Download alveolar-capillary membrane of the lung as awhole to ... and more Study notes Physiology in PDF only on Docsity!
J. Physiol. (^) (I959) 146, 572-
THE EFFECT OF AGE, BODY SIZE AND LUNG VOLUME
CHANGE ON ALVEOLAR-CAPILLARY PERMEABILITY AND
DIFFUSING CAPACITY IN MAN
BY MARGARET W. McGRATH AND M. L. THOMSON
From (^) the Department of (^) Applied Physiology, London School of Hygiene
and Tropical Medicine, London, W.C. 1
(Received 8 January 1959)
The published (^) figures for diffusing capacity for carbon monoxide (^) (D,O) in
normal human lungs show a wide scatter even for the same method. Minimum,
maximum and mean values for six or more normal, resting, seated subjects
by the single-breath (^) D,0 method have been (^) recently reported as (^) follows:
11-0, 37 5, 24-0 (Ogilvie, Forster, Blakemore & Morton, 1957); 12-0, 220,
16-8 (Bates & Pearce, 1956); 18-3, 41-9, 27-9 (Lewis, Lin, Noe & Komisaruk,
1958); 14-5, 30-1, 23-9 (Curtis, Bauer, Loomans & Rasmussen, 1958); 21-5,
37.3, 30-2 (Marks, Cugell, Cadigan & (^) Gaensler, 1957); 14-4, 33-6, 24-8 (Shep-
hard, 1958); 18-0, 39-0, 30 0 (Forster, Roughton, Cander, Briscoe & Kreuzer,
1957). This variation is to some extent due to errors of method and minor
differences in technique. Since, however, (^) Dco estimates the capacity of (^) the
alveolar-capillary membrane of the lung as a whole to transfer CO, differences
in area of membrane between persons may account for a large proportion of
the total variation.
Few authors have made any attempt to allow for body size; some have
corrected by (^) expressing D,o values (^) per unit (^) surface area (e.g. Marks et al. 1957).
The true graph, however, is almost certainly curvilinear over the full range of
surface area and therefore greater accuracy can be obtained by applying a
regression equation. This has been done by Ogilvie et al. (1957) whose equation,
discussed below, is supported by published values which have been reviewed
by Forster (1957).
The effect of age on (^) Dco has also been reviewed by Forster (1957) who found
the evidence inconclusive, although Cohn, Carroll, Armstrong, Shephard &
Riley (1954) had reported a significant relationship between age, height and
maximal (^) Do2. Evidence is (^) presented in this (^) paper that (^) D,o depends on age as well as on
body size, but that the permeability, 'k' of Krogh (1915), which also depends
LUNG PERMEABILITY AND DIFFUSING CAPACITY 573
on age, appears to be independent of body size. On this account, and because
it is more reproducible and more easily measured, it appears to be a superior
index to (^) Dco.
In addition to the wide variation for the same method, most authors (Bates
& Pearce, 1956; Marks et al. 1957; Forster et al. 1957; Shephard, 1958),
have reported higher values, ranging from 12 to 62 (^) % more for breath-holding
than for steady-state Dc,o for the same subjects. With the aim, among others
of explaining this disagreement, the effect was investigated of changes in
lung volume on k and on (^) Dco. METHODS
The permeability and diffusing^ capacity^ were measured^ in^ thirty-nine normal male persons by^ the single-breath method of Krogh (1915) using a portable box-bag apparatus (Thomson, 1958). The accuracy of this apparatus appears to be at least as great as for non-portable forms. The single-breath D0o method and the instruments^ used^ for gas analysis^ have been described^ in full by Ogilvie et al. (1957). Briefly, the subject exhales fully into the box and then inspires the test-gas mixture (He 14; CO 0-28; 0, 20; (^) N2 65.72%) from the inspirate bag. The breath is then held for 10 sec and a rapid deep expiration is made, of which the^ first^ 750 ml.^ is^ discarded^ into the box and the remainder, the alveolar sample, is collected in the expirate bag. Analysis of expirate and inspirate is then made using a katharometer for He and an infra-red analyser for^ CO. The katharometer is sensitive to CO,, which must therefore be absorbed^ before^ He determina- tion. The C00-free gas then passes through the infra-red analyser. No correction is^ therefore required when calculating k since the correction factor appears in^ numerator^ and denominator (equation (1) below). In calculating^ D0o a correction factor of 5% has been allowed for oxygen consumption after correcting the inspired volume to^ S.T.P.D.^ It^ should also be^ noted^ that^ calibra- tion of the instruments is not critical when, as in^ these^ calculations, ratios and^ not^ absolute values are required. In all but three subjects the^ tests^ were^ made^ in^ duplicate and^ repeated if duplicates differed by more than 10 %. The subjects were^ invariably seated^ in^ this^ and^ the^ follow- ing trials. In eight persons we ascertained the effect on k and D00 of altering alveolar volume (VA) over the widest range possible, i.e. from total lung capacity to residual volume (RV) + 1200 ml., comprising 750 ml. required to wash out dead space and 450 ml. for^ analysis. Determinations were carried out in duplicate on successive days, as in the first study above, but according to a prearranged plan where (^) VA was randomized with time. On the rare occasions when more than two determinations were made on one day an interval of several hours separated the additional measurements, to allow blood carboxyhaemoglobin to return to within the normal range. Krogh's permeability (k) is the time constant of the exponential decay of CO concentration; during breath-holding. ECO IC from which k = t .In (^) F-= tI FIHe.FEW min
where FINCO = initial concentration of CO in the lungs after dilution but before absorption;
FIHe HEHe^ FkCO, F0co^ are^ respectively^ concentrations of^ He^ and^ CO^ in^ inspirate^ and^ expirate; t =^ time in minutes from beginning of inspiration to time at which expired gas sample was^ col- lected. The diffusing capacity was then found from DC°^ V'^ .k BAp 47 ml/min xmm^ Hg:^ (2) where B.P. =^ barometric (^) pressure in mm (^) Hg; V =^ lung volume in millilitres derived from a 37-
LUNG PERMEABILITY AND DIFFUSING CAPACITY 575
effect of smoking.) In normal adults k may therefore be predicted from
equation (4) or D'0 from equation (3). We have calculated the residual varia-
tion about the regression in both equations and found the coefficients of varia-
tion to be 18-3 (^) % for (^) D0o and 14-9 (^) % for k. Thus the predictive accuracy is
greater for k than for Dc,.
TABLE 1. (^) Lung function measurements and physical characteristics of thirty-nine male subjects vital Age (^) Weight Height SA capacity (^) VA Subject (yr) (kg) (^) (cm) (m2) (ml.) (ml.) 1 2 3 4 5 6 7 8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Mean
15 15 17 17 17 17 18 18 22 22 23 24 27 28 30 32 32 35 36 38 38 40 43 44 49 51 53 57 58 60 62 63 63 65 65 65 66 72 75 40-
44- 57- 75- 86- 75- 65- 79- 69- 66- 64- 75- 78- 66- 72- 79- 72- 68- 70- 93- 71- 59- 66- 79- 71- 58- 72- 68- 49- 52- 70- 88- 77- 82- 64- 54- 46- 54- 74-
155 158 180 174 170 168 186 i 178 177 186 174 177 177 177 160 175 175 170 188 177 177 178 184 165 170 164 173 172 174 183 170 184 170 158 165 173 165
1- 1- 1- 1- 1- 1- 2- 1- 1- 1- 2- 1- 1- 1- 1- 1- 1- 1- 1- 2- 1- 1- 1- 1- 1- 1- 1- 1- 1- 1- 1- 2- 1- 2- 1- 1- 1- 1- 1- 1-
2750 3800 3250 3950 4280 5520 4400 5360 4800 5310 4600 5430 4900 6040 4800 5620 4200 5010 4900 6330 5070 6680 4950 5700 4250 4820 3900 5050 4180 5460 3450 4600 4100 5270 3750 4660 4050* 5820* 3500 4730 4700 6840 3550 5830 3650 5260 4300 5630 4950 6880 3650 4780 (^4300 ) 2800 3880 3150 4270 3600 4840 2450* 3960* 2800* 4540* 3800 6260 3150 4540 3350 5260 2400 3740 3530 5210 2550 4440 2500 4210 3827 5161
RV' Mean k Mean D', (ml./
(ml.) (min-) min x mm Hg) 1080 5-14 24- 780 4-31 21- 1360 4-25 30- 1060 5-40 37- 640 5-57 (^) 41- 980 4-96 37- (^780) 5-50 42- 1260 5-89 42- 900 6-34 40- 1530 5-20 42- 1650 5-56 47- 1480 4-66 34- 840 4-63 28- 1220 4-78 31- 1430 3-75 26- 1170 4.44 26- 1320 4-96 33* 1340 4-05 26- 1770* 4-11* 26-80* 1450 4-52 27- 2290 5-24 45- 2400 3-20 23- 1680 3-27 22- 1480 4-25 33- 2230 3 39 32- 1200 3-66 22- 1840 4.26 32- (^1270) 4-38 22- (^1370) 3-47 18- 1700 2-70 (^) 17- 1510* 4.15* (^) 20- 1760* 3.20* 18-83* 2580 2-89 24- 2120 3-65 23- 1970 4-15 28- 1820 4-25 20- 1850 2-26 15- 1920 3-07 17- 1700 3-20 16- 1506 4-27 28- The formsV' (mean alveolar volume) (^) RV' (mean (^) residual volume) and D'0 (^) (diffusing capacity) are used to indicate that these values are based on the (^) method of (^) single-breath He dilution.
- (^) Single observations. The (^) effect on k and (^) D0o of changing VA The above results have been obtained for different subjects, measured at maximum V,. Figure 3 shows the effect on k and D' of (^) deliberately changing
576 MARGARET W. McGRATH AND M. L. THOMSON
I E x C E E
i^8
0 10 20 30 40 50 60 70 80 90 Age (yr) Fig. 1. (^) Relationship between diffusing capacity, age and body surface area for (^) thirty-nine male subjects. SA, 2-1-1-9 (^) m2, C]; 1-9-1-7 (^) m2, A; 1-7-1-5 (^) m2, 0; 1-5-1-3 (^) m2, v. Solid (^) signs indicate smokers.
0 6 - v
o 5 0 00
4 C 4~~~~~~~~~~ 0~~~
3
0
2
10 20 30 40 50 60 70 80 Age (yr) Fig. 2. Relationship between^ permeability (k) and^ age for^ thirty-nine male^ subjects. Single observations, O; mean^ of^ two^ or more^ observations, 0. Solid^ signs indicate smokers.
578 MARGARET W. MCGRATH AND M. L. THOMSON
been drawn in Fig. 3 (lower right) together with the corresponding (Dc0, Vi), graphs (^) (D,o =^ k V'J713).^ There^ is^ theoretical^ justification,^ discussed^ below,
for the heavy-line graph which, to the left of v, has the reciprocal form
lc-3 = 6/(V2 -1), and to the right of v, the form k-3 =^ 1-38 (VA-1)*.
The linear graph k = -15V + 12, in the first phase, produces the some-
what unexpected inverted U-shaped D' graph, similar to that^ found^ in nos. 3
and 20 of Fig. 3. The reciprocal form has been prolonged to the right of v:
this appears to fit best the graphs of subject no. 9 in this range.
From a practical point of view measurements^ of k will give best^ reproduci-
bility at maximum VA; a slight departure from maximum should not intro-
duce serious error.
The form of the (^) (D'0, VA) graphs is more variable than that of the (k, VA)
graphs because the conversion involves the^ direct^ factor VA which^ is^ always
greater than unity. The curve to the left of v generally rises with V ; that^ to
the right of^ v rises more rapidly. By^ contrast^ with^ k,^ therefore,^ dependence^ of
D 0 measurements on the achievement of a maximum inspiration is^ critical.
The general form is in agreement with Shephard's (1958)^ results in two
subjects.
The scatter about the mean value of points within 1 1.^ of the maximum^ V,
attained for each subject has been calculated for k and (^) D'o so as to compare the
effect on the accuracy of measurement of these indices of^ failing^ to^ secure
maximum inspiration. Mean coefficients of variation for eight subjects were
4.6% and 8.6% for k and Do0 respectively. Although the number^ of^ points was small these were, without exception, higher for (^) D'o than for k and the
difference was, therefore, highly significant.
The eight subjects have been arranged in Fig. 3 in order of increasing age
from no. 3 to no. 35. Effects which appear to be attributable^ to^ age are,
diminution of volume range, a decline in the reproducibility of the results and,
as has already been shown in the group of thirty-nine subjects, a^ fall^ in^ k^ at
maximum VA.
Marshall (1958) showed in one subject that (^) D,0 by single-breath method^ at
VA =^ total^ lung^ capacity was^ higher than^ at^ VI^ =^ functional^ residual
capacity and concluded that the difference between steady-state and^ single-
breath (^) D,0 was due to the different lung volumes at which the tests were carried out. The mean (^) Do, of our eight subjects at^ total^ lung capacity was
34-5 and at estimated functional residual capacity was^ 26-3^ ml./ min^ x^ mm
Hg. The mean increase (31 %) is close^ enough^ to^ that of the^ values^ reported
above in the introduction (39 (^) %) to confirm this explanation.
LUNG PERMEABILITY AND DIFFUSING CAPACITY 579
DISCUSSION There is fairly good agreement between the rise of (^) D,0 with surface area given
here and that given by Ogilvie et al. (1957), where they predict Dco from sur-
face area only
Dco =^ 18-85^ x^ SA-6-8.
The regression of (^) Dco on SA from our results becomes
Dco =^ 27-12 x SA-20-26.
The agreement is better if the results of three children (aged 8, 10 and 10)
in the values of Ogilvie et al. (1957) are excluded.
Cohn et al. (1954) found a significant effect of age and height on maximal
(exercise) diffusing capacity for oxygen,
D°2 = 0-67 x Height-0.55 x Age-40 9. (5)
Since the membrane offers only about half the total resistance, as has been
shown by Roughton (1945) and Roughton & Forster (1957), the factor
D°2 =^ 1-23 Dco should not be used, and on this and other grounds the co-
efficients in equation (5) would not be expected to agree closely with those
reported in this study in equation (3).
Estimating diffusing capacity of the diseased (^) lung
On one point all authors seem to agree, that multibreath and not single-
breath VA should be used in calculating breath-holding (^) D,O. The rationale
for this is not at first sight clear. Alveolar volume by multibreath method will
exceed that obtained by single-breath He dilution by a quantity which will
depend on^ the uniformity of^ ventilation. For simplicity this may be visual-
ized as a separate volume X. The membrane of the extra volume X is avail-
able for diffusion if ventilation and perfusion could be improved and is
probably of some respiratory value, even with the existing ventilatory defect,
under the more natural, steady-state conditions. The existing practice would
make allowance for this membrane as though its permeability were equal to
that of the well-ventilated space, i.e. the measured k. (Incidentally the volume
of the anatomical dead space has been similarly credited in the modified
method (Forster et al. 1955; Ogilvie et at. 1957)). Whether this is so will depend
on the nature of the abnormality. Thus (^) D,o based on VA +X is likely to be a
maximum and on VA a minimum.
In the same circumstances k would estimate a mean permeability per unit
area of the well-ventilated space VA. This might well be a more useful index
than (^) D,0 in the abnormal lung. If, however, the total capacity to transfer gas is required, the clinician may use, for calculating (^) D,o, V[ together with that
fraction of X which is most applicable having regard to the nature of the
LUNG PERMEABILITY AND DIFFUSING CAPACITY 581
of the first phase. It can be shown that when a spherical alveolus expands in
such a way that the volume of the investing membrane remains constant,
the reduction in thickness more than compensates for the decrease in k
associated with decrease in area/unit volume. The net result is that k increases
approximately with the radius (r), or the cube root of the volume, where
membrane thickness is small compared with the radius. Thus the (k, VA)
relationship for lung volumes-greater than v has been represented in Fig. 3
by one of the family of curves
ki-C2 =^ C3(VA-C)C^ (6)
since roc (VA- C1)k. Here C3 is a constant equal to PA/H (v - C1),, found by
substituting in (6) the k, VA values common to both curve phases, i.e. at
VA =^ V.
It should be pointed out that k, like D0,, depends on gas uptake in the blood
as well as on transfer across the membrane. The above discussion is limited to
membrane effects. In general the investigation supports the theory that in any
individual there is a specific lung volume (about 4-51. for male adults) below
which the alveolar membrane collapses, its physical condition remaining
unaltered, and above which the membrane stretches.
SUMMARY
1. In thirty-nine normal males aged 15-75 there was a highly significant fall
in alveolar-capillary permeability with age (kCo = -0-038 Age + 5-8). A highly
significant multiple regression was also obtained between age, surface area
and D,o, based on single-breath helium-dilution lung capacity determination
(D,o =^ - 0-29^ Age+^24 SA-^3 4).^ Even when^ corrected^ for^ surface^ area^ the
predictive accuracy of D,0 was less than that of k.
2. In eight subjects at lung volumes (VA) up to 11. from maximum the
(keo, VA)^ relationship^ appears^ to^ be^ reciprocal, consistent with^ constant^ area,
thickness and specific permeability of the membrane.
3. Within 1 1. of maximum k1o was relatively independent of VA, as would
occur if the membrane were stretching: by contrast, D,o was critically depend-
ent on VA over this range. This adversely affects the precision of D,o deter-
mination. In calculating D,o, VA appears as a direct multiplier and therefore
the (D,o, VA) relationship shows a more variable form than does (k,0, VA).
4. In addition to being independent of body size and relatively reproducible
at maximum VA, the use of k has the advantage that it avoids the measure-
ment of lung volume with the laborious corrections for temperature, pressure
and oxygen consumption. It is concluded that k is a better index than D,o.
We are very grateful to Professor G. P. Crowden and to Dr P. (^) Armitage for (^) advice, to the staff of the School who acted as subjects, and to the University of London Central Research Fund for a (^) grant for apparatus.
582 MARGARET W. McGRATH AND M. L. THOMSON
REFERENCES BATES, D. V. & CHRISTIE, R. V. (1950). Intrapulmonary mixing of helium in health and in emphysema. Clin. Sci. 9, 17-27. BATES, D. V. & PEARCE, J. F. (1956). The pulmonary diffusing capacity; a comparison of methods of measurement and a study of the effect of body position. J. Physiol. 132, 232-238. B0JE, 0. (1933). tber die Grosse der Lungen^ diffusion des Menschen wahrend^ Ruhe^ und^ korper- liche Arbeit. Arbeitsphysiologie, 7, 157-166. CoHN, J. E., CARROLL, D. G., ARMSTRONG, B. W., SHEPARI), R. H. &^ RILEY, R. L. (1954). Maxi- mal diffusing capacity of the lung in normal male subjects of different ages. J. appl. Physiol. 6, 588-597. CURTIS, J. K., BAUER, H., LOOMANS, S. & RASMUSSEN, H. K. (1958). Pulmonary diffusing capa- city studies 2. Clinical results using a modified carbon monoxide breathholding technique. Amer. J. med. Sci. 236, 57-64. FORSTER, R. E. (1957). Exchange of gases between alveolar air and pulmonary capillary blood: pulmonary diffusing capacity. Physiol. Rev. 37, 391-452. FORSTER, R. E., CoHN, J. E., BRISCOE, W. A., BLAKEMORE, W. S. & RILEY, R. L. (1955). A modi- fication of the Krogh carbon monoxide breath holding technique for estimating the diffusing capacity of the lung; a comparison with three other methods. J. clin. Invest. 34, 1417-1426. FoRsTzR, R. E., ROUGHTON, F. J. W., CANDER, L., BRISCOE, W. A. &^ KREuZER, F. (1957). Ap- parent pulmonary diffusing capacity for CO at varying alveolar O2 tensions. J. appl. Physiol. 11, 277-289. KROGH, M. (1915). The diffusion of gases through the lungs of man. J. Physiol. 49, 271-300. LEWiS, B. M., LIN, T., NOE, F. E. & KOMISARUK, R. (1958). The measurement of pulmonary capil- lary blood volume and pulmonary membrane diffusing capacity in normal subjects; the effects of exercise and position. J. clin. Invest. 37, 1061-1070. MACKLIN, W. H. & MACKLIN, C. C. (1944). The size of human lung alveoli. Trans. Roy. Soc. Can. 38, 63-82. MARKS, A., CUGELL, D. W., CADIGAN, J. B. & GAENSLER, E. A. (1957). Clinical determination of the diffusion capacity of the lungs. Amer. J. Med. 22, 51-73. MARsTrATLT, R. (1958). Methods of measuring pulmonary diffusing capacity and their significance. Proc. R. Soc. Med. 51, 101-104. OGILVIE, C. M., FORSTER, R. E., BLAKEMORE, W. S. & MORTON, J. W. (1957). A standardized breath holding technique for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. J. clin. Invest. 36, 1-17.
ROUGHTON, F. J. W. (1945). The kinetics of the reaction CO + 2Hb:O± 02 +COHb in human
blood at body temperature. Amer. J. Physiol. 143, 609-620. ROUGHTON, F. J. W. &^ FORSTER, R. E. (1957). Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung. J. appl. Physiol. 11, 290-302. SHEPHARD, R. J. (1958). 'Breath-holding' measurement of carbon monoxide diffusing capacity. Comparison of a^ field test with steady-state and other methods of measurement. J. Physiol. 141, 408-419. THOMSON, M. L. (1958). A portable apparatus for spirometry and for measuring timed vital capacity and diffusing capacity of the human lung. J. Physiol. 142, 17-19P. WILSON, R. H., EVANS, R. L., JOHNSON, R. S. & DEMPSEY, M. E. (1954). An estimation of the effective alveolar respiratory surface and other pulmonary properties in^ normal persons. Amer. Rev. Tuberc. 70, 296-303.
