Exercise-induced arterial hypoxemia (EIH) is a frequent occurrence in the highly trained male athlete, and the mechanisms responsible for this phenomena have attracted considerable research interest (3,5,8,10,18,23,30). Earlier work suggested that at exercise levels of V˙O2 ∼ 3.0 L·min−1, the most significant contributor to widening of the alveolar arterial O2 difference [(A-a)DO2] was diffusion limitation (7). More recent studies of very highly trained athletes (V˙O2 > 5 L·min−1) during maximal exercise have demonstrated that approximately 60% of the widening (A-a)DO2 is explained by ventilation-perfusion (V/Q) mismatch (11). One possible explanation for both increased V/Q mismatch and diffusion limitation is the accumulation of interstitial pulmonary edema (4,7,26). Stress failure of the capillary endothelium has been demonstrated in animal models (28,29), and the recent observation of red cells and protein in broncholavage fluid after intense exercise in athletes supports this mechanism in humans (12).
A technique for the direct assessment of subclinical perivascular and/or peribronchial edema has yet to be established. Indirect evaluation has been made by measurement of pulmonary diffusing capacity (DL). At the onset of exercise, DL increases because of increases in membrane diffusing capacity (DM) and pulmonary capillary volume (VC) (2,6). After exercise, there is a decrease in DL below preexercise levels, which persists for a minimum of 6 h (23). This persistent decrease in DL has been used as an argument, by several investigators, for injury to the alveolar-capillary membrane and the development of interstitial edema (14,16).
We hypothesized that repetitive heavy exercise would result in a progressive increase in the severity of EIH in highly trained athletes. We further hypothesized that, if interstitial edema did develop during these repetitive exercise bouts, this would be reflected in a sequential decrease in pulmonary diffusion capacity. Therefore, the purpose of this study was to examine the effect of two bouts of heavy exercise on the development of EIH and to assess the corresponding changes in DL, DM, and VC.
Subjects. Thirteen, nonsmoking, endurance-trained male athletes (age = 27 ± 3 yr, height = 179.6 ± 5.0 cm, weight = 71.8 ± 6.9 kg) were recruited. All subjects were screened for inclusion in the study, and met the following criteria: (i) V˙O2max ≥ 60 mL·kg−1·min−1 or 5 L·min−1, and (ii) normal resting pulmonary function. The research protocol was approved by Clinical Screening Committee for Research and Other Studies Involving Human Subjects and informed consent was obtained from each subject.
Experimental protocol. Before all testing, subjects were required to avoid exhaustive exercise for 24 h; abstain from ingestion of food or fluid, except water for 2 h; and refrain from consuming alcohol and caffeine for 12 h. Pulmonary diffusing capacity was determined before the initial exercise (preexercise) and 60 min after the completion of each exercise test by the single-breath carbon monoxide diffusing capacity test (DLCO, mL·min−1·mm Hg−1). All DLCO measurements were preceded by 30 min of seated rest to ensure stabilization of measurement (2). Hemoglobin (Hb) concentration was measured before the determination of pulmonary diffusing capacity.
V˙O2max was determined using a ramp protocol beginning at 0 W, with increments of 30 W·min−1. The criteria for maximal aerobic capacity was attainment of three of the following: (i) a plateau in V˙O2 with increasing work load, (ii) Respiratory Exchange Ratio (RER) > 1.10, (iii) 90% of age-predicted maximal HR, or (iv) volitional fatigue. An active cool-down period of 7 min at a work load of 75 W was followed by 60 min of seated rest, followed by a postexercise DLCO measurement (postexercise 1). The entire testing sequence was repeated with a final DLCO measure (postexercise 2) made after 60 min of seated rest after the second maximal exercise bout.
Experimental measures. Spirometry and DLCO were measured and analyzed using a Collins Survey Tach Pulmonary Function testing unit (Warren E. Collins Inc., Braintree, MA) according to the method of Roughton and Forster (20) as modified by Ogilvie et al. (17). Duplicate trials were performed to ensure that values differed by less than 3 mL·min−1·mm Hb−1, using a test gas of 21% O2, 10% He, 0.3% CO, in a balance of N2. To partition the components of DL, the rate of disappearance of CO from a second test gas of 10% He, 0.3% CO, in a balance of O2 was measured after a 5-min wash-out period. During the wash-out period, subjects breathed a mixture of 90% O2 (balance N2) through a low resistance, nonrebreathing Hans Rudolph valve (Kansas City, MO). Plots from the two gas concentrations of 1/DL vs the inverse of the reaction rate of CO with Hb (1/θ), enabled the determination of 1/DM from the y-intercept. Values for θ pre- and postexercise were corrected for deviations from normal Hb concentration.
The maximal exercise tests were performed on an electronically braked cycle ergometer (Quinton, Excalibur; Groningen, The Netherlands). Subjects breathed room air through a low resistance, nonrebreathing valve (Hans Rudolph). Expired gases were measured and analyzed continuously by an automated gas analysis system (Rayfield; Waitsfield, VT). Arterial oxygen saturation was measured with an ear oximeter (Ohmeda Biox 3740 pulse oximeter; Louisville, CO). To improve perfusion, the ear lobe was rubbed with a vasodilator cream (Finalgon, Boehringer Ingelheim; Burlington, ON). To determine the minimum level of %SaO2 for statistical analysis, 15-s averages of %SaO2 during exercise were calculated and graphed. All external devices were integrated with an IBM computer utilizing a data collection software package (LABTECH Notebook; Laboratory Technologies Corporation; Wilmington, DE). V˙O2max for data analysis was determined by averaging the four highest consecutive 15-s values of V˙O2. HR was monitored and recorded every 15 s throughout the entire testing protocol using a portable HR monitor (Polar Vantage XL; Kempele, Finland).
Statistical analyses. A two-way multivariate repeated measures ANOVA was used to analyze the mean differences in the physiological variables between the three test conditions (preexercise, postexercise 1, and postexercise 2). When significant differences were observed, post hoc comparisons were performed using a Bonferroni t-procedure. Differences were tested for the following: DL, DM, VC, [Hb], and body mass. A matched pairs t-test was used to analyze the mean differences in the physiological variables between the two bouts of exercise. Differences were tested for the following: V˙O2max, %SaO2min, respiratory frequency (fR), maximal pulmonary ventilation (V˙E), peak power output, and maximal HR. The level of significance for each test was set at P < 0.05.
Subjects were all highly trained cyclists or triathletes (V˙O2max = 67.0 ± 3.6 mL·kg−1·min−1). All resting pulmonary function test results showed no abnormalities and represented expected values for an athletic population of healthy individuals. The mean physiological measures (± SD) are presented in Table 1. No significant differences were found between the two exercise bouts for maximum values of V˙E (P = 0.53), V˙O2 (P = 0.08), and fR (P = 0.10), or minimum values of %SaO2 (P = 0.21). There was a significant decrease (P = 0.003) in peak power output during the second exercise test.
Mean values (± SD) for pulmonary diffusing capacity for each measurement period are presented in Table 2. There was a significant difference in DL (P < 0.0001), DM (P = 0.02), and VC (P < 0.0001) between preexercise and postexercise measurements. Figure 1 depicts these changes. Diffusing capacity decreased approximately 17% below preexercise values. There was a significant 11% decrease (P < 0.05) from preexercise after the first bout of exercise and a further 6% decrease (P < 0.05) after the second bout of exercise. There was an 11% decrease (P < 0.05) in DM from preexercise to postexercise 1; however, the 2% decrease between the first and second exercise bouts was not significant (P > 0.05). The changes in VC were the greatest with a 10% reduction (P < 0.05) from preexercise measures after the first exercise bout. A further 10% reduction in VC between the first and second exercise bouts was also statistically significant (P < 0.05). Approximately 60% of diffusing capacity could be attributed to the membrane component, whereas 40% was dependent upon pulmonary capillary blood volume and the reaction rate of CO with Hb.
The individual values of %SaO2 at rest and during the two maximal tests are presented in Table 3. The 13 subjects achieved a percent saturation of 91.4 and 91.6 for each of the two maximal exercise tests, respectively. Consistent with the method of Powers et al. (18), six subjects who demonstrated a minimum %SaO2 of <91 on the initial maximal exercise test were labeled desaturators (group A); there were seven athletes with a %SaO2 > 91 who did not experience desaturation (group B). When these two groups were compared, there were no significant differences found between the groups for any of the variables measured, except for the minimum degree of arterial saturation (P = 0.0001). The minimum level of %SaO2 was approximately 3.5% lower in the desaturation group A (89.6%, Ex1; 90.0%, Ex2) compared with the nondesaturating group B (93.0%, Ex1; 92.9%, Ex2). The second exercise bout did not result in an exacerbation of EIH in either group. There was no significant relationship between the percent decrease in %SaO2 from resting to minimum exercise values and percent decrease from preexercise to postexercise values in DL, DM, or VC.
This study confirms the decrease in pulmonary diffusing capacity after short-term intense exercise and extends this relationship to a further decrease when exercise is repeated after 60 min of recovery. The reduction in DLCO after the first exercise bout is similar to the findings of other studies that have examined the effects of short-term, high-intensity exercise on postexercise pulmonary diffusing capacity (14,19,23). Similar results are also apparent after exercise of longer duration (4,13,16). Our data, which show an additional reduction in DLCO after a second exercise bout, contrast with previous findings that report reduced DLCO after one bout of maximal rowing but no further reductions after a second maximal rowing effort (8). The decrease in DLCO between studies varies between 5 and 28% and may be a function of a number of factors including the duration and intensity of exercise (13,14), changes in HR and CO (1), the time of measurement postexercise (23), and the resting state immediately before measurement (2).
Significant decreases in postexercise DM have been observed after short-term exercise (8,13) and long-term exercise (14,16). In the current study, an additional nonsignificant decrease in DM occurred after a second exercise bout, which is in agreement with Hanel et al. (8). Many authors have interpreted the decreases in DM to be as a result of perivascular and/or alveolar wall edema (14,16,21). The earliest form of pulmonary edema is characterized by engorgement of the peribronchial and perivascular spaces and is classified as interstitial edema. Thickening of the alveolar-capillary membrane causes an increase in the distance between the gas and blood phase of the lung tissue resulting in a persistent decreased membrane diffusing capacity (24). According to the Starling hypothesis for capillary fluid exchange, pulmonary edema may result from elevated capillary hydrostatic pressure, increased capillary permeability to plasma proteins, increased capillary surface area, and decreased lymphatic drainage. Alternatively, the increase in pulmonary artery pressure (PAP) during exercise may be of sufficient magnitude to cause disruption of the endothelium of the capillary wall and epithelium of the alveolar (25). Studies have reported that PAP can exceed 40 mm Hg under high-intensity exercise (22,26). The end result by either mechanism is an accumulation of pulmonary fluid.
The significant reductions in VC observed after each of the two exercise bouts are consistent with a previous study (8). The decrease in VC continues for 6 h postexercise, paralleling decreases in DLCO (23). A recent report by Hanel et al. (9) has indicated that approximately one-half of the post exercise reduction in DLCO is explained by a decrease in pulmonary blood volume. Further evidence in support of a reduction in central fluid volume after short-term maximal exercise is the finding of an increase in thoracic electrical impedance (8,19). In the present study, the calculated decrease in VC contributed approximately 40% to the decrease in DLCO observed post exercise. It is possible, however, that the changes in Vc were responsible for a portion of the decrease in DM. The fundamental relationship between the surface area available for gas exchange (A), the thickness of the alveolar capillary membrane (y), the diffusion coefficient (d) and the membrane diffusing capacity (DM) is represented by the equation: (Equation)
It is conceivable that the significant reductions in VC resulted in a decrease in capillary recruitment which would result in a fall in A. Thus, the changes observed in DM might be secondary to hemodynamic shifts.
After the first bout of maximal exercise, DLCO was characterized by a reduction in both DM and VC, whereas after the second bout of heavy work, the reduction in DLCO was due only to a further decrease in VC. It is possible, therefore, that the changes in DLCO after the two different exercise bouts were due to two different mechanisms. After the initial test, the observations could be explained by the development of interstitial edema, but after the second exercise session, the changes in DLCO might be more readily explained by a redistribution of pulmonary capillary blood. It is also feasible that any further edema that resulted from the second exercise session developed in the peribronchial and perivascular regions of the lung away from the sites where gas exchange would occur. This would help explain why DM did not fall significantly after the second bout.
The observed changes in pulmonary diffusion capacity, after the first bout of heavy work, did not appear to influence the outcome of the second maximal cycle test. There were no significant changes in minute ventilation, maximal aerobic capacity, maximal HR, respiratory frequency, or %SaO2. There was a small, but significant, reduction in peak power output during the second test, most likely due to peripheral muscle fatigue. It appears therefore, that the post exercise decreases in DLCO, VC, and DM have no physiological consequence to a second exercise session in these highly trained subjects.
Arterial oxygen Hb saturation decreased to a minimum value at maximal exercise. The Ohmeda Biox 3740 pulse oximeter has been shown to be a valid and reliable tool for assessing arterial oxygen saturation during intense exercise in subjects with high aerobic capacities. The oximeter used in this lab has been validated with arterial blood gas measures using highly trained athletes exercising at maximal aerobic capacity (r = 0.87, P < 0.001). Others have also demonstrated a strong positive relationship (r = 0.94, P < 0.05) between pulse oximeter estimates of %SaO2 and measured arterial oxygen Hb saturation over a wide range of saturation levels and exercise intensities (15).
Although we did not attempt to investigate the mechanisms responsible for the observed changes in percent saturation of Hb with oxygen, the fact that the hypoxemia was not worse after a second bout of high intensity exercise indicates that the mechanism responsible for the hypoxemia is not aggravated by repeat exercise. The data in Table 3 shows that the pattern of desaturation is highly reproducible in both the group that desaturated (group A) and in the group (B) that did not. Recovery of %SaO2 was complete in all subjects by the start of the second exercise bout, and the magnitude of the desaturation was not affected by the second V˙O2max test. Further analysis of the DLCO in these athletes indicated that the fall in pulmonary diffusion capacity was the same for both groups. Thus, athletes who do not desaturate demonstrate a parallel fall in DLCO compared with those athletes who do desaturate. This supports a hemodynamic explanation for the changes in DLCO.
The lack of a further change in %SaO2 during the second exercise task is difficult to reconcile with the progressive changes in DLCO. If the changes in pulmonary diffusion post exercise reflect the development of interstitial edema, the severity of EIH should have increased with the additional heavy work. These observations question the meaning of postexercise measurements of pulmonary diffusion capacity, and its components, relative to pulmonary gas exchange and pulmonary fluid accumulation during exercise. The fact that there was no change in EIH indicates that any edema formed was not of clinical significance; alternatively, the changes in DLCO may be related more to redistribution of blood than the development of interstitial edema. This would explain the significant decrease in DLCO and VC, but not DM, after the second exercise period. These observations are supported by the work of Hanel et al. (9), who have demonstrated a decrease in the pulmonary blood volume with redistribution to the periphery.
Supporting arguments regarding the physiological significance of the decreases in DLCO have been made by Sheel et al. (23), who have demonstrated that the post exercise decreases in DLCO are not related to fitness status. These authors argue that EIH is only observed in highly trained male athletes (18) but the postexercise decreases in DLCO are observed in subjects with a wide range of maximal aerobic capacities. Our data would support the lack of a relationship between exercise measurements of EIH and post exercise DLCO and its components.
In summary, we have documented decreases in the postexercise measurements of pulmonary diffusing capacity after short-term intense exercise and extended these findings to show a progressive decrease in DL and VC, but not DM, when a second maximal cycle task is repeated after 60 min of recovery. The athletes demonstrated a decrease in %SaO2 at the end of the first bout of exercise, but there were no further changes in %SaO2 during the second exercise task. These data indicate that the postexercise changes in pulmonary diffusion capacity are not related to the occurrence of hypoxemia during maximal exercise. Furthermore, this data cannot support the hypothesis that the lung is unable to maintain lung fluid balance during repeat exercise.
1. Andrew, G. M., and L. Baines. Relationship of pulmonary diffusing capacity (DL) and CO (QC) in exercise. Eur. J. Appl. Physiol.
2. Billiet, L. Time course of the changes in pulmonary diffusing capacity due to rest or exercise. In: Pulmonary Diffusing Capacity on Exercise.
M. Scherrer, (Ed.). Bern: Hanx Huber, pp. 187-196, 1970.
3. Caillaud, C., F. Anselme, J. Mercier, and C. Prefaut. Pulmonary gas exchange and breathing pattern during and after exercise in highly trained athletes. Eur. J. Appl. Physiol.
4. Caillaud, C., O. Serre-Cousine, F. Anselme, X. Capdevilla, and C. Prefaut. Computerized tomography and pulmonary diffusing capacity in highly trained athletes after performing a triathlon. J. Appl. Physiol.
5. Dempsey, J. A., P. G. Hanson, and K. S. Henderson. Exercise-induced arterial hypoxemia in healthy human subjects at sea level. J. Physiol.
6. Fisher, J. T., and F. J. Cerny. Characteristics of adjustment of lung diffusing capacity to work. J. Appl. Physiol. Respir. Environ. Exerc. Physiol.
7. Hammond, M. D., G. E. Gale, K. S. Kapitan, A. Ries, and P. D. Wagner. Pulmonary gas exchange in humans during exercise at sea level. J. Appl. Physiol.
8. Hanel, B., P. S. Clifford, and N. H. Secher. Restricted post exercise diffusion capacity does not impair maximal transport for O2
. J. Appl. Physiol.
9. Hanel, B., I. Teunissen, A. Rabol, J. Warberg, and N. H. Secher. Restricted postexercise pulmonary diffusion capacity
and central blood volume depletion. J. Appl. Physiol.
10. Hopkins, S. R., and D. C. McKenzie. Hypoxic ventilatory response and arterial desaturation during heavy work. J. Appl. Physiol.
11. Hopkins, S. R., D. C. McKenzie, R. B. Schoene, R. W., Glenny, and H. T. Robertson. Pulmonary gas exchange during exercise in athletes. I. Ventilation-perfusion mismatch and diffusion limitation. J. Appl. Physiol.
12. Hopkins, S. R., R. B. Schoene, W. R. Henderson, R. G. Spragg, T. R. Martin, and J. B. West. Intense exercise impairs the integrity of the pulmonary blood-gas barrier in elite athletes. Am. J. Respir. Crit. Care Med.
13. Manier, G., J. Moinard, and H. Stoicheff. Pulmonary diffusing capacity after maximal exercise
. J. Appl. Physiol.
14. Manier, G., J. Moinard, P. Techoueyres, N. Varene, and H. Guenard. Pulmonary diffusion limitation after prolonged strenuous exercise. Respir. Physiol.
15. Martin, D., S. Powers, M. Cicale, N. Collop, D. Huang, and D. Criswell. Validity of pulse oximetry during exercise in elite endurance athletes. J. Appl. Physiol.
16. Miles, D. S., C. E. Doerr, S. A. Schonfeld, D. E. Sinks, and R. W. Gotshall. Changes in pulmonary diffusing capacity and closing volume after running a marathon. Respir. Physiol.
17. Ogilvie, C. M., R. E. Forster, W. S. Blakemore, and J. W. Morton. A standardized breath holding technique for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. J. Clinic. Invest.
18. Powers, S. K., S. Dodd, J. Lawler, et al. Incidence of exercise-induced hypoxemia in elite endurance athletes at sea level. Eur. J. Appl. Physiol.
19. Rasmussen, J., B. Hanel, K. Saunamaki, and N. H. Secher. Recovery of pulmonary diffusing capacity after maximal exercise
. J. Sports Sci.
20. Roughton, F. J. W., and R. E. Forster. Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung, with special reference to true diffusing capacity of pulmonary membrane and volume of blood in the lung capillaries. J. Appl. Physiol.
21. Schaffartzik, W., J. Arcos, K. Tsukimoto, O. Mathieu-Costello, and P. D. Wagner. Pulmonary interstitial edema in the pig after exercise. J. Appl. Physiol.
22. Schaffartzik, W., D. C., Poole, T. Derion, et al. VA
/Q distribution during heavy exercise and recovery in humans: implications for pulmonary edema
. J. Appl. Physiol.
23. Sheel, A. W., K. D. Coutts, J. E. Potts, and D. C. McKenzie. The time course of pulmonary diffusing capacity for carbon monoxide following short duration high intensity exercise. Resp. Physiol.
24. Staub, N. C., H. Nagano, and M. L. Pearce. Pulmonary edema
in dogs, especially the sequence of fluid accumulation in lungs. J. Appl. Physiol.
25. Tsukimoto, K., O. Mathieu-Costello, T. Prediletto, A. R. Elliott, and J. B. West. Ultrastructural appearances of pulmonary capillaries at high transmural pressures. J. Appl. Physiol.
26. Wagner, P. D., G. E. Gale, R. E. Moon, J. R. Torre-Bueno, B. W. Stolp, and H. A. Saltzman. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J. Appl. Physiol.
27. Warren, G. L., K. J. Cureton, W. F. Middendorf, C. A. Ray, and J. A. Warren. Red blood cell pulmonary capillary transit time during exercise in athletes. Med. Sci. Sports Exerc.
28. West, J. B., O. Mathieu-Costello, J. H. Jones, et al. Stress failure of the pulmonary capillaries in racehorses with exercise-induced pulmonary hemorrhage. J. Appl. Physiol.
29. West, J. B., K. Tsukimoto, O. Mathieu-Costello, and R. Prediletto. Stress failure in pulmonary capillaries. J. Appl. Physiol.
30. Williams, J. H., S. K. Powers, and M. K. Stuart. Hb desaturation in highly trained athletes during heavy exercise. Med. Sci. Sports Exerc.
Keywords:© 1999 Lippincott Williams & Wilkins, Inc.
ARTERIAL OXYGEN SATURATION; PULMONARY DIFFUSION CAPACITY; MEMBRANE DIFFUSING CAPACITY; PULMONARY CAPILLARY VOLUME; MAXIMAL EXERCISE; PULMONARY EDEMA