| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Medicine 0931, University of California, San Diego, La Jolla, California 92093
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The analysis of the
gas in a single expirate has long been used to estimate the degree of
ventilation-perfusion (
A/
) inequality in the
lung. To further validate this estimate, we examined three measures of
A/ inhomogeneity calculated from a single
full exhalation in nine anesthetized mongrel dogs under control
conditions and after exposure to aerosolized methacholine. These
measurements were then compared with arterial blood gases and with
measurements of A/ inhomogeneity obtained
using the multiple inert gas elimination technique. The slope of the
instantaneous respiratory exchange ratio (R slope) vs. expired volume
was poorly correlated with independent measures, probably because of
the curvilinear nature of the relationship due to continuing gas
exchange. When R was converted to the intrabreath
A/ (i/), the best index was the
slope of i/ vs. volume over phase III (i/
slope). This was strongly correlated with independent measures,
especially those relating to inhomogeneity of perfusion. The
correlations for i/ slope and R slope considerably
improved when only the first half of phase III was considered. We
conclude that a useful noninvasive measurement of
A/ inhomogeneity can be derived from the
intrabreath respiratory exchange ratio.
single-breath tests; respiratory exchange ratio
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
THE ANALYSIS OF THE
GAS in a single expirate has long been used to estimate the
degree of inequality in the ventilation-to-perfusion ratio
(A/) in the lung. West and colleagues
(10) showed that an index of A/
inequality could be obtained in the lung by calculating the change in
respiratory quotient in expired gas, provided mixed venous blood gas
composition was assumed to be constant during the expiration. Later,
Guy et al. (3) used computerized data acquisition to allow
the calculation of the intrabreath respiratory exchange ratio (R) over
the course of a single vital capacity expiration. In addition, the
resulting plot of intrabreath R could then be compared with the
theoretical behavior of a perfectly mixed lung model to provide an
index of A/ inhomogeneity that was less
sensitive to ongoing gas exchange. As shown by West et al.
(10), proportional changes in respiratory quotient could
be converted to proportional changes in A/ without knowledge of the exact mixed venous blood composition if this
remained the same.
The technique has been applied to provide an index of the range of
A/ in the lung. For example, Prisk et al.
(5) used this technique to measure the range of
A/ in subjects exposed to periods of
weightlessness during spaceflight in Spacelab, and at the present time
this technique is also in use on the International Space Station.
Recently, Cremona et al. (1) used plots of intrabreath R
to determine closing volume in subjects who may have high-altitude pulmonary edema, but they did not extend the measurements to
determination of the range of A/ in these subjects.
However, despite occasional use, the technique has never been validated
against other known techniques for measuring
A/ inequality in the lung. We provide the
first such validation by measuring the intrabreath
A/ range (i/) in anesthetized dogs under normal conditions and after the inhalation of methacholine. The noninvasive measurements of intrabreath R and i/,
which are comparable to the previous method of West et al.
(10), are compared with the range of
A/ in the same animals determined using
arterial blood gases and also the multiple inert gas elimination technique (MIGET).
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Experimental details.
The study was approved by the University of California, San Diego,
Animal Subjects Committee. Nine mongrel dogs [18-24 (mean 20.6)
kg body wt] were anesthetized with pentobarbital sodium (30 mg/kg iv)
and paralyzed with pancuronium bromide (0.1 mg/kg iv). The level of
anesthesia and relaxation was maintained by incremental administration
of both drugs. A cuffed endotracheal tube (9 mm ID) was placed through
a tracheostomy. Normal arterial blood gas tensions were maintained by
adjusting the frequency of a Harvard mechanical ventilator set at a
tidal volume of 15 ml/kg. A 7-Fr Swan-Ganz catheter was inserted via
the right external jugular vein and advanced into the pulmonary artery
by using direct pressure monitoring. The femoral artery was cannulated
for sampling arterial blood (Fig. 1).
|
Methacholine administration. We used aerosolized methacholine to induce bronchoconstriction and, thus, alter the distribution of ventilation. A 1% methacholine solution was aerosolized using a jet nebulizer (Acorn II, Marquest Medical Products) and administered via the ventilator circuit for 1 min. During this time, the respiratory rate was adjusted to maintain a normal end-tidal CO2. A short period was then allowed to obtain a new quasi-steady state, and the measurements after methacholine administration were performed. On average, postmethacholine measurements were made ~27 min after methacholine administration.
MIGET. MIGET was applied as described by Gale et al. (2). The inert gas solution (SF6, ethane, cyclopropane, enflurane, ether, and acetone) was prepared in 5% dextrose (2) and infused for ~20 min at ~10 ml/min before collection of the samples. The total volume of fluid infused over the course of the study (1-2 h) was ~1 liter.
Quadruplicate 15-ml samples of mixed expired gas and duplicate 6-ml samples of pulmonary and systemic arterial blood were obtained using gastight syringes under normal conditions (before the i/
measurements) and after the stabilization subsequent to methacholine
administration (just before the i/
measurements). These samples were used to measure the steady-state
concentrations of the six inert gases using a gas chromatograph (model
5890A, Hewlett-Packard, Wilmington, DE) (9).
A/ distributions were calculated by using
MIGET in the usual fashion. Solubilities, retentions (the ratio of
arterial to mixed venous partial pressures), and excretions (the ratio
of mixed expired to mixed venous partial pressure) for the inert gases
were determined and corrected for body temperature, and
A/ distributions were calculated from the
inert gas data (8, 9). The second moment of the perfusion distribution exclusive of intrapulmonary shunt (SD) and the second moment of the ventilation distribution exclusive of dead
space (SD) were used as indicators of the degree of
A/ inequality. Dispersion of retention (DispR), excretion (DispE), and retention minus
excretion (DispR 
E) were derived directly from the
retention and excretion data (2). The residual sum of
squares was used as an indicator of the adequacy of fit of the data to
the 50-compartment model of the lung (9).
Blood gas measurements. Arterial samples (2 ml) were collected immediately after each inert gas sample and kept on ice until analyzed for PO2, PCO2, and pH using a blood gas analyzer (model IL-1306, Instrumentation Laboratories, Lexington, MA). Alveolar-arterial PO2 difference was determined using the ideal alveolar PO2 calculated using the alveolar gas equation and measured arterial PCO2 (PaCO2) and R.
i/ test maneuver.
For each test, the dogs were ventilated at a constant rate and tidal
volume until stable end-tidal gas concentrations were obtained. Data
were then collected for 
60 s to provide a measurement of
O2 consumption and CO2 production. The
technique described by Tomioka et al. (7) was used. At the
end of a normal expiration, the airway was connected to a
constant-pressure source of air at +80 cmH2O pressure, and
the animal was inflated to full lung volume. Inflation was via a flow
restrictor limiting flow to ~0.1 l/s. On reaching maximum volume, the
airway was connected to a constant-pressure source of 
80
cmH2O and deflated (again via the flow restrictor at ~0.1
l/s) to minimum lung volume. The animal was then returned to normal ventilation.
i/ analysis.
The slow vital capacity expiration was analyzed using the techniques
described by Guy et al. (3). Gas concentration data were
converted on a point-by-point basis to R using the standard alveolar
gas equation (3)
|
A/ inequality.
|
).
Appropriate choices for residual volume, , lung tissue volume
and ventilation, and gas exchange from the preceding period of quiet
breathing are made (3). Errors made in the choice of these
variables have only modest effects on the output. For example, a large
(10 Torr) error in the mixed venous PCO2
results in only a 5% error in the resulting variation in
A/ (3).
On the basis of these initial conditions, the mixed venous and alveolar
points corresponding to that A/ are
calculated using the Kelman routines (11). The model then
calculates the instantaneous R as a function of expired volume for the
same volume history as the animal. To produce a family of R isopleths,
the simulation is repeated varying (and, therefore,
A/) from its initial estimate in 10% steps
up and down. These R isopleths form a scale for
A/, inasmuch as each individual isopleth describes the behavior of a lung, free of A/
inequality, following the same lung volume history as the animal.
In this way, the measured R curve is transformed to an i/
curve by calculating the point-by-point i/ from linear
interpolation between the family of R isopleths calculated for
different A/ values. This has the advantage
of providing straight-horizontal isopleths of i/ vs.
expired volume as opposed to the curvilinear R isopleths. Figure 2
shows a sample intrabreath R curve, the isopleths of
A/, and the resultant i/ curve.
From the plot of i/ as a function of expired volume (Fig.
3), we identified the onset of phase IV
(airway closure) and measured the slope of the curve using linear
least-squares regression as a function of expired volume
(i/ slope). We determined the range of i/
(i/ range) by measuring the maximum differences in
i/ over the portion of the exhalation corresponding to phase III. Thus i/ range included any excursions in
i/ beyond that of slope itself caused by cardiogenic
oscillations and by any deviation from the least-squares fitted line.
Figure 3 illustrates these parameters. Additionally, using the
intrabreath R curve (Fig. 2), we measured the slope of R as a function
of expired volume (R slope) over the same lung volume range and also used this as a possible index of inhomogeneity of gas exchange, reasoning that, although the isopleths were curvilinear, an index based
on this curve alone would be easier to calculate than the i/-based measurements.
|
Statistical techniques. Values are means ± SE. Two-way ANOVA was performed on the results where appropriate, and in cases with significant F ratios, post hoc pairwise comparisons were made using the Bonferroni adjustment (Systat version 5). To compare responses, linear regression was used (Excel 2002, Microsoft, Redmond, WA). Significance was accepted at P < 0.05.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Gas exchange.
During control measurements performed before methacholine
administration, arterial blood gases were essentially normal, with PaCO2 of 34.5 Torr and arterial
PO2 (PaO2) of 87 Torr
(inspired O2 fraction = 0.21 in all cases).
Methacholine administration worsened gas exchange:
PaCO2 of 36.2 Torr and PaO2 of 63 Torr (Table 1). Alveolar-arterial
PO2 difference increased from 23 to 40 Torr
after methacholine administration. Table 1 lists the average values for
the gas exchange variables and other measurements for the control
period and after methacholine administration.
|
Multiple inert gas variables.
Methacholine administration resulted in a significant widening of the
A/ distribution (Table 1), with
SD increasing from 0.64 to 1.33 (Fig.
4). There were comparable increases in
DispR, DispE, and DispR E.
|
Expired gas variables.
Neither i/ range nor R slope changed significantly as a
result of methacholine administration (Table 1). i/ slope became steeper (more negative) as a result of methacholine
administration, changing from 0.092 ± 0.021 to 0.245 ± 0.089 (P < 0.10; cf. Fig. 3 with Fig.
5).
|
/ as a function of volume, we
then considered R slope and i/ slope over the first and second halves of phase III by dividing that portion of the exhalation into two halves and repeating the slope measurements (Fig. 5). When the
breath divided into the first and second halves of phase III was
considered, the effect of methacholine on i/ slope was
much greater over the first than over the second half of the breath.
Over the first half of phase III, i/ slope significantly steepened from 0.020 ± 0.025 to 0.379 ± 0.108 (P < 0.05); over the second half of the breath,
i/ slope changed from 0.149 ± 0.034 to
0.122 ± 0.076 (not significant). In contrast, R slope over the
first half of the exhalation became considerably steeper after
methacholine administration (0.014 ± 0.017 vs. 0.190 ± 0.033, P < 0.05), but R slope over the second half of
the breath was considerably flatter after exposure to methacholine
(0.325 ± 0.061 vs. 0.180 ± 0.028, P < 0.05). The major changes are shown in Fig.
6.
|
Correlations between intrabreath R, i/, and other
measures of gas exchange.
i/ slope correlated strongly with several measures of gas
exchange and A/ distribution. Table
2 lists the correlation coefficients
between i/ and blood gas and inert gas variables. i/ slope was most strongly correlated with blood gases
and with the indexes of MIGET associated with pulmonary perfusion, DispR and SD.
|
/ slope over the first half of the
breath correlates strongly with i/ slope over the entire
breath (r2 = 0.85, P < 0.05). However, i/ slope over the first half of the
breath is much less strongly correlated with i/ slope over the second half of the breath (r2 = 0.28), supporting the observation of significant nonlinearities in
i/ slope. The change in i/ slope over the
first half of phase III was the most strongly correlated with all other
variables measured in these dogs, showing significant correlations with all measures of A/ inhomogeneity,
with the exception of SD. Correlations with
i/ slope over the second half of the breath were much
weaker (Table 2).
Similarly, R slope over the first half of phase III was strongly
correlated with other measures of gas exchange. In general, R slope
over the first half of the breath correlated with the same variables as
did i/ slope over the first half of the breath, and the
correlations were of similar strength (Table 2). As was the case with
i/ slope, R slope over the second half of the breath was
poorly correlated with other measures of gas exchange.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
i/ slope.
The correlations between i/ slope and independent
measures of A/ inhomogeneity before and after
methacholine administration (Table 2) suggest that i/
slope provides a useful index of gas exchange disruption. In
particular, in this circumstance, it appears that the administration of
aerosolized methacholine resulted in different behavior of the lung
when the gas expired from the lung in the first half of the breath was
compared with the gas expired from the lung in the second half of the
breath. When we considered i/ slope calculated from only
the first half of the exhalation, we obtained better correlations with
independent measures of A/ inhomogeneity
(Table 2). Although calculation of i/ slope is somewhat
complex, i/ slope has the advantage of being a completely
noninvasive measurement and relatively simple to perform experimentally.
/ are the
concentrations of O2 and CO2 in the mixed
venous blood entering the lungs. We used an estimate of these based on
O2 consumption, CO2 production, and assumed
(the Fick principle). West et al. (10) showed that
although the value of i/ corresponding to a particular
value of R is sensitive to the composition of the mixed venous blood,
the fractional change in calculated i/ is very
insensitive. Errors in O2 and CO2 concentration
in opposite directions (such as those arising from an error in
metabolic rate or ) result in very small errors in
i/. A 10% error of this nature results in changes in
i/ of <1%, which, when coupled with the previous
observations of West et al., shows that conversion from R to
i/ is robust.
Effect of methacholine.
We used methacholine to induce bronchoconstriction and, presumably,
uneven ventilation, thus disrupting gas exchange. We wished to induce
gas exchange lesions of varying intensity to test the usefulness of the
change in i/ slope over a range of gas exchange defect
severity. PaO2 was between 93 and 44 Torr (between 93 and 69 Torr for control condition and between 83 and 44 Torr after methacholine), which shows that we were successful in inducing a range
of severity of the gas exchange defect between different animals.
Similarly, calculated alveolar-arterial PO2
difference rose from 23 to 40 Torr after methacholine administration.
That i/ slope (and especially i/ slope over
the first half of the exhalation) correlates well with the change in
PaO2 and with alveolar-arterial
PO2 difference (Table 2) is a good indication that the measurement is a useful noninvasive means by which to determine the severity of a gas exchange defect. The change in the
distribution of A/ distributions from MIGET
(Fig. 4) is similar to that reported in previous studies. In
particular, an increase in the inhomogeneity of ventilation resulting
from bronchoconstriction in dogs results in a widening of the
distribution of perfusion (an increase in SD)
through the appearance of a low A/ mode
(6).
Intrabreath changes in i/.
There were significant differences in i/ slope measured
over the first and second halves of phase III of the vital capacity expiration, in particular, in those tests after methacholine
administration (Table 1). Although it might be tempting to compare the
first and second halves of the breath, such a comparison is almost
certainly invalid when attempting to compare the data before and after
methacholine administration. Although after methacholine administration
the portion of the lung that empties first exhibits a greater degree of
A/ inhomogeneity than does the portion that
empties last, there is no means to determine whether these correspond
to the same lung regions measured before methacholine administration.
A/ inequality in
the entire lung, it is important to consider data only from phase III.
The onset of phase IV represents the point at which airway closure occurs somewhere in the lung. Thus, after that point, exhaled gas is a
reflection of only those regions still contributing to the exhalate.
However, it is not generally possible to determine which lung regions
close, especially in abnormal lungs, such as those exposed to
methacholine. Although the i/ data from phase IV may
provide insight into other aspects of A/
inhomogeneity, they are likely not useful in comparing i/
with whole lung measures such as MIGET.
It is important to recognize that the model used to generate the R
isopleths, from which i/ is calculated, is not intended to represent an actual lung. Rather, it is used to provide a series of
reference lines describing how a perfect lung, following the same lung
volume history as the test lung, would behave. The data from the test
lung are then interpreted using this scale "as if" the test
lung were a collection of such perfect lungs following the same volume
history. Although the gas arising from a particular lung unit with a
given A/ may well have significant phase III
slopes of O2 and CO2, there is no corresponding
effect on the phase III slope for R (and, hence, for i/).
This is because, on the O2-CO2 diagram, the
value of R is the slope of the line joining the inspired point and the
alveolar point. Thus any factor that contributes to variable mixtures
from these sources produces no change in R (the obvious example being
the lack of effect of dead space admixture, which has a big effect on
the O2 and CO2 trace, especially in the first
half of expiration). Thus, with changes in R (and i/), we
are measuring the effects of different lung regions with different
A/ ratios emptying at different points in the
breath. The causes of such differences in emptying cannot be inferred
from these measurements. However, it is clear from these data, and from
those of West et al. (10) and Guy et al. (3),
that the patterns that result are strikingly different between normal
and abnormal lungs.
Using R slope instead of i/ slope.
We also attempted to determine whether the slope of the R plot vs.
expired volume (R slope) could be used instead of the computationally more demanding i/ plot. As the data in Table 2
illustrate, there were virtually no significant correlations between R
slope and the independent measures of A/
inhomogeneity, nor did R slope change as a result of methacholine
administration (Table 1). We were not entirely surprised by this
finding. As Fig. 2 illustrates, even a perfectly mixed lung (the model
lines in Fig. 2) produces curvilinear R plots. Attempting to measure a
slope from a curvilinear plot such as that shown in Fig. 2
introduces considerable error into the resulting slope unrelated to a
A/ defect. This is reflected in the poor
correlation performance (Table 2).
/ slope measured over the first half of phase
III. This measurement may provide a more easily calculated surrogate to
i/ slope, albeit at the expense of not having a scale
that is as easily interpretable. The degree of curvilinearity [which
results from continuing gas exchange as the exhalation proceeds
(3, 4)] depends largely on lung volume and expired flow
rate, with smaller lung volumes and lower flow rates exhibiting greater
curvilinearity. These effects are smallest over the first half of phase
III. Thus measurements of R slope over the first half of phase III in
adult humans in which exhalation is kept to less than ~10 s may be
less prone to the curvature artifact. In that case, a stronger
correlation between the independent measures of
A/ inhomogeneity and R slope might result.
Other expired gas measurements of A/
inhomogeneity.
The most informative expired gas measurements of
A/ inhomogeneity in this study
were i/ slope and R slope over the first half of phase
III. We attempted to use the range of i/, which also
includes any contribution from cardiogenic oscillations. It might be
expected that the magnitude of the cardiogenic oscillations would be an
indicator of the range of A/
inhomogeneity in the lung. If this is the case, then it might be
thought that including this component in the measurement by measuring
the range (as opposed to the slope, which averages out any component
due to cardiogenic oscillations) would improve the noninvasive
estimation of A/ inhomogeneity. However, on
the basis of the much less robust correlations in Table 2, this appears
not to be the case.
/ slope provide a useful index of the
degree of A/ disruption in the lung. The
changes in i/ slope correlate well with more direct
measurements of A/ range obtained using MIGET
and also with changes in arterial blood gas variables. The degree of
correlation is improved when only gas from the first half of exhalation
is considered. Because i/ slope can readily be measured
noninvasively, this may prove useful in future studies of disruption to
gas exchange, especially in humans in remote environments where
noninvasive techniques are often desirable.
| |
ACKNOWLEDGEMENTS |
|---|
We acknowledge the excellent technical support of Jane Lindsay and Frank Lopez and thank Peter Wagner for useful discussions relating to these studies.
| |
FOOTNOTES |
|---|
This work was supported in part by National Aeronautics and Space Administration Contract NAS9-98124.
Present address of J. W. Reed: Dept. of Physiological Sciences, University of Newcastle Upon Tyne, Newcastle Upon Tyne, UK.
Address for reprint requests and other correspondence: G. K. Prisk, Dept. of Medicine, 0931, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0931 (E-mail: kprisk{at}ucsd.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published October 25, 2002;10.1152/japplphysiol.00662.2002
Received 19 July 2002; accepted in final form 14 October 2002.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Cremona, G,
Asnaghi R,
Baderna P,
Brunetto A,
Brutsaert T,
Cavallaro C,
Clark TM,
Cogo A,
Donis R,
Lanfranchi P,
Luks A,
Novello N,
Panzetta S,
Perini L,
Putnam M,
Spagnolatti L,
Wagner H,
and
Wagner PD.
Pulmonary extravascular fluid accumulation in recreational climbers: a prospective study.
Lancet
359:
303-309,
2002[Web of Science][Medline].
2.
Gale, GE,
Torre-Bueno JR,
Moon RE,
Saltzman HA,
and
Wagner PD.
Ventilation-perfusion inequality in normal humans during exercise at sea level and simulated altitude.
J Appl Physiol
58:
978-988,
1985
3.
Guy, HJ,
Gaines RA,
Hill PM,
Wagner PD,
and
West JB.
Computerized noninvasive tests of lung function. A flexible approach using mass spectrometry.
Am Rev Respir Dis
113:
737-744,
1976[Web of Science][Medline].
4.
Meade, F,
Pearl N,
and
Saunders MJ.
Distribution of lung function (A/) in normal subjects deduced from changes in alveolar gas tensions during expiration.
Scand J Respir Dis
48:
354-365,
1967.
5.
Prisk, GK,
Elliott AR,
Guy HJB,
Kosonen JM,
and
West JB.
Pulmonary gas exchange and its determinants during sustained microgravity on Spacelabs SLS-1 and SLS-2.
J Appl Physiol
79:
1290-1298,
1995
6.
Rubinfeld, AR,
Wagner PD,
and
West JB.
Gas exchange during acute experimental canine asthma.
Am Rev Respir Dis
118:
525-536,
1978[Medline].
7.
Tomioka, S,
Kubo S,
Guy HJB,
and
Prisk GK.
Gravitational independence of single-breath washout tests in recumbent dogs.
J Appl Physiol
64:
642-648,
1988
8.
Wagner, PD,
Naumann PF,
and
Laravuso RB.
Simultaneous measurement of eight foreign gases in blood by gas chromatography.
J Appl Physiol
36:
600-605,
1974
9.
Wagner, PD,
Saltzman HA,
and
West JB.
Measurement of continuous distributions of ventilation-perfusion ratios: theory.
J Appl Physiol
36:
588-599,
1974
10.
West, JB,
Fowler KT,
Hugh-Jones P,
and
O'Donnell TV.
Measurement of the ventilation-perfusion ratio inequality in the lung by the analysis of a single expirate.
Clin Sci
16:
529-547,
1957[Web of Science][Medline].
11.
West, JB,
and
Wagner PD.
Pulmonary gas exchange.
In: Bioengineering Aspects of the Lung, edited by West JB.. New York: Dekker, 1977, p. 361-457. (Lung Biol. Health Dis. Ser.)
This article has been cited by other articles:
|
|
 |
|
  G. K. Prisk, J. M. Fine, T. K. Cooper, and J. B. West Vital capacity, respiratory muscle strength, and pulmonary gas exchange during long-duration exposure to microgravity J Appl Physiol, August 1, 2006; 101(2): 439 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. K. Prisk, J. M. Fine, T. K. Cooper, and J. B. West Pulmonary gas exchange is not impaired 24 h after extravehicular activity J Appl Physiol, December 1, 2005; 99(6): 2233 - 2238. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
