Clinical Science (1991)80,353-358
353
Extent of pulmonary emphysema in man and its relation to
the loss of elastic recoil
M. GUGGER, G. GOULD, M. F. SUDLOW, P. K. WRAITH AND W. MACNEE
Unit of Respiratory Medicine, Rayne Laboratory, Department of Medicine (RIE),University of Edinburgh, Edinburgh, Scotland, U.K.
(Received 2 May/2 November 1990; accepted 19 November 1990)
SUMMARY
1. We assessed lung density, determined by computerized tomography, as a measure of emphysema and related
this to lung function and measurement of the elastic recoil
of the lung in normal subjects and patients with chronic
obstructive lung disease.
2. We found a sigmficant correlation between
measurements of elastic recoil pressure at 90% of total
lung capacity and both the forced expiratory volume in
1 s (r= 0.80, P < 0.001) and the transfer factor for carbon
monoxide ( r =0.70, P < 0.001). Measurements of elastic
recoil of the lung also correlated with lung density as
measured by computerized tomography scanning
(P< 0.001).
3. Multiple regression analysis demonstrated a correlation between the density of the lowest fifth percentile of
the computerized tomography lung-density histogram,
and both the natural logarithm of the shape parameter of
the pressure-volume curve (P<0.01), and the transfer
factor for carbon monoxide ( P < 0.01). However, the
mean computerized tomography lung density correlated,
in addition, with the elastic recoil pressure of the lungs at
90%of total lung capacity ( P < 0.001).
4. Since the elastic recoil pressure correlates with
computerized tomography lung density, and hence with
emphysema, and since elastic recoil pressure also correlates with the forced expiratory volume in 1 s, these
results suggest that loss of elastic recoil is one determinant
of airflow limitation in patients with chronic obstructive
lung disease.
Key words: computerized tomography scan, elastic recoil,
emphysema.
Abbreviations: AW/W, alveolar wall/unit lung volume;
COLD, chronic obstructive lung disease; CT, computerized tomography; FEV,, forced expiratory volume in 1 s;
FVC, forced vital capacity; K,,, diffusion coefficient for
Correspondence: Dr W. MacNee, Unit of Respiratory
Medicine, Department of Medicine (RIE), City Hospital,
Greenbank Drive, Edinburgh EHlO 5SB,U.K.
carbon monoxide; PL,90,
elastic recoil pressure at 90% of
total lung capacity; RV, residual volume; TL,co,transfer
factor for carbon monoxide;TLC, total lung capacity.
INTRODUCTION
In resected human lungs [l]and in animal models [2], it
has been suggested that a protease/anti-protease
imbalance leads to pulmonary emphysema, involving
destruction of elastic fibres in the lungs [3, 41. Loss of
elastic recoil should therefore occur with advancing
emphysema and may contribute to airflow limitation [5],
which influences survival [6, 71 in patients with chronic
obstructive lung disease (COLD).Whether loss of elastic
recoil or small-airway abnormalities are more important
in causing chronic airflow limitation remains controversial
[8-121. Studies in excised or post-mortem human lungs
from patients with emphysema have shown a correlation
between the loss of elastic recoil, measured in vitro, and
the extent of emphysema, assessed pathologically
[ 13-17]. This is also supported by animal studies in which
emphysema was chemically induced, where a correlation
between the mean linear intercept, a measure of the
alveolar surface area, and the elastic recoil pressure of the
lungs has been demonstrated [2,18,19]. Although Silvers
et al. [16] found a s'
cant loss of elastic recoil in early
emphysema in man, ese changes appeared to be out of
proportion to the degree of emphysema. They concluded
that the loss of elastic recoil did not result entirely from
emphysema. In another study, petty et al. [20] found that
the degree of elastic recoil, measured in vitro in human
lungs, correlated with small-airwaypathology.
There are relatively few studies in which emphysema
has been correlated with the elastic recoil of the lungs
measured in vivo, and in these studies the results are
equivocal [21-251. A major problem with such studies
has been the use of a semi-quantitative assessment of the
extent of emphysema, based on a picture-grading technique which assesses only macroscopic emphysema [26].
Lung density as measured by computerized tomography
(CT) relates to the pathological extent of microscopic
emphysema, measured in resected lungs, and therefore
Y
354
M. Gugger et al.
allows the assessment of the extent of emphysema in life
[27]. Variables derived from lung pressure-volume (P/V )
curves reflect the elastic properties of both lungs. Since
CT scanning also gives an assessment of emphysema, in
both lungs, it seems appropriate to re-examine the
relationship between lung elastic recoil, airflow limitation
and the extent of emphysema, in man, in life.
METHODS
Subjects
Twenty-four male smokers or ex-smokers presenting
with chronic bronchitis (productive cough for 3 months/
year for 2 consecutive years) and breathlessness, were
recruited from our outpatient clinic. All of the patients
had airflow limitation [forced expiratory volume in 1 s
(FEV,) 1.8 litres SD 0.2 litre; FEVJforced vital capacity
(FVC)ratio < 65%] with < 15% improvement in FEV, at
least in response to two puffs of a B,-adrenoceptor
agonist from a metered dose inhaler. Other lung diseases
were excluded by history, examination and chest X-ray. In
addition, 12 healthy non-smokers with no respiratory
symptoms and no history of atopy, asthma or other respiratory symptoms, and with normal lung function, were
studied (FEV, 3.8 litres, SD 0.2 litre) (Table 1).Informed
consent was obtained from all subjects and the study was
approved by our local ethics committee.
Respiratory function
Measurements of ventilatory capacity (FEV,, FVC),
subdivisions of lung volumes [residual volume (RV ), total
lung capacity (TLC), RV/TLC] and gas transfer for
carbon monoxide [single-breath transfer factor ( TL,,,)
and diffusion coefficient (Kco)] were measured using a
Gould 2400 system (Gould Electronics Ltd, U.K.). TL,,,
was measured by the method of Ogdvie et al. [28] with the
time of breath-holding calculated by a modification of the
technique of Jones & Meade [29]. K,, was calculated
using the alveolar volume, measured by the dilution of
helium during the single-breath manoeuvre. Predicted
values for lung volumes were derived from Crapo et al.
[30] for men and from Hall et al. [31] for women, and
predicted values of K,, for both men and women were
from Cotes [32].
Table 1. Characteristics of the 12 healthy subjects and
the 24 patients with COLD
Values are means with ranges in parentheses.
Characteristic
n
Age (yews)
FEV, (% of predicted)
RV/TLC (“/o of predicted)
TLco (YOof predicted)
Ink
PL,,,(CmH20)
Value
36
53 (22-77)
62(17-109)
135 (71-212)
70 (31-118)
-1.9(-1.2t0 -2.5)
11.9 (3.1-20.6)
Elastic recoil
Quasi-static P/ V curves were obtained with the subject
seated in a volume-displacement body plethysmograph.
Elastic recoil pressure was measured using a thin-walled
latex balloon (containing 0.5 ml of air) and a catheter
system (internal diameter 0.2 cm) by the technique of
Milic-Emili et al. [33]. Mouth flow was measured with a
Fleisch pneumotachograph (no. 3) and lung volume
changes were determined from the wedge spirometer of
the plethysmograph. The subjects took two deep breaths
to standardize the volume history before the
measurement was obtained. The P / V curves were
recorded on line with a PDP/1173 computer. Curves with
a flow rate greater than 0.2 l/s at any time during expiration
were rejected. Two to five satisfactory P / V curves, with
no oesophageal spasms, swallows or glottis closures, were
analysed in each subject. All P / V curves were inspected
by eye and any curves with obvious artefacts were
rejected before curve-fitting. In three patients only one
satisfactory P / V curve could be obtained.
Analysis of P / V curves was performed by using an
objective method recently developed in our laboratory
[34].Briefly, a cubic function was fitted to the P/V data to
define an inflection point on the P / V curve. An exponential function, as described by Colebatch et al. [35]:
where V = lung volume, P = recoil pressure and A,B and
k are constants, was then fitted to the data for volumes
above the inflection point [34]. The exponential constant
(k), its normally distributed natural logarithm (Ink) and
the elastic recoil pressure at 90% of TLC (PL,90)
were
determined [36,37].
CT scans
Whole-chest CT scans were carried out using a GE
9000 whole-body scanner with a 5 s scan time and a slice
thickness of 10 mm, using a scanner ring of 42 cm and a
beam energy of 120 kV and 200 mA. In 18 of the patients
with COLD, a whole-chest scan with cuts made at 3 cm
intervals was obtained. In the remaining six subjects with
COLD and in four of the healthy subjects, a limited scan
was obtained comprising two CT cuts, 6 and 10 cm below
the sternal notch. During the scan the subject held his
breath in inspiration. A custom-written computer
program, developed in this laboratory, was used to outline
the lung fields excluding the hilar region, from which a
frequency histogram of the lung density (measured as the
E M number) for each lung field was produced. Datr
were pooled for all CT cuts in each subject [27]. As
measures of the extent of emphysema, the mean EMI
number and the EMI number of the lowest fifth percentile
of the pooled density histogram of both lungs were calculated. Linear correlation coefficients were calculated by
using standard techniques and multiple linear regression
analyses were used for comparisons.
Correlation in vivo between emphysema and elastic recoil
RESULTS
The subjects/patientswho were studied had a wide range
of age, respiratory function and of measurements of
elastic recoil (Table 1).CT lung density also showed a
wide range of values in the subjects/patients studied. The
mean EMI number ranged from -391 to -441 (mean
-420) and the EMI of the lowest fifth percentile from
- 443 to - 493 (mean - 46 1).The aim of the study was
to recruit individuals with a wide spectrum of disease
from normal to severe airflow limitation. Therefore the
data from the healthy subjects and patients were pooled
and analysed together.
Both the mean EMI number and the EMI number of
the lowest fifth percentile from the CT lung-densityhistogram (Fig. 1)correlated significantly with Ink (Table 2).
The elastic recoil variable PL,yo
correlated Significantly
with the mean EMI number (Fig. 2) but did not correlate
sigdicantly with the lowest fifth percentile of the CT
lung-densityhistogram (Table 2). There were also si@-
-7
0,-7
- 2.44
.. ....
.
.... . . .
-2.22.0-1.8- 1.4-
16
.
.
-om1412-
.
*:
7
100
8h
v
*
.
.
I
I
0
8
3 -1.6-
cant correlations between both the mean EMI number
and the EMI number of the lowest fifth percentile and
the FEV, and TL.co (Table 2). Significant correlations
between PL,yo
and FEV, (Fig. 3), RV/TLC (Fig. 4) and
TL,co(Fig.5)were also observed.
A multiple regression analysis was carried out between
the mean EMI number or the EMI number of the lowest
fifth percentile, and the TL,Co, FEV,, RV/TLC, PL,90 and
Ink. This analysis showed that the mean EMI number
.
-390
,
I
-400 -410
.
?-
*.
2
f
0
355
#
I
I
-420
-450
-430 -440
Mean EMI no. in CT scan
.
Fig. 2. Relationship between CT lung density, measured
as the mean EMI (density)number of the CT lung-density
in 28 subjects (four normal subjects
histogram, and PL,90
and 24 patients withCOLD).r = -0.65, P<O.OOl.
0
- 11.21
.0
L
L,
-440 -450
I
-460 -470 -480 -490 -500
EMI no. of lowest 5th percentile in CT scan
Fig. 1. Relationship between CT lung density, measured
as the EMI (density) number which defines the lowest
fifth percentile of the CT lung-density histogram, and the
shape parameter of the lung P / V curve as expressed by
Ink, in 28 subjects (four normal subjects and 24 patients
with COLD). r = -0.60, P<O.OOl.
18-
EMI no. of the
lowest fifth
percentile
r
P
r
-0.58
0.65
0.59
0.56
0.57
0.50
0.55
<0.005
<0.001
0.001
<0.01
<0.01
< 0.01
<0.005
< 0.01
-0.60
0.35
0.52
0.51
0.51
0.45
0.49
0.45
P
a
a
a
a
ON
<
f
2
16a
14-
a
a
12-
1 *:
4
*
a
a
10-
8-
Mean EMI no.
.
20
h
Table 2. Correlation coefficient (r) for the relationship
between tests of lung function, elastic recoil and the
extent of emphysema measured as the mean or lowest
fifth percentile of the CT lung density histogram
Abbreviation: NS, not significant( P = 0.1).
a
261
24
a
a m
e
a
**aI
2
Ink
P,
nn
L.7"
TWO
TL,co(% of predicted)
KP,
KE: (% of predicted)
FEV,
FEV, fo/o of Dredicted)
0.50
0.001
NS
<0.01
<0.01
<0.01
<0.05
0.01
<0.01
FEV, (litres)
Fig. 3. Relationship between airways obstruction (FEV,)
in 36 subjects (12 normal subjects and 24
and PL,yo
subjects with COLD).r = 0.80, P<O.OOl.
356
M. Gugger et al.
241
26
22
**
20
.
**
0
a
-
10
**
:I, i**;
,::
*
*
**
4
2
20
30
40
,
50
60
70
80
RV/TLC (“/o)
Fig. 4. Relationship between hyperinflation (RV/TLC)
and P,,,, in 36 subjects (12 normal subjects and 24
subjects with COLD). r = - 0.80, P < 0.001.
.
2624-
...
22 -
20181614-
. . .
12v
9 10-
a:
. .
.
8-
%
derived from studies where elastic recoil was measured in
vitro, either in excised human [ 13- 171 or animal [ 18, 191
lungs. Despite finding a s i m c a n t correlation between
elastic recoil and emphysema, Berend et al. [15]
concluded that “...it seems unlikely that the presence of
emphysema can be recognised accurately in life from
standard analysis of the pressure-volume characteristics”.
There are only a few studies where measurements of
elastic recoil, made in life, have been compared with a
pathological assessment of emphysema, either in lung
tissue resected at a subsequent thoracotomy or at post
mortem [21-251. Boushy et al. [21] suggested that lung
recoil pressure could distinguish patients with mild or no
emphysema from those with severe emphysema. The
extent of emphysema was assessed in that study by the
point-counting method of Dunnill [38]. Berend et al.
[23] found a significant correlation ( r= 0.49, P < 0.05)
between k and macroscopic emphysema assessed using
‘an arbitrary scale of 0 to 100’. Nevertheless they concluded that tests of elastic recoil were not predictive of
early emphysema (scores between 0 and 50). Pare et al.
[24] also found a sigmficant correlation between both k
(r=0.35, P < O . O l ) and P,,,, (r=0.31, P<0.05)and a
pathological score of macroscopic emphysema, based on
a picture-grading system [26].As a group those with mild
emphysema were distinguishable for normal subjects by
measurements of k and P,,,,. They concluded that
minimal emphysema could be detected by an exponential
analysis of the lung P/ V curve [24]. However, inspection
of their data reveals that the correlations were weak and
the measurements had a poor discriminant value. Furthermore, in a different study population, the same group
reported no correlation between either k or maximum
elastic recoil pressure as a percentage of their predicted
values and an emphysema score [25].
A major criticism of all of these studies is that parametric linear regression analysis was used in the correlations, despite the fact that emphysema ‘grades’ or ‘scores’
do not represent a continuously distributed variable. In
addition, when parametric statistics are used, the
normally distributed variable Ink should be used rather
than k, which has a skewed distribution [39]. However,
CT scanning provides a linear measurement of physical
density [27] and therefore enables us to use parametric
statistics. Moreover, the emphysema score is a semiquantitative measure of macroscopic emphysema. We
have previously shown that measurements of CT lung
density correlate with quantitative measurements of
microscopic emphysema, as measured by the alveolar
wall/unit lung volume (AW/UV)ratio [27].
The main result of this study is that there is a significant
correlation between the extent of loss of elastic recoil and
the extent of emphysema (CT lung density), assessed in
both lungs, both measurements being made in vivo.
Multiple regression analysis of the extent of emphysema
with the elastic recoil, TL,co,FEV, and TLC indicates that
both measurements of elastic recoil and TL,coare significantly associated with the extent of emphysema (Table 2).
Since the elastic recoil correlates both with CT lung
density, and hence emphysema, and with the FEV,, these
6
4
2
0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
TL,co(nun01m i - ’ kPa-’)
Fig. 5. Relationship between TL,co and PL,,, in 36
subjects, (12 normal subjects and 24 subjects with
COLD). r = 0.70, P < 0.001.
correlated with TL,co( P < 0.005),Ink ( P < 0.02) and P,,,,
(P<O.OOl). The EM1 number of the lowest fifth percentile correlated with Ink ( P < O . O O l ) and TL,co
( P < 0.01) but not with PL,90.
DISCUSSION
Most of the evidence supporting the concept of a correlation between elastic recoil of the lungs and emphysema is
Correlation in vivo between emphysema and elastic recoil
results suggest that elastic recoil is an important
determinant of airflow limitation in patients with COLD.
It is interesting to note that Ink, the shape parameter of
the P/V curve correlates sigtllficantlywith both the mean
EMI number and the EMI number of the lowest fifth percentile of the CT lung-density histogram. In contrast,
PL,90correlates significantly only with the mean CT lung
density. In a previous study from this laboratory we
showed that the EMI number defining the lowest fifth
percentile of the CT lung-density histogram correlated
best with the mean A W / W ratio in the five 1mm X 1mm
microscopic fields with the lowest A W / W ratios, out of
the 20-35 such fields examined in the resected lung, in
each patient [27]. The EMI number of the lowest fifth
percentile of the CT lung-density histogram is therefore a
measure of the most markedly enlarged distal airspaces
(i.e.the most severe emphysema).Since Ink and the lowest
fifth percentile CT lung-density histogram correlate, then
the elasticity of the most emphysematous parts of the
lungs seems to be the important factor in determining the
shape of the P/ V curve, as expressed by Ink. On the other
hand, the mean EMI number reflects the average extent of
emphysema in the lungs. Thus the fact that the average
degree of emphysema correlates with PL,90
might suggest
that measures of absolute lung elastic recoil pressures,
such as PL,90,
are mainly a function of the elasticity of the
whole of the lungs. Of course it must be remembered from
the exponential equation described earlier that PL,90
is not
completely independent of Ink.
In order to explain the differing correlations between
PL,90rInk and CT lung density, we assume that the
severity of macroscopic emphysema vanes throughout
the lungs. Then, on deflation, a greater volume contribution would be made to the overall P / V curve from the
more elastic, emphysematous regions of the lungs, as
assessed by the lowest fifth percentile of the CT lungdensity histogram. The effect of this would be to shift the
lower portion of the P / V curve to the left, thereby making
the curve more angular, which would increase Ink. By
contrast, PL,90,
which is a measure of the upper part of the
PIV curve, would be less influenced by these severely
emphysematous areas, and more influenced by the
average elasticity of the lungs, including areas with and
without macroscopic emphysema as measured by the
mean CT lung density. However, the true interactions
between areas of lung with variable degrees of emphysema and their relative contributions to the overall elastic
properties of the lungs, are, of course, much more
complex.
In summary, this study has demonstrated that the
extent of emphysema, the elastic recoil of the lungs and
the degree of airflow limitation correlate significantlywith
each other. Therefore it seems likely that loss of elastic
recoil is one important determinant of airflow limitation
in patients with emphysema.
ACKNOWLEDGMENTS
This study was supported by the Norman Salvesen
Emphysema Research Fund. We thank Mrs C. Hendrie
357
for typing the manuscript. M.G. was the recipient of a
travelling fellowship from the Swiss National Science
Foundation.
REFERENCES
1. Damiano, V.V., Tsang, A., Kucich, U. et al. Immunolocalization of elastase in human emphysematous lungs. J. Clin.
Invest. 1986; 78,482-93.
2. Snider, G.L., Lucey, E.C., Christensen, T.G. et al.
Emphysema and bronchial secretory cell metaplasia
induced in hamsters by human neutrophil products. Am.
Rev. Respir. Dis. 1984; 129,155-60.
3. Wewels, M. Pathogenesis of emphysema. Assessment of
basic science concepts through clinical investigation. Chest
1989;95,190-5.
4. Niewoehner, D.E. Cigarrette smoking, lung inflammation,
and the developmen<of emphysema. J. Lab. Clin. Med:
1988: 111.15-27.
5 . NagG, A.,’ West, W.W. & Thurlbeck, W.M. The National
Institutes of Health Intermittent positive-pressure breathing
trial: pathology studies. II. Correlation between morphologic findings, clinical findings, and evidence of air-flow
obstruction. Am. Rev. Respir. Dis. 1985; 132,946-53.
6. Peto, R, Speizer, F.E., Cochrane, A.L. et al. The relevance
in adults of air-flow obstruction, but not of mucus hypersecretion, to mortality from chronic lung disease. Am. Rev.
Respir. Dis. 1983; 128,491-500.
7. Burrows, B., Fletcher, C.M., Heard, B.E., Jones, N.L. &
Wootliff, J.S. The emphysematous and bronchial types of
chronic airways obstruction. A clinicopathological study of
patients in London and Chicago. Lancet 1966; i, 830-5.
8. Park, S.S., Yoo, O.H., Janis, M. & Williams, M.H. Postmortem evaluation of airflow limitation in obstructive lung
disease. J. Appl. Physiol. 1969; 27,308-12.
9. Mitchell, R.S., Stanford, R.E., Johnson, J.M., Silvers, G.W.,
Dart, G. & George, M.S. The morphologic features of the
bronchi, bronchioles, and alveoli in chronic airway obstruction: a clinicopathological study. Am. Rev. Respir. Dis.
1976; 114,137-45.
10. Cosio, M., Ghezzo, H., Hogg, J.C. et al. The relations
between structural changes in small airways and pulmonaryfunction tests. N. Engl. JrMed. 1977; 298; 1277-81.
11. Greaves, IA. & Colebatch, H.J.H. Observations on the
pathogenesis of chronic airflow obstruction in smokers:
implications for the detection of ‘early’lung disease. Thorax
1986; 41,81-7.
12. Snider, G.L. Chronic obstructive pulmonary disease - a
continuing challenge. Am. Rev. Respir. Dis. 1986; 133,
942-4.
13. Niewoehner, D.E., Kleinennan, J. & Liotta, L. Elastic
behavior of postmortem human lungs: effects of aging and
mild emphysema.J. Appl. Physiol. 1975; 39,943-9.
14. Greaves, IA. & Colebatch, H.J.H. Elastic behavior and
structure of normal and emphysematous lungs postmortem.
Am. Rev. Respir. Dis. 1980; 121,127-36.
15. Berend, N., Skoog, C. & Thurlbeck, W.M. Pressure-volume
characteristics of excised human lungs: effect of sex, age and
emphysema. J. Appl. Physiol. 1980; 49,558-65.
16. Silvers, G.W., Petty, T.L. & Stanford, R.E. Elastic recoil
changes in early emphysema. Thorax 1980; 35,490-5.
17. Petty, T.L., Silvers, G.W. & Stanford, R.E. Functional
correlations with mild and moderate emphysema in excised
human lungs. Am. Rev. Respir. Dis. 1981; 124,700-4.
18. Niewoehner, D.E. & Kleinennan, J. Effects of experimental
emphysema and bronchiolitis on lung mechanics and
morphometry. J. Appl. Physiol. 1973; 35,25-31.
19. Haber, P.S., Colebatch, H.J.H., Ng, C.K.Y. & Greaves, LA.
Alveolar size as a determinant of pulmonary distensibility in
mammalian lungs. J. Appl. Physiol. 1983; 54,837-45.
358
M. Gugger et al.
20. Petty, T.L., Silvers, G.W. & Stanford, R.E. Small airway
disease is associated with elastic recoil changes in excised
human lungs. Am. Rev. Respir. Dis. 1984; 130,42-5.
21. Boushy, S.F., Aboumrad, M.H., North, L.B. & Helgason,
A.H. Lung recoil pressure, airway resistance, and forced
flows related to morphologic emphysema. Am. Rev. Respir.
Dis. 1971; 104,551-61.
22. Gelb, A.F., Gold, W.M., Wright, R.R., Bruch, H.R. & Nadel,
J.A. Physiologic diagnosis of subclinical emphysema. Am.
Rev. Respir. Dis. 1973; 107,50-63.
23. Berend, N., Woolcock, A.J. & Marlin, G.E. Correlation
between the function and structure of the lung in smokers.
Am. Rev. Respir. Dis. 1979; 119,695-705.
24. Pare, P.D., Brooks, L.A., Bates, J. et al. Exponential analysis
of the lung pressure-volume curve as a Predictor of
pulmonary emphysema. Am. Rev. Respir. Dis. 1982; 126,
54-61.
25. Bergin, C., Mueller, N., Nichols, D.M. et al. The diagnosis of
emphysema. A computed tomographic-pathologic correlation. Am. Rev. Respir. Dis. 1986; 133,541-6.
26. Thurlbeck, W.M., Dunnill, M.S., Hartong, W., Heard, B.E. &
Ryder, R.C. A
Of three methods Of
emphysema. Hum. Pathol. 1970; 1,215-26.
27. Could, GA., MacNee, W., McLean, A. et al. CT
measurements of lung density in life can quantitate distal
airspace enlargement - an essential defining feature of
human emphysema. Am. Rev. Respir. Dis. 1988; 137,
380-92.
28. Ogilvie, C.M., Forster, R.E., Blakemore, W.S. & Morton,
J.W. A standardized breathholding technique for the clinical
measurement of the diffusing capacity of the lung for carbon
monoxide. J. Clin. Invest. 1957; 36,l-17.
29. Jones, R.S. & Meade, F. A theoretical and experimental
analysis of anomalies in the estimation of pulmonary
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
diffusing capacity by the single breath method. Q. J. Exp.
Physiol. 1961; 46,131-43.
Crapo, R.O., Moms, A.H. & Gardner, R.M. Reference
spirometric values using techniques and equipment that
meet ATS recommendations. Am. Rev. Respir. Dis. 1981;
123,659-64.
Hall, A.M., Heywood, C. & Cotes, J.E. Lung function in
healthy British women. Thorax 1979; 34,359-65.
Cotes, J.E. Lung function at different stages of life, including
reference values. Lung function. 3rd ed. Oxford: Blackwell
Scientific Publications, 1975: 340-95.
Milic-Emili, J., Mead, J., Turner, J.M. & Glauser, E.M.
Improved technique for estimating pleural pressure from
esophageal balloons. J. Appl. Physiol. 1964; 19,207-1 1.
Gugger, M., Wraith, P.K. & Sudlow, M.F. A new method of
analysing pulmonary quasi-static pressure-volume curves in
normal subjects and in patients with chronic airflow
obstruction. Clin. Sci. 1990; 78,365-69.
Colebatch, H.J.H., Ng, C.K.Y. & Nikow, N. Use of an
exponential function for elastic recoil. J. Appl. Physiol.
1979; 46,387-93.
Knudson, R.J. & Kaltenborn, W.T. Evaluation of lung elastic
recoil by exponential curve analysis. Respir. Physiol. 1981;
46,29-42.
McCuaig, KJ% VeSSal, S.7 COPPh K., Wigs, B.J.R.,
Dahlby, R. & Pare, P.D. Variability in measurements of
pressure-volume curves in normal subjects. Am. Rev.
Respir. Dis. 1985; 131,656-8.
D u r n , M.S. Quantitative m&ods in the study of
pulmonary pathology. Thorax 1962; 17,320-8.
Knudson, R.J. & Kaltenborn, W.T. Evaluation of lung elastic
recoil by exponential curve analysis. Respir. Physiol. 1981;
46,29-42.