What follows is a general summary report concerning the scientific testing analyses and the group responses to the "live high - train low" altitude model.
Study Questions
Does the altitude training strategy of live high - train low
significantly enhance group racing performance at sea level?
Answer: Yes!
Personal Questions
How are you affected by acute (short term, initial) exposure to
high altitude and how does this affect your racing performance at
altitude?
How do you adapt over time to high altitude and how does this affect your racing performance at altitude?
While considering the
results, there are a number of important factors to keep in mind:
Hi-Lo altitude training is believed to work through two
mechanisms a) living at high altitude causes an increase in red
blood cells, which improves VO2max and race performance, and B)
training low allows you to maintain your interval paces near what
you would do at sea level, so that training is maintain.
While we know that the performance enhancing effects of HI-Lo (mentioned above) hold true in looking at group responses, Hi-Lo may not work for some individuals - meaning that some people will not make more red blood cells by living at 8,000ft for a month, and some people, despite driving down to 4,000 ft for interval sessions, will still run their intervals slower than sea level and may lose “fitness” as a result.
Just as important as (or even more important than) the scientific results is your own analysis of how you felt before, during, and after the Hi-Lo camp. You certainly know your own body better than anyone else.
Keep in mind the phenomenon known as the “training camp effect.” In short, when you get a group of athletes together to train, it doesn’t matter if you go to Park City, Bloomington, Eugene, or Boulder, the group will typically get faster. Another confounding variable is your state of fitness at the start of the camp. Some of you were coming off of a very long year of XC, indoors, and outdoors. Some of you were very fit coming off of nationals. Others of you were injured on and off throughout the spring, and some of you were not particularly fit at the start of camp. How much of your improvement (or decline) in VO2max or 3k time was due to the altitude training effects and how much was due to your initial, pre-camp state of fitness (or state of overstraining or injury status)?
All of these are factors you’ll have to keep in mind when you integrate the scientific information with your own perceptions concerning this summer’s altitude training camp.
Blood responses
Upon arrival altitude, your kidneys begin secreting a hormone
called erythropoietin (EPO). EPO is the hormonal signal that
tells your bones to produce red blood cells. EPO levels typically
peak sometime between 24 and 48 hours after arrival at altitude,
then begins to slowly decline over the next 4 weeks back to near
sea level values. Usually after 2-3 weeks at altitude, EPO is
still slightly elevated over pre-altitude levels. After 4 weeks
at altitude, EPO levels typically are slightly less than they
were pre-altitude (this is why 3-4 weeks is the optimal time for
an altitude camp).GRAPH
As a group, your EPO responses followed this typical pattern. The values below are the average for all 27 athletes +/- the standard deviation.
| Pre-altitude | 1 day at altitude | 18 days at altitude | Post-altitude/29 days |
| 8.6+/-2.5 mU/ml | 16.5+/-4.4 mU/ml | 9.8+/-2.2 mU/ml | 7.6+/-2.2 mU/ml |
This transient increase in EPO caused hemoglobin concentration and hematocrit to increase by just over 8% in the group from pre-altitude to post-altitude.
| Pre-Altitude | Post-Altitude | |
| Hemoglobin | 13.2+/-1.1 g/dl | 14.3+/-1.1 g/dl |
| Hematocrit | 41.0+/-3.3% | 42.7+/-3.1% |
Most individual EPO and hemoglobin responses showed similar patterns as the group responses, however there were some individuals who had “extreme” responses. For example, one athlete actually had their EPO level decline from pre-altitude to 1 day at altitude, and 2 athletes had very vigorous EPO responses (one subject’s EPO concentration went up over 300% after 1 day at altitude). Similar individual variation in the response to altitude was seen in the hemoglobin measures. Three athletes had their hemoglobin values go down after the altitude camp, and 3 athletes had a > 20% increase in their hemoglobin levels.
When looking at your individual EPO data, keep these things in mind:
Having a pre-altitude EPO level that is a lot bigger or smaller than the mean group value probably isn’t important......
For example, the average pre-altitude EPO level for the group was 8.6. If you happened to have an EPO concentration of 15 (or on the other hand 4), it probably doesn’t mean much. What is important is the CHANGE in EPO concentration after 2-days, and how fast it declines over the next 3-4 weeks. In general, it is believed that the bigger the short term “bump” in EPO and the slower decline in EPO levels from 2 days to 18 days, the bigger stimulus you provide to your bones to make more red blood cells.
......What is more important is the change in EPO or hemoglobin over time.
In evaluating your change in EPO over time, there are two ways to look at it. One is to look at the absolute change in EPO concentration (for example, going from an EPO level of 8.3 mU/ml pre-altitude to 17.2 mU/ml after 2 days at altitude). The other way is to look at the percent change from baseline. Using the example above, if your EPO went from 8.3 to 17.2, then you got a 107% increase in EPO concentration. Which is the more appropriate way to look at it? We believe that the body perceives hormonal stimuli as a percent change from a baseline level. It’s kind of like if you run 100 miles per week, adding 10 more miles per week doesn’t change the “stimulus” very much. However, if you run 30 miles per week, adding 10 miles per week is a big deal.
The same argument made above for EPO, also should apply for hemoglobin. Here’s an example:
Note that athlete #2 had a 3 mU/ml increase in EPO after 1 day at altitude, where athlete #1 had a slightly smaller (but basically similar) 2.5 mU/ml increase in EPO concentration. However, because the “bump” in EPO was only 27% for athlete #2, his increase in hemoglobin was only 0.4 g/dl (or 2.6%). The body of athlete #1 sensed a much bigger (51%) increase in EPO, and her hemoglobin went up 0.8 g/dl, or 6.4% - over 2 times that of athlete #1.
Therefore, look at the graph displaying your EPO “profile” carefully. If the slope of the line showing your initial EPO response after 1 day at altitude is similar to or steeper than the group, you probably got (and can expect to get with future altitude training) a nice blood adaptation that will help performance. If the slope of line of the initial EPO response is a lot smaller than the group, then either a) you might want to try living higher than 8,000 ft to see if you can get the EPO response to improve, or b) it might be possible that altitude training may not work for you.
When looking at your individual hemoglobin data, keep these things in mind:
Hemoglobin is not "red cell mass"
Hemoglobin is a concentration measure, meaning it is dependent on the actual amount of hemoglobin in relation to what it is dissolved in (i.e. plasma). Therefore, how hydrated you were (or how your total body plasma volume may have changed) will determine how well a change in hemoglobin concentration represents changes in the total volume or number of red blood cells. This change in the number of red cells is what is physiologically relevant to improving performance, and hemoglobin concentration is only a marker of that change (but probably a good one).GRAPH
Performance markers
The two primary performance based markers that we have are the treadmill test and the 3,000m races.
Pre-altitude to post-altitude changes As a group, from pre-altitude to post-altitude camp, improvements were seen in both VO2max (about a 3% improvement) and sea level 3,000m time (about a 1% improvement).GRAPH
Of the 27 athletes that went on the trip, we included 26 in the group VO2max data (above table). One athlete’s data was excluded due to what we feel was excessive fatigue/overtraining incurred prior to the camp, preventing even modest training. For the group performance data we included only those who ran 3,000m. The difference in 3,000m performance for the whole group of 25 athletes was significantly improved by 4.6 seconds. For the purposes of a secondary analysis, we excluded three other sets of individual performance data because of (what we feel were) significant disruptions in training due to illness or injury near the end of the camp. We do not feel their 3,000m performances fairly reflect what adaptation or lack of adaptation they might have had. This leaves a total of 22 athletes data sets for analysis of performance which are presented in the following table. The average improvement was 5.7 seconds and is statistically significant.
So overall we had approximately a 1% improvement in performance and a 3% improvement in VO2max. In this summer’s group it is a little hard to say this was due to altitude exclusively. However, in previous summers where college runners participated, we accounted for a “training camp effect” and normalization of iron stores by bringing them all in for a sea level camp 6 weeks before going to altitude. We also excluded a bias to want to get better at altitude by randomly assigning the athletes to either sea level or altitude groups. In those previous summers we obtained essentially the same results as this year, namely a 1.4% improvement in performance and a 4% improvement in VO2max. As you can see in the attached figures, the amount of improvement in maximal oxygen uptake and performance is independent of the initial starting point. Thus, we think that the predominant reason for the group’s improvement was the HiLo training paradigm.
As you also can see from the above tables, VO2max can be presented in two ways. The first is called relative VO2max, which has units of ml/kg/min. This is the way that most of you frequently hear VO2max being presented. Relative VO2max takes into account the person’s body weight, so that values can be compared between people. The other way VO2max is presented is called absolute VO2max, which has units of 1/min. This measure looks at the total volume of oxygen that you consume in one minute, independent of body weight. Since a larger person (say a male football lineman) has more muscle mass than a smaller person (lets say a female XC runner), the football player will have a much larger absolute VO2max (1/min) than the female XC runner. However, when we take into account body weight, the XC runner has a much higher relative VO2max (ml/kg/min) than the male football lineman.
Why is the important? Well sometimes only looking at relative VO2max can be deceiving. For example, since relative VO2max takes into account body weight, a person could get a higher number from one test to the next just by losing weight - even if their “aerobic fitness” didn’t improve. Similarly, gaining weight would have the opposite effect - meaning that an improvement in aerobic fitness might be obscured by only looking at relative VO2max if the person gained weight.
Notice that, as a group, weight went up from pre-camp to post-camp. This is probably due to a number of factors. Since blood has weight, some of the weight gain could be due to an increase in the volume of red blood cells after altitude exposure. The pre-altitude measure occurred right after nationals when many of you were “peaked” and probably were at your low weight for the year. There also was access to good food at the camp, which you may or may not have at home.
So when looking at the improvements in VO2max, it is probably more revealing to look at absolute VO2max from pre- to post-altitude, rather than what your initial level was and how it compares to the rest of the group. In short, everyone had a high VO2max, and VO2max within the elite population is a poor predictor of performance. What that means is, if athlete A has a VO2max of 70 and athlete B has a VO2max of 75, it’s anybody’s guess as to who will win on the track. On the other hand, if athlete A’s VO2max is 35 and athlete B’s VO2max is 80, the answer is a little clearer. Another example: in testing done previously with track team athletes at Indiana University, the highest VO2max measured was an 81, on a male who consistently ran between #7 and #10 on the IU XC team. One of the lowest VO2max’s of the men’s team was a 62, by someone who was consistently #3 or #4 on the team. The take home message being that VO2max is not the end-all be-all that determines race performance. Other factors such as running economy, lactate threshold, anaerobic capacity, etc. also have a strong influence on racing performance. Finally, if you have ever had your VO2max measured in other laboratories, it may or may not be comparable to the number that you got in Indian’s lab. There is approximately a 1% day-to-day variation within individuals and a 1% accuracy of measurement error to consider. The protocol that we used (with the race pace and hypoxic testing prior to the max test) may have influenced your VO2max to a small degree. Also, the method we used at Indiana (taking a full minute average) versus that used in some other labs (measuring shorter rime periods) sometimes gives slightly smaller values. However, our primary interest was how VO2max changed with altitude training, so the measurement method used is not important (as long as it was measured the same way both times - which it was).
The group improvements in VO2max with Hi-Lo training are very compelling. A time trial, on the other hand, isn’t as “robust” of a test - meaning it could be influenced by many factors - this is why laboratory testing is helpful and complimentary. As things went, the group improvements was about 4 seconds, which over 3,000 m at the elite level is no small amount. In fact, only 8 of the 27 people ran slower, and some individuals had huge improvements in performance (an average of 17.6 seconds in five athletes). The question is, how much of that improvement is due to the effects of Hi-Lo and how much of it is due to some other factors which influence race performance? Factors such as fitness levels at the start of the camp, errors in pacing, colds or injuries could have affected the pre- or post-altitude time trials.
So what’s the take home message? Does Hi-Lo training work? For a group, the answer is an unqualified yes. more importantly, does Hi-Lo training work for you? As a rule, you have to consider all the factors - not just your race performances. Just because your 3k time didn’t get better, doesn’t mean that Hi-Lo doesn’t work for you. Think about all the factors that influenced both races: mental state, the weather, injuries, pacing/tactics, etc. Also think about how long it took to get your “wheels” back after coming down off the mountain. Consider that just because you ran faster doesn’t mean that Hi-Lo works better than sea level training for you. It is possible that even if we would have gone to San Diego to the Olympic Training Center for a month, some of you would have run pre- to post-camp. The question is - how much faster? Would it have been more with Hi-Lo? In previous work by Dr. Levine and Dr. Stray-Gundersen that included controls, the response was almost identical to your response. The point is make sure that you consider all the factors, not just the change in your 3k time.
Change in 3k time at high altitude
An interesting question is can we predict, based on your sea-level treadmill test, who was going to have the biggest drop in 3k performance at altitude? Based on previous work, the answer was probably going to be yes - and it was. This is potentially an important result, if you might be going to a race at altitude (like the 1998 World XC Championships in Marakesh).GRAPH
Group average 3,0000m time slowed from 8:45.2 to 9:33.7, a difference of 48.5 seconds. The decline in performance was essentially equal between men (8:19.9 to 9:08.3, 48.4 seconds) and women (9:33.2 to 10:21.8, 48.6 seconds). It was much windier at the first Park City race than the first Bloomington race, but a majority of the decline in performance was (obviously) due to the effects of altitude.
Altitude affects performance by decreasing the content of oxygen in the arterial blood. During exercise, the “arterial saturation” or percent of hemoglobin in the arterial blood that has an oxygen bound to it, is dependent on my factors: ventilation, the speed at which the blood moves through the lung, lung diffusing capacity, and others. The end result is that, even at sea level, arterial saturation during heavy exercise can vary widely. For example, in the group of 27 athletes, arterial saturation during race pace at sea-level ranged from a low of 85.9% to a height of 95.4%. Theoretically, when hypoxia (altitude) is superimposed on exercise at race pace, athletes with a low arterial saturation to begin with have a much smaller “reserve” to defend oxygen delivery to the working muscles. We would predict that those of you with low arterial saturation values during sea-level exercise would have a larger decline in both VO2 and 3k time at altitude - and both of those relationships are true. Therefore, based on a simple sea-level treadmill test, we would advise some of you to avoid races at altitude, as you will likely be affected to a greater extent than the rest of the field. Similarly, there are some of you that are not going to be affected as much as everyone else, and you would likely have an advantage over the rest of the field racing at altitude. USATF also gives a 3% altitude adjustment on qualifying for nationals for events 1,500m or over, run at altitudes above 4,000ft. Two athletes had less than a 5% decrement in performance at 7,000ft, and many of you had a decrement of between 5% and 7%. At an altitude between 4,000 and 5,000ft, many of you should have less than a 3% decrement in performance - therefore in practical terms, you’re given a few seconds head start on the qualifying mark before you even start the race. In a year when you might be battling an early season injury or illness, some of you might have an easier time getting a qualifier at altitude (with the 3% adjustment) versus sea level.
What about acclimatizing to altitude? Will that help improve race performance at altitude? Most of you already can see by your own performances that the answer is yes. What are the factors involved in predicting who will acclimatize best? That analysis is a little more tricky, and will take a little bit of time.
HVR - the breathing test
The results here were very interesting. Previous research that elite distance runners demonstrate a “blunted” response to hypoxia compared to untrained individuals. What this means is that an untrained person will respond to hypoxia by vigorously increasing ventilation, trying to maintain the content of oxygen in the arterial blood. As a rule, most elite athletes do not increase ventilation markedly in response to hypoxia, allowing their arterial saturation to fall rather quickly.
As a group your response was typical of elite athletes with a very “blunted” or low ventilatory response to hypoxia (although there were 5 of you who had more “normal” responses). The $64,,000 question is why? Two theories: a) It’s genetic, meaning that people who have this low response to hypoxia will likely be better distance athletes. Therefore if you take a group of elite distance athletes, they’ll likely be low responders. b) It’s a result of long-term training. One way runners judge how hard they’re running is by how hard they’re breathing. So over years of running, they may “teach” themselves to depress ventilation and not respond to lower levels of oxygen in the blood.
Believe it or not, it looks like the answer is a) it’s genetic. One study measured a group of athletes and their parents and siblings, all of who were non-athletes. Then they also measured a group of normal non-athletes. What they found is that the athletes and their relatives (even though the relatives were not athletes) had low responses, while the control group of non-athletes had a high response. OK - what does this mean for you? Surprisingly, about 80% of you had a significant (and in some cases quite a large) increase in your HVR values after the altitude camp. How does this change potentially affect your race performance? Also how does your baseline HVR level potentially affect your individual degree of adaptation to altitude? For now, nothing obvious jumps out, but with more analysis time the answer may be clearer.
Summary
Hopefully you found this initial summary and analysis helpful. There is a mountain of data still left to go through and several relationships still left to examine. Several other key individual relationships between your data, altitude acclimatization, and race performance will likely be discovered. Also with time to look at everyone’s individual data, it may be possible to determine what specific factors may be limiting your performance at sea-level, and what changes you may want to consider in your training to offset those limitations. For now, hopefully you’ll carefully consider the analysis concerning the response to Hi-Lo, and if it works for you, give this strategy some consideration in your future training plans.
