Journal of Applied Physiology

Point: Positive effects of intermittent hypoxia (live high:train low) on exercise performance are mediated primarily by augmented red cell volume

Benjamin D. Levine, James Stray-Gundersen

For nearly half a century, athletes have used “altitude training” to enhance sea level performance. Both altitude acclimatization and hypoxic exercise have been proposed as mediating this enhancement. However, hypoxic exercise impairs training quality (18, 19) and, in the absence of acclimatization, does not augment performance (30). The “living high-training low” model was therefore developed (19) and demonstrated to be effective for athletes of all abilities (4, 18, 26).

So which of the myriad aspects of altitude acclimatization (16) might be responsible for improving performance of athletes at sea level? Rigorous use of accepted scientific principles must be applied to determine cause and effect. We would like to propose “Levine and Stray-Gundersen’s Postulates” (a modification of Nobel laureate Robert Koch “Koch’s Postulates”) to determine the etiology of performance enhancement with altitude exposure, as follows.

First, the response (improvement in V̇o2 max and performance) must be present when the mechanism (increase in erythrocyte volume) is present. Corollary: when no increase in erythrocyte volume is present, there is no increase in V̇o2 max and no improvement in aerobic performance.

Second, the mechanism must be isolatable and demonstrated to have an unequivocal relationship to altitude exposure and improved performance.

Third, when the mechanism is manipulated independently (without altitude exposure), then the same improvement in physiological parameters and performance must occur. Corollary: in the presence of altitude exposure, when the specific mechanism is inhibited, then the outcome is prevented.

In the original publication of the “live high-train low” model, we demonstrated clearly that exposure for >20 h/day to 2,500 m altitude for 4 wk led to an increase in erythrocyte volume, an increase in V̇o2 max, and improved performance in an event (5,000 m time trial) that is dependent on high rates of oxygen transport (18). In contrast, a control group exposed to identical training, but living at sea level, improved neither erythrocyte volume, V̇o2 max, nor performance. Although we have consistently emphasized that the “low-altitude training” component of the high-low model is essential to allow high rates of oxygen flux, which maintain the muscle structure and function required for success of an endurance athlete (27), we will focus exclusively on the “altitude acclimatization” component for the purposes of this debate.

To further define the mechanisms underlying the improvement in performance with altitude training, all the altitude-living athletes from our previous studies (18, 28) were divided into two groups based on only one criterion: those who improved their race time by more than the group mean (“responders”) and those that got worse (“nonresponders”; Ref. 4). There were no differences between these groups with respect to numerous physiological variables that might influence acclimatization to altitude (4).

Rather, the key distinguishing feature was that the responders had a greater increase in erythropoietin concentration with acute altitude exposure, which remained elevated for a more prolonged period of time. Indeed, the erythropoietin increase in the responders after 2 wk at altitude was equivalent to the peak response in the nonresponders, in whom erythropoietin had returned to baseline. This difference in erythropoietin response patterns was clearly physiologically significant and not a chance occurrence; the responders had an increase in erythrocyte volume and increased V̇o2 max, whereas the nonresponders did not. Furthermore, the increase in V̇o2 max was exactly what would be predicted from change in blood volume and hemoglobin concentration (31): predicted increase 248 ml/min − actual increase 245 ml/min(4). This derivation model was confirmed prospectively in another population (4, 26).

Although our results have led us to focus on the erythropoietic pathway, this was not an exclusive hypothesis at the beginning of our experiments. For example, in large numbers of runners (n > 100), running economy never changed (18, 19); anaerobic capacity never changed (16, 18, 19); muscle biopsies did not increase in buffer capacity or oxidative enzymes (27). Thus the weight of evidence has led us inexorably toward the primary effect of altitude acclimatization, given an adequate exposure, on sea level performance in competitive athletes being on the erythropoietic pathways.

But is there other evidence that altitude exposure is erythropoietic? Indeed, this evidence is extensive and quite compelling, particularly when the exposure is high enough and sustained for a long enough period of time. For example, cross-sectional studies in North (33) and South America (12, 20, 23) have demonstrated that there is an elevated red cell mass in natives of high altitude that is proportional to the altitude of residence and oxyhemoglobin saturation (12, 33). When sea level natives ascend acutely to altitude, there is a large increase in iron turnover that begins immediately on exposure (6, 11, 20). Most convincingly, direct examination of the bone marrow during acute high-altitude exposure has documented a dramatic increase in erythroid cell lines, from 20.0% at sea level to 40.5% after 1 mo at 4,300 m (11, 20). Accelerated erythropoiesis has also been confirmed in elite athletes at more moderate altitudes (7, 10, 24, 32). Thus despite rare exceptions (8), the evidence from multiple research groups has confirmed that moderate altitude exposure for nearly 24 h/day increases the red cell mass even in elite athletes.

But do lesser durations of exposure also increase the red cell mass? Clearly very short duration exposures of even extreme altitude are not sufficient to accomplish this goal (14). However, 12-16 h/day of normobaric hypoxia for 3 (3) or 4 wk (15, 21, 22) closely replicates the results observed in the field studies with an increase in both hemoglobin mass and V̇o2 max. In contrast, our opponents, using only 8–10 h/day of normobaric hypoxia (2,500–3,000 m) for 10–21 days failed to demonstrate an increase in hemoblobin mass or V̇o2 max (1). We would submit that this exposure is insufficient to stimulate a sustained accelerated erythropoiesis—the “dose” of altitude is simply too low (17).

Recent advances in the basic science of hypoxia response pathways may explain this apparent “threshold” phenomenon (25). For example, the principal transcriptional activator of gene expression in hypoxic cells is hypoxia-inducible factor-1α (HIF-1α). Under well-oxygenated conditions, HIF-1α is hydroxylated and binds to the Von Hipple-Lindau factor, which targets the entire complex for ubiquitin degradation. This process is so rapid, that in the presence of oxygen, HIF-1α has one of the shortest half-lives of any known protein (13). Moreover, when altitude natives or sojourners return to sea level, there is a suppression of erythropoietin (4, 6, 18, 20), a reduction in iron turnover and erythroid cell lines (11, 20), and a marked decrease in red cell survival time (20), termed “neocytolysis.” Both the rapid destruction of HIF-1α and neocytolysis may compromise the ability of short duration (<12–16 h/day) hypoxia to increase the red cell mass.

So what about the last of “Koch’s postulates”? Can the red cell mass be manipulated independently from altitude exposure and obtain the same effect? Clearly this is so. For example, increasing the red cell mass directly by blood doping or indirectly by injecting erythropoietin improves V̇o2 max, laboratory-based performance (2, 5), and success in international competition (29). Furthermore, low-dose erythropoietin injection increases the erythrocyte volume to a degree that is virtually identical to that acquired by 4 wk of altitude exposure to 2,500 m, data that have been obtained in collaboration with our opponents (17). Finally, when the increase in erythrocyte volume at altitude is inhibited in athletes by iron deficiency (9, 19) or infection (27), V̇o2 max does not increase and performance is not augmented.

In summary, the evidence demonstrates that given adequate exposure (living high enough, long enough, for enough hours per day), altitude is clearly erythropoietic even in elite athletes and leads to an increase in erythrocyte volume/red cell mass, V̇o2 max, and performance in endurance sport. To our knowledge, there are no other effects of altitude acclimatization (including all the alternatives proposed by our opponents) that can be manipulated independently and demonstrated to improve athletic performance over a sustained period of time. The magnitude of the response at altitude is qualitatively and quantitatively similar to that induced by isolated manipulation of the red cell mass (low-dose Epo injection), and the outcome is prevented if the erythropoietic process is impaired by iron deficiency or infection. Thus we would contend that all of Koch’s Postulates have been fulfilled for determining a cause-and-effect relationship between erythropoiesis and success of altitude training.