Relationship between arterial oxygen desaturation and ventilation during maximal exercise MOTOHIKO MIYACHI AND IZUMI TABATA Department of Physiology and Biomechunics, Nutionul Shiromizu, Kunoyu, Kugoshimu 891-23, Japan MIYACHI,
M~TOHIKO,
AND IZUMI TABATA.R&~~~&QI
be-
tweenarterial oxygendesaturationand ventilation during muximal exercise.J. Appl. Physiol. 73(6): 2588-2591, 1992.---The purpose of the present study was to investigate the contribution of ventilation to arterial 0, desaturation during maximal exercise.Nine untrained subjectsand 22 trained long-distance runners [age 18-36 yr, maximal U2 uptake (vo2,,) 48-74 ml min-’ 4kg-l] volunteered to participate in the study. The subjectsperformed an incremental exhaustive cycle ergometry test at 70 rpm of pedaling frequency, during which arterial O2 saturation (Sa,,) and ventilatory data were collectedevery minute. Sao, was estimated with a pulse oximeter. A significant positive correlation was found between Sa,, and end-tidal PO, (PETE,; r = 0.72, ? = 0.52, P < 0.001) during maximal exercise. Thesestatistical results suggestthat -50% of the variability of Sao, can be accounted for by differences in PETIT, which reflects alveolar PO,. Furthermore, PETE, was highly correlated with the ventilatory equivalent for 0, (VE/~O~; r = 0.91, P < O.OOl),which indicates that PETIT could be the result of ventilation stimulated by maximal exercise,Finally, Saozwas positively related to vEli70, during maximal exercise(r = 0.74, ? = 0.55,P < 0.001).Therefore, one-half of the arterial 0, desaturation occurring during maximal exercise may be explained by lesshyperventilation, specifically for our subjects,who demonstrated a wide range of trained states. Furthermore, we found an indirect positive correlation between Sao, and ventilatory responseto CO, at rest (r = 0.45,P < 0.05), which wasmediated by ventilation during maximal exercise. These data also suggest that ventilation is an important factor for arterial 0, desaturation during maximal exercise. l
arterial oxygen saturation; ventilatory responseto carbon dioxide
thatarterialO,partialpressure (PaoJ and arterial 0, hemoglobin saturation (Sao,) are maintained at near resting levels during maximal voluntary exercise at sea level (1, 6). However, several studies (3, 4, 7, 8, 13, 14, 18) recently demonstrated a reduction of PaoZ or Sao, during heavy exercise (exerciseinduced hypoxemia or exercise-induced arterial 0, desaturation). The reduction of Pa,, or Sa,, during heavy exercise may be related to an increase in the alveolar-to-arterial PO, difference (A-aDo,) induced by diffusion limitation and ventilation-perfusion inequality (3-5, 7, ‘17). In addition, Dempsey et al. (3, 4) and Powers et al. (13, 14) reported that exercise-induced hypoxemia and desaturation were observed in subjects with lower alveolar 0, partial pressure (PA& and less hyperventilation. We hypothesize that less hyperventilation and an inITISGENERALLYACCEPTED
2588
Institute
of Fitness and Sports,
creasing A-aDo, play a major role in the genesis of arterial hypoxemia and 0, desaturation during exercise. Although the mechanisms that regulate ventilation during maximal exercise are unclear, several studies (9, 10) reported significant relationships between ventilatory responses to hypercapnia at rest and ventilation during maximal exercise. If ventilation during maximal exercise is a major factor in arterial 0, desaturation, as we hypothesize, the ventilatory response to hypercapnia at rest should be related to Sa,, during maximal exercise. Therefore, we considered it worthwhile to investigate the relationship between ventilation and Sa,, during maximal exercise and the ventilatory response to CO, in a large number of subjects. MATERIALS
AND METHODS
Subjects. Nine untrained subjects and 22 trained longdistance runners volunteered to participate in the experiments. All were healthy adult males with no history of lung disease. Descriptive data (means t SD) are as follows: age = 23 t 4 yr, body weight = 58.8 t 6.5 kg, maximal 0, uptake (TO 2max) = 62 t 7 ml mine1 kg-l STPD. The subjects were informed of the nature of the experiments and gave informed consent before testing. Procedure. The study consisted of two experiments. The subjects first performed a ventilatory response to CO, test at rest and then completed a maximal exercise test. Environmental conditions remained relatively constant during the experiments: temperature = 2O-25”C, barometric pressure = 740-755 Torr. Ventilatory response to CO, test. The ventilatory response to CO, was measured by the rebreathing method of Ohkuwa et al. (11) and Ohyabu et al. (12) with a slight modification. After arrival at the laboratory the subjects rested for 20 min while seated on a chair. Subsequently, a mouthpiece was connected to a rebreathing bag and noseclips were fitted on the subjects. The subjects rebreathed a hyperoxic and hypercapnic gas mixture from the bag through the mouthpiece for 3.0-3.5 min. The expired ventilation volume (VE) during rebreathing was measured with a pneumotachograph (model RM 200, Minato Medical Science) attached between the mouthpiece and the rebreathing bag. All gas volumes were corrected to BTPS conditions. To measure the end-tidal CO, partial pressure (PET,,, ) , the expired gas was continuously withdrawn from the mouthpiece through a sampling capillary tube connected to a gas analyzer (model MGA 1100, Perkin-Elmer). The VE and PET,,, data were
0161-7567/92 $2.00 Copyright 0 1992 the American Physiological Society
l
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ARTERIAL
DESATURATION
stored on a breath-by-breath basis and averaged every 30 s. Each rebreathing bag was filled with 5 liters of 5% CO,-95% 0,. The slope of the %-PET,,, line (S) was calculated by least-squares regression analysis. The equation VE = S (PETIT, - B) was used for determination of the ventilatory response to CO,. To confirm the S of each subject, duplicate determinations were conducted on different days in 12 of the subjects. Exercise test. The incremental cycle exercise began at a work rate of 103 W (70 rpm), and the power output was increased by 17 W/min until the subject could not maintain the fixed pedaling frequency. The subjects were encouraged during the ergometer test to exercise for as long as possible. Saoz, heart rate (HR), and a rating of perceived exertion (RPE) were monitored minute by minute during exercise. 0, uptqke ($$), VE (BTPS), the ventilatoryequivalentforo, (VE/VO, ,BTPS/STPD) ,PET~*~, and the end-tidal partial pressure of 0, (PETE,) were monitored during the last 30 s of each increase in the work rate after the RPE reached 18. Sa,, was estimated with a pulse oximeter (model OLV1200, Nihon Kohden) on the tip of the right forefinger. The accuracy of pulse oximetry has been proved by several studies (l&16,19). The subjects were advised not to strain while moving their right hands to allow us to monitor the pulse waves correctly and to measure Sh, accurately. RPE was obtained using the modified Borg scale (2). Subjects breathed through a low-resistance two-way valve (model 2700, Hans Rudolph), and the expired air was collected in Douglas bags. Expired 0, and CO, gas concentrations were measured by mass spectrometry (model MGA 1100; Perkin-Elmer), and gas volume was determined using a dry gas meter (model NDS2A-T, Shinagawa). All gas volumes were corrected to BTPS conditions. To measure PETITand PET,,,, a small amount of expired gas (60 ml/mm) was sampled continuously by the gas analyzer through a sampling capillary connected to the mouthpiece. Data analysis. The maximal value of $70, during the exercise test was called peak irO,. Typical data for all other parameters included only those values occurring at peak VO, during the exercise test. The values are expressed as means t SD. The difference between the first and second measurements of S in duplicate determinations was determined by the paired t test. The correlation coefficients among the parameters were determined by the Pearson correlation. The level of significance was established at P < 0.05. RESULTS Ventilutory response to CO, test. PET,,, increased linearly (5.9 t 1.1 Torr/min) during the rebreathing test. However, in 29 subjects, i7E did not change significantly during the 1st min of the rebreathing tests. After that, it increased linearly with the. increase in PETIT,. In the remaining two subjects, VE increased linearly with throughout the rebreathing test. The mean pmm, was 47.4 (range 42.3-52.4) Torr during the PETCO, first 30 s of the rebreathing test and 66.7 (range 62.672.3) Torr during the last 30 s. The mean TE was 10.8 (5.8-20.1) llmin BTPS during the first 30 s and 32.9 (20.6-
AND VENTILATION
2589
47.8) l/min BTPS during the last 30 s. The mean value of S in all subjects was 1.18 1 min-’ Torr-l BTPS. The reliability of S was tested by duplicate determinations for each subject in 12 cases. The mean values of the first and second measurements were 0.98 t 0.44 and 1.00 t 0.35 1 min-l Torr-l, respectively. There was no significant difference and a high correlation between the first and second measurements (r = 0.92). Exercise test. Table 1 shows the changes in Sa,, of all subjects during exercise. Saoz did not change during light exercise until 5-6 min, when, for some subjects, it decreased progressively as the intensity increased. Finally, the Saoz at which peak VoZ was obtained ranged from 92 to 97%. The peak VO, ranged from 46.5 to 73.6 ml min-’ kg? Peak Vg, was observed at exhaustion in 27 subjects. In the remaining four subjects, VO, leveled off before exhaustion. Table 2 shows the mean t SD and maximum and minimum values of peak TO, and the Saao2,VE, vE/ vo, 9 PETco, 9PETE,, and HR at which peak Vo, was obtained during the exercise test. Relationships among parameters. Table 3 presents the correlation matrix among all parameters. There were positive significant relationships among Sa,, , PET,, f and bE/iro, (see Fig. 1 for relationship between Sa,, and vE/vO&. 8 at rest was significantly related to ventilation during maximal exercise (expressed as VE/~O~, VE, PETE,, or PET,,~). We found a significant positive correlation between Saoz during maximal exercise and S at rest, although the correlation coefficient was quite weak (r = 0.45). l
l
l
l
l
l
DISCUSSION
In the present study, we found a positive correlation between Sa,, and PET,, (? = 0.52) during maximal exercise. This result statistically shows that ~50% of the variability of Sao, can be accounted for by differences in PET*,, which reflects PACT.Furthermore, PETE, was highly correlated with i7E/vO, (r = 0.91), which indicates that PETIT is the result of ventilation stimulated by maximal exercise. Finally, Sa,, was positively related to vE/ VO, during maximal exercise (Fig. 1; ? = 0.55). Therefore, -50% of the variability in arterial 0, desaturation during maximal exercise can be explained by less hyperventilation, specifically in our subjects. Hopkins and McKenzie (7) reported that Pao, was not related to PAo2 (r = 0.51, P = 0.08) but was closely related to A-aDo, (r = 0.98, P < 0.001) during heavy exercise. They concluded that the increase in A-aDo, plays a major role in genesis of arterial hypoxemia and desaturation during maximal exercise with a lesser contribution of ventilation. We consider that this difference may be explained by the difference in the trained state of the subjects, i.e., they used only trained subjects, whereas we used subjects from a wide range of trained states. Trained subjects breathe less per unit of metabolic rate than untrained subjects (4,9,10); therefore the variability of ventilation in the study of Hopkins and McKenzie was narrower than ours. In fact, the variability in ~EP?o~ (29.5-36.4) for their subjects (7) corresponds to ~35% of the variability in vE/vO, (29.5-49.3) for our subjects. Powers et al.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
2590 TABLE
ARTERIAL
1, Individual
Sa, responses
DESATURATION
AND
at rest and during exercise Exercise
Subj No.
Rest
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 Mean tSD * Arterial
2
3
4
5
6
7
8
9
10
11
12
13
9% 9% 9% 99 97 97 9% 99 99 97 99 99 9% 9% 99 99 9% 9% 98 98 9% 98 98 99 98 9% 9% 97 9% 98 9%
97 97 97 97 96 97 98 97 97 96 97 98 97 95 9% 98 96 97 97 99 97 98 96 97 97 96 98 97 9% 97 97
96 9% 97 97 96 97 96 97 97 95 97 9% 97 96 9% 94 95 97 9% 9% 97 95 97 9% 97 97 97 96 98 97 97
97 97 99 97 97 97 96 9% 97 96 9% 9% 9% 9% 9% 94 97 97 97 98 97 96 96 98 97 97 99 96 97 9% 97
9% 96 97 96 97 96 96 98 98 96 97 9% 97 9% 9% 96 96 96 96 97 97 97 95 98 97 97 9% 96 97 97 97
9% 97 9% 97 96 97 98 97 97 96 98 9% 9% 97 97 96 97 96 96 97 96 95 95 9% 97 97 97 96 97 97 97
96 97 98 97 97 97 9% 97 98 96 97 97 9% 96 97 97 95 96 96 96 96 96 95 97 97 97 97 95 97 96 97
97 96 9% 96 96 96 97 96 98 96 96 96 9% 97 96 95 94 96 95 96 95 97 94 97 97 96 96 95 95 95 97
97 97 97 96 96 96 97 96 98 96 95 97 97 96 97 96 93 96 95 95 95 94 94 97 96 95 96 96 95 95 96
96* 96 97 97 96 97 97 96 96 94 95 96 96 95 95 93 93 94 95 94 95 94 94 97 96 95 95 95 95 96 95
96 96 96 95* 97* 97 95 96 93 94* 96 96 95* 95 95 92 93* 94 94 94* 94* 93 97 95* 94 95 94 95 95 95*
96 97 96*
96* 96
96*
97* 96 95* 93*
96 95*
95
98.1 kO.6
97.1 t0.8
96.8 t1.0
97.2 a.0
96.9 to.9
96.9 -to.9
96.7 HI*%
96.1 kl.1
95.9 zkl.1
95.3 3-1.2
94.8 H.2
saturation
(Saoz,
%) at which
peak
O2 uptake
(13, 14) reported that PETE, tended to be lower and PET,,, higher in subjects with desaturation during maximal exercise. Dempsey (3) noted that PAN, was ~110 Torr in subjects with greater hypoxemia. These previous results support the present finding. We suggest that about half of desaturation is due to a lower PACT (less hyperventilation), and the other half may be due to a wider A-aDo,, specifically in our subjects. Martin et al. (9, 10) demonstrated that ventilation during maximal exercise is proportional to the ventilatory response to hyoxia or hypercapnia at rest. There was 2: Mean, maximum, and minimum values at peak Vu2 during exercise test and S at rest TABLE
%,J 7-6 VEIV~~ (BTPS/STPD) VU,, ml . mine1 kg-’ VE, 1 min-’ kg-’ PETco2, Torr PET*, , Torr HR, beats/min S, 1 min-’ Torr-’ BTPS l
l
l
l
Time, min
1
0, hemoglobin
l
VENTILATION
Mean t SD
Minimum
Maximum
94.5t1.9 38.2t5.2 61.9t6.8 2.35tO.31 34.lt3.7 113.7t3.9 1%5_+11*3 1.171-0.48
92 29.5 46.5 1.71 28.0 105.7 165 0.33
97 49.3 73.6 2.84 42.7 121.1 202 2.33
n = 31. TjE/k$, ventilatory equivalent fur 0,; voz, expired ventilation vulume; PETIT,and PETo,, end-tidal HR, heart rate; S, slope of $7~ - PETE% relationship.
0, uptake; 93, PCO,and PO,;
95* 96* 95 95 92
95 94 92*
95 94 94
94* 94*
92* 97* 95 94 95* 93 94 94*
92
94.8 tl.4
94* 93*
92* 92* 94
93*
94.0 tl.6
94.2 -+1.3
was observed.
a wide range of ventilation during maximal exercise in our subjects. We also found a positive relationship between ventilation during maximal exercise (expressed as VE/~O~ or VE) and S at rest. Furthermore, PETIT,, which reflects arterial PQ, was negatively correlated to S. However, S was obtained in the PCO, range above resting, whereas Pcoz during maximal exercise was below resting. Furthermore, these correlations were quite weak (? = 0.2-0.4); therefore, other factors should be considered as contributors to the variability of ventilation during maximal exercise. We found a significant positive correlation between Saoz during maximal exercise and S at rest, although the correlation coefficient was quite weak (r = 0.45). Also, TABLE
3. Correlation matrix among all parameters Sa0g
sao2 VE/VO, vo, VE PET,*, pE%z HR s
1.00
VE/iTOz VQ2 0.74” 1.00
-0.70” -0.43* 1.00
VE PETcoz PETIT 0.16 0.62* 0.43* 1.00
-0.65* -0.%9* 0.24 -0.71” 1.00
0.72” 0.91* -0.34 0.62* -0.97* 1.00
HR
S
0.21 0.20 -0.03 0.14 -0.33 0.35 1.00
0.45” 0.62* -0.22 0.45* -0.54* 0.55* 0.10 1.00
n = 31. *P < 0.05.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ARTERIAL
DESATURATION
In summary, we demonstrated that Sap, had a positive relationship with PET,, and VE/~O, during maximal exercise in our subjects. These data suggest that -50% of the arterial 0, desaturation during maximal exercise was induced by less hyperventilation, specifically in those subjects with a wide range of trained states.
98 96 S&2
64
94
Present address of M. Miyachi: Dept. of Health and Sports Sciences, Kawasaki University of MedicaI Welfare, 288 Matsushima, Kurashiki City, Okayama 701-01, Japan. Present address and address for reprint requests: I. Tabata, Lab. for Exercise Physiology, Div. of Health Promotion, National Institute of Health and Nutrition, 1-23-1 Toyama, Shinjuku-ku, Tokyo, 162 Japan.
92 901 25
2591
AND VENTILATION
I
30
I
I
35
40
I
45
1
50
Received 18 May 1990; accepted in final form 23 June 1992.
i/E/co2
FIG. 1. Relationship between arterial .02 hemoglobin saturation (Sao,) and ventilatory equi.valent for 0, (VE/%,; r = 0.74, ?. = 0.55, P < 0.001). Sao, and VE/Q were the data at which peak VO, was obtained in exercise tests. Correlation shows that 50% of arterial 0, desaturation during maximal exercise can be explained by less hyperventilation in our subjects.
there were statistically significant relationships between Sa, and ventilation during maximal exercise and between ventilation during maximal exercise and S in our subjects. These relationships led us to examine the relationships among Sa,, , ventilation during maximal exercise, and S at rest. So, we performed a stepwise regression analysis among Sao?, VE/VO,, and S. Between Sa,, and vE/%2, the simple correlation coefficient was 0.74, and among Saoz, vE/vg:!, and S, the multiple correlation coefficient was also 0.74. Moreover the standard partial regression coefficients of vE/%, and S were 0.57 and only -0.02, respectively. These statistical results suggest that S at rest has no independent effect on Sao, during maximal exercise. In other words, ventilation during maximal exercise is an intermediary factor between S at rest and Sa,, during maximal exercise. Martin et al. (9, 10) also reported a significant positive relationship between ventilation during maximal exercise and the ventilatory response to hypercapnia at rest. These data support our main finding that -50% of arterial 0, desaturation is due to less hyperventilation during maximal exercise. One methodological concern is the use of the pulse oximeter to measure Sa,,. Although this instrument has been shown to provide valid and reliable assessment of trends in Sa,, at rest and during exercise, the standard error of the estimate is LO-2.0% (15, 16, 19). This random error might affect Sa,, in the present study. The other concern is the validity and reliability of the S measurement. The ventilatory response to CO, was measured by the rebreathing method of Ohkuwa et al. (11) and Ohyabu et al. (12) with a slight modification. The S values in these studies do not seem to differ from the S values in the present study. Therefore the value of S of the present study may be valid for the Japanese population. In addition, the reliability of S in the present study was examined by duplicate determinations for 12 subjects. There was a high correlation (r = 0.92) and no significant difference between the first and second measurements. These results agree well with those of Ohkuwa et al. and confirm the reliability of S in the present study.
REFERENCES 1. ASMUSSEN, E., AND M. NIELSEN. Alveolo-arterial gas exchange at. rest and during work at different 0, tensions. Acta Physiol. Stand. 50: 153-166,196O. 2. BORG, G. A. V. Psychophysical bases of perceived exertion. IWed. Sci Sports Exercise 14: 377-381, 1982. 3. DEMPSEY, J. A. Is the lung built for exercise? Med. Sci. Sports Exercise 18: 143-155,19&6. 4. DEMPSEY, J. A., P. G. HANSON, AND K. S. HENDERSON. Exerciseinduced arterial hypoxemia in healthy human subjects at sea level. J. Physiol. Lord. 355: 161-175, 1984. 5. GALE, G. E., J. R. TORRE-BUENO, R. E. MOON, H. A. SALTZMAN, AND P. D. WAGNER. Ventilation-perfusion inequality in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol.
58: 978-988,
1985.
6. HESSER, C. M., AND G. MATELL. Effect of light and moderate exercise on alveolar-arterial 0, tension difference in man. Acta Physiol. Scund. 63: 247-256,1965. 7. HOPKINS, S. R., AND D. C. MCKENZIE. Hypoxic ventilatory response and arterial desaturation during heavy exercise. J. Appl. Physid. 67: 1119-l 124, 1989. 8. LAWLER, J., S. K. POWERS, AND D. THOMPSON. Linear relationship between VO, maxand vogrnax decrement during exposure to acute hypoxia. J. Appl. Physiol. 64: 1486-1492, 1988. 9. MARTIN, B. J., K. E. SPARKS, C. W. WILLICH, AND J. V. WEIL. Low exercise ventilation in endurance athletes. Med. Sci. Sports Exercise 11: 181-185, 1979. 10. MARTIN, B. J.) J. V. WEIL, K. E. SPARKS, R. E. MCCULLOUGH, AND R. F. GROVER. Exercise ventilation correlates positively with ventilatory chemosensitiveness. J. Appl. Physiol. 45: 557-564, 1978. 11. OHKUWA, T., N. FUJITSUKA, T. UTSUNO, AND M. MIYAMURA. Ventilatory response to hypercapnia in sprint and long-distance swimmer. Eur. J. Appl. Physiol. Occup. Physiol. 43: 235-241, 1980. 12. OHYABU, Y., A. USAMI, I. OHYABU, Y. ISHIDA, C. MIYAGAWA, T. ARAI, AND Y. HONDA. Ventilatory and heart rate chemosensitivity in track-and-field at,hletes. Eur. J. Appl. Physiol. Occup. Physiol. 59: 460-464,199O. 13. POWERS, S. K., S. DODD, J. LAWLER, M. KIRTLEY, T. MCKNIGHT, AND S. GRINTON. Incidence of exercise induced hypoxemia in elite endurance athletes at sea level. Eur. J. Appl. Physiol. Occup. Physiol. 58: 298-302, 1988. 14. POWERS, S. K., S. DODD, J. WOODYARD, R. E. BEADLE, AND G. CHURCH. Hemoglobin saturation during incremental arm and leg exercise. Br. J. Sports Med. 18: 212-216, 1984. 15. RIES, A., J. FARROW, AND J. CLAUSEN. Accuracy of two ear oximeters at rest and during exercise in pulmonary patients. Am. Rev. Respir. Dis. 132: 685-689, 1985. 16. TAYLOR, M. B., AND J. G. WHITMAN. The accuracy of pulse oximeters. Anesthesia 43: 229-232, 1988. 17. TORRE-BUENO, J. R., P. D. WAGNER, H. A. SALTZMAN, G. E. GALE, AND R. E. MOON. Diffusion limitation in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol. 58: 9&9-995,1985. 18. WILLIAMS, J. H., S. K. POWERS, AND M. K. STUART. Hemoglobin desaturation in highly trained athletes during heavy exercise. 1Med. Sci. Sports Exercise 18: 168-173, 1986. 19. YELDERMAN, M., AND W. NEW. Evaluation of pulse oximetry. Anesthesiology 59: 349-352, 1983.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
M~TOHIKO,
AND IZUMI TABATA.R&~~~&QI
be-
tweenarterial oxygendesaturationand ventilation during muximal exercise.J. Appl. Physiol. 73(6): 2588-2591, 1992.---The purpose of the present study was to investigate the contribution of ventilation to arterial 0, desaturation during maximal exercise.Nine untrained subjectsand 22 trained long-distance runners [age 18-36 yr, maximal U2 uptake (vo2,,) 48-74 ml min-’ 4kg-l] volunteered to participate in the study. The subjectsperformed an incremental exhaustive cycle ergometry test at 70 rpm of pedaling frequency, during which arterial O2 saturation (Sa,,) and ventilatory data were collectedevery minute. Sao, was estimated with a pulse oximeter. A significant positive correlation was found between Sa,, and end-tidal PO, (PETE,; r = 0.72, ? = 0.52, P < 0.001) during maximal exercise. Thesestatistical results suggestthat -50% of the variability of Sao, can be accounted for by differences in PETIT, which reflects alveolar PO,. Furthermore, PETE, was highly correlated with the ventilatory equivalent for 0, (VE/~O~; r = 0.91, P < O.OOl),which indicates that PETIT could be the result of ventilation stimulated by maximal exercise,Finally, Saozwas positively related to vEli70, during maximal exercise(r = 0.74, ? = 0.55,P < 0.001).Therefore, one-half of the arterial 0, desaturation occurring during maximal exercise may be explained by lesshyperventilation, specifically for our subjects,who demonstrated a wide range of trained states. Furthermore, we found an indirect positive correlation between Sao, and ventilatory responseto CO, at rest (r = 0.45,P < 0.05), which wasmediated by ventilation during maximal exercise. These data also suggest that ventilation is an important factor for arterial 0, desaturation during maximal exercise. l
arterial oxygen saturation; ventilatory responseto carbon dioxide
thatarterialO,partialpressure (PaoJ and arterial 0, hemoglobin saturation (Sao,) are maintained at near resting levels during maximal voluntary exercise at sea level (1, 6). However, several studies (3, 4, 7, 8, 13, 14, 18) recently demonstrated a reduction of PaoZ or Sao, during heavy exercise (exerciseinduced hypoxemia or exercise-induced arterial 0, desaturation). The reduction of Pa,, or Sa,, during heavy exercise may be related to an increase in the alveolar-to-arterial PO, difference (A-aDo,) induced by diffusion limitation and ventilation-perfusion inequality (3-5, 7, ‘17). In addition, Dempsey et al. (3, 4) and Powers et al. (13, 14) reported that exercise-induced hypoxemia and desaturation were observed in subjects with lower alveolar 0, partial pressure (PA& and less hyperventilation. We hypothesize that less hyperventilation and an inITISGENERALLYACCEPTED
2588
Institute
of Fitness and Sports,
creasing A-aDo, play a major role in the genesis of arterial hypoxemia and 0, desaturation during exercise. Although the mechanisms that regulate ventilation during maximal exercise are unclear, several studies (9, 10) reported significant relationships between ventilatory responses to hypercapnia at rest and ventilation during maximal exercise. If ventilation during maximal exercise is a major factor in arterial 0, desaturation, as we hypothesize, the ventilatory response to hypercapnia at rest should be related to Sa,, during maximal exercise. Therefore, we considered it worthwhile to investigate the relationship between ventilation and Sa,, during maximal exercise and the ventilatory response to CO, in a large number of subjects. MATERIALS
AND METHODS
Subjects. Nine untrained subjects and 22 trained longdistance runners volunteered to participate in the experiments. All were healthy adult males with no history of lung disease. Descriptive data (means t SD) are as follows: age = 23 t 4 yr, body weight = 58.8 t 6.5 kg, maximal 0, uptake (TO 2max) = 62 t 7 ml mine1 kg-l STPD. The subjects were informed of the nature of the experiments and gave informed consent before testing. Procedure. The study consisted of two experiments. The subjects first performed a ventilatory response to CO, test at rest and then completed a maximal exercise test. Environmental conditions remained relatively constant during the experiments: temperature = 2O-25”C, barometric pressure = 740-755 Torr. Ventilatory response to CO, test. The ventilatory response to CO, was measured by the rebreathing method of Ohkuwa et al. (11) and Ohyabu et al. (12) with a slight modification. After arrival at the laboratory the subjects rested for 20 min while seated on a chair. Subsequently, a mouthpiece was connected to a rebreathing bag and noseclips were fitted on the subjects. The subjects rebreathed a hyperoxic and hypercapnic gas mixture from the bag through the mouthpiece for 3.0-3.5 min. The expired ventilation volume (VE) during rebreathing was measured with a pneumotachograph (model RM 200, Minato Medical Science) attached between the mouthpiece and the rebreathing bag. All gas volumes were corrected to BTPS conditions. To measure the end-tidal CO, partial pressure (PET,,, ) , the expired gas was continuously withdrawn from the mouthpiece through a sampling capillary tube connected to a gas analyzer (model MGA 1100, Perkin-Elmer). The VE and PET,,, data were
0161-7567/92 $2.00 Copyright 0 1992 the American Physiological Society
l
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ARTERIAL
DESATURATION
stored on a breath-by-breath basis and averaged every 30 s. Each rebreathing bag was filled with 5 liters of 5% CO,-95% 0,. The slope of the %-PET,,, line (S) was calculated by least-squares regression analysis. The equation VE = S (PETIT, - B) was used for determination of the ventilatory response to CO,. To confirm the S of each subject, duplicate determinations were conducted on different days in 12 of the subjects. Exercise test. The incremental cycle exercise began at a work rate of 103 W (70 rpm), and the power output was increased by 17 W/min until the subject could not maintain the fixed pedaling frequency. The subjects were encouraged during the ergometer test to exercise for as long as possible. Saoz, heart rate (HR), and a rating of perceived exertion (RPE) were monitored minute by minute during exercise. 0, uptqke ($$), VE (BTPS), the ventilatoryequivalentforo, (VE/VO, ,BTPS/STPD) ,PET~*~, and the end-tidal partial pressure of 0, (PETE,) were monitored during the last 30 s of each increase in the work rate after the RPE reached 18. Sa,, was estimated with a pulse oximeter (model OLV1200, Nihon Kohden) on the tip of the right forefinger. The accuracy of pulse oximetry has been proved by several studies (l&16,19). The subjects were advised not to strain while moving their right hands to allow us to monitor the pulse waves correctly and to measure Sh, accurately. RPE was obtained using the modified Borg scale (2). Subjects breathed through a low-resistance two-way valve (model 2700, Hans Rudolph), and the expired air was collected in Douglas bags. Expired 0, and CO, gas concentrations were measured by mass spectrometry (model MGA 1100; Perkin-Elmer), and gas volume was determined using a dry gas meter (model NDS2A-T, Shinagawa). All gas volumes were corrected to BTPS conditions. To measure PETITand PET,,,, a small amount of expired gas (60 ml/mm) was sampled continuously by the gas analyzer through a sampling capillary connected to the mouthpiece. Data analysis. The maximal value of $70, during the exercise test was called peak irO,. Typical data for all other parameters included only those values occurring at peak VO, during the exercise test. The values are expressed as means t SD. The difference between the first and second measurements of S in duplicate determinations was determined by the paired t test. The correlation coefficients among the parameters were determined by the Pearson correlation. The level of significance was established at P < 0.05. RESULTS Ventilutory response to CO, test. PET,,, increased linearly (5.9 t 1.1 Torr/min) during the rebreathing test. However, in 29 subjects, i7E did not change significantly during the 1st min of the rebreathing tests. After that, it increased linearly with the. increase in PETIT,. In the remaining two subjects, VE increased linearly with throughout the rebreathing test. The mean pmm, was 47.4 (range 42.3-52.4) Torr during the PETCO, first 30 s of the rebreathing test and 66.7 (range 62.672.3) Torr during the last 30 s. The mean TE was 10.8 (5.8-20.1) llmin BTPS during the first 30 s and 32.9 (20.6-
AND VENTILATION
2589
47.8) l/min BTPS during the last 30 s. The mean value of S in all subjects was 1.18 1 min-’ Torr-l BTPS. The reliability of S was tested by duplicate determinations for each subject in 12 cases. The mean values of the first and second measurements were 0.98 t 0.44 and 1.00 t 0.35 1 min-l Torr-l, respectively. There was no significant difference and a high correlation between the first and second measurements (r = 0.92). Exercise test. Table 1 shows the changes in Sa,, of all subjects during exercise. Saoz did not change during light exercise until 5-6 min, when, for some subjects, it decreased progressively as the intensity increased. Finally, the Saoz at which peak VoZ was obtained ranged from 92 to 97%. The peak VO, ranged from 46.5 to 73.6 ml min-’ kg? Peak Vg, was observed at exhaustion in 27 subjects. In the remaining four subjects, VO, leveled off before exhaustion. Table 2 shows the mean t SD and maximum and minimum values of peak TO, and the Saao2,VE, vE/ vo, 9 PETco, 9PETE,, and HR at which peak Vo, was obtained during the exercise test. Relationships among parameters. Table 3 presents the correlation matrix among all parameters. There were positive significant relationships among Sa,, , PET,, f and bE/iro, (see Fig. 1 for relationship between Sa,, and vE/vO&. 8 at rest was significantly related to ventilation during maximal exercise (expressed as VE/~O~, VE, PETE,, or PET,,~). We found a significant positive correlation between Saoz during maximal exercise and S at rest, although the correlation coefficient was quite weak (r = 0.45). l
l
l
l
l
l
DISCUSSION
In the present study, we found a positive correlation between Sa,, and PET,, (? = 0.52) during maximal exercise. This result statistically shows that ~50% of the variability of Sao, can be accounted for by differences in PET*,, which reflects PACT.Furthermore, PETE, was highly correlated with i7E/vO, (r = 0.91), which indicates that PETIT is the result of ventilation stimulated by maximal exercise. Finally, Sa,, was positively related to vE/ VO, during maximal exercise (Fig. 1; ? = 0.55). Therefore, -50% of the variability in arterial 0, desaturation during maximal exercise can be explained by less hyperventilation, specifically in our subjects. Hopkins and McKenzie (7) reported that Pao, was not related to PAo2 (r = 0.51, P = 0.08) but was closely related to A-aDo, (r = 0.98, P < 0.001) during heavy exercise. They concluded that the increase in A-aDo, plays a major role in genesis of arterial hypoxemia and desaturation during maximal exercise with a lesser contribution of ventilation. We consider that this difference may be explained by the difference in the trained state of the subjects, i.e., they used only trained subjects, whereas we used subjects from a wide range of trained states. Trained subjects breathe less per unit of metabolic rate than untrained subjects (4,9,10); therefore the variability of ventilation in the study of Hopkins and McKenzie was narrower than ours. In fact, the variability in ~EP?o~ (29.5-36.4) for their subjects (7) corresponds to ~35% of the variability in vE/vO, (29.5-49.3) for our subjects. Powers et al.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
2590 TABLE
ARTERIAL
1, Individual
Sa, responses
DESATURATION
AND
at rest and during exercise Exercise
Subj No.
Rest
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 Mean tSD * Arterial
2
3
4
5
6
7
8
9
10
11
12
13
9% 9% 9% 99 97 97 9% 99 99 97 99 99 9% 9% 99 99 9% 9% 98 98 9% 98 98 99 98 9% 9% 97 9% 98 9%
97 97 97 97 96 97 98 97 97 96 97 98 97 95 9% 98 96 97 97 99 97 98 96 97 97 96 98 97 9% 97 97
96 9% 97 97 96 97 96 97 97 95 97 9% 97 96 9% 94 95 97 9% 9% 97 95 97 9% 97 97 97 96 98 97 97
97 97 99 97 97 97 96 9% 97 96 9% 9% 9% 9% 9% 94 97 97 97 98 97 96 96 98 97 97 99 96 97 9% 97
9% 96 97 96 97 96 96 98 98 96 97 9% 97 9% 9% 96 96 96 96 97 97 97 95 98 97 97 9% 96 97 97 97
9% 97 9% 97 96 97 98 97 97 96 98 9% 9% 97 97 96 97 96 96 97 96 95 95 9% 97 97 97 96 97 97 97
96 97 98 97 97 97 9% 97 98 96 97 97 9% 96 97 97 95 96 96 96 96 96 95 97 97 97 97 95 97 96 97
97 96 9% 96 96 96 97 96 98 96 96 96 9% 97 96 95 94 96 95 96 95 97 94 97 97 96 96 95 95 95 97
97 97 97 96 96 96 97 96 98 96 95 97 97 96 97 96 93 96 95 95 95 94 94 97 96 95 96 96 95 95 96
96* 96 97 97 96 97 97 96 96 94 95 96 96 95 95 93 93 94 95 94 95 94 94 97 96 95 95 95 95 96 95
96 96 96 95* 97* 97 95 96 93 94* 96 96 95* 95 95 92 93* 94 94 94* 94* 93 97 95* 94 95 94 95 95 95*
96 97 96*
96* 96
96*
97* 96 95* 93*
96 95*
95
98.1 kO.6
97.1 t0.8
96.8 t1.0
97.2 a.0
96.9 to.9
96.9 -to.9
96.7 HI*%
96.1 kl.1
95.9 zkl.1
95.3 3-1.2
94.8 H.2
saturation
(Saoz,
%) at which
peak
O2 uptake
(13, 14) reported that PETE, tended to be lower and PET,,, higher in subjects with desaturation during maximal exercise. Dempsey (3) noted that PAN, was ~110 Torr in subjects with greater hypoxemia. These previous results support the present finding. We suggest that about half of desaturation is due to a lower PACT (less hyperventilation), and the other half may be due to a wider A-aDo,, specifically in our subjects. Martin et al. (9, 10) demonstrated that ventilation during maximal exercise is proportional to the ventilatory response to hyoxia or hypercapnia at rest. There was 2: Mean, maximum, and minimum values at peak Vu2 during exercise test and S at rest TABLE
%,J 7-6 VEIV~~ (BTPS/STPD) VU,, ml . mine1 kg-’ VE, 1 min-’ kg-’ PETco2, Torr PET*, , Torr HR, beats/min S, 1 min-’ Torr-’ BTPS l
l
l
l
Time, min
1
0, hemoglobin
l
VENTILATION
Mean t SD
Minimum
Maximum
94.5t1.9 38.2t5.2 61.9t6.8 2.35tO.31 34.lt3.7 113.7t3.9 1%5_+11*3 1.171-0.48
92 29.5 46.5 1.71 28.0 105.7 165 0.33
97 49.3 73.6 2.84 42.7 121.1 202 2.33
n = 31. TjE/k$, ventilatory equivalent fur 0,; voz, expired ventilation vulume; PETIT,and PETo,, end-tidal HR, heart rate; S, slope of $7~ - PETE% relationship.
0, uptake; 93, PCO,and PO,;
95* 96* 95 95 92
95 94 92*
95 94 94
94* 94*
92* 97* 95 94 95* 93 94 94*
92
94.8 tl.4
94* 93*
92* 92* 94
93*
94.0 tl.6
94.2 -+1.3
was observed.
a wide range of ventilation during maximal exercise in our subjects. We also found a positive relationship between ventilation during maximal exercise (expressed as VE/~O~ or VE) and S at rest. Furthermore, PETIT,, which reflects arterial PQ, was negatively correlated to S. However, S was obtained in the PCO, range above resting, whereas Pcoz during maximal exercise was below resting. Furthermore, these correlations were quite weak (? = 0.2-0.4); therefore, other factors should be considered as contributors to the variability of ventilation during maximal exercise. We found a significant positive correlation between Saoz during maximal exercise and S at rest, although the correlation coefficient was quite weak (r = 0.45). Also, TABLE
3. Correlation matrix among all parameters Sa0g
sao2 VE/VO, vo, VE PET,*, pE%z HR s
1.00
VE/iTOz VQ2 0.74” 1.00
-0.70” -0.43* 1.00
VE PETcoz PETIT 0.16 0.62* 0.43* 1.00
-0.65* -0.%9* 0.24 -0.71” 1.00
0.72” 0.91* -0.34 0.62* -0.97* 1.00
HR
S
0.21 0.20 -0.03 0.14 -0.33 0.35 1.00
0.45” 0.62* -0.22 0.45* -0.54* 0.55* 0.10 1.00
n = 31. *P < 0.05.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
ARTERIAL
DESATURATION
In summary, we demonstrated that Sap, had a positive relationship with PET,, and VE/~O, during maximal exercise in our subjects. These data suggest that -50% of the arterial 0, desaturation during maximal exercise was induced by less hyperventilation, specifically in those subjects with a wide range of trained states.
98 96 S&2
64
94
Present address of M. Miyachi: Dept. of Health and Sports Sciences, Kawasaki University of MedicaI Welfare, 288 Matsushima, Kurashiki City, Okayama 701-01, Japan. Present address and address for reprint requests: I. Tabata, Lab. for Exercise Physiology, Div. of Health Promotion, National Institute of Health and Nutrition, 1-23-1 Toyama, Shinjuku-ku, Tokyo, 162 Japan.
92 901 25
2591
AND VENTILATION
I
30
I
I
35
40
I
45
1
50
Received 18 May 1990; accepted in final form 23 June 1992.
i/E/co2
FIG. 1. Relationship between arterial .02 hemoglobin saturation (Sao,) and ventilatory equi.valent for 0, (VE/%,; r = 0.74, ?. = 0.55, P < 0.001). Sao, and VE/Q were the data at which peak VO, was obtained in exercise tests. Correlation shows that 50% of arterial 0, desaturation during maximal exercise can be explained by less hyperventilation in our subjects.
there were statistically significant relationships between Sa, and ventilation during maximal exercise and between ventilation during maximal exercise and S in our subjects. These relationships led us to examine the relationships among Sa,, , ventilation during maximal exercise, and S at rest. So, we performed a stepwise regression analysis among Sao?, VE/VO,, and S. Between Sa,, and vE/%2, the simple correlation coefficient was 0.74, and among Saoz, vE/vg:!, and S, the multiple correlation coefficient was also 0.74. Moreover the standard partial regression coefficients of vE/%, and S were 0.57 and only -0.02, respectively. These statistical results suggest that S at rest has no independent effect on Sao, during maximal exercise. In other words, ventilation during maximal exercise is an intermediary factor between S at rest and Sa,, during maximal exercise. Martin et al. (9, 10) also reported a significant positive relationship between ventilation during maximal exercise and the ventilatory response to hypercapnia at rest. These data support our main finding that -50% of arterial 0, desaturation is due to less hyperventilation during maximal exercise. One methodological concern is the use of the pulse oximeter to measure Sa,,. Although this instrument has been shown to provide valid and reliable assessment of trends in Sa,, at rest and during exercise, the standard error of the estimate is LO-2.0% (15, 16, 19). This random error might affect Sa,, in the present study. The other concern is the validity and reliability of the S measurement. The ventilatory response to CO, was measured by the rebreathing method of Ohkuwa et al. (11) and Ohyabu et al. (12) with a slight modification. The S values in these studies do not seem to differ from the S values in the present study. Therefore the value of S of the present study may be valid for the Japanese population. In addition, the reliability of S in the present study was examined by duplicate determinations for 12 subjects. There was a high correlation (r = 0.92) and no significant difference between the first and second measurements. These results agree well with those of Ohkuwa et al. and confirm the reliability of S in the present study.
REFERENCES 1. ASMUSSEN, E., AND M. NIELSEN. Alveolo-arterial gas exchange at. rest and during work at different 0, tensions. Acta Physiol. Stand. 50: 153-166,196O. 2. BORG, G. A. V. Psychophysical bases of perceived exertion. IWed. Sci Sports Exercise 14: 377-381, 1982. 3. DEMPSEY, J. A. Is the lung built for exercise? Med. Sci. Sports Exercise 18: 143-155,19&6. 4. DEMPSEY, J. A., P. G. HANSON, AND K. S. HENDERSON. Exerciseinduced arterial hypoxemia in healthy human subjects at sea level. J. Physiol. Lord. 355: 161-175, 1984. 5. GALE, G. E., J. R. TORRE-BUENO, R. E. MOON, H. A. SALTZMAN, AND P. D. WAGNER. Ventilation-perfusion inequality in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol.
58: 978-988,
1985.
6. HESSER, C. M., AND G. MATELL. Effect of light and moderate exercise on alveolar-arterial 0, tension difference in man. Acta Physiol. Scund. 63: 247-256,1965. 7. HOPKINS, S. R., AND D. C. MCKENZIE. Hypoxic ventilatory response and arterial desaturation during heavy exercise. J. Appl. Physid. 67: 1119-l 124, 1989. 8. LAWLER, J., S. K. POWERS, AND D. THOMPSON. Linear relationship between VO, maxand vogrnax decrement during exposure to acute hypoxia. J. Appl. Physiol. 64: 1486-1492, 1988. 9. MARTIN, B. J., K. E. SPARKS, C. W. WILLICH, AND J. V. WEIL. Low exercise ventilation in endurance athletes. Med. Sci. Sports Exercise 11: 181-185, 1979. 10. MARTIN, B. J.) J. V. WEIL, K. E. SPARKS, R. E. MCCULLOUGH, AND R. F. GROVER. Exercise ventilation correlates positively with ventilatory chemosensitiveness. J. Appl. Physiol. 45: 557-564, 1978. 11. OHKUWA, T., N. FUJITSUKA, T. UTSUNO, AND M. MIYAMURA. Ventilatory response to hypercapnia in sprint and long-distance swimmer. Eur. J. Appl. Physiol. Occup. Physiol. 43: 235-241, 1980. 12. OHYABU, Y., A. USAMI, I. OHYABU, Y. ISHIDA, C. MIYAGAWA, T. ARAI, AND Y. HONDA. Ventilatory and heart rate chemosensitivity in track-and-field at,hletes. Eur. J. Appl. Physiol. Occup. Physiol. 59: 460-464,199O. 13. POWERS, S. K., S. DODD, J. LAWLER, M. KIRTLEY, T. MCKNIGHT, AND S. GRINTON. Incidence of exercise induced hypoxemia in elite endurance athletes at sea level. Eur. J. Appl. Physiol. Occup. Physiol. 58: 298-302, 1988. 14. POWERS, S. K., S. DODD, J. WOODYARD, R. E. BEADLE, AND G. CHURCH. Hemoglobin saturation during incremental arm and leg exercise. Br. J. Sports Med. 18: 212-216, 1984. 15. RIES, A., J. FARROW, AND J. CLAUSEN. Accuracy of two ear oximeters at rest and during exercise in pulmonary patients. Am. Rev. Respir. Dis. 132: 685-689, 1985. 16. TAYLOR, M. B., AND J. G. WHITMAN. The accuracy of pulse oximeters. Anesthesia 43: 229-232, 1988. 17. TORRE-BUENO, J. R., P. D. WAGNER, H. A. SALTZMAN, G. E. GALE, AND R. E. MOON. Diffusion limitation in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol. 58: 9&9-995,1985. 18. WILLIAMS, J. H., S. K. POWERS, AND M. K. STUART. Hemoglobin desaturation in highly trained athletes during heavy exercise. 1Med. Sci. Sports Exercise 18: 168-173, 1986. 19. YELDERMAN, M., AND W. NEW. Evaluation of pulse oximetry. Anesthesiology 59: 349-352, 1983.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on November 19, 2018. Copyright © 1992 the American Physiological Society. All rights reserved.
