Every breath you take: acclimatisation at altitude

The major challenge of going to altitude is related to the hypoxic environment; and of all the acute compensatory responses to hypobaric hypoxia at altitude, hyperventilation is usually considered the most critical for adequate acclimatisation. Hypocapnia minimises the inspired to alveolar difference in the partial pressure of oxygen, and through the concurrent alkalosis shifts the oxyhaemoglobin dissociation curve towards the left, both of which facilitate an increased oxygen uptake in the lungs at low partial pressures of oxygen. However, hypocapnia-induced cerebral arteriolar vasoconstriction as well as the left-shifted oxygen dissociation curve may reduce oxygen availability to the brain.

The first direct measurements of cerebral blood flow in humans during acute hypoxia were conducted more than 60 years ago. About 40 years ago, Severinghaus et al. (1966) reported the first data on how cerebral blood flow varies at altitude. In general, this as well as several consecutive studies indicate that cerebral oxygen metabolism is well protected even at substantial hypoxia; however, the finding that neurological deficits at altitude occur more frequently in those with the most pronounced hypoxic ventilatory response (Hornbein et al. 1989) indicates a conflict between ventilatory control and systemic haemodynamic regulation on one side, and cerebral haemodynamics on the other side (West, 1990). Thus, the respiratory and cerebral protective mechanisms, their interaction and their variation over time at altitude still warrant further exploration.

The study by Fan et al. (2010) published in a recent issue of The Journal of Physiology provides interesting new data to aid in the interpretation of how altitude affects the relationship between ventilation, respiration and cerebrovascular haemodynamics. In accordance with previous reports, the ventilatory recruitment threshold (the at which an increase in ventilation is observed from baseline) was reduced, and the ventilatory CO₂ sensitivity (the slope between and minute ventilation above the ventilatory threshold) was increased, consistent with an augmented CO₂ response. Additionally, the authors used transcranial Doppler ultrasound to measure cerebrovascular reactivity, defined as the percentage change in cerebral blood flow velocity per mmHg change in . Both the hypocapnic (measured during paced voluntary hyperventilation while breathing ambient air) and the hypercapnic (measured by rebreathing of 7% CO₂ in 93% O₂) cerebrovascular CO₂ reactivity increased at altitude; the latter correlated with the arterial pH. In contrast, no association was found between the ventilatory and the cerebrovascular response to CO₂.

The authors suggest that alterations in pH buffering may partly account for the observed changes in ventilatory and cerebrovascular CO₂ responsiveness, and that the concomitant increase in both indicates a common underlying factor. However, it may also be reasonable to infer that whereas hypoxia underlies the increased ventilatory response to CO₂ at altitude, it does not mediate the increased cerebrovascular response to CO₂, at least not by the same signalling pathways as those mediating the change in ventilation. Thus, there was no correlation between the absolute values of or change in cerebrovascular response and the ventilatory response to CO₂, the latter of which is well known to be a function of . To further address how cerebrovascular CO₂ reactivity varies at altitude, I have incorporated data from the present as well as a previous study by the authors (Ainslie et al. 2007) (Table 1). With considerable reservation for both the separation in time and the different methodologies employed, the collated data indicate that cerebrovascular CO₂ reactivity varies in a manner different from that of ventilatory CO₂ reactivity. First, although hypocapnic CO₂ reactivity appears to be substantially increased only at the highest altitude (5050 m), this increase may be a spurious finding related to additional vasoconstriction induced by a hyperventilation-induced increase in (with volunteers breathing ambient air), which would be detectable only at sufficiently severe pre-existing hypoxia. Secondly, in contrast to the measurement of hypocapnic CO₂ reactivity, hypercapnic CO₂ reactivity was measured at hyperoxia, thus eliminating any confounding effect of oxygenation-induced changes in cerebrovascular tone. Hypercapnic CO₂ reactivity appears to be equally increased at 1400, 3800, and 5050 m, indicating that the severity of hypoxia does not regulate the response, in contrast to that observed for the ventilatory CO₂ response. Thus, it would appear that whereas the ventilatory response to CO₂ is progressively augmented at increasing altitude, the cerebrovascular CO₂ response is characterized by a ceiling effect that becomes operative from a relatively low altitude. In other words, the cerebrovascular and the ventilatory responses to CO₂ may be regulated differentially at altitude, at least when the direct influence of hypoxia is eliminated.

Table 1.

Cerebrovascular carbon dioxide reactivity at different altitudes

Approximate altitude (m)	0	1400	3840	5050
(mmHg)	105 ± 11	74 ± 3	54 ± 3	44 ± 3
Cerebrovascular CO2 reactivity (% mmHg−1)
Hypocapnia	2.0 ± 0.6	3.1 ± 0.7	1.9 ± 0.6	4.2 ± 1.0
Hypercapnia*	2.9 ± 1.1	5.5 ± 0.7	5.3 ± 0.7	4.8 ± 1.4

Values at 0 and 5050 m are from Fan et al. (2010) (n= 17). Values at 1400 and 3840 m are from Ainslie et al. (2007) (n= 5). Cerebrovascular reactivity was measured as percentage change in cerebral blood flow velocity (middle cerebral artery) per mmHg change in . All values are given as means ±s.d. *Conducted at hyperoxia.

How can these findings be substantiated and elaborated upon? Incorporating data from several experiments is notoriously associated with severe problems related to different methods and study groups. This issue could be resolved by measuring ventilatory and cerebrovascular hypo- and hypercapnic CO₂ reactivity during acclimatisation in one single group of volunteers at several steps of increasing altitude. It might also be helpful to measure both hypocapnic and hypercapnic CO₂ reactivity at both levels of , i.e. breathing ambient air and during acute hyperoxia, to single out the direct effect of hypoxia. These and further mechanistic studies may help us understand the interaction between ventilatory control and cerebral haemodynamics, which may be of benefit in terms of preserving brain function not only in commuters, trekkers and climbers at altitude, but also in patients with chronic hypoxia, such as those with chronic lung disease, or with acute hypoxia, such as during critical illness.