Stimulation of the motor cortex using transcranial magnetic stimulation (TMS) and electromyographic (EMG) recordings have shown corticospinal excitability to be depressed following exercise (Brasil-Neto et al. 1993). We have further examined the central fatigue processes in response to discrete rhythmic exercise of an upper arm muscle on one side of the body to investigate whether this fatigue-related depression spreads to non-exercising muscles of the same limb or to muscles in the contralateral limb.

With local ethical approval and informed consent, eight healthy volunteers (7 males; aged 19-25 years) were seated with their arms relaxed on horizontal armrests and bilateral surface electromyographic (EMG) recordings taken from the first dorsal interosseous (FDI) and biceps brachii (BB) muscles. TMS was applied using a MagStim 200 stimulator connected to a 9 cm circular coil centred over the vertex. A train of six stimuli was delivered each minute, alternating between trials producing stimulation of right arm (coil A-side up) and left arm (coil B-side up) muscles. The stimulus intensity was set to 1.2 X threshold for evoking responses in the relaxed BB muscle. Four trials were completed before starting the exercise protocol; 30 s after cessation of exercise a further 60 trials were conducted. Before exercise a 3.5 kg weight was strapped to the wrist. Subjects first completed a heavy exercise protocol by producing right-armed biceps curls, to a tone repeating at a frequency of 0.8 Hz, until exhaustion. On a second occasion, at least 3 days later, six of the subjects completed a light exercise protocol by performing the same exercise routine for 25 % of the time previously achieved. The amplitude of motor evoked potentials (MEPs) was measured in all four muscles at 2 min intervals while relaxed before and after exercise.

After the heavy exercise protocol the exercised BB showed depressed MEP amplitudes (ANOVA, P < 0.05) falling to a mean level (± S.E.M.) of 25.8 ± 5.0 % of the pre-exercise values. Mean MEP amplitudes werre still depressed (P < 0.05), but had recovered to 58.3 ± 15.0 % 60 min post-exercise, indicating central fatigue processes were still operating an hour after exercise. The non-exercised BB showed significant (ANOVA, P < 0.05) MEP depression 9.5 min post-exercise (60.6 ± 4.9 %), which returned to pre-exercise levels after 40 min. No significant changes were seen in the FDI of either hand. Following the light exercise protocol a lower level of MEP depression, lasting 10 min (ANOVA, P < 0.05), was seen in the exercised BB (53.4 ± 10.1 %), whilst none was seen in the non-exercised BB or either FDI.

We conclude that post-exercise MEP depression seen in non-exercised BB represents spread of cortical depression, possibly via homotopic inhibitory connections (Meyer et al. 1995) in the corpus callosum. The presence of contralateral MEP depression only at higher levels of exercise suggests that this spread occurs when central fatigue processes have reached a substantial level. The spread of post-fatigue depression to the other cortex may operate similarly to the transcortical facilitation observed during unilateral contraction of hand muscles (Stedman et al. 1998).