The hypothesis that high-intensity (HI) intermittent exercise impairs mitochondrial function was investigated with different microtechniques in human muscle samples. Ten male students performed three bouts of cycling at 130% of peak O2 consumption (V·O2,peak). Muscle biopsies were taken from the vastus lateralis muscle at rest, at fatigue and after 110 min recovery. Mitochondrial function was measured both in isolated mitochondria and in muscle fibre bundles made permeable with saponin (skinned fibres). In isolated mitochondria there was no change in maximal respiration, rate of adenosine 5'-triphosphate (ATP) production (measured with bioluminescence) and respiratory control index after exercise or after recovery. The ATP production per consumed oxygen (P/O ratio) also remained unchanged at fatigue but decreased by 4% (P<0.05) after recovery. In skinned fibres, maximal adenosine 5'-diphosphate (ADP)-stimulated respiration increased by 23% from rest to exhaustion (P<0.05) and remained elevated after recovery, whereas the respiratory rates in the absence of ADP and at 0.1 mM ADP (submaximal respiration) were unchanged. The ratio between respiration at 0.1 and 1 mM ADP (ADP sensitivity index) decreased at fatigue (P<0.05) but after the recovery period was not significantly different from that at rest. It is concluded that mitochondrial oxidative potential is maintained or improved during exhaustive HI exercise. The finding that the sensitivity of mitochondrial respiration to ADP is reversibly decreased after strenuous exercise may indicate that the control of mitochondrial respiration is altered.