The effects of concurrent hypoxic/endurance training on mitochondrial respiration in permeabilized fibers in trained athletes were investigated. Eighteen endurance athletes were divided into two training groups: normoxic (Nor, n = 8) and hypoxic (H, n = 10). Three weeks (W1-W3) of endurance training (5 sessions of 1 h to 1 h and 30 min per week) were completed. All training sessions were performed under normoxic [160 Torr inspired Po(2) (Pi(O(2)))] or hypoxic conditions ( approximately 100 Torr Pi(O(2)), approximately 3,000 m) for Nor and H group, respectively, at the same relative intensity. Before and after the training period, an incremental test to exhaustion in normoxia was performed, muscle biopsy samples were taken from the vastus lateralis, and mitochondrial respiration in permeabilized fibers was measured. Peak power output (PPO) increased by 7.2% and 6.6% (P < 0.05) for Nor and H, respectively, whereas maximal O(2) uptake (Vo(2 max)) remained unchanged: 58.1 +/- 0.8 vs. 61.0 +/- 1.2 ml.kg(-1).min(-1) and 58.5 +/- 0.7 vs. 58.3 +/- 0.6 ml.kg(-1).min(-1) for Nor and H, respectively, between pretraining (W0) and posttraining (W4). Maximal ADP-stimulated mitochondrial respiration significantly increased for glutamate + malate (6.27 +/- 0.37 vs. 8.51 +/- 0.33 mumol O(2).min(-1).g dry weight(-1)) and significantly decreased for palmitate + malate (3.88 +/- 0.23 vs. 2.77 +/- 0.08 mumol O(2).min(-1).g dry weight(-1)) in the H group. In contrast, no significant differences were found for the Nor group. The findings demonstrate that 1) a 3-wk training period increased the PPO at sea level without any changes in Vo(2 max), and 2) a 3-wk hypoxic exercise training seems to alter the intrinsic properties of mitochondrial function, i.e., substrate preference.