Hindlimb vascular responses to sympathetic augmentation during acute anemia

The effect of increased sympathetic activity on skeletal muscle blood flow during acute anemic hypoxia was studied in 16 anesthetized dogs. Sympathetic activity was altered by clamping the carotid arteries bilaterally below the carotid sinus. One group (n = 8) was beta blocked by administration of propranolol (1 mg/kg); a second group (n = 8) was untreated. Venous outflow from the left hindlimb was isolated for measurement of blood flow and O2 uptake (VO2). After a 20-min control period, both carotid arteries were clamped (CC) for 20 min followed by a 20-min recovery period. The sequence was repeated after hematocrit was lowered to about 15% by dextran exchange for blood. Prior to anemia, CC did not alter cardiac output or limb blood flow in either group. After induction of anemia, hindlimb resistance was higher with CC in the beta block than in the no block group. Both limb blood flow and VO2 fell in the beta-block group with CC during anemia. Beta block also prevented the additive increases in whole body VO2 seen with CC and induction of anemia. The data showed that the increased vasoconstrictor tone that was obtained with beta block during anemia was successful in redistributing the lower viscosity blood away from resting skeletal muscle, even to the point that muscle VO2 was decreased.     

Muscle perfusion and oxygenation during local hyperoxia

Ventilation with O2 was previously shown to decrease whole-body and hindlimb muscle O2 uptake (VO2) in anesthetized dogs, particularly during anemia. To determine whether this was a purely local effect of hyperoxia (HiOx), we pump perfused isolated dog hindlimb muscles with autologous blood made hyperoxic (PO2 greater than 500 Torr) in a membrane oxygenator while the animals were ventilated with room air. Both constant-flow and constant-pressure protocols were used, and half the dogs were made anemic by exchange transfusion of dextran to hematocrit (Hct) approximately 15%. Thus there were four groups of n = 6 dogs each. A 30-min period of HiOx was preceded and followed by similar periods of perfusion with normoxic blood. In HiOx all four groups showed increased leg hindrance, increased leg venous PO2, and no significant changes in leg O2 inflow. Limb blood flow and VO2 decreased approximately 20% in HiOx with constant-pressure perfusion, regardless of Hct. In the constant-flow protocol, leg VO2 in HiOx was maintained by the anemic animals and actually increased in the normocythemic group. We conclude that HiOx directly affected vascular smooth muscle to cause flow restriction and maldistribution. Constant flow offset these effects, but the increased limb VO2 may have been a toxic effect. Anemia appeared to exaggerate the microcirculatory maldistribution caused by HiOx.     

Respiratory and cardiovascular responses to acute hypoxia and hyperoxia in internally pipped chicken embryos.

During the first day of hatching, the developing chicken embryo internally pips the air cell and relies on both the lungs and chorioallantoic membrane (CAM) for gas exchange. Our objective in this study was to examine respiratory and cardiovascular responses to acute changes in oxygen at the air cell or the rest of the egg during internal pipping. We measured lung (VO2(lung)) and CAM (VO2(CAM)) oxygen consumption independently before and after 60 min exposure to combinations of hypoxia, hyperoxia, and normoxia to the air cell and the remaining egg. Significant changes in VO2(total) were only observed with combined egg and air cell hypoxia (decreased VO2(total)) or egg hyperoxia and air cell hypoxia (increased VO2(total)). In response to the different O2 treatments, a change in VO2(lung) was compensated by an inverse change in VO2(CAM) of similar magnitude. To test for the underlying mechanism, we focused on ventilation and cardiovascular responses during hypoxic and hyperoxic air cell exposure. Ventilation frequency and minute ventilation (V(E)) were unaffected by changes in air cell O2, but tidal volume (V(T)) increased during hypoxia. Both V(T) and V(E) decreased significantly in response to decreased P(CO2). The right-to-left shunt of blood away from the lungs increased significantly during hypoxic air cell exposure and decreased significantly during hyperoxic exposure. These results demonstrate the internally pipped embryo's ability to control the site of gas exchange by means of altering blood flow between the lungs and CAM.     

Effect of hyperoxia on substrate utilization during intense submaximal exercise

Six trained males [mean maximal O2 uptake (VO2max) = 66 ml X kg-1 X min-1] performed 30 min of cycling (mean = 76.8% VO2max) during normoxia (21.35 +/- 0.16% O2) and hyperoxia (61.34 +/- 1.0% O2). Values for VO2, CO2 output (VCO2), minute ventilation (VE), respiratory exchange ratio (RER), venous lactate, glycerol, free fatty acids, glucose, and alanine were obtained before, during, and after the exercise bout to investigate the possibility that a substrate shift is responsible for the previously observed enhanced performance and decreased RER during exercise with hyperoxia. VO2, free fatty acids, glucose, and alanine values were not significantly different in hyperoxia compared with normoxia. VCO2, RER, VE, and glycerol and lactate levels were all lower during hyperoxia. These results are interpreted to support the possibility of a substrate shift during hyperoxia.     

Regional hemodynamic responses to hypoxia and hypermetabolism in polycythemic dogs

Normovolemic polycythemia did not improve the ability of either resting muscle or gut to maintain O2 uptake (VO2) during severe hypoxia because of the adverse effects of increased viscosity on blood flow to those regions. The present study tested whether increased metabolic demand would promote vasodilation sufficiently to overcome those effects. We measured whole body, muscle, and gut blood flow, O2 extraction, and VO2 in anesthetized dogs after increasing hematocrit to 65% and raising O2 demand with 2,4-dinitrophenol (n = 8). We also tested whether regional denervation (n = 8) and hypervolemia (n = 6) affected these responses. After raising hematocrit and metabolism, the dogs were ventilated with air, with 9% O2-91% N2, and again with air for 30-min periods. Reduced blood flow and increased O2 demand, caused by increased blood viscosity and 2,4-dinitrophenol, respectively, increased O2 extraction so that muscle VO2 was nearly supply limited in normoxia. Denervation showed that vasoconstriction had increased in gut and muscle with hypoxia onset but this was overcome after 15 min. By then, muscle was receiving a major portion of cardiac output, whereas gut showed little change. With hypervolemia cardiac output increased in hypoxia but neither gut nor muscle increased blood flow in those experiments. Because regional and whole body VO2 fell in all groups during hypoxia to the same extent found earlier in normocythemic dogs, any real benefit of polycythemia under the conditions of these experiments was dubious at best.     

Effects of hyperoxia on the metabolic response to cold of the newborn rat.

We asked what effects hyperoxia may have on the metabolic response to cold of the newborn rat. Whole body gaseous metabolism (VO2 and VCO2) was measured in 2-day old rats by open flow respirometry at ambient temperatures (Tamb) between 40 and 20 degrees C, changed at a rate of 0.5 degrees C/min during normoxia and hyperoxia (100% O2 breathing). In normoxia, the thermoneutral range was very narrow, at Tamb = 33-35 degrees C. A decrease in Tamb at first stimulated VO2; a further drop in Tamb below 28 degrees C reduced metabolic rate. The metabolic response to cold was not sufficient to maintain body temperature (Tb). In hyperoxia average values of VO2 were above the normoxic values at all Tamb, but the difference was mostly apparent at low Tamb; at 20 degrees C, hyperoxic VO2 averaged 73% more than in normoxia. This metabolic increase determined a significant but small rise of Tb. We conclude that in the 2-days-old rat hyperoxia has a stimulatory effect on metabolism which is Tamb-dependent, being much more apparent in the cold. This supports the concept that the normoxic VO2 of the newborn is limited by the supply of O2. However, the fact that in the cold, even in hyperoxia, VO2 did not reach very high values, and Tb was not maintained, suggests that not only O2 availability, but also the rate of O2 utilization limits the aerobic metabolic response of the newborn.     

The role of aortic chemoreceptors during acute anemia

The importance of aortic chemoreceptors in the circulatory and metabolic responses during acute anemia was studied in anesthetized dogs. Data were obtained from nine dogs in which the aortic chemoreceptors were surgically denervated prior to induction of anemia, and from seven sham-operated dogs. Cardiac output (QT), limb blood flow (QL), limb and whole body oxygen uptake (VO2) were determined at normal hematocrit (Hct) and at 30 min of anemia (Hct = 13%) produced by isovolemic dextran-for-blood exchange. At 30 min of anemia, QT was increased from 91 to 186 mL . kg-1 . min-1 (p less than 0.01) and from 99 to 153 mL . kg-1 . min-1 (p less than 0.01) in the sham and denervated groups, respectively. The increase in QT during anemia was less (p less than 0.05) in the aortic-denervated series. Limb flow was also increased during anemia in both groups (p less than 0.01); the mean value of 89 mL . kg-1 . min-1 in the denervated group was less than that of 130 mL . kg-1 . min-1 observed in the sham animals (p less than 0.05). Whole body VO2 decreased (p less than 0.05) in the denervated group at 30 min of anemia; limb VO2 was maintained at the preanemic control value in both groups. The data indicate that during acute anemia the aortic chemoreceptors contribute to the increase in QT.     

Alteration by hyperoxia of ventilatory dynamics during sinusoidal work

The effects of hyperoxia on ventilatory and gas exchange dynamics were studied utilizing sinusoidal work rate forcings. Five subjects exercised on 14 occasions on a cycle ergometer for 30 min with a sinusoidally varying work load. Tests were performed at seven frequencies of work load during air or 100% O2 inspiration. From the breath-by-breath responses to these tests, dynamic characteristics were analyzed by extracting the mean level, amplitude of oscillation, and phase lag for each six variables with digital computer techniques. Calculation of the time constant (tau) of the ventilatory responses demonstrated that ventilatory kinetics were slower during hyperoxia than during normoxia (P less than 0.025; avg 1.56 and 1.13 min, respectively). Further, for identical work rate fluctuations, end-tidal CO2 tension fluctuations were increased by hyperpoxia. Ventilation during hyperoxia is slower to respond to variations in the level of metabolically produced CO2, presumably because hyperoxia attenuates carotid body output; the arterial CO2 tension is consequently less tightly regulated.     

Ventilation in newborn rats after gestation at simulated high altitude

Pregnant rats were kept at a simulated altitude of 4,500 m (PO2 91 Torr) for the whole of gestation and returned to sea level 1 day after giving birth. During pregnancy, body weight gain and food intake were approximately 30% less than in controls at sea level. Measurements were made on the 1-day-old (HYPO) pups after a few hours at sea level. In normoxia, ventilation (VE) measured by flow plethysmography was more (+17%) and O2 consumption (VO2) measured by a manometric method was less (-19%) than in control (CONT) pups; in HYPO pups VE/VO2 was 44% greater than in CONT pups. In acute hyperoxia, VE/VO2 of HYPO and CONT pups decreased by a similar amount (15-20%), indicating some limitation in O2 availability for both groups of pups in normoxia. However, VE/VO2 of HYPO pups, even in hyperoxia, remained above (+34%) that of CONT pups. HYPO pups weighed slightly less than CONT pups, their lungs were hypoplastic, and their hearts were a larger fraction of body weight. An additional group of female rats was acclimatized (8 days) to high altitude before insemination. During pregnancy, body weight gain and food intake of these females were similar to those of pregnant rats at sea level. Measurements on the 1-day-old pups of this group were similar to those of HYPO pups. We conclude that newborn rats born after hypoxic gestation present metabolic adaptation (low VO2) and acclimatization (high VE/VO2), possibly because of hypoxemia. Maternal acclimatization before insemination substantially alters maternal growth in hypoxia but does not affect neonatal outcome.     

Effects of alpha -adrenergic blockade during acute anemia

Studies were carried out in seven anesthetized paralyzed dogs to examine the importance of alpha -adrenergic tone in the cardiovascular responses during acute anemia. Data were obtained 1) at normal hematocrit (Hct), 2) during anemia produced by isovolemic hemodilution with dextran (Hct, 13-15%), 3) during anemia after alpha -blockade (alpha -bl) with phenoxybenzamine (3 mg/kg), and 4) following volume expansion during anemia with a red blood cell dextran solution. Cardiac output (QT), limb and total body oxygen uptake (VO2), and limb blood flow (QL) were determined. Both QT and QL increased during anemia (P less than 0.01), whereas limb resistance (RL) and total peripheral resistance (TPR) were decreased (P less than 0.01). No further change in either RL or TPR occurred with alpha -blockade anemia, but both QT and QL decreased (P less than 0.01). Whole-body VO2 increased during anemia and then declined with alpha -bl and anemia. Following volume expansion during anemia with alpha -bl, QT, QL, and whole-body VO2 increased. We conclude that alpha -adrenergic sympathetic tone to capacitance vessels is essential for the cardiac output increased during anemia, but has little or no effect on resistance vessels and hence distribution of peripheral blood flow.     

Muscle and blood ammonia and lactate responses to prolonged exercise with hyperoxia

Investigations using nonsteady-state and fatiguing exercise protocols have demonstrated a strong relationship between ammonia and lactate metabolism and have suggested a cause and effect relationship between these two variables. We investigated the lactate-ammonia response using prolonged exercise and inspiration of hyperoxic gas (60% O2-40% N2). The exercise consisted of either 70-75% maximal O2 uptake (VO2 max) for 40 min (series 1, n = 6) or 75-80% VO2max for 30 min (series 2, n = 6) with the subjects inspiring room air on one occasion and hyperoxia in the other test. In both series blood ammonia rose continuously throughout the exercise regardless of the inspired gas treatment; in contrast blood lactate did not increase after 10 min with room air, and with hyperoxia blood lactate was reduced. Muscle lactate and ammonia (series 2; vastus lateralis) had responses similar to the blood data. The data demonstrated no apparent lactate-ammonia relationship with prolonged exercise or in response to hyperoxia, suggesting that ammonia production can be independent of lactate metabolism. The data also suggest that type I fibers can be a major source of ammonia in humans.     

Effects of hypoxia on metabolic rate of conscious adult cats during cold exposure

Oxygen consumption (VO2) and shivering movements were recorded in adult, conscious cats in a thermoneutral (24-27 degrees C) and in a cold (3-8 degrees C) environment during normoxia, hypoxia, or hyperoxia for 55 min. In the cold environment, VO2 correlated with shivering index (SI) under conditions of normoxia or ambient hypoxia (FIO2 = 0.12). During normoxia, VO2 was 63% higher in the cold than the thermoneutral environment. Ambient hypoxia acutely reduced VO2 in cold and thermoneutral environments, the decrement being greater for the former than the latter. Similarly, the variation in VO2 for unit change in SI was greater in hypoxia than normoxic conditions, suggesting that hypoxia influenced nonshivering as well as shivering components of cold-induced VO2. Hypoxia induced by CO (FICO = 0.002) also reduced VO2 and SI, a result that is consistent with previous results indicating that carotid body chemoreceptors do not mediate the suppression of shivering by ambient hypoxia. Hyperoxia increased VO2 and SI in the cold, and the effects of both hypoxia and hyperoxia in the cold were antagonized by increasing FICO2 to 0.03. The results demonstrate that hypoxia suppresses VO2 in the cold by reducing the intensity of shivering and, probably, by an action on metabolic rate that is unrelated to cold-induced calorigenesis.     

Effect of hyperoxia and hypoxia on leg blood flow and pulmonary and leg oxygen uptake at the onset of kicking exercise

The purpose of this study was to examine the interactions of adaptations in O2 transport and utilization under conditions of altered arterial O2 content (CaO2), during rest to exercise transitions. Simultaneous measures of alveolar (VO2alv) and leg (VO2mus) oxygen uptake and leg blood flow (LBF) responses were obtained in normoxic (FiO2 (inspired fraction of O2) = 0.21), hypoxic (FiO2 = 0.14), and hyperoxic (FiO2 = 0.70) gas breathing conditions. Six healthy subjects performed transitions in leg kicking exercise from rest to 48 +/- 3 W. LBF was measured continuously with pulsed and echo Doppler ultrasound methods, VO2alv was measured breath-by-breath at the mouth and VO2mus was determined from LBF and radial artery and femoral vein blood samples. Even though hypoxia reduced CaO2 to 175.9 +/- 5.0 from 193.2 +/- 5.0 mL/L in normoxia, and hyperoxia increased CaO2 to 205.5 +/- 4.1 mL/L, there were no differences in the absolute values of VO2alv or VO2mus across gas conditions at any of the rest or exercise time points. A reduction in leg O2 delivery in hypoxia at the onset of exercise was compensated by a nonsignificant increase in O2 extraction and later by small increases in LBF to maintain VO2mus. The dynamic response of VO2alv was slower in the hypoxic condition; however, hyperoxia did not affect the responses of oxygen delivery or uptake at the onset of moderate intensity leg kicking exercise. The finding of similar VO2mus responses at the onset of exercise for all gas conditions demonstrated that physiological adaptations in LBF and O2 extraction were possible, to counter significant alterations in CaO2. These results show the importance of the interplay between O2 supply and O2 utilization mechanisms in meeting the challenge provided by small alterations in O2 content at the onset of this submaximal exercise task.     

Oxygen uptake and blood flow in canine skeletal muscle during moderate and severe anemia

The metabolic and cardiovascular adjustments of the whole body and skeletal muscle were studied during moderate and severe acute anemia. In 15 anesthetized dogs, venous outflow from the gastrocnemius-plantaris muscle group was isolated. Cardiac output (QT) muscle blood flow (QM), total body and muscle oxygen uptake (VO2) were determined during a control period, and at 30 and 60 min of either (i) moderate anemia (n = 8) in which the mean hematocrit (Hct) was 25% or (ii) progressive anemia (n = 7) in which the mean Hct values were 25% at 30 min and 16% at 60 min of anemia. Muscle VO2, QT, and QM were increased in both groups at 30 min of anemia. By 60 min, QT and QM declined to preanemic control values in the moderate anemia group; whole body VO2 was maintained at the control level. Arterial oxygen transport was the same in the two groups at both 30 and 60 min of anemia despite the difference in Hct at 60 min. Muscle VO2 showed a further and similar rise in both groups between 30 and 60 min of anemia. These data show that the rise in muscle VO2 during acute anemia was not directly proportional to the degree of the hematocrit reduction. Further, the findings suggest that the muscle VO2 response was related to the decrease in arterial oxygen transport.     

Regulation of canine skeletal muscle and hindlimb blood flow in acute anemia

Redistribution of blood flow away from resting skeletal muscles does not occur during anemic hypoxia even when whole body oxygen uptake is not maintained. In the present study, the effects of sympathetic nerve stimulation on both skeletal muscle and hindlimb blood flow were studied prior to and during anemia in anesthetized, paralyzed, and ventilated dogs. In one series (skeletal muscle group, n = 8) paw blood flow was excluded by placing a tourniquet around the ankle; in a second series (hindlimb group, n = 8) no tourniquet was placed at the ankle. The distal end of the transected left sciatic nerve was stimulated to produce a maximal vasoconstrictor response for 4-min intervals at normal hematocrit (Hct.) and at 30 min of anemia (Hct. = 14%). Arterial blood pressure and hindlimb or muscle blood flow were measured; resistance and vascular hindrance were calculated. Nerve stimulation decreased blood flow (p less than 0.05) in the hindlimb and muscle groups at normal Hct. Blood flow rose (p less than 0.05) during anemia and was decreased (p less than 0.05) in both groups during nerve stimulation. However, the blood flow values in both groups during nerve stimulation in anemic animals were greater (p less than 0.05) than those at normal Hct. Hindlimb and muscle vascular resistance fell significantly during anemia and nerve stimulation produced a greater increase in vascular resistance at normal Hct. Vascular hindrance in muscle, but not hindlimb, was less during nerve stimulation in anemia than at normal Hct.(ABSTRACT TRUNCATED AT 250 WORDS)     

Critical O2 transport values at lowered body temperatures in rats

Whole-body O2 uptake (VO2) in rats was reported not to increase when total O2 transport (TOT = cardiac output X arterial O2 concentration) was increased above normal ranges when body temperature was kept at 38 degrees C (J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 53: 660-664, 1982). Similar experiments were performed to see if hypothermic rats at 34 degrees C would increase VO2 with an increased TOT in an effort to generate heat. Anesthetized rats were ventilated with 9 or 12% O2 (hypoxia), room air (normoxia), and O2 (hyperoxia) to vary TOT from 52.6 to 6.6 ml X kg-1 X min-1. VO2 was measured in a closed-circuit, double servospirometer system. Although VO2 was significantly lower at 34 degrees C than the values previously found at 38 degrees C with normoxia and hyperoxia, there was no increase with increasing values of TOT. In spite of a lower plateau value of VO2 at 34 degrees C, the critical value of TOT below which VO2 could not be maintained was nearly the same as at 38 degrees C (22 ml X kg-1 X min-1). The reason for this was that O2 was less completely extracted as TOT was lowered below the critical value in the hypothermic animal. Some of the difficulty in extracting O2 at the tissues was probably due to the decrease in P50 (PO2 at 50% saturation) that occurs with decreased body temperature.     

Metabolism during normoxia, hyperoxia, and recovery in newborn rats.

Aerobic metabolism (oxygen consumption, VO2, and carbon dioxide production, VCO2) has been measured in newborn rats at 2 days of age during normoxia, 30 min of hyperoxia (100% O2) and an additional 30 min of recovery in normoxia at ambient temperatures of 35 degrees C (thermoneutrality) or 30 degrees C. In normoxia, at 30 degrees C VO2 was higher than at 35 degrees C. With hyperoxia, VO2 increased in all cases, but more so at 30 degrees C (+20%) than at 35 degrees C (+9%). Upon return to normoxia, metabolism readily returned to the prehyperoxic value. The results support the concept that the normoxic metabolic rate of the newborn can be limited by the availability of oxygen. At temperatures below thermoneutrality the higher metabolic needs aggravate the limitation in oxygen availability, and the positive effects of hyperoxia on VO2 are therefore more apparent.     

Maximal oxygen consumption in relation to subordinate traits in lines of house mice selectively bred for high voluntary wheel running.

We studied relations between maximal O2 consumption (VO2 max) during forced exercise and subordinate traits associated with blood O2 transport and cellular respiration in four lines of mice selectively bred for high voluntary wheel running (S lines) and their four nonselected control (C) lines. Previously, we reported VO2 max of 59 females at three Po2 (hypoxia = 14% O2, normoxia = 21%, hyperoxia = 30%). Here, we test the hypothesis that variation in VO2 max can be explained, in part, by hemoglobin concentration and Po2 necessary to obtain 50% O2 saturation of Hb (an estimate of Hb affinity for O2) of the blood as well as citrate synthase activity and myoglobin concentration of ventricles and gastrocnemius muscle. Statistical analyses controlled for body mass, compared S and C lines, and also considered effects of the mini-muscle phenotype (present only in S lines and resulting from a Mendelian recessive allele), which reduces hindlimb muscle mass while increasing muscle mass-specific aerobic capacity. Although S lines had higher VO2 max than C, subordinate traits showed no statistical differences when the presence of the mini-muscle phenotype was controlled. However, subordinate traits did account for some of the individual variation in VO2 max. Ventricle size was a positive predictor of VO2 max at all three Po2. Blood Hb concentration was a positive predictor of VO2 max in S lines but a negative predictor in C lines, indicating that the physiological underpinnings of VO2 max have been altered by selective breeding. Mice with the mini-muscle phenotype had enlarged ventricles, with higher mass-specific citrate synthase activity and myoglobin concentration, which may account for their higher VO2 max in hypoxia.     

Single and multigenerational responses of body mass to atmospheric oxygen concentrations in Drosophila melanogaster : evidence for roles of plasticity and evolution

Greater oxygen availability has been hypothesized to be important in allowing the evolution of larger invertebrates during the Earth’s history, and across aquatic environments. We tested for evolutionary and developmental responses of adult body size of Drosophila melanogaster to hypoxia and hyperoxia. Individually reared flies were smaller in hypoxia, but hyperoxia had no effect. In each of three oxygen treatments (hypoxia, normoxia or hyperoxia) we reared three replicate lines of flies for seven generations, followed by four generations in normoxia. In hypoxia, responses were due primarily to developmental plasticity, as average body size fell in one generation and returned to control values after one to two generations of normoxia. In hyperoxia, flies evolved larger body sizes. Maximal fly mass was reached during the first generation of return from hyperoxia to normoxia. Our results suggest that higher oxygen levels could cause invertebrate species to evolve larger average sizes, rather than simply permitting evolution of giant species.     

Effects of incomplete pulmonary gas exchange on VO2 max

Recent evidence suggests that heavy exercise may lower the percentage of O2 bound to hemoglobin (%SaO2) by greater than or equal to 5% below resting values in some highly trained endurance athletes. We tested the hypothesis that pulmonary gas exchange limitations may restrict VO2max in highly trained athletes who exhibit exercise-induced hypoxemia. Twenty healthy male volunteers were divided into two groups according to their physical fitness status and the demonstration of exercise-induced reductions in %SaO2 less than or equal to 92%: 1) trained (T), mean VO2max = 56.5 ml.kg-1.min-1 (n = 13) and 2) highly trained (HT) with maximal exercise %SaO2 less than or equal to 92%, mean VO2max = 70.1 ml.kg-1.min-1 (n = 7). Subjects performed two incremental cycle ergometer exercise tests to determine VO2max at sea level under normoxic (21% O2) and mild hyperoxic conditions (26% O2). Mean %SaO2 during maximal exercise was significantly higher (P less than 0.05) during hyperoxia compared with normoxia in both the T group (94.1 vs. 96.1%) and the HT group (90.6 vs. 95.9%). Mean VO2max was significantly elevated (P less than 0.05) during hyperoxia compared with normoxia in the HT group (74.7 vs. 70.1 ml.kg-1.min-1). In contrast, in the T group, no mean difference (P less than 0.05) existed between treatments in VO2max (56.5 vs. 57.1 ml.kg-1.min-1). These data suggest that pulmonary gas exchange may contribute significantly to the limitation of VO2max in highly trained athletes who exhibit exercise-induced reductions in %SaO2 at sea level.(ABSTRACT TRUNCATED AT 250 WORDS)