Contents

Effects of exercise-induced arterial hypoxemia on limb muscle fatigue and performance

Summary

1. Reductions in arterial O₂ saturation (-5 to -10% SaO₂ < rest) occur over time during sustained heavy intensity exercise in a normoxic environment, due primarily to the effects of acid pH and increased temperature on the position of the HbO₂ dissociation curve.

2. We prevented the desaturation incurred during exercise at ∼90% O_2 MAX via increased F_IO₂ (.23 to .29) and showed that exercise time to exhaustion was increased.

3. We used supramaximal magnetic stimulation (1 – 100 Hz) of the femoral nerve to test for quadriceps fatigue. We used mildly hyperoxic inspirates (F_IO₂ .23 to .29) to prevent O₂ desaturation. We then compared the amount of quadriceps fatigue incurred following cycling exercise at SaO₂ 98% vs. 91% with each trial carried out at identical work rates and for equal durations.

4. Preventing the normal exercise-induced O₂ desaturation prevented about one-half the amount of exercise-induced quadriceps fatigue; plasma lactate and effort perception were also reduced. In a subset of less fit subjects who showed only minimal arterial hypoxemia during sustained exercise (SaO₂ ∼95%), breathing a mildly hypoxic inspirate (F_IO₂ .17; SaO₂ ∼88%) exacerbated the quadriceps fatigue.

5. We conclude that the normal exercise-induced O₂ desaturation during heavy intensity endurance exercise contributes significantly to exercise performance limitation in part because of its effect on locomotor muscle fatigue.

Exercise-Induced Arterial Hypoxemia (EIAH)

EIAH is defined as a reduction in arterial O₂ saturation (SaO₂) and occurs for a variety of reasons. During short-term incremental exercise in some highly trained subjects arterial PO₂ may fall secondary to an excessively widened alveolar to arterial PO₂ difference and in the absence of significant hyperventilation.¹ If this EIAH is prevented (via increased F_IO₂), O_2 MAX is increased.² During constant load, high intensity cycling or running exercise sustained to the point of exhaustion, SaO₂ falls progressively over time due primarily to a time- (and intensity-) dependent metabolic acidosis and rising body temperature, which shifts the O₂ dissociation curve to the right (see Figure 1). In some highly fit subjects (especially during running exercise) a reduced PaO₂ will also contribute to a reduced SaO₂³ (see Figure 2). Preventing this desaturation by adding small amounts of hyperoxic inspired gas mixtures (.23 - .30 F_IO₂) induces an increase in exercise time to exhaustion (See Figure 3). Furthermore, if the O₂ desaturation is exacerbated by acutely reducing F_IO₂ or ascending to high altitudes, exercise time to exhaustion is further reduced (Figure 3).

Figure 1. EIAH during heavy intensity, constant load cycling exercise in 11 fit young adult male cyclists. The O₂ desaturation was due primarily to a time-dependant metabolic acidosis (pH ∼7.17) and rise in temperature (∼ +2°C) as PaO₂ was 80 – 90 mmHg.

Figure 2. Arterial O₂ saturation during constant load, high intensity treadmill running to exhaustion in 17 fit young women (F_IO₂ .21). Mean values are shown for those with a low PaO₂ (∼75 mmHg) (closed circles) and those who maintained a high PaO₂ (85 – 90 mmHg) (open circles) throughout. Dashed lines refer to subjects who did not show hypoxemia during exercise (mean values, open symbols) and solid lines refer to those with hypoxemia (closed symbols, mean values). The PaCO₂ was higher and the alveolar to arterial PO₂ difference wider in the low PaO₂ group. The time-dependant fall in SaO₂ beyond the first two minutes of running was due to the rise in temperature (+2.2°C) and fall in arterial pH (∼7.25 pH). (After Wetter et al., 2001.³)

Figure 3. Effects of EIAH on time to exhaustion at a fixed, high intensity work rate. Note in a normoxic environment with an end-exercise arterial O₂ saturation which averaged 91% (see Figure 2) time to exhaustion was about 13 minutes. Preventing this reduction in SaO₂ (via F_IO₂ .23 to .29) allowed the subjects to exercise at least 16% longer, whereas reducing F_IO₂ below normoxic levels caused moderate (at F_IO₂ .17 and 87% SaO₂) and then marked (at .13 F_IO₂ and 75% SaO₂) reductions in exercise time to exhaustion.

We asked the fundamental question, “Why does arterial hypoxemia—either the 6 to 10% reduction in SaO₂ induced by prolonged heavy exercise in a normoxic environment or the more severe O₂ desaturation encountered during prolonged heavy exercise at high altitudes—curtail performance time?” Is this curtailment strictly a result of reduced O₂ transport to working locomotor muscle leading to “peripheral” end-organ fatigue? This peripheral fatigue effect is certainly a reasonable hypothesis given the evidence that hypoxemia will reduce Ca²⁺ reuptake and release in the sarcoplasmic reticulum, thereby decreasing cross-bridge activation and force output;⁴ and this effect may occur through a number of mechanisms, including accumulation of lactate and hydrogen ions, inorganic phosphate and/or free radical production.⁵ A recent study showed indirect myoelectric evidence of severe hypoxic effects on locomotor muscle fatigue during cycling.⁶

Alternatively, the long-held concept of a “central governor” limiting motor recruitment of working muscle such that the function of vital organs is protected may explain exercise limitation in the presence of hypoxemia.⁷ Hence, this latter hypothesis would require reflex inhibition of central motor output to locomotor muscles in order to protect against impending failure of vital organs,⁷ and/or the occurrence of lung edema.⁸ Limiting the duration and/or magnitude of cerebral hypoxia in order to preserve cerebral aerobic metabolism may present yet another potential source of central inhibition of locomotor muscle recruitment. A recent study prevented EIAH during maximal rowing exercise and observed an increase in brain oxygenation but no change in muscle oxygenation, thereby implying that changing SaO₂ had little effect on muscle O₂ transport, per se.⁹ Indeed, the classic studies of John Sutton, Jack Reeves and colleagues in Operation Everest II predicted a major role for non-peripheral factors in limiting exercise performance during the simulated ascent of Everest.^10,11

General Methods

Eleven subjects were studied of above average aerobic fitness (O_2 MAX = 44 – 69 ml/kg/min, ages 19 – 33 years). We used supramaximal magnetic stimulation of the femoral nerve before and after cycling exercise to determine if indeed locomotor muscle fatigue, per se, was induced by changing levels of arterial oxygenation during high intensity exercise in normoxic and in hypoxic environments. This procedure consisted of paired, supra-maximal stimuli delivered over a range of frequencies (1 – 100 Hz), achieved by varying the duration of the inter-stimulus interval. The quadriceps force output in response to supramaximal nerve stimulation was shown to be highly reproducible (coefficient of variation < ± 6%) both within- and between-days. Evoked potentials in response to nerve stimulation were measured from the quadriceps muscle EMG; their magnitude remained unchanged from baseline to post exercise conditions ensuring that the motor input to the muscle was supramaximal and equal before and after the cycling exercise. Superimposition of a supramaximal twitch on a maximum voluntary quadriceps contraction produced an average force output that averaged 7% of the potentiated twitch value at rest, indicating that subjects did not fully activate their quadriceps via voluntary effort.

Experiment A. Preventing EIAH in a Normoxic Environment

Subjects cycled at a fixed workload at an intensity that averaged 90% of their peak maximal work rate, until they could no longer maintain a target pedaling frequency. Arterial blood was obtained periodically and magnetic stimulation was applied at baseline and at intervals from 2.5 to 70 minutes following exercise. Then the subjects returned and repeated the experiment only with supplemental inspired O₂ (.23 - .29 F_IO₂) added in amounts that were just sufficient to prevent EIAH, i.e., SaO₂ was maintained at resting levels (∼98%). On this second day subjects exercised at power outputs and for durations that were identical to those under control (F_IO₂ .21) conditions. Thus the only difference between the two exercise conditions was the SaO₂, i.e., 91% vs. 98%.

Figure 4. Cycling exercise to exhaustion in normoxia caused a reduction in force output of the quadriceps in response to supramaximal femoral nerve stimulation which averaged one-third below baseline. When the hypoxemia was prevented (.27 F_IO₂) and the exercise carried out for an identical time and work rate as at F_IO₂ .21, quadriceps fatigue was reduced by more than 50%. When EIAH was made greater by mild environmental hypoxia (F_IO₂ .17), quadriceps fatigue was enhanced. Finally exercise to exhaustion in normoxia and severe hypoxia (F_IO₂ .13) are shown. Note that severe hypoxemia caused exercise to stop in less than one-third the time as in normoxia; however the amount of quadriceps fatigue was only one-half that in normoxia.

The key fatigue findings are summarized in Figure 4. Note that exercise in normoxia, which caused a progressive desaturation to 91% SaO₂ (range = 87 – 93%), resulted in a reduction of force output immediately following exercise at all stimulation frequencies (1 – 100 Hz) that averaged 33% below baseline and returned gradually to baseline levels over 70 minutes of recovery. When the EIAH was prevented and SaO₂ held at resting levels, the reduction in force output was still significant but only about one-half that which occurred under control conditions in the presence of EIAH. Thus, the prevention of EIAH, per se, significantly reduced the amount of quadriceps fatigue induced by the exercise; it also significantly lowered the absolute level and rate of rise of arterial blood lactate concentration over the final half of the exercise, and reduced the rate of rise of effort perception for both limb discomfort and dyspnea (data not shown). Finally, using the twitch stimulation superimposed on the maximum voluntary contraction, we observed that voluntary activation of the quadriceps was reduced from 93% during the pre-exercise resting baseline to 85% following exercise in normoxia; and when desaturation was prevented, voluntary activation fell less than half this amount (93% at baseline to 90% immediate post-exercise).

These findings demonstrate that the arterial O₂ desaturation that normally accompanies heavy intensity sustained exercise in a normoxic environment contributes significantly to locomotor muscle fatigue. In turn, we think it reasonable to conclude that the lessening of local muscle fatigue with the prevention of O₂ desaturation contributes to an enhancement of exercise performance. Nevertheless, we cannot claim a true cause-effect relationship, because we are unable to determine how these data obtained during supramaximal nerve stimulation in recovery translate precisely into the subjects’ capability for sustaining a given (likely sub-maximal) power output during the preceding exercise.

While these data clearly implicate a significant effect of reduced O₂ transport on locomotor muscle fatigue and on exercise performance, they do not rule out an effect of O₂ desaturation on reducing motor output to the locomotor muscles during exercise i.e. “central fatigue”.¹² Indeed the finding that exercise significantly reduced voluntary activation of the quadriceps, and that this was largely relieved by preventing O₂ desaturation, indirectly implicates a contribution from “central fatigue” to hypoxemic effects on exercise limitation. A major outstanding problem with interpretation of these tests is whether the change in force output with the superimposed twitch, as conducted in the resting subject during recovery, truly represents “central inhibition” of the volitional force produced during the preceding rhythmic exercise task. To date there is no direct evidence—pro or con—of an effect of arterial hypoxemia on reflex inhibition of central motor output to locomotor muscles during exercise. Certainly the reduced rates of rise of effort perceptions during exercise when EIAH was prevented might also have contributed to exercise performance limitation and may be classified as “central” fatigue (or “symptom limited”). However since much of the cause of enhanced effort perceptions in the presence of hypoxemia likely originated from intensified sensory feedback input from fatiguing, acidic muscles, then this type of “central” fatigue is causally linked to “peripheral” fatigue.

Experiment B. Effect of Hypoxic-Induced Moderate Hypoxemia

This experiment was conducted in those subjects who experienced minimal O₂ desaturation (∼95%) during the exercise in normoxia. A similar design was used as in experiment A, in that the effect on quadriceps fatigue was compared following exercise of identical work rates and durations. In these subjects an F_IO₂ of .17 reduced the mean exercise SaO₂ to 88% and significantly increased the amount of quadriceps fatigue by 20 – 25% over that observed at F_IO₂ .21 (SaO₂ 95%) (see Figure 4). Furthermore the moderate reductions in SaO₂ below 90% increased the rate of rise of blood lactate and effort perceptions during the exercise. So again, as with the prevention of EIAH in a normoxic environment, the further reduced SaO₂ in a mildly hypoxic environment was linked to performance limitation by means of O₂ transport-induced reductions in the force output of the locomotor muscles in response to supramaximal motor nerve stimulation.

We propose that the effects of EIAH on locomotor muscle (peripheral) fatigue mechanisms were due to reductions in muscle O₂ transport, which in turn would reduce muscle capillary PO₂ and mitochondrial PO₂. Since the work rates in our study required a O₂ very close to O_2 MAX preventing the O₂ desaturation also raised mean O₂ about 5% (at end-exercise). Thus, subjects were exercising at a slightly lower relative work intensity which would account for at least some of the reduction in lactate production and fatigue.

Summary

The schematic diagram in Figure 5 outlines the various types of contributions to curtailment of performance experienced in the presence of arterial hypoxemia. Listed are peripheral muscle fatigue secondary to reduced O₂ transport to muscle and two types of “central” factors, namely conscious effort perception and reflex inhibition, which might limit performance by reducing motor output to the working locomotor muscles. Our results show that for both levels of hypoxemia, its effect on limiting performance time was consistently associated with significant peripheral (i.e. locomotor muscle) fatigue. We especially emphasize that even in a normoxic (i.e. sea-level) environment, the 6 to 10% arterial O₂ desaturation normally produced during heavy intensity, sustained, exercise in healthy subjects is sufficient to significantly exacerbate locomotor muscle fatigue. An additional contribution to exercise limitation occurs from the two types of “central” influences inhibiting motor output to the limb muscles during exercise. One of these "central" factors, i.e. conscious effort perception, is strongly influenced by peripheral muscle fatigue, per se. The other, “reflex” inhibition, has not been measured directly during whole body exercise. A significant contribution from one or more of these “central” influences is likely to be present during exercise at all levels of arterial hypoxemia. It is also likely that the relative contributions of these “peripheral” and “central” mechanisms—and their interactive effects—to exercise performance will depend upon the exercise intensity, severity of hypoxemia and even the fitness of the subject.

Figure 5. Schematic of the “peripheral” and “central” fatigue influences on hypoxemic-induced limitations to exercise performance. We found peripheral locomotor muscle fatigue to be induced by all levels of arterial hypoxemia studied—including the EIAH which occurs during heavy sustained exercise in normoxia. Indirect evidence also implicates “central” fatigue contributions—especially in severe environmental hypoxia-induced arterial hypoxemia.

Acknowledgements

References

1.  Dempsey JA, Wagner PD. Exercise-induced arterial hypoxemia. J. Appl. Physiol. 1999; 87(6):1997-2006.

2.  Harms CA, McClaran SR, Nickele GA, Pegelow DF, Nelson WB, Dempsey JA. Effect of exercise-induced arterial O₂ desaturation on VO_2 max in women. Med. Sci. Sports Exerc. 2000; 32(6):1101-8.

3.  Wetter TJ, St Croix CM, Pegelow DF, Sonetti DA, Dempsey JA. Effects of exhaustive endurance exercise on pulmonary gas exchange and airway function in women. J. Appl. Physiol. 2001; 91(2):847-58.

4.  Duhamel TA, Green HJ, Sandiford SD, Perco JG, Ouyang J. Effects of progressive exercise and hypoxia on human muscle sarcoplasmic reticulum function. J. Appl. Physiol. 2004; 97(1):188-96.

5.  Fitts RH. Cellular mechanisms of muscle fatigue. Physiol. Rev. 1994; 74(1):49-94.

6.  Taylor AD, Bronks R, Smith P, and Humphries B. Myoelectric evidence of peripheral muscle fatigue during exercise in severe hypoxia: some references to m. vastus lateralis myosin heavy chain composition. Eur. J. Appl. Physiol. 1997; 75:151-159.

7.  Hill AV. Muscular exercise, lactic acid, and the supply and utilisation of oxygen. Parts VII-VIII. Proc. R. Soc. Br. 1924; 97:155-176.

8.  Pickar JG, Hill JM, Kaufman MP. Stimulation of vagal afferents inhibits locomotion in mesencephalic cats. J. Appl. Physiol. 1993; 74(1):103-10.

9.  Nielsen HB, Boushel R, Madsen P, Secher NH. Cerebral desaturation during exercise reversed by O₂ supplementation. Am. J .Physiol. 1999; 277(3 Pt 2):H1045-52.

10.  Garner SH, Sutton JR, Burse RL, McComas AJ, Cymerman A, Houston CS. Operation Everest II: neuromuscular performance under conditions of extreme simulated altitude. J. Appl. Physiol. 1990; 68(3):1167-72.

11.  Sutton JR, Reeves JT, Wagner PD, Groves BM, Cymerman A, Malconian MK, Rock PB, Young PM, Walter SD, Houston CS. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J. Appl. Physiol. 1988; 64(4):1309-21.

12.  Gandevia SC. Spinal and supraspinal factors in human muscle fatigue. Physiol. Rev. 2001; 81:1725-1789.