Journal of Applied Physiology

Counterpoint: Hypobaric hypoxia does not induce different responses from normobaric hypoxia

Remi Mounier, Julien V. Brugniaux

Studies on hypoxia are performed by lowering ambient oxygen partial pressure (Po2) either by reducing the barometric pressure (hypobaric hypoxia) or by lowering the O2 fraction [normobaric hypoxia at the prevailing barometric pressure (PB)]. Upon reflection we can see that many landmark studies including the Silver Hut expedition or the American medical research expedition to Everest (AMREE) were conducted at terrestrial high altitude (HA). However, simulated altitude has progressively replaced field experiments to a point where nowadays the majority of research is conducted in the laboratory environment. For a variety of reasons, ease of use being arguably the most important, most of these studies are conducted in normobaric hypoxia rather than hypobaric hypoxia. The counterargument by Millet et al. (9) supports the idea that the physiological responses induced by hypobaric or normobaric hypoxia are different, whereas this Counterpoint will present evidence arguing that these physiological responses are indeed equivalent.

Semantic considerations.

The first remark we can make is semantic. Hypoxia is defined as a reduction in the amount of oxygen (O2) available to any cell, tissue, or organism (21) and in that respect is independent of changes in PB. Hypoxia can be either continuous or intermittent; continuous hypoxia being generally encountered during high altitude exposure, i.e., hypobaric hypoxia. On the other hand, intermittent or transient hypoxia as experienced under various clinical conditions, such as obstructive sleep apnea (OSA) or stroke, is always characterized by hypoxic/ischemic episode(s) irrespective of the ambient pressure. These two conditions also highlight the two extremes of the spectrum of hypoxic levels, OSA representing a systemic hypoxia whereas stroke is more local.

Interchangeability between normobaric and hypobaric hypoxia.

The carotid bodies, located at the bifurcation between the internal and external carotid arteries, are oxygen sensors. As such, they respond to a wide range of arterial partial pressure of O2 (PaO2; ≈100–30 mmHg) (16, 17). Another unique feature is that they respond almost instantaneously to a drop in PaO2. Because of this brisk response inducing an increase in ventilation (16), various tests have been designed to investigate their sensitivity. For instance, the now classical test originally proposed by Weil et al. (23) has inspired a variety of duplications that take advantage of normobaric hypoxia in laboratory set-ups for the specific determination of the hypoxic ventilatory response (HVR). HVR has been proposed to predict exercise ventilation in hypoxia (19) or acute mountain sickness (AMS) (18), which is a neurological disorder characterized by headache as a primary clinical outcome occurring after 6 or more hours of exposure to high altitude/hypoxia (6). With reference to the latter though, it is noteworthy that there is also ample evidence suggesting that the broad interindividual variability precludes reliable interpretation (1). On the basis of a meta-analysis from Burtscher et al. (3), it appears that arterial oxygen saturation (SaO2), determined upon exposure to acute simulated altitude between 2,300 and 4,200 m, is a more accurate predictor of AMS susceptibility. The link between HVR and high altitude pulmonary edema (HAPE), a condition occurring essentially during mountaineering expeditions, at terrestrial HA has also been considered. In a review by Bärtsch et al. (1), the authors highlighted that a low HVR is a predictor of susceptibility to HAPE. It has been estimated that approximately 100 million employees alone (without counting leisure activities) are working every year in hypoxic conditions (7). Because of the prevalence of the aforementioned conditions, AMS in particular, rapidly developing upon exposure to high altitude, counterprotection measures such as preacclimatization involving normobaric hypoxia have been developed (2, 8). Although the evidence is still equivocal (2), it has been proposed that few sessions at night at an altitude simulating the target “field” altitude can be sufficient (8). On the other hand, it has also been suggested that a more thorough protocol involving 1–4 h of daily exposures for 1–5 wk is required to stimulated adaptation (2).

Arguably one of the most studied adaptations to hypoxia relates to accelerated red blood cell production. This response is initiated by the secretion of erythropoietin (Epo) upon regulation by the transcription factor hypoxia-inducible factor-1 (HIF-1) (4). The magnitude of the Epo response has been demonstrated to be altitude dependent (5). Although this study from Ge et al. (5) used hypobaric hypoxia, the authors, as well as others (4), acknowledged that the increase in Epo is of similar magnitude in response to hypobaric or normobaric hypoxia (providing that the inspired Po2 is equivalent). Our group confirmed this observation over a 3-h normobaric hypoxic exposure (3,000 m) during which serum Epo concentration increased significantly (11). It is, however, noteworthy that the increase in Epo is also time dependent as highlighted by Pialoux et al. (15) who observed a progressive rise in plasma Epo from 2 to 12 h normobaric hypoxia exposure (end-tidal Po2 held constant at 60 mmHg for all subjects) (15).

It's all about oxygen sensing.

It appears the human body has O2 sensors located in different places not only restricted to the carotid bodies, leading to both acute and chronic adaptations. Indeed, all nucleated cells in the body can sense and potentially respond to different levels of Po2 and induce physiological responses at different time scales. For instance, the kidneys are sensitive to a drop in PaO2, but at much lower level of oxygen pressure than the carotid bodies because the Po2 in the kidney can naturally be as low as 10 mmHg in the renal medulla (14). As previously discussed, the timeframe of the response is also different, inasmuch as erythropoiesis is much slower than the ventilatory response (days vs. seconds). The beauty of the system is such that the human body actually possesses O2 sensors responding to a very wide range of changes in Po2 with a different timeframe, allowing the body to cope with emergency situations as well as developing long-term strategies permitting life-long exposure in O2-depleted environments. Indeed, under conditions of reduced oxygen pressure, HIF-1 regulates the expression of more than 70 genes mediating the adaptive responses beyond simply hematopoiesis (20). The organ-dependent (e.g., brain, kidney, liver, and heart) variation in HIF-1 expression at various levels of hypoxia has been elegantly reviewed by Stroka et al. (22). As our group recently demonstrated, this key adaptive protein is expressed in the leukocytes as well as in skeletal muscle during exposure to both acute (10, 11, 13, 15) and chronic normobaric hypoxia (11, 12).

To our knowledge, no studies in the literature have provided convincing arguments supporting the idea that the physiological or pathophysiological responses induced by chronic hypobaric or normobaric hypoxia are indeed different. As noted by Kupper et al. (7), the physiological differences between normobaric and hypobaric hypoxia are too small to be clinically relevant. Finally, no robust hypothesis could reasonably be proposed to explain the putative physiological differences between these two modalities of hypoxia.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.