Share this article on:

Ventilatory sensitivity to carbon dioxide: the influence of exercise and athleticism


Medicine & Science in Sports & Exercise: June 1996 - Volume 28 - Issue 6 - p 685-691
Basic Sciences/Regulatory Physiology: Original Investigations

Endurance training reduces the ventilatory response to a given level if work, and there is evidence that endurance athletes possess attenuated chemosensitivity at rest; but it is unclear whether attenuation persists during exercise. We compared the carbon dioxide sensitivity (S) of endurance-trained (ETG), sprint-trained (STG), and control subjects (CG), at rest and during cycle ergometry. Steady-state carbon dioxide (CO2) inhalation was employed; ventilatory parameters were measured using an ultrasonic flowmeter linked to a computer. CO2 concentrations were measured at the mouth using an infrared CO2 analyzer or mass spectrometer. Mean resting CO2 sensitivity of the ETG was significantly lower than that of the STG (P < 0.05), but not the CG(P < 0.058). S increased from rest to exercise in all endurance-trained subjects, but the responses of the STG and CG were varied. Compared to rest, mean S was significantly higher during exercise for the ETG, but not for the STG or CG. S was the same in all groups during exercise. During air breathing exercise all subjects were mildly hypercapnic. The ETG showed the greatest rise in mean alveolar PCO2, but this could not be attributed to attenuated chemosensitivity since responsiveness during exercise was identical in all three groups.

School of Sport & Exercise Sciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UNITED KINGDOM, and Department of Human Sciences, Loughborough University of Technology, Loughborough, Leicestershire LE11 3TU, UNITED KINGDOM

Submitted for publication May 1995.

Accepted for publication March 1996.

Address for correspondence: Dr. A. K. McConnell, School of Sport & Exercise Sciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. E-mail:

Endurance training has been reported to induce a reduction in the ventilatory response to a given level of work (34). A number of studies have reported a correlation between the magnitude of the ventilatory response to exercise (Δ˙VE/ΔVCO2) and the ventilatory sensitivity to inhaled carbon dioxide(Δ˙VE/ΔPCO2), both at rest(10,20,22,29) and during exercise(22). Indeed, McConnell et al. (22) noted a stronger correlation between Δ˙VE/ΔVCO2 and CO2 sensitivity (S) during exercise than at rest. The physiological significance of S remains unresolved, but it has been argued that these data support a role for S in determining the magnitude of the exercise hyperpnea(22,29). It is reasonable to hypothesize that in endurance events, it is desirable to conserve energy by restraining the hyperpnea and allowing PCO2 to rise; such a strategy is unnecessary in short events such as sprinting. Under these conditions, one might expect the S of endurance athletes to be lower during exercise than in sprinters or untrained subjects.

The available literature has been unable to establish whether there is a clear and consistent difference in the S of trained or untrained subjects, either at rest or during exercise. Several studies have found endurance trained athletes “tend” to have a lower resting S than untrained subjects (29,31,32), but the data are inconclusive and are contradicted by reports that endurance-trained athletes have a similar S to untrained subjects(19,26,27). The literature comparing athletes from different sporting disciplines is similarly confusing, but there is some evidence that a reduced S is a feature of endurance training, rather than power or sprint-based training (4,24).

There remains no definitive answer to the question of whether exercise alters CO2 sensitivity. There have been reports that in endurance-trained athletes, S will always fall from rest to exercise(25), or always rise (5). It is unclear whether the tendency for endurance athletes to possess reduced S at rest persists during exercise; the available evidence suggests that there is little difference between the S of endurance and non-endurance athletes during exercise (20), but the literature is far from consistent.

The aim of the present study was to compare the ventilatory response to inhaled CO2 of endurance-trained, sprint-trained, and untrained(control) subjects, at rest and during moderate exercise. Using rigorous criteria for the categorization of subjects, we wished to determine whether individuals in these three categories possess differing responsiveness to CO2, and/or exercise.

Back to Top | Article Outline



Thirty subjects were studied (15 male and 15 female); all subjects were healthy nonsmokers with no previous history of respiratory complaints. Subjects were recruited into one of three groups: 1) an endurance-trained group (ETG), 2) a sprint-trained group (STG), or 3) a control group (CG). There were 10 subjects in each group: 5 males and 5 females. The ETG consisted of subjects involved regularly in competitive, endurance-based athletic events, the STG consisted of subjects involved regularly in competitive, power-based athletic events. Each subject had been free from injury and training regularly for some time. The sporting standard of each subject varied, but all competed for the University first team or represented their county or country in their chosen event. None of the subjects in the CG were involved in regular training or representative sport of any kind. University Ethics Committee approval and informed consent were obtained.

Back to Top | Article Outline

Study Design

The study was in two parts: measurement of ˙VO2max and the˙VO2-work rate relationship. Prior to measurement of CO2 sensitivity (S), subjects attended the laboratory twice: 1) to determine˙VO2max (ml·min-1·kg-1), and 2) to establish a steady-state ˙VO2-work rate relationship.

Apparatus. Exercise was performed on an electro-magnetically braked cycle ergometer (Siemens 380B). Minute ventilation (˙VE), oxygen consumption (˙VO2), and carbon dioxide production(˙VCO2) were measured using one of two automated systems:

Oxycon 4. Subjects breathed via a mouthpiece attached to a low-dead-space one-way valve. Volume of expired gas was measured by dry gas meter; oxygen and carbon dioxide concentrations were measured by paramagnetic and infrared analyzers, respectively (Mijnhardt). The gas analyzers were calibrated before each experiment using standard gases analyzed previously by the Lloyd-Haldane technique. Calibration of the gas meter was checked periodically using repeated evacuations of a 1-1 syringe pump. Heart rate was monitored continuously by three-lead ECG (Cardiorater, CR7). Print-outs of ventilatory and gas exchange parameters were obtained at 30-s intervals.

Computerized breath-by-breath system. Subjects breathed via an ultrasonic flowmeter (Birmingham Flowmetrics Ltd.). Respired gas concentrations were measured at the mouth by mass spectrometer (Airspec MGA 2000). Signals from the measurement apparatus were analyzed by computer(Macintosh IIcx), which calculated ventilation and gas exchange breath-by-breath in realtime. Calibration was performed before each experiment: the flowmeter by repeated evacuations of a 1-1 syringe pump; the mass spectrometer using standard gases analyzed previously by the Lloyd-Haldane technique. Heart rate was monitored continuously by three-lead ECG.

Protocol. Determination of ˙VO2max, An incremental exercise test was performed on a cycle ergometer. The starting work rate depended upon the gender, mass, and sporting standard of the subject; 180 W was the starting work rate for athletic male subjects, and 140 W for sedentary male subjects; 120 W for athletic female subjects, and 80 W for sedentary female subjects. The work rate of ergometer was increased every 2 min, by 40 W for men and 20 W for women, until the subject was unable to continue cycling due to exhaustion. By using this protocol a subject's˙VO2max was achieved within approximately 10 to 15 min. To confirm whether subjects had achieved their ˙VO2max they were asked, after a 5 min rest, to cycle at a supramaximal load for a maximum of 2 min, the aim being to demonstrate a plateau in oxygen consumption, defined as an increase of less than 2 ml·min-1·kg-1, with an increase in work rate. This increase from the subject's final work rate was 40 W for men and 20 W for women.

˙VO2-work rate relationship. The second laboratory trial was used to establish the subject's steady-state ˙VO2(ml·min-1·kg-1) at four submaximal work rates. Subjects cycled at each submaximal work rate for a minimum of 5 min; steady-state was defined as a change of less than 2 ml·min-1·kg-1 in ˙VO2 in the last 30 s of exercise. The first three work rates were performed consecutively, after which the subject rested for between 2-5 min before cycling at the fourth and final work rate. Linear regression analysis was used to derive an equation relating ˙VO2 to work rate for each subject; the work rate equivalent to 30% of the subject's ˙VO2max was then calculated. This work rate was used for measurement of S during exercise.

Back to Top | Article Outline

Measurement of CO2 Sensitivity

Apparatus. Carbon dioxide sensitivity was measured at rest and during the steady state of cycle ergometer exercise at 30% of˙VO2max.

The ventilatory response to CO2 was measured using an open-circuit apparatus by a steady-state method. Air was blown via a flowmeter through a wide-bore tube (inspiratory line) at a rate sufficient to exceed the subject's peak inspiratory flow. Subjects inspired and expired via an ultrasonic flowmeter attached to a side arm on the inspiratory line. Subjects were interfaced to the flowmeter using a mouthpiece, which was used in the usual way with a nose clip. Pure CO2 was bled into the inspiratory line at its proximal end to obtain constant inspired concentrations of 0% (room air), 2%, 3%, and 4% CO2 in air. The composition of the inspirate was monitored continuously via a chart recording and adjusted by the experimenter as necessary.

The signal from the flowmeter was analyzed by computer (Amstrad PC1512). Real-time breath-by-breath analysis was carried out and all breaths occurring within the measurement period were averaged. The flowmeter was calibrated volumetrically before and after each experiment using a 1-1 syringe.

Respired CO2 concentration was measured continuously at the mouth by mass spectrometer (Airspec MGA 2000), which sampled from the manifold of the mouth-piece. The output was recorded continuously on a chart recorder (Gould BS 272). The CO2 analyzer was calibrated before and after each experiment using standard gases analyzed previously by the Lloyd-Haldane technique. An ECG displayed continuously via three chest electrodes and an ECG monitor.

Protocol. Each gas mixture (room air, 2%, 3%, and 4% CO2), was administered for 14 min in total. For the first 8 min the subject sat on the cycle ergometer breathing through the mouthpiece; minute ventilation(˙VE) and PaCO2 were measured for the last 2 min. The subject then cycled for 6 min at 30% of ˙VO2max (determined during the second visit) breathing exactly the same concentration of gases. Minute ventilation and PaCO2 were measured during the last minute of exercise. Between each inhalation period the subject dismounted the ergometer and rested for a minimum of 5 min. The order in which room air and the three CO2 concentrations were administered was randomized.

Subjects arrived at the laboratory having fasted for at least 2 h, during this time they had not consumed any tea, coffee, or alcohol. All the subjects remained unaware of the exact purpose of the experiment. The athletic subjects continued to train normally in the evenings.

Data analysis. S was calculated using the following equation:Equation

where S is the slope of the CO2 response curve, and therefore CO2 sensitivity (e.g., the change in ˙VE per unit change in PaCO2) and B are the x-axis intercept of the extrapolated CO2 response curve; the PaCO2 at which ˙VE is theoretically zero.

The mean alveolar partial pressure of CO2 (PaCO2) was used as an estimate of arterial PCO2 (PaCO2) at rest and during exercise(30). PaCO2 was obtained breath-by-breath from the inspired and expired CO2 concentrations recorded continuously on the chart recorder and derived using the “graphical method”(35). This method of estimating PaCO2 has been validated by a number of studies (30). The breath-by-breath values obtained during the sampling period were averaged to give a mean value.

Hey plot. The Hey plot is a plot of ˙VE against VT; the relationship between these two variables is characterized by the following equation (12):Equation

where m is the slope of the regression line and k is the intercept of the extrapolated regression line on the VT axis. Regression analysis was applied to the plot of ˙VE against VT for each subject, both at rest and during exercise.

Back to Top | Article Outline

Statistical Analysis

Standard statistical methods were used. In all tables the arithmetic mean, standard deviation (SD), and range of values are reported. Tests of statistical significance were made by one-way analysis of variance and Fisher's least square difference. The Pearsonian coefficient of skewness was used to determine whether the data were normally distributed about the mean.

Linear regression analysis was used to define the relationship between two variables, and statistical significance was tested using the correlation coefficient. In some of the illustrations, bar charts are used to illustrate the difference between means, and here the standard error (SEM) has been used to show the variation around the mean. The significance level was set at 5%.

Back to Top | Article Outline



There were no significant differences in the age and stature of the three groups, but the mass and B.S.A. of the STG were significantly higher than those of the ETG. The mean ˙VO2max of the ETG was significantly larger than that of the STG and CG; the ˙VO2max of the STG was significantly larger than that of the CG (Table 1).

Back to Top | Article Outline

Normocapnic ˙VE and PaCO2

There were no significant differences among the three groups in˙VE at rest, or the rise in ˙VE from rest to exercise. There was no significant difference among the three groups in their PaCO2 at rest. During exercise, PaCO2 rose from its resting value in all subjects; the increase for the ETG was significantly larger than that of the STG and CG (Table 2).

Back to Top | Article Outline

CO2 Sensitivity (S)

At rest, the S of the ETG was significantly lower than that of the STG, but just failed to be significantly lower than that of the CG (P < 0.058) (Table 3). All endurance-trained subjects demonstrated an increase in S from rest to exercise; the responses of the STG and CG were variable. The ETG had a significantly larger change in mean S from rest to exercise than either the STG or CG, and a significantly larger S during exercise than at rest (Fig. 1). There was no significant change in mean S from rest to exercise for either the STG or CG. There was no significant difference among the three groups in their value of S during exercise.

Back to Top | Article Outline

The X-Axis Intercept (B)

At rest, B was significantly higher for the STG than for either the ETG or CG (Table 3). During exercise there were no significant differences among the three groups in their values for B. Both the STG and CG had significantly lower values for B at rest compared to exercise. The change in B from rest to exercise was significantly smaller for the ETG than either the STG or CG (Fig. 2).

Back to Top | Article Outline

Breathing Pattern

The Hey plot. The ETG had a lower mean value for parameter m and a larger mean value for parameter k than either the STG or CG, both at rest and during exercise, but these differences were not statistically significant. There was no significant correlation between CO2 sensitivity and parameter m, either at rest or during exercise.

Back to Top | Article Outline


The resting CO2 sensitivity of endurance-trained athletes (ETG) was significantly lower than that of either power-based athletes (STG), or untrained subjects (CG). However, during exercise there was little difference in the mean S of the three groups. Exercise stimulated variable changes in S in the STG and CG, but always induced an increase in ETG. Thus, the data confirm the suggestion that endurance-trained athletes do possess attenuated resting CO2 sensitivity, but demonstrate that this does not persist during exercise. In addition, air breathing exercise was mildly hypercapnic in all subjects, with the ETG exhibiting the largest increase in PaCO2. This finding supports the hypothesis that endurance athletes optimize breathing efficiency by restraining their ventilatory response and permitting a slight hypercapnia. However, the data do not support a role for CO2 sensitivity in this adaptation.

Back to Top | Article Outline

CO2 Sensitivity at Rest

Our finding that endurance-trained athletes had the lowest resting S is in agreement with the majority of previous investigators(3,25,31,32). The difference was statistically significant between the ETG and STG, but just beyond the 5% level of significance for the CG (P < 0.058). It has been suggested that S is inversely related to ˙VO2max, and that a low resting S may be a feature of endurance training (3). However, when the data from our three groups were pooled, we found no evidence of a relationship between S and ˙VO2max.

Although the consensus within the literature appears to be that endurance-trained individuals “tend” to have a lower resting S, some previous studies have found little difference between athletic and untrained subjects (19,27). There are two possible explanations for this inconsistency. Firstly, the categorization of the athletic groups has not always been clear, with endurance and non-endurance-trained subjects being placed into the same group. Our data suggest that categorization is an important consideration, as a lower value for resting S appears to be a feature of endurance-, but not power-based athletes. A further source of the discrepancy may be the method of CO2 administration used. The studies noting no influence of athleticism upon S have all employed the rebreathing technique. The value of CO2 response curves derived using this method have been questioned(6,15). Jacobi et al. (16) have suggested that the rebreathing method generally produces higher values for CO2 sensitivity, and that these reflect a true and larger˙VE response to higher PCO2 values. However, Pianosi et al.(28) have recently reported that in their hands, there is no statistically significant difference in the gradient of the CO2 response (S) derived using the two methods. Nevertheless, it is possible that one or both of these methodological differences between studies may explain the diversity of their findings.

Back to Top | Article Outline

CO2 Sensitivity during Exercise

Perhaps the most intriguing observation from the present study was the finding that during exercise, the ˙VE-PaCO2 relationship became virtually identical in our three groups. There remains no agreement within the literature upon the influence of exercise on S(1,2,4-9,13-17,20-23,25,26,36). It has been suggested that the rebreathing method is more susceptible to error during exercise because the higher ˙VO2 can lead to a rapid shrinkage of the rebreathing bag; this can alter the linearity of the rise in PCO2, disturbing the equilibrium in PCO2 between bag and chemoreceptors (11). It is interesting to note that only studies employing the rebreathing method have found that all subjects demonstrate a decrease in CO2 sensitivity from rest to exercise(4,25). Duffin et al. (9) have suggested that the fall in CO2 sensitivity during exercise is an artifact of the rebreathing method itself. The high inspired CO2 concentrations associated with the technique and the increased metabolic rate of exercise lead fairly quickly to a maximum ˙VE. This can produce a flattening off in the CO2 response curve at the higher PCO2 values, as ˙VE fails to increase with further rises in PaCO2.

Two previous studies have compared the S of athletes and non-athletic control subjects during exercise, and the results are contradictory(18,25). Lally et al. (18) employed a steady-state method and observed no significant difference in the mean CO2 sensitivity of endurance-trained runners and control subjects. In contrast, Miyamura et al. (25) employed the rebreathing method and observed that marathon runners had a significantly lower CO2 sensitivity than control subjects. To our knowledge, only one previous study has compared the CO2 sensitivity of endurance and non-endurance-trained athletes during exercise (20). Martin et al. (20) employed a steadystate method and observed no significant difference between the two groups; this finding is in agreement with our own observations. It is possible that the findings of Miyamura et al. (25) are unique to the rebreathing method.

In the present study, endurance athletes always demonstrated an increase in CO2 sensitivity from rest to exercise, whereas the other two groups demonstrated a heterogeneous pattern of change. In a previous study, we have examined the plasticity of S from day to day. We found that the increase in S displayed by endurance-trained subjects was a consistent feature of their response to exercise (33). Our data support the notion that changes in CO2 sensitivity from rest to exercise are present in a normal healthy population, but indicate that the magnitude and direction of change varies between sprint- and endurance-trained individuals and may be linked to athleticism. It seems that an important factor in the pattern observed in endurance athletes is their lower resting value for S; the underlying mechanism and functional significance of which remains to be elucidated.

The lack of agreement within the literature on the influence of exercise upon CO2 sensitivity can probably be accounted for by the intersubject variability, together with differences in methodology, protocol, and selection of subjects. We believe that our data have rationalized some of these inconsistencies by identifying an important factor in determining the influence of exercise upon S, viz., participation in endurance-based sports.

There has been some speculation that where exercise induces changes in S, the magnitude of the change is a function of work rate(1,6,15,16,21,36). A potential weakness of many previous studies has been the use of identical work rates for all subjects. Under such conditions, it is possible that differing magnitudes of change in CO2 sensitivity may simply be a reflection of a subject's physical capacity to exercise at a given absolute work rate. In the present study, subjects exercised at 30% of their ˙VO2max, i.e., the same `relative' work rate. The ETG exercised at a higher mean absolute work rate than the other two groups, but we do not believe that difference in work rate could account for the differences that we have observed among the three groups. The endurance-trained women exercised at a lower mean absolute work rate (51.0 ± SD 9.62 W) than either the men of the STG (80.0± SD 13.7 W), or the CG (65 ± SD 14.6 W), yet the non-endurance-trained men demonstrated variable changes in S from rest to exercise, whilst the endurance-trained women all demonstrated an increase. Furthermore, the endurance-trained men exercised at a significantly higher work rate than the women, but the largest change in CO2 sensitivity was recorded in a woman (105%). Finally, we observed no correlation between work rate and the size of the change in CO2 sensitivity from rest to exercise. It therefore seems extremely unlikely that level of work played a significant role in determining the magnitude or direction of change in S from rest to exercise.

Back to Top | Article Outline

Hypercapnia during Exercise

The endurance athletes displayed the largest rise in PaCO2(ΔPaCO2) from rest to exercise in normocapnia. We do not believe that this was an artifact of work rate, as there was no significant correlation between work rate and ΔPaCO2. Furthermore, the endurance-trained women exercised at a lower mean work rate than the men in the other two groups, but demonstrated a greater ΔPaCO2. These observations suggest that the ETG adopted a breathing strategy that minimized respiratory work by allowing PaCO2 to rise. The ventilatory demands of a given work intensity may be reduced in two ways. Firstly, tidal volume(VT) may be raised in order to reduce the VD/VT ratio; under these conditions ˙VE is reduced, but PaCO2 remains unchanged. Alternatively, ˙VE may be reduced without any alteration in VT; under these conditions, PaCO2 must rise. Our data suggest that the endurance athletes adopted the latter strategy, since they showed a larger increase in PaCO2 during normocapnic exercise, but no significant difference in their pattern of breathing (as defined by the Hey relationship). The mechanism responsible for this phenomenon remains the subject of speculation, but our data indicate that it cannot be attributed to a blunted chemosensitivity during exercise.

Back to Top | Article Outline


Data from the present study support the notion that endurance athletes adopt a breathing strategy that minimizes respiratory work. On the basis of resting CO2 sensitivity data, one might have attributed this to the lower value of S observed at rest in the ETG. However, during exercise the characteristics of the CO2 response curves (S and B) of the three groups became virtually identical. These data argue against a role for chemosensitivity in the minimization of respiratory work. We believe that our findings have also helped to rationalize some of the contradictions that exist in respect of CO2 sensitivity, exercise, and endurance training.

Figure 1-Mean change in CO2 sensitivity (Δs) from rest to exercise for the three groups studied. ETG = endurance-trained group; STG = sprint-trained group; CG = control group.

Figure 1-Mean change in CO2 sensitivity (Δs) from rest to exercise for the three groups studied. ETG = endurance-trained group; STG = sprint-trained group; CG = control group.

Figure 2-Mean change in the x-axis intercept (Δb) from rest to exercise for the three groups studied. ETG = endurance-trained group; STG = sprint-trained group; CG = control group.

Figure 2-Mean change in the x-axis intercept (Δb) from rest to exercise for the three groups studied. ETG = endurance-trained group; STG = sprint-trained group; CG = control group.

Back to Top | Article Outline


1. Asmussen, E. and M. Nielsen. Ventilatory response to CO2 during work at normal and low oxygen tensions. Acta Physiol. Scand. 39:27-35, 1957.
2. Bannister, R. G., D. J. C. Cunningham, and C. G. Douglas. The CO2 stimulus to breathing in severe exercise. J. Physiol.(Lond.) 125:90-117, 1954.
3. Byrne-Quinn, E., J. V. Weil, E. I. Sodal, G. F. Filley, and R. F. Grover. Ventilatory control in the athlete. J. Appl. Physiol. 30:91-98, 1971.
4. Clark, T. J. H., and S. Godfrey. The effect of CO2 on ventilation and breath-holding during exercise and while breathing through an added resistance. J. Physiol. (Lond.) 201:551-556, 1969.
5. Clark, J. M., R. D. Sinclair, and J. B. Lennox. Chemical and nonchemical components of ventilation during hypercapnic exercise in man.J. Appl. Physiol. 48:1065-1076, 1980.
6. Cummin, A. R. C., J. Alison, M. S. Jacobi, V. I. Iyawe, and K. B. Saunders. Ventilatory sensitivity to inhaled CO2 around the control point during exercise. Clin. Sci. 71:17-22, 1986.
7. Cunningham, D. J. C., B. B. Lloyd, and J. M. Patrick. The relation between ventilation and end-tidal PCO2 in man during moderate exercise with and without CO2 inhalation. J. Physiol. (Lond.) 619:104-106, 1963.
8. D'Angelo, E. and G. Torelli. Neural stimuli increasing respiration during different types of exercise. J. Appl. Physiol. 30:116-121, 1971.
9. Duffin, J., R. R. Bechbache, R. C. Goode, and S. A. Chung. The ventilatory response to carbon dioxide in hyperoxic exercise.Respir. Physiol. 40:93-105, 1980.
10. D'Urzo, A. D., K. R. Chapman, and A. S. Rebuck. Effect of inspiratory resistive loading on control of ventilation during progressive respiratory exercise. J. Appl. Physiol. 62:134-140, 1987.
11. Flenley, D. C. and P. M. Warren. Ventilatory responses to O2 and CO2 during exercise. Ann. Rev. Physiol. 45:415-426, 1983.
12. Hey, E. N., B. B. Lloyd, D. J. C. Cunningham, M. G. M. Jukes, and D. P. G. Bolton. Effects of various respiratory stimuli on the depth and frequency of breathing in man. Respir. Physiol. 1:193-205, 1966.
13. Hickam, J. B., W. W. Proyor, E. B. Pag, and R. J. Atwell. Respiratory regulation during exercise in unconditioned subjects.J. Clin. Invest. 30:503-516, 1951.
14. Hulsbosch, M. A. M., R. A. Binkhorst, and H. Th. Folgering. Interaction of CO2 and positive and negative exercise stimuli on ventilation in man. Pflugers Arch. 394:16-20, 1982.
15. Jacobi, S. M., V. I. Iyawe, C. P. Patil, A. R. C. Cummin, and K. B. Saunders. Ventilatory responses to inhaled carbon dioxide at rest and during exercise in man. Clin. Sci. 73:177-182, 1987.
16. Jacobi, M. S., C. P. Patil, and K. B. Saunders. Transient, steady state and rebreathing responses to carbon dioxide in man, at rest and during light exercise. J. Physiol. 41:85-96, 1989.
17. Kelly, M. A., G. R. Owens, and A. P. Fishman. Hypercapnic ventilation during exercise, effects of exercise methods and inhalation techniques. Respir. Physiol. 50:75-85, 1982.
18. Lally, D. A., F. W. Zechman, and R. A. Tracy. Ventilatory response to exercise in divers and non divers. Respir. Physiol. 20:117-119, 1974.
19. Mahler, D. A., E. D. Moritz, and J. Loeke. Ventilatory responses at rest and during exercise in marathon runners. J. Appl. Physiol. 52:388-392, 1982.
20. Martin, B. J., J. V. Weil, K. E. Sparks, R. E. McCullough, and R. F. Govner. Exercise ventilation correlates positively with ventilatory chemoresponsiveness. J. Appl. Physiol. 45:557-564, 1978.
21. Mason, R. G. and S. Lahiri. Chemical control of ventilation during hypoxic exercise. Respir. Physiol. 22:241-262, 1974.
22. McConnell, A. K., E. S. G. Semple, and C. T. M. Davies. Ventilatory responses to exercise and carbon dioxide in elderly and younger humans. Eur. J. Appl. Physiol. 66:332-337, 1992.
23. Mercier, J., M. Ramonatxo, and C. Prefaut. Hyperpnea and CO2 sensitivity of the respiratory centres during exercise. Eur. J. Appl. Physiol. 59:411-415, 1990.
24. Miyamura, M. and K. Ishida. Adaptive changes in hypercapnic ventilatory response during training and detraining. Eur. J. Appl. Physiol. 60:353-359, 1990.
25. Miyamura, M., T. Yamashina, and Y. Honda. Ventilatory responses to CO2 rebreathing at rest and during exercise in untrained subjects and athletes. Jpn. J. Physiol. 26:245-254, 1976a.
26. Miyamura, M., H. Th. Folgering, R. A. Binkhorst, F. D. J. Smolders, and F. Kreuzer. Ventilatory response to CO2 at rest and during positive and negative exercise in normoxia and hyperoxia.Pflugers Arch. 364:7-15, 1976b.
27. Ohyabu, Y., A. Usami, I. Ohyabu, Y. Ishida, C. Miyagawa, T. Arai, and Y. Honda. Ventilatory and heart rate chemosensitivity in track-and-field athletes. Eur. J. Appl. Physiol. 59:460-464, 1990.
28. Pianosi, P., D. Grondin, K. Desmond, A. L. Coates, and J. V. Aranda. Effect of caffeine on the ventilatory response to inhaled carbon dioxide. Respir. Physiol. 95:311-320, 1994.
29. Rebuck, A. S. and J. Read. Patterns of ventilatory response to carbon dioxide during recovery from serve asthma. Clin. Sci. 52:388-392, 1971.
30. Robbins, P. A., Conway, J., Cunningham, D. A., Khamnei, S., and Paterson, D. J. A comparison of indirect methods for continuous estimation of arterial PCO2 in men. J. Appl. Physiol. 68:1727-1731, 1990.
31. Saunders, N. A., S. R. Leeder, and A. S. Rebuck. Ventilatory response to carbon dioxide in young athletes: a family study.Am. Rev. Respir. Dis. 113:497-502, 1976.
32. Scoggin, C. H., R. D. Doekel, M. H. Kryger, C. W. Zwillich, and J. V. Weil. Familial aspects of decreased hypoxic drive in endurance athletes. J. Appl. Physiol. 44:464-468, 1978.
33. Semple, E. S. G. and McConnell, A. K. Factors influencing day-to-day variability of human ventilatory sensitivity to carbon dioxide. Proceedings of the 5th Meeting on the Control of Breathing and Its Modelling Perspective, Y. Honda et al., (Eds.). New York: Plenum Press, 1992, pp. 151-154, 1992.
34. Taylor, R. and N. Jones. The reduction by training of CO2 output during exercise. Eur. J. Cardiol. 9:53-62, 1979.
35. Ward, S. A. and B. J. Whipp. Ventilatory control during exercise with increased external dead space. J. Appl. Physiol. 48:225-231, 1980.
36. Weil, J. V., E. Byrne-Quinn, I. E. Sodal, J. S. Kline, R. E. McCullough, and G. F. Filley. Augmentation of chemosensitivity during mild exercise in normal man. J. Appl. Physiol. 33:813-819, 1972.


©1996The American College of Sports Medicine