Share this article on:

Evidence for an inadequate hyperventilation inducing arterial hypoxemia at submaximal exercise in all highly trained endurance athletes


Medicine & Science in Sports & Exercise: May 2000 - Volume 32 - Issue 5 - pp 926-932
BASIC SCIENCES: Original Investigations

DURAND, F., P. MUCCI, and C. PRÉFAUT. Evidence for an inadequate hyperventilation inducing arterial hypoxemia at submaximal exercise in all highly trained endurance athletes. Med. Sci. Sports Exerc., Vol. 32, No. 5, pp. 926–932, 2000.

Purpose: The majority of highly trained endurance athletes with a maximal oxygen uptake greater than 60 mL·min−1·kg−1 develop exercise-induced hypoxemia (EIH). Yet some of them apparently do not. The pathophysiology of EIH seems to be multifactorial, and one explanatory hypothesis is a relative hypoventilation. Nevertheless, conflicting results have been reported concerning its contribution to EIH. The aim of this study was to compare the cardiorespiratory responses to maximal exercise of highly trained endurance athletes demonstrating the same aerobic capacity without EIH (N athletes) and with EIH (H athletes).

Methods: Ten N athletes and twelve H athletes performed an incremental exercise test. Measurements of arterial blood gases and cardiorespiratory parameters were performed at rest and during exercise.

Results: All athletes presented a significant decrease in PaO2 (P < 0.05) from rest up to 80% V̇O2max associated with an increase in PaCO2, both findings consistent with a relative hypoventilation. Then the H athletes, who had a greater training volume per week and a higher second ventilatory threshold than the N athletes (respectively, 17 ± 1.1 vs 13.1 ± 0.7 h·wk−1; 91.8 ± 1.7 vs 86.1 ± 1.8% V̇O2max), presented a continuous PaO2 decrease up to V̇O2max. This was associated with a widening (Ai-a)DO2.

Conclusion: This study showed that a relative hypoventilation, probably induced by a high level of endurance training, induced hypoxemia in all athletes. However, a nonventilatory mechanism, perhaps related to the volume of training, seemed to affect gas exchanges beyond the second ventilatory threshold in the H athletes, thereby enhancing EIH.

It has been generally accepted that performance of intense dynamic exercise in highly trained endurance athletes is associated with reduced arterial pressure in O2 (PaO2) and an increased ideal alveolar-arterial PO2 difference (Ai-a)DO2 (10,12,27,32). This exercise-induced hypoxemia (EIH) has been well documented in young athletes, but the pathophysiology is still unclear. Three potential mechanisms are thought to contribute to EIH: 1) a lack of compensatory hyperpnea, 2) a VA/Q mismatching, and/or 3) a diffusion limitation. Studies have suggested that the inadequate hyperventilation acts essentially for submaximal exercise (28) and that the VA/Q mismatching and the diffusion limitation appear for near-maximal levels of exercise (16,28,32).

One intriguing finding has not yet been studied, however. The majority of athletes who have a maximal oxygen uptake (V̇O2max) greater than 60 mL·min1·kg1 and who have been endurance training for several years develop this EIH. Yet some identically trained athletes do not exhibit it, and in some cases these athletes have shown a frank decrease in PaO2 at submaximal workload and then an increase at maximal exercise during incremental testing (10; C. Préfaut and F. Durand, personal observations). It is possible, therefore, that near maximal exercise some physiological process either further impairs or improves arterial hypoxemia, resulting in no exercise-induced hypoxemia in some athletes. Few reports have compared endurance-trained athletes with the same aerobic capacity with and without the development of EIH, and studies that have examined the ventilatory hypothesis showed conflicting results about its contribu-tion to EIH. Some studies indicated that an inadequate hyperventilation entirely explains EIH (10,22). Another showed that this hypothesis holds only at submaximal exercise and thus explains only a part of EIH (26).

One hypothesis that would explain the divergent data in the literature is that a lack of compensatory hyperpnea exists in all endurance athletes but only at submaximal exercise. To better understand the mechanisms underlying exercise-induced hypoxemia, this study compared the cardiorespiratory responses with an incremental exercise of two groups of highly trained endurance athletes with the same aerobic capacity: one demonstrating exercise-induced hypoxemia, the other one not.

Laboratoire de Physiologie des Interactions, Hôpital A. de Villeneuve, Montpellier, FRANCE

Submitted for publication September 1998.

Accepted for publication June 1999.

Address for correspondence: Fabienne Durand, Laboratoire de Physiologie des Interactions, Service EFR, Hôpital Arnaud de Villeneuve, 34295 Montpellier Cedex 5, France. E-mail:

©2000The American College of Sports Medicine