Purpose:To investigate the interaction between the development of peripheral locomotor muscle fatigue, muscle recruitment and performance during repeated-sprint exercise (RSE).Method:In a single-blind, randomised and cross-over design, ten male team-sport athletes performed two RSE (fifteen 5-s cycling sprints interspersed with 25 s of rest; power self-selected) in normoxia and in acute moderate hypoxia (FIO2 0.138). Mechanical work, total electromyographic intensity (summed quadriceps electromyograms, RMSsum) and muscle (vastus lateralis) and pre-fontal cortex near-infrared spectroscopy (NIRS) parameters were calculated for every sprint. Blood lactate concentration ([Lac-]) was measured throughout the protocol. Peripheral quadriceps fatigue was assessed via changes in potentiated quadriceps twitch force ($Δ$Qtw,pot) pre- versus post-exercise in response to supra-maximal magnetic femoral nerve stimulation. The central activation ratio (QCAR) was used to quantify completeness of quadriceps activation.Results:Compared with normoxia, hypoxia reduced arterial oxygen saturation (-13.7%, P=0.001), quadriceps RMSsum (-13.7%, P=0.022), QCAR (-3.3%, P=0.041) and total mechanical work (-8.3%, P=0.019). However, the magnitude of quadriceps fatigue induced by RSE was similar in the two conditions ($Δ$Qtw,pot: -53.5% and -55.1%, P=0.71). The lower cycling performance in hypoxia occurred despite similar metabolic (muscle NIRS parameters and blood [Lac-]) and functional (twitch and M-wave) muscle states.Conclusion:Results suggest that the central nervous system regulates quadriceps muscle recruitment and, thereby, performance to limit the development of muscle fatigue during intermittent, short sprints. This finding highlights the complex interaction between muscular perturbations and neural adjustments during sprint exercise, and further supports the presence of pacing during intermittent sprint exercise. © 2013 Billaut et al.