Introduction
It is generally acknowledged that maximal oxygen consumption (V˙O2max) and exercise performance are reduced at altitude. In contrast to normoxia (N) altitude environment is characterized by a lower PO2leading to reduced alveolar O2partial pressure (PAO2). To some extent reduced PAO2can be compensated for, e.g. by hyperventilation. Endurance-trained athletes seem to be particularly sensitive to hypoxia (H). Some authors reported dec-rements in V˙O2maxeven at low altitude (∼600 m) [10, 22] where-by a great individual variability was observed [3,12,14,15, 22]. From this it seems to be advantageous for mountaineers and ath-letes training or competing at altitude to know their individual
response to altitude. Hypoxia susceptibility can be tested by per-formance tests under normoxic and hypoxic conditions, whereby the hypoxia-dependent decrement can be measured directly.
Several studies tried to identify the factors related to the decline of V˙O2max(∆V˙O2max) in acute H. It has been demonstrated that people with high V˙O2max in N [3, 7, 8,14,15,17, 22], low arterial oxygen saturation in N (SaO2) at exhaustion [5,13], low hypoxic SaO2at exhaustion [7,15], high decrement in SaO2at exhaustion from N to H [8,10], high anaerobic lactate threshold [14], high lean body mass [22], and low ventilatory equivalent for O2[8] are more susceptible to H. Although functional parameters (e.g. ventilation or V˙O2max) are held responsible for some of the varia-Abstract
We examined the effect of normobaric hypoxia (3200 m) on maximal oxygen uptake (V˙O2max) and maximal power output (Pmax) during leg and upper-body exercise to identify functional and structural correlates of the variability in the decrement of V˙O2max(∆V˙O2max) and of maximal power output (∆Pmax). Seven well trained male Nordic combined skiers performed incremen-tal exercise tests to exhaustion on a cycle ergometer (leg exer-cise) and on a custom built doublepoling ergometer for cross-country skiing (upper-body exercise). Tests were carried out in normoxia (560 m) and normobaric hypoxia (3200 m); biopsies were taken fromm. deltoideus.∆V˙O2maxwas not significantly dif-ferent between leg (– 9.1 ± 4.9 %) and upper-body exercise (– 7.9 ± 5.8 %). By contrast, Pmaxwas significantly more reduced during leg exercise (– 17.3 ± 3.3 %) than during upper-body exer-cise (– 9.6 ± 6.4 %, p < 0.05). Correlation analysis did not reveal
any significant relationship between leg and upper-body exer-cise neither for∆V˙O2maxnor for∆Pmax. Furthermore, no relation-ship was observed between individual∆V˙O2maxand∆Pmax. Anal-ysis of structural data ofm. deltoideusrevealed a significant cor-relation between capillary density and ∆Pmax (R = – 0.80, p = 0.03), as well as between volume density of mitochondria and ∆Pmax (R = – 0.75, p = 0.05). In conclusion, it seems that V˙O2maxand Pmaxare differently affected by hypoxia. The ability to tolerate hypoxia is a characteristic of the individual depending in part on the exercise mode. We present evidence that athletes with a high capillarity and a high muscular oxidative capacity are more sensitive to hypoxia.
Key words
Hypoxia · exercise testing · V˙O2max· maximal power · muscle mor-phology
T
rai
ning
&
T
est
ing
1
Affiliation Department of Anatomy, University of Bern, Bern, Switzerland
Correspondence Dr. Michael Vogt · Department of Anatomy · University of Bern · Baltzerstr. 2 · 3012 Bern · Switzerland ·
Phone: + 41316 3184 68 · E-mail: vogt@ana.unibe.ch
Accepted after revision:Accepted after revision: March 25, 2005
Bibliography Int J Sports Med © Georg Thieme Verlag KG · Stuttgart · New York · DOI 10.1055/s-2005-865652 · Published online 2005 · ISSN 0172-4622 M. Angermann
H. Hoppeler M. Wittwer C. Däpp H. Howald M. Vogt
bility in the decline of V˙O2max, several authors pointed out that muscular characteristics (e.g. mitochondrial density or capillar-ity) have also to be taken into account [3, 20 – 22]. This assump-tion is supported by mathematical models evaluating the inter-action of ventilation, cardiac output, O2-transport, and muscular characteristics in N and H [6, 27]. Moreover, it has been demon-strated extensively that exercise in H leads to muscle structural modifications [12]. However, there is currently no experimental data linking hypoxia sensitivity to muscle tissue characteristics.
Whereas impairment of V˙O2maxin H was extensively examined in the past, less attention was paid to the decrement of Pmaxin H (∆Pmax). Most studies examining the effect of H on V˙O2maxand Pmax used cycle-ergometry [8,10,14,15,17, 21, 22] or treadmill running [3, 5, 23], thus focusing on the lower limbs. Only few reports exist about arm [24] or upper-body exercise (e.g. rowing) [20].
The purpose of the present investigation was to evaluate the ef-fect of normobaric H (3200 m) on V˙O2maxand Pmaxduring leg and upper-body exercise in well trained athletes. Furthermore we tried to identify performance parameters (ranking based on cross-country skiing performance, V˙O2max, Pmax, anaerobic lactate threshold) and muscle tissue characteristics (fibre type distribu-tion, capillary per fibre ratio, capillaries per fibre area, fibre area, mitochondrial volume density, intramyocellular lipid content), which could be related to the hypoxia-dependent decrement of V˙O2maxand Pmax. It was hypothesized that H would reduce V˙O2max to a greater extent than Pmax[8,15, 20, 21, 23]. Furthermore, we assumed that∆V˙O2maxand∆Pmaxwould be the same for both ex-ercise modes (leg vs. upper-body exex-ercise). It was expected that athletes with good performance values would be more suscepti-ble to H.
Materials and Methods
Subjects
Seven well trained male Nordic combined skiers gave their in-formed consent to participate in this study. With the exception of one all were members of the A-team of the Swiss Ski Federa-tion. Mean body weight and lean body mass were 70.1 ± 6.5 and 66.3 ± 6.0 kg, respectively. Mean body height was 178 ± 6.0 cm and age 21.4 ± 2.5 years. Blood analysis revealed a hemoglobin concentration of 147.4 ± 6.8 g/l and a haematocrit of 45.0 ± 2.8 %. The study was approved by the Ethical Committee of the Canton of Bern (KEK-Bern), Switzerland.
Performance tests
Two different incremental step-tests to exhaustion were per-formed to determine maximal power output (Pmax) and maximal oxygen consumption (V˙O2max): one on an electromagnetically braked cycle-ergometer (Ergoline 800 S, Ergoline GmbH, Bitz, Germany) and the other on a custom-built double-poling ergom-eter for cross-country skiers (for details see Angermann et al. [1, 2]). In short, on the double-poling ergometer the athletes were driving a chain with their poles. The chain was connected to a normal cycle ergometer through which the load was set. Dou-ble-poling mainly involved the muscles of the arms, the shoul-ders (to move the poles back), and the trunk. Each athlete was tested under normoxic (560 m) and simulated hypoxic (3200 m,
FiO2= 14.6 %) conditions during leg and upper-body exercise. To simulate hypoxic conditions inspired air was diluted with nitro-gen (“Altitrainer 200”, Sport and Medical Technology, Geneva, Switzerland) [26]. All tests were carried out in a three-day period, beginning with tests in H, followed by one day rest after which tests in N were performed. Upper-body ergometry (morning) and cycle ergometry (afternoon) were separated at least by 5 hours.
Cycle ergometry (leg exercise) started after 2 min rest followed by 3 min warm-up with a load of 100 W (N) or 40 W (H). There-after the load was increased by 30 W every second minute until the subjects were unable to maintain a cadence of 60 RPM.
Double-poling ergometry (upper-body exercise) also started with 2 min rest followed by 3 min warm-up. Initial exercise loads of 60 W (N) or 40 W (H) were increased by 20 W every 2 minutes to loads just above anaerobic lactate threshold (determination of anaerobic lactate threshold see ref. [16]). Blood lactate (Lactate Pro, Arkray Factory Inc., Shiga, Japan) was measured after each step during a 30-sec break. After passing the anaerobic threshold the athletes were allowed a 2.5-min break before they started their final bout to exhaustion at 60 W below that particular load. They were encouraged to exercise continuously while the load was increased by 20 W every 30 sec. This particular protocol was chosen in order to keep the exercise duration reasonably short to allow for V˙O2maxto be reached. The total load the ath-letes had to overcome during double-poling ergometry consisted of the predetermined external load in addition to the resistance (friction) of the ergometer. Preliminary tests revealed that the re-sistance caused by friction was close to 70 W. Power output data presented in this paper represent external load only.
Measurement of heart rate (Accurex Plus, Polar Electro Finland Oy, Kempele, Finland), arterial oxygen saturation (Oxymeter: PULSOX-3 i, Minolta, Osaka, Japan), and breath-by-breath anal-ysis of expired air (Oxycon alpha, Jäger GmbH, Würzburg, Ger-many) was done continuously.
Analysis of muscle structure
Prior to the performance tests fine-needle biopsies were taken fromm. deltoideus.One part of the biopsy was immediately
fro-zen in isopentane cooled with liquid nitrogen. Cryosections from this part were used for analysis of fibre-type composition deter-mined by ATPase staining [4]. The other part of the biopsy was fixed in a 6.25 % solution of glutaraldehyde in 0.1 M sodium caco-dylate buffer adjusted to 430 mOsm with NaCl (total osmolarity of the fixative: 1200 mOsm, pH 7.4). The specimens were then postfixed in a 1 % solution of osmium tetroxide in 0.06 M veronal acetate buffer (total osmolarity: 386 mOsm, pH 7.4). They were contrasted with 0.5 % uranyl acetate in 0.05 M maleate buffer, pH 5.0. After dehydration in increasing concentrations of ethanol (70 – 100%), they were passed through propylene oxide and em-bedded in Epon. For stereological analysis, we cut 2 randomly chosen blocks from each biopsy. Ultrathin sections (50 – 70 nm) were cut and double-contrasted. The orientation of the sections was transverse or slightly oblique with regard to the fibre axis. For estimation of capillary number and fibre cross-sectional area we used a final magnification of × 1850. A final magnification of × 24 000 was used for estimation of the volume densities of mi-tochondria, intramyocellular lipid, myofibrils, and residual
coplasmic components per volume of muscle fibre. Morphome-try was done using standard procedures [29].
Statistical analysis
Data are presented as means ± SD. Differences in performance values between N and H and between upper-body and leg exer-cise were analyzed using a Wilcoxon test for paired samples. Dif-ferences between∆Pmax and ∆V˙O2max in both exercise modes were evaluated by 2-way ANOVA with Tukey HSD post-hoc test. Univariate linear regression analyses with∆Pmaxand∆V˙O2maxas
dependent variables were performed. The level of statistical sig-nificance was set at p < 0.05, results with p < 0.10 were interpret-ed as tendencies. The software package STATISTICA for Windows Version 6.1 (Statsoft Inc., Hamburg, Germany) was used for sta-tistical analysis.
Results
Maximal performance values of upper-body and leg exercise in N and H are shown in Table1. Pmaxwas significantly reduced by H
in both exercise modes. Absolute values recorded during upper-body and leg work cannot be compared due to technical differ-ences in the design of the respective ergometers. V˙O2maxreached
slightly higher values during leg ergometry and was significantly decreased by H compared to N. The V˙O2-measurement of one
cycle ergometer test N was dismissed due to malfunction of the metabolic cart. Maximum values for both blood lactate and heart rate indicated that the subjects had reached the point of exhaus-tion, but there was no systematic difference between either N and H or upper-body and leg performance. Ventilation at exhaus-tion was significantly higher with leg compared to upper-body exercise. Oxygen equivalent was significantly higher under leg vs. upper-body exercise and significantly lower in N compared to H in the upper-body performance test. SaO2significantly
de-creased under H in both exercise modes.
Effect of hypoxia on V˙O2maxand Pmaxduring upper-body
and leg exercise
No statistically significant difference between the hypoxia-in-duced reduction of Pmax(∆Pmax) and of V˙O2max(∆V˙O2max) was
ob-served during upper-body exercise (– 9.6 ± 6.4 % and – 7.9 ± 5.8 %,
respectively). In contrast, during leg exercise under hypoxic con-ditions Pmaxwas reduced almost twice as much as V˙O2max(∆Pmax
– 17.3 ± 3.3 % vs.∆V˙O2max– 9.1 ± 4.9 %, p < 0.05).∆V˙O2maxwas sim-ilar for upper-body and leg exercise, whereas the decrement of maximal power output in H was significantly greater for leg compared to upper-body exercise. There was no significant rela-tionship between∆V˙O2maxand∆Pmaxwhen measured during leg (R = 0.53, p = 0.28) or during upper-body exercise (R = 0.06, p = 0.90). Correlations between leg and upper-body exercise for both∆V˙O2max(R = – 0.09, p = 0.86) and∆Pmax(R = 0.34, p = 0.46)
were low.
Muscular factors related to∆V˙O2maxand∆Pmaxduring upper-body exercise
Fibre type composition and morphometric data of m. deltoideus are shown in Table2. Correlation analysis (Table3) revealed that a high capillary density is significantly related to a high decre-ment of Pmaxin H. Furthermore, a high∆Pmaxtended to correlate with high mitochondrial volume density, high type 2 a fibre con-tent, and cross-country skiing performance (rank). Significantly greater decrements of V˙O2maxunder hypoxic conditions were
ob-served in athletes with high intramyocellular lipid contents.
Discussion
In order to study athletes with well trained arm and leg muscles we recruited a group of Nordic combined skiers competing at in-ternational level. This limited the number of available athletes and resulted in a low variability of performance parameters (e.g. V˙O2max, Pmax). The small sample size and the low variation in
var-iables studied reduced the power of correlation analysis so that we were unable to perform multiple regression analysis [22]. A further limitation of the present investigation is that for techni-cal reasons the upper-body exercise protocol was discontinuous, whereas the leg exercise protocol was uninterrupted.
A major result of the study is that V˙O2maxand Pmaxwere affected
differently by acute normobaric hypoxia corresponding to 3200 m. In contrast to the literature we found larger relative dec-rements of Pmaxcompared to V˙O2maxduring leg exercise under
hypoxic conditions. Another important finding is that during Table1Maximal performance values of upper-body and leg exercise tests in normoxia (560 m) and normobaric hypoxia (3200 m)
Upper-body performance test Leg performance test p2
560 m 3200 m p1 560 m 3200 m p1 560 m 3200 m
Pmax[W/kg] 3.4 ± 0.2 3.1 ± 0.3 * 5.4 ± 0.2 4.4 ± 0.2 * n.a. n.a.
V˙O2max[ml/min/kg] 53.6 ± 4.2 49.3 ± 3.4 * 57.3 ± 3.7 52.5 ± 3.0 * * (*)
La [mmol/L] 10.2 ± 1.3 14.1 ± 1.4 * 14.0 ± 1.6 14.2 ± 2.1 ns * ns
Hf [bpm] 192 ± 10 190 ± 10 ns 193 ± 7 188 ± 8 * ns (*)
V˙E [L/min] 144.4 ± 15.6 148.0 ± 14.3 ns 183.2 ± 21.7 174.0 ± 18.3 * * *
EQ O2[unitless] 38.5 ± 3.0 42.9 ± 2.7 * 44.8 ± 4.1 47.3 ± 2.7 (*) * *
SaO2[%] 93.2 ± 2.5 75.0 ± 5.9 * 92.8 ± 1.4 70.1 ± 5.2 * ns *
Mean values ± SD; La, blood lactate; Hf, heart rate; V˙E, ventilation; EQ O2, oxygen equivalent; SaO2, minimal oxygen saturation; p1, significance of differences between 560 m and 3200 m; p2, significance of differences between upper-body and leg exercise; * p < 0.05; (*) p < 0.10; ns, not significant; n.a., not applicable; n = 7 except for V˙O2maxin leg performance test in normoxia (n = 6).
T
rai
ning
&
T
est
ing
upper-body exercise athletes with high muscle capillarity and high muscle mitochondrial density show larger losses of maxi-mal power output when O2supply is reduced. They are therefore more susceptible to hypoxia.
Comparison of∆V˙O2maxand∆Pmax
In general the measured∆V˙O2maxwas less pronounced compared to other studies with similarly trained athletes exposed to sim-ilar altitudes [14,15, 21, 25]. Previous work showed that∆V˙O2max caused by H is proportional to V˙O2max determined in N [3, 8, 14,15,17, 22] and therefore it is larger in endurance-trained ath-letes than in sedentary people [7]. We measured lower absolute V˙O2maxvalues in N compared to others [14,15, 21, 25], and this could explain the smaller∆V˙O2maxobserved in our study. Previ-ously it was found that successful high altitude climbers do not have extraordinarily high V˙O2maxin N [17, 30]. Because individ-uals with high V˙O2maxin N work on the steeper part of the oxygen equilibrium curve, any fall in inspired O2partial pressure would lead to a more pronounced reduction of oxygen saturation [7] and V˙O2max. Therefore, having a very high V˙O2maxmust not be an advantage at least in the case of high altitude climbers. As a prac-tical consequence our group of Nordic combined skiers should theoretically perform quite well at altitude compared to their sea-level performance.
During upper-body exercise,∆V˙O2maxwas not significantly differ-ent from∆Pmax, but during leg exercise H affected Pmaxmore than V˙O2max. This finding is in contrast to the literature where most studies with trained subjects show that compared to∆V˙O2max maximal power output and/or performance is equally [9] or less influenced by H [8,15, 20, 21, 23]. It was demonstrated that the amount of activated muscle mass is an important factor influ-encing the relationship between V˙O2maxand Pmaxand their hyp-oxia-dependent decrease [24]. In untrained subjects the hypox-ia-dependent (12 % O2,∼4300 m) loss of maximal aerobic power (∆V˙O2max) is greater than the loss in maximal power output
(∆Pmax) during two-leg exercise (– 28.2 % vs. – 19.6 % for∆V˙O2max and∆Pmax, respectively), but these values are different for arm and shoulder exercise (– 5.9 % vs. – 5.2 %), one-leg exercise (– 7.8 % vs. – 9.0 %) or arm exercise (– 4.6 % vs. – 12.4 %). The au-thors propose that during maximal exercise smaller muscles (arms) are more difficult to perfuse and therefore maximal exer-cise is limited primarily by the intrinsic power of muscles rather than by O2supply [24]. In accordance with our study these re-sults show that V˙O2maxand Pmaxcan differently be affected by H depending on the muscle mass recruited during the exercise task.
Another possible explanation for the differences between
∆V˙O2maxand∆Pmaxin leg exercise could be provided by the con-cept of the “central governor” [13,18]. This concon-cept postulates that during exercise in H central drive is reduced as a conse-quence of the lower inspired O2fraction, which would lead to re-duced muscle recruitment. During an incremental exercise test V˙O2increases almost linearly with increasing load but O2 con-sumption usually levels off some time before the termination of the test, determining V˙O2max. On the other hand power output in-creases continuously until the end of the test where Pmaxis mea-sured. Thus, V˙O2maxis often reached before Pmax. According to the concept of central governor, reduced muscle recruitment during maximal exercise in H could affect Pmaxto a larger extent than V˙O2max, but it is questionable whether quadriceps muscles are fully activated during incremental exercise.
Table2Fibre type composition and morphometric data of
m.del-toideus
Parameter Value
type 1 fibres 69 ± 11%
type 2 a fibres 23 ± 9%
type 2 x fibres 8 ± 12%
NN(c, f) 1.8 ± 0.4
NA(c, f) 424 ± 100 mm–2
a(f) 4551 ± 1798 µm2
Vv(mc, f) 5.1 ± 1.1%
Vv(ms, f) 1.1 ± 0.5%
Vv(mt, f) 6.2 ± 1.5%
Vv(li, f) 0.34 ± 0.20%
Vv(fi, f) 80.1 ± 2.9%
Mean values ± SD; NN(c, f): capillaries per muscle fibre; NA(c, f): capillaries per fibre area; a(f): fibre cross sectional area; Vv(mc, f): central mitochondria volume density; Vv(ms, f): subsarcolemmal mitochondria volume density; Vv(mt, f): total mitochon-dria volume density; Vv(li, f): intramyocellular lipid content; Vv(fi, f): myofibrils per fibre volume
Table3Correlations between∆V˙O2maxor∆Pmax and performance
values of upper-body exercise, fibre type composition, and
morphometric data ofm.deltoideus, respectively
∆V˙O2max ∆Pmax
R p R p
Performance values in normoxia
– V˙O2max – 0.53 0.22 – 0.13 0.78
– Pmax – 0.19 0.68 0.02 0.96
– LT – 0.11 0.81 0.57 0.18
– rank 0.35 0.45 – 0.74 0.05
Fibre type composition
– type 1 fibres – 0.21 0.66 – 0.12 0.80 – type 2 a fibres 0.46 0.31 – 0.69 0.08
– type 2 x fibres – 0.13 0.77 0.61 0.14 Morphometric data
– NN(c, f) 0.18 0.70 – 0.28 0.54 – NA(c, f) – 0.41 0.36 – 0.80 0.03
– a(f) 0.45 0.31 0.58 0.17
– Vv(mc, f) – 0.58 0.17 – 0.75 0.05 – Vv(ms, f) – 0.51 0.25 – 0.37 0.40 – Vv(mt, f) – 0.60 0.15 – 0.68 0.09
– Vv(li, f) – 0.78 0.04 – 0.27 0.56 R, correlation coefficient; for abbreviations see Tables1and2.Bold figures indicate a significant (p < 0.05) relationship or a tendency (p < 0.10)
T
rai
ning
&
T
est
ing
It has been shown previously that V˙O2maxcorrelates well with
rowing time in N (R = – 0.90) but not in H (R = – 0.48) [20]. The authors of that study summarized that H had a greater influence on V˙O2maxthan on exercise performance. This finding suggests
that the relationship between V˙O2maxand Pmaxis lost in H. Our
results support this result because we did not find a correlation between∆Pmaxand∆V˙O2maxneither for leg nor for upper-body
exercise. Thus, maximal aerobic power and maximal power out-put seem to be affected differently by H.
Finally, influences by the different test protocols for leg and upper body exercise in our study must also be considered. Faster increments in the workload during upper-body exercise may have led to an extension in the anaerobic part of the exercise test, thus influencing the relationship between∆V˙O2maxand∆Pmax.
Comparison between leg and upper-body exercise
Our results indicate that average∆V˙O2maxis similar for leg and
upper-body exercise in Nordic combined skiers. V˙O2maxduring
upper-body exercise was between 92 % and 95 % of the V˙O2max
measured during leg exercise. This suggests that the upper body exercise mode (double-poling) used in the present study is parable to whole body exercise at least for trained Nordic com-bined skiers. As similar V˙O2maxvalues were reached in both
exer-cise modes, it is not surprising that there is no significant differ-ence in∆V˙O2maxcaused by H. We conclude that the tested Nordic
combined skiers were well trained in the upper body and that upper body exercise involves nearly a similar amount of muscle mass compared to leg exercise on a bicycle ergometer.
Although average∆V˙O2maxwas quite similar for leg and
upper-body exercise, correlation analysis revealed that significant indi-vidual differences exist for the two exercise modes. Measure-ments identified athletes with a high decrement of V˙O2max
in-duced by H on the cycle ergometer, but with a low decrement of V˙O2maxon the double-poling ergometer, and vice versa. If the
hy-poxia-dependent decrease of V˙O2maxis a general characteristic of
a person we would expect a correlation between leg and upper-body exercise. Our results do not support this conclusion and therefore challenge the idea of the existence of a general ability of an individual to tolerate H at maximal exercise. In the other hand, our findings indicate that the ability to tolerate H depends in part on the exercise mode.
In contrast to∆V˙O2maxH-induced∆Pmaxwas significantly larger
during leg than upper body exercise. In a study of Harms et al. [11] it was shown that the work of breathing (Wb) can have sig-nificant effects on leg blood flow and V˙O2maxduring maximal
ex-ercise. Harms found that with increased Wb whole body V˙O2max
and cardiac output remains unchanged but leg blood flow and leg O2consumption are reduced. In contrast, both whole body
V˙O2maxand cardiac output are reduced when Wb is lower while
leg blood flow and leg O2consumption are increased [11]. In our
study, the maximal oxygen equivalent (EQ O2) was higher during
leg compared to upper-body exercise both under N and H, indi-cating indirectly that Wb per liter O2was higher during leg
exer-cise on the bicycle ergometer. This might have compromised maximal power output in H via reduced leg blood flow. In other words, due to increased Wb in H, leg blood flow and therefore
maximal power output is reduced while V˙O2max remains the
same.
Muscular factors related to∆V˙O2maxand∆Pmaxfor upper-body exercise
Classical physiological performance parameters (V˙O2max, Pmax,
anaerobic lactate threshold) measured in N did not explain the variability of∆V˙O2maxand∆Pmaxduring upper-body exercise
(Ta-ble3).This result was rather unexpected because most studies us-ing leg exercise found a good correlation between V˙O2maxin N and
∆V˙O2max[3, 8,14,15,17, 22] while only a few did not [5, 9]. Data
an-alysis of studies showing good correlations between V˙O2maxand
∆V˙O2max [15,17, 22] revealed a wide range of V˙O2max values as
trained and untrained subjects were involved in the same experi-ment. Obviously, a wide range in measured V˙O2maxor∆V˙O2max
al-lows for easier detection of a relationship between the two varia-bles. When the focus is on a sub-group of similarly (endurance) trained subjects, as was the case in our study, a relationship be-tween V˙O2maxand∆V˙O2maxcannot be demonstrated in these
stud-ies. This fact challenges the explanatory power of V˙O2max
measur-ed in N on∆V˙O2maxfor a homogeneous group of trained subjects.
To our knowledge the present study is the first investigating the relation of ∆V˙O2maxand ∆Pmax to muscle morphology. A novel
finding is that athletes with muscles adapted to endurance train-ing (high capacity to supply and to utilise O2) exhibit higher
dec-rements of Pmaxin H during upper-body exercise. This was shown
by the correlations between ∆Pmaxand muscle capillarity and
density of mitochondria ofm. deltoideus,respectively (Fig.1). It is generally believed that peripheral (muscular) factors such as perfusion, diffusion, and mitochondrial capacity modulate the hypoxia-dependent decrease of V˙O2max and Pmax [3, 6, 21, 22,
24, 27], but so far no direct experimental evidence existed. Re-gression analysis displayed in Fig.1 points to a significant or nearly significant correlation between capillarity and∆Pmax, and
between mitochondrial volume density and∆Pmax, respectively.
Although this result is strongly influenced by the one single ath-lete displaying rather poor oxidative capacities ofm. deltoideus,
Fig.1 ∆Pmaxand∆V˙O2maxof upper-body exercise in acute hypoxia in dependence of volume density of total mitochondria (Vv [mt, f]) and of capillaries per fibre area (NA [c, f]).
T
rai
ning
&
T
est
ing
his example supports the theory that well trained athletes be-come more sensitive to H compared to untrained subjects.
Some authors calculated the different contributions of the lung, the circulatory system, and the muscles to the limitation of V˙O2max[6, 27, 28]. They concluded that during exercise in H mus-cular factors become more important in limiting V˙O2max, when small muscle groups are exercised or for well trained athletes having a high V˙O2max.
In contrast to∆Pmaxno correlations were found between∆V˙O2max and capillarity or mitochondrial density. This can partly be ex-plained by the data structure of∆V˙O2maxand the small sample size. We have to keep in mind that athletic performance is a com-plex output of interactions between different body functions. From this systemic point of view, our study examined only a small part of the system and when we interpret the results we have to realize that there exists substantial individual variation in com-pensatory mechanisms, especially during exercise in hypoxia.
Conclusion
Maximal O2uptake and maximal power output are affected dif-ferently by hypoxia. However, it remains an open issue to which extent they are reduced at altitude. Our results question the paradigm that the ability to tolerate hypoxia (altitude) depends only on the athlete’s individuality and training status. The mode of exercise has also to be taken into account. Furthermore our findings expand the knowledge about the relation between pe-ripheral factors and the hypoxia-dependent decrease of Pmax and V˙O2maxin trained athletes. The results provide evidence that the impairment of exercise performance in acute hypoxia, corre-sponding to an altitude of 3200 m, is more pronounced in ath-letes with muscles well adapted to endurance training.
Acknowledgements
This study was funded by the Swiss Sports Commission (ESK). We thank Christoph Lehmann for constructing the upper-body ergometer, Franziska Graber for processing and analyzing the bi-opsies, and Ruth Vock and Silvia Schmutzfor their assistance in the preparation of the manuscript.
References
1Angermann M, Hoppeler H, Däpp C, Wittwer M, Vogt M. Effekte eines intermittierenden Höhentrainings auf die langlaufspezifische Ober-körperleistungsfähigkeit bei Nordisch Kombinierten. Leistungssport 2004; 3: 35 – 41
2Angermann M, Lehmann C, Hoppeler H, Däpp C, Vogt M.
Oberkörper-ergometrie: Spezifische Leistungsdiagnostik für Langläufer und Nor-disch Kombinierer. SchweizZ Sportmed Sporttraumatol 2003; 51: 168 – 173
3Billat VL, Lepretre PM, Heubert RP, Koralsztein JP, Gazeau FP. Influence of acute moderate hypoxia on time to exhaustion at vVO2maxin unac-climatized runner. Int J Sports Med 2003; 24: 9 – 14
4Brooke MH, Kaiser K. The “myosin adenosine triphosphatase” sys-tems: the nature of their pH liability and sulfhydryl dependence. J Histochem Cytochem 1970; 18: 670 – 672
5Chapman RF, Emery M, Stager JM. Degree of arterial desaturation in normoxia influences V˙O2maxdecline in mild hypoxia. Med Sci Sports Exerc 1999; 31: 658 – 663
6Di Prampero PE. Factors limiting maximal performance in humans. Eur J Appl Physiol 2003; 90: 420 – 429
7Ferretti G, Moia C, Thomet JM, Kayser B. The decrease of maximal
gen consumption during hypoxia in man: a mirror image of the oxy-gen equilibrium curve. J Physiol 1997; 498: 231 – 237
8Gavin TP, Derchak PA, Stager JM. Ventilation’s role in the decline in
V˙O2maxand SaO2 in acute hypoxic exercise. Med Sci Sports Exerc 1998; 30: 195 – 199
9Gore CJ, Hahn AG, Scroop GC, Watson DB, Norton KI, Wood RJ, Campell
DP, Emonson DL. Increased arterial desaturation in trained cyclists during maximal exercise at 580 m altitude. J Appl Physiol 1996; 80: 2204 – 2210
10Gore CJ, Little SC, Hahn AG, Scroop GC, Norton KI, Bourdon PC, Wool-ford SM, Buckley JD, Stanef T, Cambell DP, Watson DB, Emonson DI. Reduced performance of male and female athletes at 580 m altitude. Eur J Appl Physiol 1997; 75: 136 – 143
11Harms CA, Wetter TJ, McClaran SR, Pegelow DF, Nickele GA, Nelson WB, Hanson P, Dempsey JA. Effects of respiratory muscle work on car-diac output and its distribution during maximal exercise. J Appl Phys-iol 1998; 85: 609 – 618
12Hoppeler H, Vogt M. Muscle tissue adaptions to hypoxia. J Exp Biol 2001; 204: 3133 – 3139
13Kayser B. Exercise starts and ends in the brain. Eur J Appl Physiol
2003; 90: 411 – 419
14Koistinen P, Takal T, Martikkala V, Leppäluoto J. Aerobic fitness influ-ences the response of maximal oxygen uptake and lactate threshold in acute hypobaric hypoxia. Int J Sports Med 1995; 26: 78 – 81 15Lawler J, Powers SK, Thompson D. Linear relationship between V˙O
2max and V˙O2maxdecrement during exposure to acute hypoxia. J Appl Phys-iol 1988; 64: 1486 – 1492
16Levine BD, Stray-Gundersen J. “Living high-training low”: effect of moderate altitude acclimatization with low-altitude training on per-formance. J Appl Physiol 1997; 83: 102 – 112
17Martin D, O’Kroy J. Effects of acute hypoxia on the V˙O
2maxof trained and untrained subjects. J Sports Sci 1993; 11: 37 – 42
18Noakes TD, Peltonen JE, Rusko HK. Evidence that a central governor regulates exercise performance during acute hypoxia and hyperoxia. J Exp Biol 2001; 204: 3225 – 3234
19OelzO, Howald H, Di Prampero PE, Hoppeler H, Claassen H, Jenni R, Buehlmann A, Ferretti G, Brueckner JC, Veicsteinas A, Gussoni M, Cer-retelli P. Physiological profile of world-class high-altitude climbers. J Appl Physiol 1986; 60 (5): 1734 – 1742
20Peltonen JE, Rantamäki J, Niittymäki PT, Sweins K, Viitasalo JT, Rusko HK. Effects of oxygen fraction in inspired air on rowing performance. Med Sci Sports Exerc 1995; 27: 573 – 579
21Peltonen JE, Tikannen HO, Rusko HK. Cardiorespiratory responses to exercise in acute hypoxia, hyperoxia, and normoxia. Eur J Appl Physiol 2001; 85: 82 – 88
22Robergs RA, Quintana R, Parker DL, Frankel CC. Multiple variables ex-plain the variability in the decrement in V˙O2maxduring acute hypo-baric hypoxia. Med Sci Sports Exerc 1998; 30: 869 – 879
23Roberts AD, Daley PJ, Martin DT, Hahn A, Gore CJ, Spence R. Sea level
V˙O2maxfails to predict V˙O2maxand performance at 1800 m altitude. Med Sci Sports Exerc 1998; 30: S111
24Shepard RJ, Bouhlel E, Vandewalle H, Monod H. Muscle mass as a
fac-tor limiting physical exercise. J Appl Physiol 1988; 64: 1472 – 1479 25Ventura N, Hoppeler H, Seiler R, Binggeli A, Mullis P, Vogt M. The
re-sponse of trained athletes to six weeks of endurance training in hyp-oxia or normhyp-oxia. Int J Sports Med 2003; 24: 166 – 172
26Vogt M, Puntschart A, Geiser J, Zuleger C, Billeter, Hoppeler H. Molecu-lar adaptations in human skeletal muscle to endurance training under simulated hypoxic conditions. J Appl Physiol 2001; 91: 173 – 182 27Wagner PD. A theoretical analysis of factors determining V˙O
2maxat sea level and altitude. Resp Physiol 1996; 106: 329 – 343
28Wagner PD. New ideas on limitations on V˙O
2max. Exerc Sport Sci Rev 2000; 28: 10 – 14
29Weibel ER. Stereological Methods. Vol. 1. Practical Methods for
Biolog-ical Morphometry. New York: Academic Press, 1979
30West JB, Boyer SJ, Graber DJ, Hackett PH, Maret KH, Milledge JS, Peters RM, Jr., Pizzo CJ, Samaja M, Sarnquist FH. Maximal exercise at extreme altitudes on Mount Everest. J Appl Physiol 1983; 55: 688 – 698
