (circle system) 54
context-sensitive half-time 188, 189, 190,
191
coronary arterial pressure gradient 128, 129
coronary artery flow and arterial pressure 128,
129
coronary autoregulation 129, 130
coronary blood flow, effects of anaesthetic
agents 129, 131
coronary circulation 128, 129, 130, 131
cricoid pressure, amount of force to apply
18
critical damping 70, 71 see also damping
critical temperature of gases 4, 5
Cushings reflex (reflex bradycardia) 124
cycles (sine waves) 28
193
Index
enflurane
MAC 182, 183
uptake 56, 57
ESPVR (end-systolic pressurevolume
relationship) 120, 121
exponential decay curve 30, 31, 36, 37, 38, 39
exponential decay process
half-life 30, 31, 38
time constant ( ) 36, 37, 38
time to finish 38
exponential growth curve 30, 31, 32, 33, 34
fentanyl 179, 190, 191
fetal haemoglobin, affinity for O2 136
first-order kinetics 168, 169, 170
flow 8
laminar 8, 9, 10, 11
turbulent 12, 13
flow measurement
gases 96, 97
liquids 96, 97, 98, 99
flow velocity
factors affecting (laminar flow) 8, 10
HagenPoiseuille equation 10
fluid deficit, and respiratory swing 112,
113
force 18, 19
Fourier analysis 28
FrankStarling relationship 116, 117, 118
FRC (functional residual capacity)
and closing volume 142, 143
measurement 104, 105, 106, 107
frequency (sine waves) 28
frequency histogram 40
frequency response (natural resonant
frequency) 72
fuel cell method (oxygen content
194

GABA1 receptor 184
gas analysers 80, 81, 82, 83, 84, 85
gas compression
ideal gas 2, 3
liquefaction 4, 5
real gas compression 4, 5, 6
volume, pressure and temperature 2, 3
gas cylinders, pressure inside 52, 53
gas equations 158, 159, 160
Bohr equation 158, 159, 160
Nunn equation 159, 160
universal gas equation 2
gas exchange (O2 and CO2 ) 158, 159
gas flow, in the circle system 54, 55, 56,
57
gas laws
Boyles law 2, 3
Charles law 2, 3
GayLussacs law 2, 3
gas pressure
decompression volume calculation 52
units of measurement and conversions 52
gas R line 158, 159, 160
gas supply, pressure/volume relationship 52,
53
gases
adiabatic compression and decompression
6, 7
critical temperature 4, 5
flow measurement 96, 97
Gaussian (normal) distribution 40, 41
GayLussacs law 2, 3
glycine receptors 184
growth factor 186
growth rate constant 30
haemoglobin
factors affecting oxygen affinity 134, 135,
136
fetal 136
light absorbances of different forms 76, 77,
78, 79
mean corpuscular concentration 134
oxyhaemoglobin dissociation curve 134,
135, 136
haemorrhage
classification of severity 122, 123
compensatory responses to 122, 123
effects of increasing severity 122, 123
HagenPoiseuille equation 10, 72, 126, 144
half-life 38
drug concentration in plasma 188, 189,
190, 191
for drug elimination 30, 31, 38
halothane
effects on coronary blood flow 131
effects on myocardial oxygen demand
131
MAC 182, 183
uptake 56, 57, 173
harmonics (sine waves) 28
head injury
cerebral perfusion pressure 126
effects on intracranial pressure 126
haematoma and/or cerebral oedema 126
hypercapnia 126, 127
hypocapnia under anaesthesia 126
loss of cerebral blood flow autoregulation
126
infrared absorption spectrophotometry 86,

87, 88, 89, 90, 91
infrared CO2 monitors 86, 87, 88, 89, 90, 91
inotropy, effects on pressurevolume loop
120, 121
integration 26, 27
area under the curve (AUC) 98, 99, 1001,
101
exponential decay curve 38, 39
intracardiac electrical potentials, harmonics
28
intracranial pressure
and cerebral elastance 124, 125
effects of head injury 126
intravascular pressure waveforms 110, 111,
112, 113
intravenous infusion duration, and drug
elimination rate 188, 189, 190, 191
inverse (reciprocal) relationship 26, 27
IPPV (intermittent positive pressure
ventilation)
cardiovascular effects 112, 113, 114, 115

saturation exponential 32, 33
use of capnography 90
iso shunt lines 156, 157
isobestic point 76, 77
isoerg (line of equal work) 22, 23
isoflurane
MAC 182, 183
uptake 56, 57, 173
isothermic compression and decompression
4, 5
isotherms 2, 3, 4, 5
Index
heart
autoregulation of perfusion 129, 130
heart muscle power 24, 25
see also cardiac cycle; myocardium
heart failure
effects on pressurevolume loop 120, 121
improvement of heart function 120, 121
LVEDP 112
heat energy, and change of phase 14
helium and oxygen mixture 12
helium wash-in method, to measure FRC
104, 105
humidification 16, 17
hydrogen ion (H+ ) concentration 92, 93, 94,
95
and cerebral blood flow 126
and O2 affinity for haemoglobin 134, 135,
136
conversion to pH 92, 93
see also pH
hypercapnia 64, 90, 91, 164, 165
effects on cerebral blood flow 126, 127
effects on hypoxic ventilatory response
162
hypocapnia 90, 91
effects on cerebral blood flow 126, 127
hypotension, differential diagnosis 90
hypovolaemia
effects of IPPV 114
hypoxia 154
hypoxic ventilatory response 162, 163
hysteresis 68, 69
J-shaped distribution 42
joule (J) 22
katharometer 104
laboratory tests
false negatives 48, 49, 50, 51
false positives 48, 49, 50, 51
location of cut-off point 48, 49, 50, 51
measure of usefulness 50, 51
sensitivity 48, 49, 50, 51
specificity 48, 49, 50, 51
LambertBeer law 76
laminar flow 8, 9, 10, 11
square function 28, 29
latent heat (definition) 14
latent heat of vaporization 6, 14, 15, 16,
17
LD50 180, 181
left ventricle
and systemic circulation 118
compliance and elastance 118, 119
LVEDP (left ventricular end-diastolic
pressure) 112, 118, 119, 120, 121
LVEDV ( left ventricular end-diastolic
volume) 118, 119, 120, 121
volume and pressure relationship 118, 119,
120, 121
leu-enkephalin 180
ligand-gated ion channel 184, 185
linear relationship 26, 27
linearity principle 68, 69
lipid solubility of inhalational anaesthetic
agents 182, 183
liquefaction of compressed gas 4, 5
liquids, flow measurement 96, 97, 98, 99
log-normal distribution 42, 43
logarithm and the logarithmic curve 32, 33,
34
logarithmic scales 34
pH measurement 92, 93
lung compliance 64
195
Index
196
lung dead space 150, 151, 152, 153

causes of changes in 152
lung filling
constant flow generator 64, 65
constant pressure ventilator 32, 33
constant pressure generator 62, 63
filling rate constant 32, 33
lung volumes 104, 105
and partial pressures of respiratory gases
158, 159, 160
lung volumes measurement 104, 105, 106,
107
closing volume 106, 107
dead space volume 106, 107
distribution of inspired air 106
emphysema 106, 107
functional residual capacity (FRC) 104,
105, 106, 107
helium wash-in method 104, 105
nitrogen washout method 106, 107
single breath analysis of expired nitrogen
106, 107
lungs
compliance 138, 139, 140, 141
distribution of blood flow 150, 151
elastic properties 138, 139, 140, 141
West zones 150, 151
LVEDP (left ventricular end-diastolic
pressure) 112, 118, 119, 120, 121
LVEDV (left ventricular end-diastolic
volume) 118, 119, 120, 121
MAC (minimum alveolar concentration)
(volatile anaesthetic agents) 182, 183
Mapleson A (Magill) breathing system
58, 59
mass and weight 18
mass spectrometry, respiratory gas analysis
84, 85
mean (average) 40, 42, 43
measurement
amplitude distortion 68, 70
baseline drift 68, 69, 70
basic concepts 68, 69, 70, 71
calibration of instruments 70
criteria for accurate measurement 68, 69,
70, 71
critical damping 70, 71
drift 68, 69, 70
dynamic accuracy 68, 70, 71
hysteresis 68, 69
linearity principle 68, 69
noise 70
phase distortion 68, 70, 71
physiological reactance 68
sensitivity drift 68, 69, 70
signal-to-noise ratio 70
static accuracy 68, 69
median (middle value) 42, 43
mitochondrial oxygen tension 154, 155
mitochondrial poisoning, tissue hypoxia 154
mitral valve disease, LVEDP 112
mode (most frequent value) 42, 43
morphine 178, 179, 180
muscarinic receptors 184
myocardium
blood flow rate 128, 131
ischaemia 129
oxygen consumption 128, 131
performance, effects of anaesthetic agents
129, 131
myogenic control of cerebrovascular tone
126
nalorphine 178, 179
naloxone 178, 179, 180
nervous system, effects on coronary vascular
resistance 130
neurogenic control of cerebrovascular tone
126
neurotropic factor 186
newton (N) 18
nicotinic receptor 184
nitrogen washout from the body 56
nitrogen washout method, to measure FRC
106, 107
nitrous oxide
adiabatic compression and decompression
6, 7
critical temperature 4
isothermic compression and
decompression 4, 5
MAC 182, 183
rate of uptake 172, 173
specific latent heat 14, 15
uptake in circle system 54, 55, 56, 57
noise (false signal) 70
non-competitive antagonists 180, 181
norepinephrine 180, 181
normal (Gaussian) distribution 40, 41
obesity, effects on thoracic compliance 140
opioid agonists and partial agonists 178, 179
opioid antagonists 178, 179
ordinate (y-axis) 26
oscillatory movement, sine waves 28, 29
oxygen
affinity for haemoglobin 134, 135, 136
alveolar partial pressure 158, 159, 160
binding to haemoglobin 80, 81, 82, 83
critical temperature 4
pA2 (measure of partial agonism) 180

parabola 28, 29
paramagnetic method, respiratory gas analysis
82, 83, 84
pascal (Pa) 18
pathological shunt fraction 156, 157
pethidine 178, 179
pH 92, 93, 94, 95
and hydrogen ion concentration 92, 93, 94,
95
normal physiological pH range 92
measurement 92, 93, 94, 95
pH meter 94, 95
see also hydrogen ion (H+ ) concentration
pharmacodynamics
agonist drugs 176, 178, 179, 180
antagonist drugs 176, 178, 179, 180, 181
binding of drugs to receptor sites 176
doseresponse curves 176, 177, 178, 179,
180, 181
drug affinity 176
drug efficacy (intrinsic activity) 176, 177,
178, 180
drug potency 176, 178, 179, 180, 182, 183
ED50 180, 181
LD50 180, 181
see also drugs
pharmacokinetics
compartmental models 170, 171
elimination rate constant (k) 168
first-order (linear) kinetics 168, 169, 170
speed of inhalational induction 174
uptake of inhalational anaesthetic agents

172, 173
see also drug elimination; drugs
phase, change of 14
phase distortion 68, 70, 71
phase shift, sine waves 28
phenoxybenzamine 180, 181
phentolamine 180, 181
phospholipase receptor 184
physiological shunt 142, 143, 152, 154, 156,
157
physiological ventilationperfusion matching
150
plethysmograph 102, 103
pneumotachograph 96, 97, 102
posture, and lung compliance 140, 141
power (P) 24, 25
pregnant women, ventilatory response to
carbon dioxide 164, 165
presentation of data see data presentation
pressure (p) 18, 19
pressure, volume and temperature 2, 3
pressure gradients, and airway resistance 144
pressure-reducing valves (anaesthetics) 20,
21
pressure transducer 102
pressurevolume loops, cardiac cycle 120,
121
propofol 190, 191
pulmonary artery catheterization 110, 111,
112, 113
pulmonary artery occlusion pressure
(PAOP), and LVEDV 118
pulmonary artery occlusion pressure
waveform 112
pulmonary artery pressure waveform 110,
111
pulmonary hypertension, effects of IPPV
114
pulmonary perfusion, distribution 150, 151
pulse oximeter
LambertBeer law 76
principle 76, 77, 78, 79
sources of error and interference 78
pumps, defining work output 22, 23
radius (of a tube), effects on flow velocity 8,
9, 10, 11
rebreathing system see circle system
receptor types
multisubunit ligand-gated ion channel 184,
185
transmembrane enzyme receptor 186, 187
reciprocal (inverse) relationship 26, 27
Index
partial pressure and cerebrovascular

resistance 126, 127
partial pressure in solution 80, 81, 82, 83
ventilatory response 162, 163
oxygen cascade (oxygen transport) 154, 155
oxygen concentration (definition) 80
oxygen content measurement 80
blood gas analysers 80, 81, 82, 83, 84
electrochemical method (fuel cell) 82, 83,
74
manometric (van Slyke) method 80, 81, 84
mass spectrometry 84, 85
paramagnetic method 82, 83, 84
polarographic method (Clark electrode) 82,
83, 84
respiratory gas analysers 83, 83, 84, 85
oxygen saturation 76, 77, 78, 79, 80
oxygen tension measurement 80, 81, 82, 83
oxygen therapy, to compensate for shunt
fraction 156, 157
oxyhaemoglobin dissociation curve 134, 135,
136
197
Index
198
rectangular hyperbola 26, 27

remifentanil
context-sensitive half-time 190, 191
elimination rate constant 168, 169
exponential decline in plasma
concentration 30, 31
respiratory gas analysers 83, 83, 84, 85
respiratory system
airway (non-elastic) resistance 144, 145
airways closure (closing volume) 142,
143
airways resistance measurement 102, 103
compliance 138, 139, 140, 141
dynamic compliance 146, 147
elastic properties of the chest wall 138, 139,
140, 141
elastic properties of the lungs 138, 139, 140,
141
mechanical properties of the chest 102,
103
resting position 138, 139
static compliance of the chest 102
Reynolds number (Re) 12
right atrial pressure waveform 110, 111
right ventricle and pulmonary circulation
118
right ventricular pressure waveform 110,
111
ROC (receiver operating characteristic) curve
48, 49, 50, 51
Rotameter 10, 96, 97
saturated vapour pressure 4
saturation exponential curve 32, 33
saturation of air with water vapour 16, 17
sensitivity (of a test) 48, 49, 50, 51
sensitivity drift 68, 69, 70
sevoflurane 172, 173
shock, use of inhalational anaesthetic agents
174
shunt fraction (physiological shunt) 152, 154,
156, 157
causes of changes in 152
compensation for 152
compensation with oxygen therapy 156,
157
effects of ageing 142, 143
pathological 156, 157
signal-to-noise ratio 70
sine waves 28, 29
sinusoid waveform 28, 29
skewed distribution 42, 43
skin, responses to blood loss 122, 123
sleep, effects on hypoxic ventilatory response
162
sleep apnoea, hypoxic ventilatory response

162
specific latent heat 14
specificity (of a test) 48, 49, 50, 51
spirometer 102
square function, parabola 28, 29
Starling resistor effect 150
static accuracy of measurement 68, 69
statistics
bimodal distribution 42, 43
J-shaped distribution 42
log-normal distribution 42, 43
mean (average) 40, 42, 43
measures of central tendency 40, 42
median (middle value) 42, 43
mode (most frequent value) 42, 43
mode of distribution 40, 41, 42, 43
normal (Gaussian) distribution 40, 41
sample 40
skewed distribution 42, 43
standard deviation (SD) 40, 41
U-shaped distribution 42
variability 40
STP (standard temperature and pressure) 2
sympathetic nervous system control, and
effects of IPPV 114
syringe emptying, force required 18, 19
T-piece breathing systems 60, 61
tachycardia, due to blood loss 122, 123
temperature, and O2 affinity for haemoglobin
134, 135, 136
temperature, volume and pressure 2, 3
temperature change, physical changes caused
by 14
tests see laboratory tests
thermal dilution method, cardiac output
measurement 98, 99, 1001, 101
thermal energy 22
thermodynamics, first law of 14
thiopentone 190, 191
time constant ( ) 36, 37, 38
transmembrane enzyme receptor 186,
187
tube radius, effects on flow velocity 8, 9, 10,
11
tubes, resistance to flow 8, 9, 10, 11
turbulent flow
factors affecting 12, 13
Reynolds number (Re) 12
U-shaped distribution 42
ultrasonic flowmeter 96
urine output, response to blood loss 122, 123
ventilatory response to carbon dioxide 164,

165
ventilatory response to oxygen 162, 163
viscosity, effects on flow velocity 8
volume, pressure and temperature 2, 3
Index
Valsalva manoeuvre 114, 115

van Slyke method (oxygen content
vaporizers 16, 54, 56, 57
ventilationperfusion relationship
alveolar ventilation and its distribution 148,
149
dead space gas 148
physiological dead space effect 150, 151,
152, 153
physiological shunt 152
pulmonary perfusion and its distribution
150, 151
ventilation/perfusion inequality 152, 153
ventilation-perfusion ratio 152, 153
ventilators
constant flow generator 64, 65
constant pressure generator 62, 63
saturation exponential 32, 33
water, specific latent heat 14, 15

watt (W) 24
weight and mass 18
West zones of the lung 150, 151
work (W) 22, 23
work and energy 22, 23
Wright peak flow meter 96, 97
x-axis (abscissa) 26
y-axis (ordinate) 26
199

Basic Science For Anaesthetists

Uploaded by

Copyright:

Available Formats

Basic Science For Anaesthetists

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Basic Science For Anaesthetists

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

Who wrote the book?

Who wrote the book?

BASIC SCIENCE FOR ANAESTHETISTS

BASIC SCIENCE FOR

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo

Part 1: Physics, mathematics, statistics, anaesthetic

4. Heat, vaporization and humidification

5. Simple mechanics 1: mass, force, pressure

6. Simple mechanics 2: work and power

8. Exponentials 1: the curves

9. Exponentials 2: properties of exponential decay curve

10. Descriptive statistics

11. Presentation of data

12. Receiver operating characteristic curve

14. The circle system

17. Lung filling with automatic lung ventilators

Part 2: Clinical measurement

2. Electromanometers, frequency response and

3. Pulse oximeter principle

4. Oxygen content and oxygen tension measurement

6. [H+], pH and its measurement

7. Principles of measurement of volume and flow in

8. Cardiac output measurement by thermal dilution

9. Measurement of the mechanical properties of the

10. Lung volumes and their measurement

Part 3a: Physiology: the cardiovascular system

3. Control of cardiac output, regulation of cardiac

4. Cardiac cycle: pressure-volume relationships

5. Blood pressure and blood volume relationship

6. Cerebral blood flow

Part 3b: Physiology: the respiratory system

2. Cardiovascular effects of intermittent positive

2. Respiratory mechanics 1: Static properties, factors

3. Respiratory mechanics 2: Dynamic properties,

5. Oxygen cascade, oxygen therapy and shunt fraction

6. Gas R line, solution of the ventilation/perfusion

7. Ventilatory response to oxygen

8. Ventilatory response to carbon dioxide

2. Uptake and distribution of inhalational anaesthetic

3. Pharmacodynamic effects of drugs

4. Minimum alveolar concentration and lipid solubility

5. Receptor types, molecular action of anaesthetics

ABBREVIATIONS and SYMBOLS

List of abbreviations and symblos

absolute temperature (K)

List of figure captions

33. Positive skew distribution of hydrogen ion concentration in

List of figure captions

69. Principal components of a capnograph.

List of figure captions

practising anaesthetists who undertake teaching and wish to remain connected

PREFACE TO THE SECOND EDITION

Gas compression, relationship of

Figure 2. Avogadros hypothesis.

Gas compression, relationship of volume, pressure and temperature

Figure 1. Three gas laws: Boyles, Charles and GayLussac.