© 2002 American Thoracic Society
Effects of Increasing Carboxyhemoglobin on the Single Breath Carbon Monoxide Diffusing CapacityDivision of Respirology, Department of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada Correspondence and requests for reprints should be addressed to Brian L. Graham, Division of Respirology, Room 546 Ellis Hall, Department of Medicine, University of Saskatchewan, Saskatoon, SK, S7N OW8 Canada. E-mail: brian.graham{at}sk.lung.ca
Although carboxyhemoglobin (COHb) is often increased in smokers, American Thoracic Society recommendations for adjusting the single breath carbon monoxide (CO) diffusing capacity (DLCOSB) for COHb remain optional. Using a previously described 3-equation technique, we measured DLCOSB and an index of diffusion inhomogeneity (DI) in 10 healthy, nonsmoking subjects who performed DLCOSB maneuvers both before and after increasing COHb. CO backpressure (FACO) was measured from the exhaled gas of a standardized deep breath of room air that immediately preceded each DLCOSB and was validated by measurement of FACO from an identical "sham" maneuver without inhaling CO. Without adjustments for FACO or COHb, DLCOSB decreased with increasing COHb. This effect persisted when DLCOSB was adjusted only for FACO, but it was eliminated with further adjustment for the anemia effect of increasing COHb. The anemia adjustment was proportional to the fractional COHb. DI, adjusted for FACO, was unaffected by increasing COHb. We conclude that DLCOSB must be adjusted for both the buildup of CO backpressure and the anemia effect of increasing COHb. Adequate corrections of DLCOSB can be implemented using FACO measured during a standardized deep breath immediately preceding the DLCOSB maneuver. Current American Thoracic Society recommendations for DLCOSB standardization do not adequately compensate for COHb.
Key Words: pulmonary diffusing capacity • carbon monoxide • carboxyhemoglobin • smoking • diffusion index
Although the conventional single breath diffusing capacity (DLCOSB) test has been accepted as a standard noninvasive test to assess the integrity of the pulmonary vasculature, numerous pitfalls limit its utility (1). One factor is the effect of carbon monoxide (CO) in cigarette smoke, which raises carboxyhemoglobin (COHb) to as high as 10–15% in current smokers, whereas values are usually 1–2% in nonsmokers (2). However, most regression values for DLCOSB are derived from the study of groups of lifetime nonsmokers (3), so that the use of these regressions in smokers underestimates the percent of predicted DLCOSB values, unless additional adjustments are made for increases in COHb. Increasing COHb reduces DLCOSB because capillary [CO], assumed in conventional methods to be zero, is substantially increased in smokers, leading to increases in CO backpressure (FACO) in the alveolar gas at the commencement of DLCOSB maneuvers. The assumption that FACO is zero causes an overestimation of the driving pressure for CO across the air–blood barrier, leading to an underestimation of DLCOSB (4). In addition, with increasing COHb, some of the hemoglobin (Hb) becomes tightly bound by CO. Therefore, the overall amount of Hb available for further binding by the fresh CO from the test gas is decreased, and hence, there is less uptake of CO. This effect is similar to having a reduced Hb. Therefore, it is often termed the "anemia" effect (5–7). We previously described a more accurate and precise measurement of DLCOSB, using the 3-equation single breath method, which was relatively maneuver-independent and allowed more accurate measurement of both anatomic dead space and absolute lung volume (8–10). To minimize the effects of previous volume history (11, 12), each DLCOSB maneuver was immediately preceded by a standardized deep breath of room air (10). During the slow exhalation phase of the deep breath, we routinely measured the expired [CO] from a virtual sample of all of the alveolar gas and assumed this to be the FACO that would exist at the beginning of inspiration of the actual DLCOSB maneuver (10). In the calculation of DLCOSB, this measured FACO was used rather than assuming it to be zero. In a previous study in smokers with relatively normal lung function, we also described an index of diffusion inhomogeneity (DI) (10, 13), which correlated with pack years of smoking when adjusted for FACO (13), but the potential anemia effects of increasing COHb were not examined. The purpose of the present study was to validate the use of the direct measurement of FACO in the deep breath of room air immediately preceding each DLCOSB maneuver for the compensation of DLCOSB and DI for the effects of increasing CO backpressure. The acute effects of increasing COHb on both DLCOSB and DI were examined using maneuvers with varying inspired volumes and breath-hold times, which were performed both before and after deliberately increasing COHb by breathing CO-rich gas. For measurement of CO backpressure, the European Thoracic Society recommends a breath-hold time of 20 seconds for adequate equilibration of CO in capillary blood and alveolar gas. Standard DLCOSB tests use a breath-hold time of 10 seconds. The 3-equation method can be implemented with no breath holding (0 s). We used all three times, which cover the extremes of the single breath maneuver, to determine whether the COHb correction was dependent on breath-hold time. We examined the accuracy of FACO measurement obtained from a standardized deep breath by comparing it with FACO measured during an identical "sham" DLCOSB maneuver, with room air substituted for the diffusion test gas. After accounting for the effects of increasing FACO, we determined the residual anemia effect on DLCOSB by assuming that increases in COHb would not fundamentally alter the volume or the distribution of red cells in the pulmonary capillary bed, and therefore, DLCOSB, when appropriately adjusted, would not change with increasing COHb. To allow adjustment of DLCOSB for this effect, without directly measuring COHb, we also developed a correlation of the measured FACO with COHb.
Effect of CO backpressure on calculations of DLCOSB For conventional methods of calculating DLCOSB, the Krogh equation, as modified by Ogilvie and coworkers (14), was written as follows:
Therefore, Fco0 = (FcoIFTS)/FTI.
If the CO backpressure is not zero, the ambient alveolar CO will be increased, and in a steady state the mean alveolar CO partial pressure will approximately equal the mean intracapillary CO partial pressure. Using FACO as the measure of CO backpressure, the Krogh equation becomes:
Equation 2 becomes:
If FACO is known, then Equation 5 can be substituted into Equation 4 to allow for an analytic correction for the CO backpressure:
Equation 6 illustrates that if FACO is assumed to be zero when in fact it is not, DLCOSB will be underestimated. The amount of underestimation is a nonlinear function of the CO backpressure, the DLCOSB, the lung volumes, and flow rates. However, Equation 6 will be subject to the errors introduced by the assumption of instantaneous inspiration and expiration, which have been documented previously (8). The foregoing considerations apply to conventional methods of measuring DLCOSB. We used the 3-equation method to calculate DLCOSB, which analytically compensates for changes in CO uptake during inhalation and exhalation, rather than assuming that all CO uptake occurs during breath holding (10). In the 3-equation method, it is not necessary to assume that FACO is zero. The method uses a continuous calculation of the difference in CO partial pressure across the alveolar–capillary membrane, based on continuous measurements of air flow and gas concentrations throughout the maneuver. The CO driving pressure is reduced by the CO backpressure, which is measured just before the inhalation of test gas. Hence, the method analytically compensates for increases in FACO.
Effect of COHb on calculations of DLCOSB
Morphometric measurements of DLCO in dogs indicate that The other inherent assumption is that the COHb in mixed venous blood, [COHb]v, is zero. If instead, [COHb]v is measured and a more appropriate assumption is made, i.e., that the amount of Hb participating in gas exchange is reduced by the fraction of COHb, FCOHb, then: DLCO corrected = DLCO measured · [Hb] standard/([Hb] measured (1 - FCOHb)).
If an individual has [Hb] measured = [Hb] standard, then the correction would be:
This first-order correction is slightly more than increasing DLCO by 1% for every 1% increase in COHb. For 10% COHb, the correction should be 11%. However, if 1/Dm were not zero, the correction required for the presence of COHb would be less.
We studied 10 lifetime nonsmoking adults (nine males, one female). The mean (± 1 SD) age was 39 ± 8.5 years, height was 176 ± 6 cm, and weight was 79 ± 9 kg. All subjects were free of recent respiratory symptoms and had normal forced expired volume in 1 second (105 ± 10% predicted) and normal forced vital capacity (104 ± 12% predicted).We measured DLCOSB, DI, and indices of ventilation inhomogeneity using previously described techniques and equipment (10, 11, 20). Both before and after increasing COHb, four DLCOSB maneuvers were performed (Figure 1) in the same order approximately 10 minutes apart. Vital capacity single breath maneuvers consisted of slow inhalation from residual volume (RV) to total lung capacity (TLC), with either 0 or 20 seconds of breath holding and slow exhalation to RV. Submaximal maneuvers consisted of slow inhalation from functional residual capacity (FRC) to one-half inspiratory capacity (IC), with either 0 or 9 seconds of breath holding and slow exhalation to RV. All inspired and expired flows were maintained at 0.5 L/second. Each DLCOSB maneuver was immediately preceded by a standardized deep breath of room air (11, 21), consisting of slow inhalation to TLC, 5 seconds of breath holding, and slow exhalation to the prescribed lung volume of the ensuing DLCOSB maneuver (RV or FRC) (Figure 2) . Immediately before each of the four actual DLCOSB maneuvers, a "sham" maneuver (Figure 2) was also performed, which was identical to the actual DLCOSB maneuver, except that the subject inhaled room air rather than diffusion test gas containing CO.
For the sham maneuvers, FACO was measured using the mean CO concentration in the exhalation phase of the single breath maneuver corresponding to the alveolar gas that would have been analyzed in the actual DLCOSB maneuver and compared with FACO values measured during the standardized deep breath (Figure 2). COHb was increased by having subjects perform a series of 7–12 slow vital capacity inhalations of air containing 0.3% CO, with 20 seconds of breath holding at least 5 minutes apart. We monitored CO in the exhaled gas to assure that the target of approximately 50–80 ppm CO was achieved but not exceeded. Venous blood was drawn and COHb measured (CIBA Corning 2500 CO-oximeter, Medfield, MA) before starting the maneuvers, after completion of the baseline maneuvers, and after the second set of maneuvers with COHb. Linear regression analysis of FACO measurements from the standardized deep breath, observed nearest the time of venous sampling for COHb, was used to determine the relationship between increases in FACO in the alveolar gas and increases in COHb in venous blood. For each maneuver, DLCOSB was calculated first, without any adjustment for FACO or COHb; second, with adjustment only for CO backpressure; and third, with adjustment for both CO backpressure and the anemia effect. Paired Student's t tests were used to compare the CO backpressure measured from the sham maneuver with that measured from the deep breath and to assess the effect of increasing COHb on DLCOSB and DI.
COHb was 1.2 ± 0.4% (mean ± 1 SD) at baseline before any DLCOSB testing. Hb was 159 ± 1.3 g/L and did not change throughout the study. For comparable maneuvers, there were no significant differences (Table 1) from control subjects, after increasing COHb, in variables that reflect the performance of the respective maneuvers (Vinsp, Vexp, tinsp, texp, tBH, Vmax). There also were no differences in variables that reflect the distribution of ventilation (Emix or Sn) (Table 1).
For all the baseline maneuvers, FACO measured from the standardized deep breath was identical to that measured in the sham maneuver (Figure 3) , but increased slightly from 4 ± 8 ppm for the first maneuver in the sequence, to 12 ± 5 ppm for the second maneuver, to 17 ± 10 ppm for the third maneuver, and to 16 ± 5 ppm for the fourth maneuver (Figure 3). This change in FACO reflects the buildup of CO with repeated diffusing capacity testing. After completion of the baseline DLCOSB maneuvers, COHb had increased to 3.2 ± 0.4%.
FACO measured from the standardized deep breath increased markedly (Figure 3) after deliberately increasing COHb. The values were 63 ± 8 ppm for the first maneuver in the sequence, 68 ± 12 ppm for the second maneuver (not statistically different from the first maneuver), 65 ± 10 ppm for the third maneuver, and 64 ± 10 ppm for the fourth maneuver. When COHb was deliberately increased, FACO was also similar for the sham versus the standardized deep breath maneuvers (Figure 3), except for the FRC to one-half IC maneuver with 9 seconds of breath holding. In this case, FACO was slightly higher during the sham maneuver (75 ± 12 ppm) compared with the deep breath (64 ± 10 ppm; p < 0.001). The COHb had increased to 11.2 ± 1.1% after deliberately increasing COHb and after completing all the DLCOSB maneuvers.
Linear regression analysis of FACO measured from the maneuvers nearest the collection of the blood samples, and COHb in venous blood, both before and after deliberately increasing COHb (Figure 4) , revealed a mean slope of 5.6 ppm increase in FACO per percent increase in COHb (r 2 = 0.91). To adjust DLCOSB for increases in COHb, we therefore used the following equation:
For all maneuvers, DLCOSB was significantly lower after deliberately increasing COHb when no adjustments were made for either FACO or COHb (Table 2, Figure 5A) . After adjusting for the measured FACO only, DLCOSB for all maneuvers continued to be significantly increased with increasing COHb (Table 2, Figure 5B). The residual anemia effect of increasing COHb for all DLCOSB maneuvers was approximately a 1% decrease in DLCOSB for every 1% increase in COHb. Adjustment of DLCOSB for the anemia effect using Equation 8 eliminated the difference in DLCOSB due to increases in COHb for all the DLCOSB maneuvers (Table 2, Figure 5C). After correcting for both backpressure and the anemia effect, the mean difference in DLCOSB between comparable maneuvers before and after increasing COHb was 0.02 ± 0.57%. DI was not significantly affected by increases in COHb for all maneuvers when adjusted for FACO (Figure 6) . Because DI is measured from the ratio of diffusing capacities, it is mathematically unaffected by correction for the anemia effect.
Using the results of the linear regression analysis, COHb could be estimated from the FACO of each maneuver. After the deliberate increase in COHb, the mean increase in COHb between comparable maneuvers was 9.5 ± 2.6%, whereas the mean decrease in DLCOSB between comparable maneuvers was 19.3 ± 6.8% without correcting for either FACO or COHb. After correcting for FACO only, the mean decrease in DLCOSB between comparable maneuvers was 8.0 ± 3.8%, indicating that the backpressure was responsible for 58.5% and the anemia effect for 41.5% of the decrease in DLCOSB.
Various adjustments in DLCOSB have been proposed to account for the acute effects of increasing COHb. In their description of the single breath method, Ogilvie and coworkers (14) recommended using the Haldane equation (22), which indirectly estimates the CO partial pressure in capillary blood from the measurement of COHb, and the relative affinity of Hb for CO and O2. They felt that the effects of increasing COHb were sufficiently small, so that routine correction of DLCOSB was not necessary (14). As a result, DLCOSB is not usually adjusted for increases in COHb in most studies of current smokers (23), and recent recommendations for a standard technique still do not currently mandate adjustments for COHb (4). The practice of asking smokers to refrain from smoking for 24 hours before DLCOSB testing has been employed (24), but because abstention from smoking for this time period is difficult, success requires verification with COHb measurements. Although regressions for DLCOSB, without adjustment for FACO or COHb, have been published for current smokers (25, 26), this approach does not distinguish the technical effects of adjusting DLCOSB for increasing COHb from the potential acute and cumulative effects of the chemicals and particulates in cigarette smoke on the pulmonary vasculature, which may vary depending on the time of testing from the last cigarette smoked and the degree of individual susceptibility to the toxins in cigarette smoke (27). FACO measured from the control deep breath was also compared with FACO from a sham maneuver without any CO in the inhaled test gas, so that the back diffusion and equilibration of CO in alveolar air could be accurately assessed under the exact conditions found in the actual single breath maneuver. We found concordance in FACO measurements between the sham maneuver and the deep breath in all circumstances (Figure 3), except for the FRC to one-half IC maneuver after 9 seconds of breath holding when COHb had been deliberately increased. In this case, FACO was slightly but statistically significantly higher in the sham versus the deep breath. Although the reason for this difference is uncertain, it might be explained by the considerable differences between the two maneuvers, which include a lower inspired volume, a longer breath-hold time, and lower lung volume from which the sample for measurement of FACO was collected in the sham maneuver versus the deep breath. The lower lung volume and the longer breath-hold time would be expected to lower the mean capillary pH and raise the mean capillary [CO2] in the sham versus the standardized deep breath. These chemical changes in capillary blood could explain the higher CO backpressure in the sham maneuver. Increases in FACO for any given COHb are predicted when [CO2] increases in capillary blood because increasing [CO2] shifts the COHb dissociation curve to the right, thus decreasing the pH, which in turn decreases the relative affinity of Hb for CO and O2 (22). These observations confirm that measurement of FACO from the immediately preceding standardized deep breath can accurately assess CO backpressure, under most circumstances. We used full vital capacity maneuvers that minimize ventilation and diffusion heterogeneities in the lung as well as submaximal maneuvers and short breath-hold times that accentuate any heterogeneity in the distribution of either ventilation or diffusing capacity (20). Under all conditions tested, the combined adjustments for CO backpressure and the anemia effect adequately compensated DLCOSB for the increased COHb. The technique we used for measuring CO backpressure in this study was similar to the method of Jones and coworkers (28), who found that CO backpressure increased with breath-hold time for forced vital capacity maneuvers. Because optimal equilibration between air and blood required 20 seconds of breath holding in their study (28) and because the European Respiratory Society recommends measurement of FACO after 20 seconds of breath holding (29) to correct for CO backpressure, we included a vital capacity maneuver with 20 seconds of breath holding in the present study. We found that FACO from the sham vital capacity maneuvers with breath-hold time of 20 seconds was not different from FACO measured from the sham vital capacity maneuvers with breath-hold time of 0 seconds or from the standardized deep breath with breath-hold time of 5 seconds (Figure 3). Hence, a 5-second period of breath holding for the standardized deep breath provides sufficient time for CO to equilibrate between air and capillary blood in normal subjects. The lack of an effect of breath-hold time on FACO in this study may be explained by the fact that we employed slow (0.5 L/second) inhalation and exhalation maneuvers, which extended both inspiratory and expiratory time to approximately 8–10 seconds (Table 1), thus allowing greater time for equilibration of [CO] between air and blood even for maneuvers with short periods of breath holding. However, baseline DLCOSB values, when adjusted for both FACO and COHb, were decreased (p < 0.01) for the 20-second versus the 0-second breath-hold vital capacity maneuvers (Figure 5A). The decrease in DLCOSB with 20 seconds of breath holding may be explained by the high mean alveolar total barometric pressure during extended periods of breath holding, which lowers the cardiac output and thus reduces the pulmonary capillary blood volume (30).
Alternative approaches to adjusting for increasing COHb have been reported. The Haldane relationship and the direct measurement of COHb in venous blood have been used to estimate FACO (5, 6, 27). Thereafter, by applying the method of Cotes and coworkers (31), adjustments were made for the anemia effect. This method requires the use of assumed values for Dm/(
The residual effect of increasing COHb on DLCOSB (after correction for CO backpressure) has been explained previously by a decrease in effective Hb with increasing COHb, thus effectively reducing the amount of Hb available for CO binding (6, 18, 19, 31). However, the observed decrease in DLCOSB with increasing COHb in this study was greater than that predicted by Cotes and coworkers (31). The magnitude of this effect is consistent with the hypothesis that as a first approximation, DLCOSB is directly proportional to Hb in normal subjects (see THEORY). Several considerations could explain this finding. Firstly, our analysis assumed that acutely increasing COHb did not fundamentally alter either the structures composing the air–blood interface, the distribution of the inspired gas, cardiac output, pulmonary vascular pressures, or distribution of capillary blood in the lung. This assumption could be questioned because recent studies indicate that CO itself is a novel signaling agent that may potentially cause peripheral vasodilatation (35) and therefore lower DLCOSB, as previously reported for a potent vasodilator, nitroglycerin (36). Secondly, increases in [CO] may also cause acute bronchodilatation in animals (37). However, bronchodilatation is unlikely to explain our results because we previously found that DLCOSB was insensitive to changes in the distribution of ventilation (20) and because direct measurements of Sn and Emix in this study did not change with increasing COHb for any of the DLCOSB maneuvers (Table 1). Thirdly, Dm/( Irrespective of which of these factors is important, it is clear that DLCOSB testing must include adjustments for both FACO and for an anemia effect. Our results are similar to those of Sansores and coworkers (27), who also examined the acute effects of increasing COHb on DLCOSB in healthy subjects and found that DLCOSB was reduced after COHb was increased, even after correcting for FACO using the Haldane equation, and after adjustments were also made for the anemia effect using the method of Cotes and coworkers (31). In the study by Sansores and coworkers, the residual effect could be eliminated by assuming that the binding affinity of CO for Hb was reduced when COHb increased (27). Mohsenifar and Tashkin (5) derived an empiric adjustment for both the anemia effect of increases in COHb and the increases in FACO subjects by analyzing the effect of breathing high ambient concentrations of CO in normal subjects. They proposed that DLCOSB, when unadjusted for FACO, could be corrected by increasing it by 1% for every 1% increase in COHb. In our study, a 9.5% increase in COHb decreased DLCOSB by 19.3%. However, as reported by Sansores and coworkers (27), reanalysis of the data of Mohsenifar and Tashkin revealed an effect of increasing COHb, which persisted even after applying their FACO correction for DLCOSB but which could be eliminated after they applied a correction for the anemia effect. Both our results and those of Sansores and coworkers (27) are not consistent with the current ATS recommendations for a standard technique (4), which propose either increasing DLCOSB by 1% for every 1% increase in COHb, or adjusting DLCOSB for increases in FACO, but do not require both corrections. However, to account for the observed effect of increasing COHb acutely in normal subjects in the present study, we had to both correct for FACO, which compensated for about 60% of the decrease, and apply an adjustment for the anemia effect, which compensated for the remaining 40% of the decrease in DLCOSB.
In clinical applications and especially in epidemiologic studies, it would be preferable to adjust for the effects of increasing COHb without the need to collect venous blood. To accomplish this, we estimated COHb from the relationship between FACO and COHb (Figure 4). The adjustment for FACO can be accomplished analytically either using the 3-equation method or using Equation 5 rather than the Krogh equation for conventional methods. For either method of measuring DLCOSB, the adjustment for the anemia effect requires that the DLCOSB be increased by slightly more than 1% for every 5.6 ppm increase in FACO, according to Equation 8. The direct measurement of Hb in venous blood further improves precision, permitting a correction for absolute reductions in Hb as well as "COHb anemia" using:
In normal subjects, we have also recently described changes in DLCOSB measured from four equal alveolar gas samples (40). The variation among these four DLCOSB values was larger for FRC to one-half IC maneuvers versus FRC to TLC maneuvers, disappeared with short periods of breath holding, and correlated with simultaneously measured improvements in ventilation inhomogeneity (40). This suggests that dynamic gas transport mechanisms in the lung periphery contribute a significant resistance to CO gas transport in normal subjects during short breath holding maneuvers. We also developed an index of the degree to which the DLCOSB from the four equal alveolar gas samples deviated from DLCOSB measured from the entire gas sample by expressing this as the root-mean-square difference (10, 13). Because DI was increased in smokers with otherwise normal lung function and correlated with age and pack years of smoking, it may provide an index to measure early structural changes in the lung due to smoking (13). In previous applications of this technique (10, 13), adjustment was made for CO backpressure in calculating DI. In this study, we wished to determine to what extent DI was influenced by deliberately increasing COHb. Increasing COHb had no effect on DI when DLCOSB measurements from the four equal alveolar gas samples (13) were adjusted for CO backpressure. This confirms that the increase in DI observed in smokers with normal forced exhaled flow rates (13) was independent of increases in COHb. Further adjustment for the anemia effect was not required because DI was the ratio of DLCOSB measured from small alveolar gas samples. DI is not influenced by adjustment for the anemia effect of increasing COHb because the four DLCOSB values from the four equal alveolar gas samples are normalized by expressing them as a percent of DLCOSB measured from the entire gas sample. In summary, DLCOSB calculations must consider both backpressure and the anemia effect of COHb. FACO measured in alveolar gas exhaled from a standardized deep breath of room air immediately preceding each DLCOSB maneuver can be used to estimate backpressure in the calculation of DLCOSB and also to estimate COHb for the adjustment of DLCOSB for the anemia effect. DI measurements must be corrected for backpressure but are not affected by the anemia effect. DI may therefore be useful in the early detection of the effects of cigarette smoke on the lung. Whereas it is reasonable to expect that the compensation for backpressure using FACO will be valid both in normal subjects and in patients with lung disease, the adjustment for COHb could be affected by changes in the distribution of ventilation or membrane conductance that occur in chronic obstructive pulmonary disease and other lung diseases. Hence, the adjustment for the anemia effect, whether using our method or conventional methods, may not be valid for patients with lung disease.
The authors are grateful to the Saskatchewan Lung Association, the John Cameron Moorhead Foundation, and the Medical Research Council of Canada for providing funding for this research.
This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form August 16, 2001; accepted in final form March 13, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
Vc, where Vc is the pulmonary capillary blood volume, and 
