Ventilatory responsiveness to hypoxia (HVR) has been reported to be different between highly trained endurance athletes and healthy sedentary controls. However, a linkage between aerobic capacity and HVR has not been a universal finding. The purpose of this study was to examine the relationship between HVR and maximal oxygen consumption (V̇o2 max) in healthy men with a wide range of aerobic capacities. Subjects performed a HVR test followed by an incremental cycle test to exhaustion. Participants were classified according to their maximal aerobic capacity. Those with a V̇o2 max of ≥60 ml·kg−1·min−1 were considered highly trained (n = 13); those with a V̇o2 max of 50–60 ml·kg−1·min−1 were considered moderately-trained (n = 18); and those with a V̇o2 max of <50 ml·kg−1·min−1 were considered untrained (n = 24). No statistical differences were detected between the three groups for HVR (P > 0.05), and the HVR values were variable within each group (range: untrained = 0.28–1.61, moderately trained = 0.23–2.39, and highly trained = 0.08–1.73 l·min·%arterial O2 saturation−1). The relationship between HVR and V̇o2 max was not statistically significant (r = −0.1723; P > 0.05). HVR was also unrelated to maximal minute ventilation and ventilatory equivalents for O2 and CO2. We found that a spectrum of hypoxic ventilatory control is present in well-trained endurance athletes and moderately and untrained men. We interpret these observations to mean that other factors are more important in determining hypoxic ventilatory control than physical conditioning per se.
- control of breathing
- hypoxic ventilatory response
maximal o2 uptake (V̇o2 max) is an indication of the capacity to both transport and utilize O2. It is well known that exercise training can increase V̇o2 max and improve the ability to perform strenuous endurance exercise. Most of the determinants of V̇o2 max can be positively influenced by exercise training (see Refs. 28, 29 for review), with the exception of the pulmonary system (4, 30). The raison d'être of exercise ventilation is to increase the level of alveolar ventilation to match O2 consumption and to maintain CO2 in proportion to muscular power output. To this end, ventilatory control must be tightly regulated. The regulation of breathing during muscular work has been the subject of much work and speculation, and the controlling mechanisms are complex and multifactorial (5). The concept that ventilatory responsiveness to chemical stimuli is different between highly trained (HT) endurance athletes and healthy sedentary controls was first suggested by Byrne-Quinn et al. (2). They observed a significant correlation between the hypoxic ventilatory response (HVR) and V̇o2 max (r = −0.768) in a group of athletes (n = 13; V̇o2 max = 60.1 ml·kg−1·min−1) and nonathletes (n = 10; V̇o2 max = 40.2 ml·kg−1·min−1). This idea was extended by Martin et al. (19), who found that “low responders” to hypoxia had a lower exercise ventilation, suggesting that a low exercise ventilation may be a link between chemoresponsiveness and athleticism.
In the present study, we questioned the purported association between HVR and V̇o2 max for two reasons. First, there have been no studies that have clearly identified a physiological mechanism that would explain why physical training would result in a lowering of chemosensitivity. Second, we questioned the magnitude of the previously reported relationship between HVR and V̇o2 max. Many studies investigating HVR and its potential relationship to V̇o2 max have utilized relatively small sample sizes. Furthermore, subjects have been grouped into categories that have typically had a narrow range of aerobic capacities. For example, in the original work by Byrne-Quinn et al. (2), the mean difference between athletes and nonathletes was ∼ 20 ml·kg−1·min−1. This presents a potential statistical problem whereby the sample used to calculate the correlation coefficient could be influenced by divergent groups, which can inflate the magnitude of the correlation. Based on the above summary, the purpose of this study was to examine the relationship between ventilatory sensitivity to hypoxia at rest and O2 consumption and ventilation at maximal exercise in a large sample of healthy men with a wide range of maximal aerobic capacities.
All experimental procedures and protocols were approved by the Clinical Screening Committee for Research of the University of British Columbia and conformed to the Declaration of Helsinki. All subjects provided written, informed consent before participating in the investigation.
Healthy male subjects (n = 55) volunteered to participate in the present study. Of these subjects, 37 had participated in previous studies conducted in our laboratory (8, 9). Subjects had normal cardiopulmonary function and were excluded from participation if they had been diagnosed with asthma, sleep apnea, had a history of smoking, or were hypertensive (systolic blood pressure >140 mmHg; diastolic blood pressure >90 mmHg). Subjects were life-long residents of sea level and had not sojourned to high altitude (>3,000 m) in the year before testing. None of the subjects participated in breath-hold diving, since this has been suggested to affect ventilatory responses to hypoxia and hypercapnia (7). Participants were classified according to their maximal aerobic capacity, as our laboratory previously described (25). Those with a V̇o2 max of ≥60 ml·kg−1·min−1 were classified as HT (n = 13), those with a V̇o2 max of 50–60 ml·kg−1·min−1 were classified as moderately trained (MT; n = 18); and those with a V̇o2 max of <50 ml·kg−1·min−1 were considered untrained (UT; n = 24). HT subjects were provincial- or national-caliber road and mountain cyclists who participated in cycle training (5–7 days/wk) and competed at the provincial, national, or international level. MT subjects were recreationally active but did not participate in structured aerobic training. The UT group was comprised of healthy nonathletic men.
HVR was assessed using previously described methods (13). Each HVR test began with a minimum of 10 min of rest to ensure stable baseline measures. Subjects wore a sealed face mask, and inspiratory flow signals (Hans-Rudolph HR800, Kansas City, MO) were integrated to provide instantaneous volume values. Subjects breathed room air from a mixing chamber, and 100% N2 was gradually added to the inspiratory circuit to evoke a gradual drop in arterial oxyhemoglobin saturation (SaO2) to 75% over an ∼5-min period using a finger-pulse oximeter (model 3740, Ohmeda, Louisville, CO). Isocapnia was maintained during the test by the manual addition of CO2 to the inspiratory circuit. Resting end-tidal Pco2 (model CD-3A CO2 analyzer, AEI Technologies, Pittsburgh, PA) was determined at the beginning of each day and maintained throughout experimentation. Inspired O2 fraction was determined by analyzing gas sampled from the proximal side of the inspiratory valve (S-3A, AEI Technologies). Inspired minute ventilation was plotted as a function of SaO2, and the slope of the linear regression was taken to represent the HVR. All data were acquired using an analog-to-digital converter (Powerlab/16SP ML 795, ADInstruments, Colorado Springs, CO) interfaced with a computer. Data were sampled at 200 Hz and stored for subsequent analysis using commercially available software (Chart version 5.02, ADInstruments).
Determination of V̇o2 max.
On completion of HVR testing, subjects performed an incremental cycle test to exhaustion. Subjects wore a nose clip and breathed through a mouthpiece connected to a nonrebreathing valve (model 2700B, Hans-Rudolph, Kansas City, MO). Subjects performed a self-selected cycling warm up (minimum of 5 min) on an electronically braked cycle ergometer (Excalibur Sport, Lode, Gronigen, Netherlands). Subjects performed a progressive exercise test starting at 0 W, which increased in a ramp fashion of 30 W/min until volitional exhaustion. Expired gases and ventilatory measures were measured continuously via open-circuit spirometry, and values were averaged over 15-s epochs. Heart rate was continuously assessed by telemetry (Polar Vantage XL, Kempele, Finland) and recorded every 15 s. All subjects cycled until volitional exhaustion and fulfilled at least two of the following criteria for V̇o2 max: 1) heart rate ≥220 − age, 2) respiratory exchange ratio ≥1.10, and 3) no further increase in O2 consumption with increasing workload.
Data were compared using analysis of variance procedures (Statistica 6.1, Stat Soft, Tulsa, OK). When significant F ratios were detected, Tukey's test was applied post hoc to ascertain where the differences resided. Pearson product-moment correlations were implemented to determine linear relationships between selected dependent variables. The level of significance was set at P < 0.05 for all statistical comparisons. All data are presented as means ± SD.
Physical characteristics are shown in Table 1. Subjects were similar for height, mass, and body surface area. The MT group was slightly but significantly older than the UT group. As expected, the HT group had a lower body mass index relative to the UT group.
Mean values for maximal exercise data are shown in Table 2. By design, V̇o2 max and peak power were higher in the HT group compared with UT and MT subjects. Also by design, our subject selection produced a large range of V̇o2 max values (27.6–72.7 ml·kg−1·min−1). Minute ventilation was higher in the MT and HT groups relative to the UT group, and this was unchanged when corrected for body surface area. At maximal exercise, ventilatory equivalents for CO2 production and O2 consumption were highest in the UT group relative to MT and HT, with no differences between MT and HT (see Fig. 1). During submaximal exercise, the ventilatory equivalent for O2 consumption was higher in the UT relative to MT and HT at 150 and 200 W.
A sample trace from one HVR test is shown in Fig. 2. As progressive hypoxia was administered and SaO2 fell, we were able to maintain end-tidal Pco2 within 1–2 Torr of resting values for all subjects. Shown in Fig. 3 are the mean HVR values for each group. No statistical differences were detected between the three groups (P = 0.333). The HVR values were variable within each group (range: UT = 0.28–1.61; MT = 0.23–2.39; HT = 0.08–1.73 l·min·%SaO2−1). When HVR was corrected for body surface area or body mass index, the similarity between the three groups remained (P > 0.05; data not shown). Although HVR can be influenced by body size, any differences in body size between the three groups did not appear to influence our results. The relationship between HVR and relative V̇o2 max was not statistically significant (r = −0.1723; P = 0.208; see Fig. 4). When the relationship between HVR and relative V̇o2 max was examined for each group, no significant correlations were detected for UT (r = −0.0528; P = 0.806), MT (r = 0.1989; P = 0.429), or HT (r = 0.1947; P = 0.524). HVR was also unrelated to maximal minute ventilation and ventilatory equivalents for O2 and CO2 (Table 3).
The purpose of this investigation was to examine the relationship between ventilatory sensitivity to hypoxia and maximal ventilation and O2 consumption. The existing literature is conflicting in this regard. Several studies have concluded that HVR and V̇o2 max are linked, whereas others have concluded otherwise. The principal findings of this study are that 1) ventilatory responsiveness to hypoxia is unrelated to V̇o2 max and 2) maximal exercise ventilation does not correlate with HVR. Our findings are consistent with the concept that HVR is unrelated to sea-level V̇o2 max or maximal exercise ventilation.
V̇o2 max and HVR.
Cross-sectional studies have suggested that endurance athletes have a blunted HVR compared with mountaineers or sedentary controls (2, 20, 23). Although these studies demonstrated a statistically significant relationship, the functional importance in terms of V̇o2 max and exercise ventilation remains obscure. It has been postulated that a lower HVR may mean less ventilation during exercise, and the reduced ventilatory demand may produce less dyspnea (24). Although this is an attractive possibility, it has not been directly tested. The results of the present study show no clear association between HVR and V̇o2 max. The interplay between environment (i.e., conditioning) and potential genetic influences on HVR and V̇o2 max is complex and not fully understood (1, 12, 14–16, 24). Although we did not make any measures that would permit us to comment directly on any potential genetic influence on HVR, we did find that a spectrum of hypoxic ventilatory control is present in HT endurance athletes and MT and UT men. We interpret these observations to mean that other factors are more important in determining hypoxic ventilatory control than physical conditioning per se. In addition to our own findings, three lines of evidence support this conclusion. First, Scoggin et al. (24) measured HVR in nonathletic healthy parents and siblings of long-distance runners and compared them to nonathletic controls. The HVR was decreased to a similar extent in runners and their relatives compared with the nonathletic controls. Second, identical twin studies suggest that when the effect of physical activity is removed, genetic factors predominate in determining HVR (14, 15). This is further supported by recent work that has implicated ancestry as a main explanation for the blunted response to sustained hypoxia and lower exercise ventilation of Quechua altitude natives (1). Third, Levine et al. (16) found that intense sea-level exercise training and the resultant increase in V̇o2 max (∼15%) had no effect on HVR.
Our experimental design, using a relatively large sample size and wide range of V̇o2 max values, failed to show a relationship between HVR and V̇o2 max. Our findings, coupled with those of others (14–16, 24), suggest that a high or low HVR has no functional relationship to maximal aerobic capacity. Although we believe that our interpretation of our own findings and other existing data is appropriate, it is important to highlight the inconsistencies in the literature. Katayama et al. (12) noted a significant decrease in mean HVR (pre = 0.43 l·min−1·%SaO2−1; post = 0.25 l·min−1·%SaO2−1) after modest sea-level endurance training (70% V̇o2 max for 30 min/day, 5 days/wk for 2 wk) that elicited a small but statistically significant increase in V̇o2 max (∼4 ml·kg−1·min−1). To our knowledge, this is the only study that has demonstrated that a training-induced increase in V̇o2 max is associated with a subsequent lowering of HVR. However, physiological mechanism(s) to explain why training is associated with a decrease in HVR are lacking. It is difficult to reconcile the findings of Katayama et al. (12) and those of Levine et al. (16). What are the potential mechanisms that could underlie the relationship between a decrease in HVR and training-induced improvements in V̇o2 max as reported by Katayama et al. (12) and proposed by others (2)? To our knowledge, no specific link between HVR and V̇o2 max has been identified (31, 32), and a full discussion of the potential mechanisms is beyond the scope of our study, but this question does warrant some discussion. It is possible that repeated exposure to humoral stimuli with exercise training may alter the gain of the carotid chemoreceptors or the processing of chemoafferent signals within the medulla. As an example, carotid chemoreflex sensitivity is enhanced in rabbits with chronic heart failure, which is related to the inhibition of neuronal nitric oxide synthase (17, 27). Exercise training attenuates, and thus normalizes, peripheral chemoreflex function (26). However, this response to exercise training does not appear to occur in animals without chronic heart failure. The mechanism by which exercise training alters carotid body nitric oxide synthesis is not clear, and this has certainly not been established in humans. Instead, based on the available data and the fact that a physiological link has not been identified, we favor the concept that resting HVR and V̇o2 max are unrelated.
Exercise ventilation and HVR.
The age-old physiological problem of what factors regulate exercise ventilation is still a topic of considerable uncertainty. In healthy humans exercising at sea level, the ventilatory control system operates to maintain arterial Po2 and Pco2 and pH at near resting levels over a wide range of work rates up to ∼60–70% V̇o2 max. Most humans hyperventilate beyond this intensity, and arterial Pco2 can be lowered by >10 Torr during near-maximal work. Conceptual paradigms to explain the regulation of exercise ventilation generally include three components: 1) a central medullary rhythm generator/integrator, 2) neural inputs into this integrator from higher locomotor areas of the central nervous system and from the periphery, and 3) regulation of the distribution of efferent motor output to the muscles of respiration (5). That the carotid chemoreceptors are an important input to the control of exercise ventilation has long been debated. Germane to the present study are the results of those investigations that have shown that the strength of the peripheral HVR is related to sea-level exercise ventilation (18, 19, 21). These studies imply that ventilation during maximal exercise is proportional to the peripheral chemoresponse to hypoxia at rest where endurance athletes with a low HVR tend to have a lower exercise ventilation. In the present study, we showed no association between measures of exercise ventilation and HVR (see Table 3). In contrast to the above-mentioned studies, our findings support the hypothesis that the net output from the carotid chemoreceptors by alterations to arterial levels of chemical stimuli is not a primary drive to exercise ventilation. Consistent with this idea is that, in the carotid body-denervated pony, the hyperventilatory response to exercise is actually greater (22). It should also be noted that, although claims of association between HVR and exercise ventilation have been made, the correlation coefficient values have typically been modest. For example, Miyachi and Tibata (21) examined UT subjects and trained long-distance runners with comparable V̇o2 max values as those obtained in the present study. The relationship between HVR and minute ventilation (r = 0.45) and minute ventilation/CO2 production (r = 0.62) were statistically significant but can be considered weak to modest. We found that HVR was unrelated to maximal minute ventilaton/O2 production or minute ventilation/CO2 production. We did, however, note that minute ventilaton/O2 production was highest in the UT group during submaximal exercise relative to MT and HT subjects (see Fig. 1). Other studies have reported that aerobic training reduces minute ventilaton/O2 production during submaximal exercise (3). In our study, HVR values across the three groups were comparable and do not appear to be related to maximal or submaximal ventilatory equivalents.
Given the above, it is difficult for us to conceive of a cause-and-effect relationship between HVR and sea-level ventilatory control during exercise. Rather, as summarized by Dempsey et al. (5), the role of the carotid chemoreceptors during exercise is to “fine tune” alveolar ventilation instead of providing the primary stimulus.
Critique of methods.
In consideration of our results, there are methodological concerns that warrant discussion. First, we determined the HVR using an isocapnic method whereby the linear regression relating ventilation to SaO2 represents the HVR. We evoked a gradual drop in SaO2 by adding 100% N2 to the inspiratory circuit over an ∼5-min period, whereas others used longer time periods and constant saturation values. Our laboratory has previously described the reproducibility of our ability to measure HVR (13), and it is in excellent agreement with values from other studies (33). However, there are other methods of determining the HVR where a hyperbolic plot of ventilation in relation to alveolar Po2 is generated. Here, the HVR is defined as value A, which is a mathematical expression of the hyperbolic function. Given that both isocapnic techniques quantify peripheral chemoreceptor drive, our findings should be similar regardless of the method employed. Furthermore, we have recently demonstrated significant increases in HVR after 12 days of intermittent hypoxic exposure (8). This suggests that our HVR measurement technique is sensitive enough to detect changes and should be appropriate to detect physiological differences between groups. However, methodological differences could have influenced our findings, and we cannot completely rule out this possibility. Second, subjects in the present study performed cycle exercise. Several of the studies with which we have compared our results used treadmill running. It is possible that feedforward and feedback ventilatory control is different between exercise modalities, and this could influence the integrated ventilatory response. However, given that our findings are in good agreement with those of others (14–16, 24), it would seem that the mode of exercise would have, at most, a minor effect. Last, it is possible that some of our HT subjects developed exercise-induced arterial hypoxemia and may have been exposed to occasional hypoxia during exercise. We do not believe that this influenced our results for two reasons. Our laboratory has previously shown that the relationship between exercise-induced arterial hypoxemia and HVR is nonsignificant (9, 11), whereas others have reported only modest correlations between end-exercise SaO2 and HVR (10). Derchak et al. (6) found the correlation between exercise-induced arterial hypoxemia and HVR to be significantly higher in male runners who did not develop expiratory flow limitation (EFL) (r = 0.92) compared with those who did develop EFL (r = 0.49). They interpreted this to mean that EFL attenuated the expression of ventilatory responsiveness and, in those with EFL, the “drive” to ventilate (as indicated by HVR) can be rendered irrelevant to some extent. The presence of EFL, or lack thereof, in our subjects may have influenced our results, and we cannot exclude this possibility.
In conclusion, we determined the HVR and V̇o2 max in healthy men with a wide range of maximal aerobic capacities. We found that HVR was variable between subjects and was unrelated to V̇o2 max or exercise ventilation. Our findings suggest that the physiological importance of a high or low HVR measured at rest is minimal in terms of sea-level exercise ventilatory control or V̇o2 max.
This study was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canadian Foundation for Innovation. A. W. Sheel was supported by a scholar award from the Michael Smith Foundation for Health Research (MSFHR) and a New Investigator award from the Canadian Institutes of Health Research. J. A. Guenette was supported by a NSERC Undergraduate Student Research Award. M. S. Koehle was supported by a NSERC Canada Graduate Scholarship and a Senior Graduate Studentship from MSFHR.
We thank our subjects for enthusiastic participation.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society