Limits of human lung function at high altitude

The lungs are the first interface between oxygen in the environment and the metabolic machinery of the body. Both at rest and at high levels of work, successful gas exchange at the alveolar–capillary membrane is necessary to achieve an adequate supply of oxygen to the tissues and for the elimination of carbon dioxide. To achieve this goal, the mechanical function of the lungs, chest wall and respiratory muscles must move a large enough volume of air to supply oxygen to the body. Furthermore, gas exchange at the alveolar–capillary membrane must allow transfer of oxygen to the blood, which includes diffusion of oxygen from the air to the hemoglobin in the red blood cells. In addition, the drive to breathe from the central nervous system must be finely tuned to increase ventilation instantaneously as metabolic demand increases. Finally, the sense of breathlessness, or dyspnea, which may inhibit attempts to increase work, must achieve a compromise with the metabolic demand so that work is not impaired.

At high altitude, respiration extracts utilizes a high proportion of the overall energy expenditure. In spite of a slight decrease in the work of breathing resulting from the lower density of the ambient air at high altitude, much greater volumes of air are necessary to supply enough oxygen to the body from atmospheric air, in which the level of oxygen is reduced. Delivery of oxygen is further impaired by a diffusion limitation of oxygen from the air to the blood, which increases with altitude. At extreme altitudes, a disabling sense of dyspnea is compounded by cerebral hypoxia, which may further limit exercise. Climbers to these heights have reported taking as many as 10 breaths per step as the rate of ascent progressively and tortuously slows. Reinhold Messner wrote of his first ascent of Mount Everest (8828m, barometric pressure 33.3kPa) without supplemental oxygen that ‘I am nothing more than a single narrow, gasping lung, floating over the mists and summits’. Thus, we come back to the lung as the primary and essential organ for human function and survival at high altitude without which the first step up could not be taken.

The purpose of this manuscript is to review work that has been carried out on the control of ventilation, the mechanics of respiration and gas exchange that pertain to the limits of exercise performance at high altitude. Both field and laboratory work will be drawn upon to gain insight into the understanding of lung function under the stress of high altitude.

For any given energy expenditure, the ventilation (V̇e_BTPS; lmin⁻¹) increases proportionately with altitude. Since barometric pressure decreases, there is less oxygen per volume of gas than at sea level. Thus, supplying the same amount of oxygen for any given rate of work requires that the lungs must take in a greater total volume of air. A number of field and low-altitude (Sutton et al., 1988; West et al., 1983) studies (in both hypobaric and normobaric hypoxia) have demonstrated this phenomenon on repeated occasions.

Data from the 1981 American Medical Research Expedition to Everest (AMREE) obtained on Mount Everest (Fig.1) and the 1985 Operation Everest II (OEII) hypobaric chamber experiments demonstrated a marked increase in ventilation up to altitudes of 6300m (barometric pressure approximately 46kPa) which, when expressed as ventilatory equivalent (V̇e_BTPS/V̇_O2,STPD), where V̇_O2 is the rate of oxygen uptake, show that for any given level of work V̇e is approximately four times greater at this altitude than at sea level. At 200W of work during AMREE (at 6300m), the mean ventilation for all subjects was 207lmin⁻¹ (btps) compared with sea-level values at that workload of approximately 65lmin⁻¹ (West et al., 1983). On OEII at comparable simulated altitudes, V̇e was approximately 160lmin⁻¹ and 89lmin⁻¹ respectively (Sutton et al., 1988). The lower V̇e may reflect less ventilatory acclimatization in OEII because of the shorter time at altitude. On AMREE, maximum V̇e on the simulated summit in two subjects was substantially lower (approximately 160lmin⁻¹), which may reflect a lower metabolic demand from a decreased workload and/or a rather blunted ventilatory response in one of the subjects. The ventilatory responses at intermediate altitudes and workloads are correspondingly lower, but much higher than at sea level. Thus, the demand on lung mechanics and respiratory muscles is greater at high altitude and may impose a limit to exercise.

The ventilatory response to exercise and hypoxia is complex and not completely understood. In spite of this limitation, hypoxemia, which stimulates the carotid body, effects an increase in ventilation. This response is proportional to the individual’s inherent hypoxia ventilation response (HVR) (Schoene et al., 1984; Schoene et al., 1990), to the altitude and to the state of acclimatization incurred by a sojourn at high altitude. Although the actual increase in the sensitivity of the carotid body to hypoxemia during the course of acclimation is not fully understood, it appears that each individual’s HVR, as measured at sea level, is roughly proportional to the ventilatory response to exercise at high altitude (Schoene et al., 1984; Schoene et al., 1990). In other words, subjects with a low HVR at rest measured at sea level have a relatively more blunted ventilatory response to exercise at high altitude than those with a high HVR, and vice versa. A large ventilatory response to exercise conveys a greater respiratory alkalosis and higher arterial oxygen saturation and may confer a better climbing performance at very high altitudes (Schoene et al., 1984) (Fig.2). However, individuals with very large HVRs may incur a greater work of breathing at extreme altitude. Thus, there may be a trade-off between increased arterial blood oxygenation and the increased work of breathing in these individuals.

As we consider the limitations to exercise imposed by high levels of ventilation, either at sea level or at high altitude, we must first consider the effect of air density on the work of breathing. Theoretically, at high altitude, the decreased gas density could substantially decrease the resistive load of the convection of gas during the movement of air at high levels of exercise. This alleviation of the work of breathing must be balanced against the increase in ventilatory demands from the necessary hypoxic stimulation of ventilation during exercise. A recent study by Cibella et al. (Cibella et al., 1999) in a small number of subjects showed that, in spite of the lower gas density, the net result was that there was still a greater energy expenditure upon breathing at high altitude for any given work load than at low altitude. Thus, it still costs more to breathe at high altitude.

At high levels of exertion, there is a finite amount of energy that can be expended. The delivery of oxygen by the circulation must be apportioned carefully to optimize the balance between the energy utilized to produce work and the work output itself. More specifically, exercising at high altitude, which requires a high level of ventilation, extracts a fair amount of energy to achieve that degree of ventilation. The respiratory muscles require a substantial proportion of the overall maximum rate of oxygen consumption (V̇_O2max), not only at high altitude but also at sea level. At low altitude, Hanson et al. (Hanson et al., 1982) exercised 15 highly trained runners to maximum and decided that the net effects of the hyperventilatory response to high levels of exercise were an advantage in that the improved gas exchange and respiratory alkalosis outweighed the expense of some inefficiency and ventilatory work during exhaustive exercise.

The next avenue of investigation looked at the limits of air flow during high levels of exertion at sea level in highly trained athletes and the work necessary to generate the pressure differential for this degree of air flow. Johnson et al. (Johnson et al., 1992) studied athletes at exhaustive levels of exercise and found that the flow limitation in the airways incurred a substantial metabolic cost. Subsequent investigations focused on actual respiratory muscle fatigue during high levels of work. This same investigative group (Johnson et al., 1993) found that in highly trained athletes the diaphragm functioned quite well up to approximately 85% of V̇_O2max, after which the transdiaphragmatic pressure, a reflection of the force generated from the diaphragm, decreased by approximately 25%. This finding suggests diaphragmatic fatigue at levels of work greater than 85% of V̇_O2max and impairment for prolonged high level exercise. In these studies and others (Marciniuk et al., 1994), respiratory muscle fatigue, particularly from the diaphragm, was not apparent at levels of exercise of 70% of V̇_O2max or less. Trying to exercise for prolonged periods above this level may extract too great a price in the overall balance of energy, a delicate situation at high altitude.

Investigators then began to consider the fraction of the overall cardiac output required to supply the muscles of respiration. Did this ‘cardiac steal’, dedicated to the muscles of respiration, limit blood flow to the exercising muscles in such a way that work output was impaired. Babcock et al. (Babcock et al., 1995) used supramaximal bilateral phrenic nerve stimulation at a number of frequencies in humans at 86–92% of V̇_O2max for approximately 13min. They found a consistent and significant fall of approximately 26% in the transdiaphragmatic pressures at all levels of work. They felt that there was competition for blood flow between the locomotor muscles and the diaphragm that eventually led to a steal of blood flow from the diaphragm and subsequent muscle fatigue. Cibella et al. (Cibella et al., 1999) tried to calculate the mechanical efficiency of breathing at low and high (5050m) altitude and estimated that the oxygen cost of breathing at high altitude and sea level amounted to 26 and 5.5% of V̇_O2max respectively (Fig.3, Fig.4). They felt that the work of breathing at high altitude may limit external work unless mechanical efficiency was extremely high.

A number of older and more recent studies have tried to measure the effect of an increased amount of respiratory muscle work on the blood flow to the working muscles of locomotion. Peripheral vascular resistance in the exercising muscles is increased as one approaches maximal exercise. This phenomenon may, in fact, decrease the blood flow to these muscles while an increase in respiratory muscle blood flow is observed (Fig.5) (Barclay, 1986; Harms et al., 1997; Wetter et al., 1999).

In summary, although because of the decreased density of the gas, the work of breathing at any given workload decreases progressively with increasing altitude, the net effect is an overall greater work of breathing, which may substantially impair energy supply to the locomotor muscles.

High altitude imposes a diffusion limitation of oxygen from the air to the blood (Fig.6). This diffusion limitation is secondary to several factors: (i) a lower driving pressure for oxygen from the air to the blood, (ii) a lower affinity of hemoglobin for oxygen on the steep portion of the oxygen/hemoglobin curve, and (iii) a decreased and inadequate time for equilibration of oxygen as the red blood cell traverses the pulmonary capillary. As mentioned above, the decreased alveolar partial pressure of oxygen at high altitude is in part minimized by ongoing ventilatory acclimation. This response to acclimation, however, is never adequate to overcome the ambient hypobaria. But, because of the large volume of gas that is required to supply an adequate quantity of oxygen, especially during the high metabolic demands of exercise, the lung is well designed to minimize the diffusion limitation in that the alveolar–capillary interface provides a large surface area for gas exchange that is instantly adaptable (West and Mathieu-Costello, 1992). Oxygen flux depends on the demand for oxygen, and the greater the ratio of diffusion capacity to perfusion the more successful is the loading of oxygen to the blood.

Arterial oxygen saturation thus decreases with exercise at high altitude, as has been documented in a number of human studies in both field and laboratory settings (Gale et al., 1985; Torre-Bueno et al., 1985; Wagner et al., 1986; Wagner et al., 1987; West et al., 1983) (Fig.7). Using sophisticated evaluations of gas exchange (the multiple inert gas elimination technique, MIGET), investigators have been able to apportion the decline in arterial oxygen saturation by looking at ventilation/perfusion relationships as well as diffusion characteristics (Fig.8). During acute hypobaric hypoxia to simulate an altitude of 4700m, Gale et al. (Gale et al., 1985), Torre-Bueno et al. (Torre-Bueno et al., 1985) and Wagner et al. (Wagner et al., 1986) showed both a ventilation/perfusion inequality and a diffusion limitation in the lung, both of which accounted for worsening hypoxemia with increasing work loads at higher altitudes. In Operation Everest II, the assimilated ascent to 8828m (P_O2, approximately 33.3kPa) demonstrated that the ventilation/perfusion heterogeneity persisted to a modest degree while the diffusion limitation of oxygen became greater with increasing altitude (Wagner et al., 1987) (Fig.8). The increase in diffusion limitation was thought to be secondary to the marked decrease in driving pressure for oxygen from the air to the blood at these extreme simulated altitudes. The ventilation/perfusion heterogeneity was attributed to interstitial edema from high intervascular pressures, which correlated with the increase in pulmonary artery pressures. Much controversy exists regarding the effect on gas exchange of increased intervascular pressures during exercise at high altitude. During acute hypoxia, extravasation of fluid from the intra- to the extravascular space has been documented both at high altitude and at sea level in human and equine experiments (Hopkins et al., 1997; Pascoe et al., 1981; Seaman et al., 1995; Whitwell and Greet, 1984).

In summary, lung function at high altitude is accompanied by both impairment of gas exchange and diffusion limitation of oxygen from the air to the blood. The net effect is worsening hypoxemia, which is proportional both to the increase in energy expenditure and to high altitude.

All accounts of climbing to extreme altitude convey that the climbers are overcome by extreme dyspnea. This sense of dyspnea, per se, inhibits further exercise or at least makes it laboriously painful and slow with increasing altitudes. At the same time, with attempts to exercise to a greater degree, climbers describe an almost euphoric haze of mental acuity, which further limits exercise and may be secondary to depression of central nervous system output resulting from brain hypoxia. Further work is necessary to measure the effects of high altitude upon exercise and cerebral blood flow, which may affect performance.

In conclusion, the lung is an elegant gatekeeper between environmental hypoxia and physical performance at high altitude. Because of the necessity of moving large quantities of air during exercise at altitude, the success of this task requires intact and functional lung mechanics, which are driven by central respiratory drive. An impairment of flow and a mechanical limitation may both be encountered, especially at extreme altitude. This process is facilitated by ongoing ventilatory acclimation, which is secondary to progressively increasing carotid body sensitivity to hypoxia. In spite of impressive lung mechanics and air flow, total body function is further impaired by arterial oxygen desaturation with increasing exercise and altitude, which is secondary both to the ventilation/perfusion heterogeneity and to the diffusion limitation of oxygen from the air to the blood. Further limitation is encountered from an extreme sense of dyspnea as well as depression of central nervous system output resulting from brain hypoxia.