EXECISE TRAINING AT HIGH ALTITUDE: Does It Improve Endurance Performance At Sea Level?

John R. Sutton, M.D., DSc.
Faculty Health Sciences University of Sydney Lidcombe,
NSW Australia Member,
Sports Medicine Review Board
Gatorade Sports Science Institute

Key Points
1. As altitude increases, the barometric pressure and the oxygen partial pressure decrease. This means that any athletic performance critically dependent on oxygen for metabolism, i.e., any event lasting longer than roughly two minutes, might be adversely affected. By contrast, because the air at high altitude is less dense than at sea level, jumping, sprinting, and other activities in which air resistance needs to be overcome might be enhanced when competing at high altitude.
2. The most important physiological adaptations to living at high altitude are increased ventilation of the lungs, increased blood hemoglobin, and enhanced extraction of oxygen by the tissues. Maximal cardiac output is not usually affected by altitude.
3. Although acclimatization to or residence at high altitude is required for optimal endurance performance at high altitude, the value of high altitude training for endurance performance at sea level has yet to be proven.
4. Nutrition for athletes at altitude can be a major problem, and weight loss will be significant at moderate to high altitudes. The amount of weight loss is generally dependent on the absolute altitude.

INTRODUCTION
High altitude can have dramatic effects on sport performance. At the 1968 Mexico City Olympics (altitude = 2,300 m) the decreased air resistance contributed to improved sprinting and jumping performances; Bob Beamon's Mexico City long jump record stood for nearly a quarter of a century, and is the classic example. The decreased air resistance at Mexico City undoubtedly helped Australian Ralph Dobell to set an Olympic record in the 800 m run. However, at longer race distances, performances were progressively worse in Mexico City than at lower altitudes. An additional observation in Mexico City was that in the endurance races most of the top performers had lived most of their lives at high altitudes or they had trained for long periods at high altitude. The defeat of Jim Ryun, the favorite and world record holder in the 1,500 m by Kip Keno of Kenya and the defeat of Ron Clark, the world record holder in the 5,000 m and 10,000 m, again by high altitude residents, were clear examples of the importance of living and training at high altitude. In fact, Ron Clark finished 6th in the 10,000 m in a state of collapse. The winning time was two minutes slower than his world record! The first five placegetters were athletes who were residents of high altitude or had lived there for some considerable time. Prior to the Mexico City Olympics, Griffith Pugh studied six international middle distance runners at Mexico City over a period of four weeks of acclimatization. He conducted time trials, as well as ergometer studies for VO2max. At this altitude a mean reduction in VO2max of 14.6% compared with sea level was demonstrated on day two, but by day 27 VO2max had improved and was now 9.5% less than sea level values. In the time trials for the one mile run, times were increased by 3.6% in the first week and 1.5% in the fourth week when compared with sea level performance. Thus, the improvement associated with acclimatization was 2.1% (or 4.9 seconds). By contrast, in a three mile time trial the increase compared with sea level was 8.5% on the fourth day and 5.7% on the 29th day. Thus, there was a gain of 2.8% (or 20 seconds) in performance as a result of acclimatization. Similar observations were made by research teams headed by Faulkner and Buskirk, clearly demonstrating the importance of acclimatization to altitude for sea level athletes wishing to compete at altitude. The performances of endurance athletes at Mexico City who normally lived at sea level but who had trained for at least one month at high altitude improved progressively, the longer the stay at altitude. Nevertheless, none of these sea-level dwellers performed in Mexico City as well as did the life-long residents at high altitude. It seemed abundantly clear on the basis of the Mexico City results that high altitude exposure was critical to success in aerobic endurance events at high altitude. But the benefit of high altitude training does not necessarily translate to improved performance at sea level. Some of the principles of high altitude training, the available research on the effects of altitude training on sea level performance, and important unresolved questions related to this issue will be reviewed in this issue of Sports Science Exchange.

PHYSIOLOGICAL COMPONENTS OF ENDURANCE PERFORMANCE
What is it about training at high altitude that might be expected to enhance performance? Because endurance performance depends largely on aerobic (oxygen dependent) metabolism and very little oxygen is stored in the body, the presumed benefit of high altitude training must include effects on the transport and utilization of oxygen. Oxygen transport and utilization requires a series of closely integrated links involving the lungs and breathing, the transfer of oxygen from the pulmonary alveoli to the blood, the carriage of oxygen by the blood in association with hemoglobin, the pumping of oxygenated blood around the body by the heart, the distribution of the oxygenated blood to the working muscle and, finally, the extraction of the oxygen and its use in metabolism by working muscles. The maximum ability to deliver, extract and utilize oxygen is referred to as one's maximal oxygen uptake or VO2max. Figure 1 demonstrates the key components of VO2max; it can be readily seen that to maximize oxygen transporting ability one needs to optimize all of these links. For a critical evaluation of the key components in the oxygen transport chain and how each of these may limit VO2max, the reader is referred to the reviews of Dempsey et al., (1988), Saltin (1988), and Sutton (1992).

Figure 1. Oxygen Transport

.                   Ventilation Hb O2 Affinity Cardiac Output SV HR BP Peripheral Circulation Muscle Blood Flow Capillary Density Diffusion O2 Extraction Metabolism Substrate Delivery Muscle Mass (Fiber size & number) Energy Stores Myoglobin Mitochondria

Figure 2 illustrates how maximal oxygen uptake decreases progressively with increasing altitude (Cymerman et al., 1989). Factors such as ventilation and the central ventilatory drive, which do not limit VO2max at sea level, may well limit performance at extreme altitude.

PHYSIOLOGICAL RESPONSES AND ADAPTATIONS TO HIGH ALTITUDE
Residence at high altitude and exercise at altitude will have significant but variable impacts on each component of the oxygen transport chain (Sutton et al., 1988; Terrados et al., 1988). Ventilation of the lungs increases and is one of the earliest responses to ascent to altitude. Heart rate and cardiac output during rest and submaximal exercise also increase rapidly, and oxygen extraction at the tissue level will reach its maximum rate during exercise at altitude. Later, the number of red blood cells, which contain hemoglobin and are responsible for carrying oxygen, also increases; this adaptive increase continues for a considerable time (Table 1). The earliest increase in hemoglobin concentration on exposure to altitude is due primarily to a decrease in plasma volume, rather than a true increase in the manufacture of new red cells (erythropoiesis). The activities of key enzymes involved in aerobic metabolism in muscle will also increase on ascent to altitude (Green et al., 1989) as will the capillary density and mitochondrial density in skeletal muscles (MacDougall et al., 1991). However, much of the increase in capillary density and mitochondrial density are not true increases in absolute capillaries or number of mitochondria but instead are due to a loss of muscle mass (MacDougall et al., 1991).

The key adaptations to extremely high altitude (> 6,000 m) seem to be increased ventilation, increased hemoglobin concentration, and enhanced maximal oxygen extraction (Sutton et al., 1988). By contrast, neither the transfer of oxygen from the lung alveoli to the capillary blood nor the cardiac output appear to be maximally extended by exposure to high altitude (Sutton et al., 1988).

If one studies athletes who stay for a prolonged time at less extreme altitude, e.g., 4,300 m, curious results are observed. For example, VO2max will not be increased, although endurance performance while remaining at that altitude will be enhanced (Majer et al., 1974; Wolfel et al., 1991). Wolfel et al., (1991) demonstrated that ventilation continues to increase with the prolonged stay at altitude, a physiological phenomenon called ventilatory acclimatization. This results in a lower alveolar PCO2and an increase in the alveolar oxygen concentration. Provided there are no problems with the lungs or with diffusion of oxygen into the blood, the increased alveolar oxygen content leads to an increased arterial PO2. When this increased arterial PO2is coupled with an increased hemoglobin concentration, the total amount of oxygen carried by the blood, i.e., the arterial blood oxygen content (CaO2), will increase significantly. Strangely, however, the body, as if to compensate for the increased blood oxygen content, reduces the cardiac output and the blood flow to the legs so that the total amount of oxygen delivered (CaO2x C.O.) to the whole body is actually unchanged following acclimatization (Wolfel et al., 1991).

CAN ALTITUDE TRAINING AFFECT SEA LEVEL PERFORMANCE?
Athletes who resided permanently at high altitude not only did well in Mexico City; many also excelled in endurance events held at sea level! These observations provoked one of the most asked questions that resulted from the Mexico City experience, i.e., Will athletes perform better at sea level if they are raised at high altitude or even trained at high altitude? As residence at high altitude causes adaptations in all aspects of the oxygen transport chain, it might be expected, if these adaptations last long enough after return to sea level, that they would be beneficial for performance at sea level.

The early studies on this question gave variable results. Balke et al., (1965) studied five male athletes who trained for 10 days at 2,300 m. On return to sea level VO2max had increased from 3.5 L/min to 3.75 L/min. Furthermore, a one mile time trial improved from 5.29 mins to 5.13 mins. However, as can be seen from their times, these subjects were not exceptional athletes. Faulkner et al., (1967) trained five athletes for 14 days at 2,300 m and showed an 8.5% improvement in VO2max, as well as improvements in one and two mile track times. Perhaps the most compelling study of all was that by Daniels and Oldridge (1970), who trained six worldclass middle distance athletes at 2,300 m and exposed them to altitudes approximating 3,300 m for several hours each day. Among the athletes were American world record holders, including the then 1,500 m and mile record holder, Jim Ryun. After spending 14 days training at altitude Ryun returned to sea level to set the world record of 3.51.3 min for the mile. Five of the other six athletes also achieved personal best performances upon return to sea level. After spending an additional 14 days at altitude Ryun returned to sea level and set a new 1,500 m world record. After a second sojourn at altitude five of six athletes again produced best performances. Thus the scene was set: not only was training at altitude vital for sea level athletes to compete at altitude, altitude training also seemed worthwhile for sea level athletes wishing to perform well at sea level.

However, there were several equivocal negative studies including that by Buskirk et al. (1967), who studied six Penn State athletes before and after up to 63 days at 4,000 m. Although VO2max was not significantly changed, running times for both one and two miles were slower (N= 2). Grover and Reeves (1967) studied five elite high school athletes who spent 20 days at 3,100 m and found a decrease in VO2max in four of the five athletes when they returned to sea level.

Shortcomings of Previous Research
Positive studies such as that of Daniels and Oldridge convinced the athletic community and certainly many coaches that altitude training was definitely an advantage for sea level performance. As a result, many Olympic trainers around the world have accepted this argument and have promoted altitude training. Unfortunately, most previous studies lacked appropriate control groups. Even the most recent studies by Terrados et al. (1988,1990), Rahkila and Rusko (1982), Levine et al. (1992), and Levine and Stray-Gundersen (1992) have significant control deficiencies. More recent studies have attempted to use control groups and to be more objective in their assessment of performance. Nevertheless, there remain serious difficulties with the experimental designs that preclude an unequivocal evaluation of training at altitude. For example, Karvonen et al. (1986) studied three elite sprinters for three weeks at 1,850 m; a control group of six similar athletes remained at sea level. With this small number of subjects it was difficult to show any significant difference between groups. Another example of a study limited by sample size was that of Terrados et al. (1988), who had eight road cyclists divided into two groups. One group was studied at sea level and the other group for three or four weeks at an altitude of 2,300 m (three weeks for two subjects and four weeks for the other two). Apparently there was significant improvement at the conclusion of the study in both groups i.e., a 22% improvement in the sea level group and a 33% improvement in the altitude group.

Inadequate selection of subjects and failure to randomly assign subjects to treatment groups are two common flaws in many of the altitude training studies:

1. Inadequate selection of subjects: For observations to be relevant to performances in top class athletes, the subjects should be top class athletes themselves who are at their peak of training at the time of the study. It is especially difficult to execute studies in which elite athletes are used as their own controls and therefore expected to be in peak competitive condition on two separate occasions within a relatively brief period.
2. Lack of randomized subject assignment: Subjects should be randomly assigned to altitude or sea level training groups. Some of the more recent and elaborately designed studies have included at least four different types of training that require random assignment of subjects to training regimens, i.e., (1) training at high altitude and sleeping at altitude, (2) training at high altitude and sleeping at low altitude, (3) training at sea level and sleeping at sea level, (4) sleeping at sea level but training at altitude. Following training, all of these regimens are then evaluated for their effect on sea level performance (B.D. Levine, personal communication).

There are other important considerations in study design that may influence the outcome of the study and therefore the validity of any conclusions. Some of these considerations are as follows:

1. The optimal duration at altitude. Should it be two, three, four, or more weeks?
2. The optimal altitude at which training should occur. Training at altitudes much below 1,800 m will not have significant physiological effects, but training above 3,000 m will decrease performance and training quality and quantity at that altitude. It is thus possible that the athletes could become effectively "detrained" while "training" at altitude!
3. Time between leaving altitude and competing at sea level. Should this interval be 24 hours or up to a week or more?
4. The optimal pattern for tapering. When, where, and for how long should training load be decreased prior to competition?
5. The need for acclimatization to heat or cold. This was particularly highlighted for endurance performance at the Barcelona Olympics when athletes needed to be heat acclimatized. It is clear that two desires, i.e., to train at altitude but yet to heat acclimatize, coupled with tapering, were competing demands for the athletes. Trying to optimize these competing demands is extremely difficult, and there is virtually no scientific evidence on which to base a decision.

NUTRITION AT ALTITUDE
Nutrition at altitude can be a major problem. This is more so for climbers at extreme altitude than athletes training at 2,000 - 3,000 m. Nevertheless, with increased altitude and increased ventilation, fluid is lost via the respiratory tract, often an unsuspected route for fluid loss. In addition, lean body mass and fat will be lost after prolonged periods at high altitude; the magnitude of the loss and the rate at which the loss occurs will depend on the altitude. For instance, in Operation Everest II there was a mean decrease in weight of 7.4 kg (Rose et al., 1988). In a more recent study, Butterfield and colleagues (1990) showed a marked increase in basal metabolic rate on first arrival at altitude and the virtual impossibility of preventing weight loss (although by meticulous attention to detail over the subsequent days and weeks, the weight loss was minimized).

SUMMARY
The Mexico City Olympics demonstrated unequivocally that to perform well in aerobic endurance events lasting longer than about two minutes at high altitude, one needs to be adequately acclimatized or, better still, be born and raised at altitude. The suggestion that training at altitude will benefit sea level performance, while theoretically sound, remains to be fully explored. No unequivocal scientific evidence exists to support or refute the claim. Endurance performance requires optimal integration of all components of the oxygen transport system that are stimulated at altitude. Theoretically, if such changes occur with training at altitude and are maintained upon return to sea level, the advantage may be real. Many additional questions remain. What is the optimal altitude at which one should train? What is the optimal duration of training at altitude and the best interval between leaving altitude and competing at sea level? These dilemmas are further compounded when the athlete may need to taper his/her exercise training regimen as well as acclimatize to heat and other environmental variables.

Q & A

I'm training for an upcoming race at high altitude. What should I eat in the days preceding the race?

Consider using the carbohydrate loading regimen. The higher your pre-exercise muscle glycogen level, the greater your endurance potential. When done properly, carbohydrate loading can increase your muscle glycogen stores by 50 to 100%. having greater glycogen stores at higher altitudes should be advantageous because energy production from carbohydrate is faster than from fat and requires less oxygen. Some other tips to remember include preventing dehydration by increasing fluid consumption, avoiding caffeine and alcohol, and knowing the warning signs of altitude sickness.

Ellen Coleman, MA, MPH, RD, Sports Nutritionist, The Sports Clinic

Do athletes training at high altitudes have an advantage?

While controversial, there is recent evidence that competitive runners who move to a moderate altitude (about 2,500 m) for 4 weeks, but who train at a lower altitude (about 1,250 m), improve performance. Upon return to sea level, maximal oxygen uptake increased (5%) and time to complete a 5,000 m run decreased (about 13 seconds). However, there were no improvements in 5,000 m run time for runners who lived and trained at moderate altitude or who lived and trained at sea level.

Lawrence Spriet, Ph.D, Professor, Human Biology & Nutritional Sciences, University of Guelph, Ontario, Canada