Introduction. The fatigue following explosive contractions (≈ 200 ms in duration) is involved in most actions, especially sporting gestures (such as shot put, vertical jumps, and sprinting during kayaking, cycling and running) are based on explosive contractions of short duration (< 200 ms). Previous studies investigated the fatiguing effect of rapid isometric contractions maintained for longer duration (1 – 5 s). Thus, so far the fatiguing effect of purely explosive isometric (which is the most ecological contraction modality) are unknown. The early phase of explosive contractions is mainly dependent on central mechanisms such as motor unit recruitment and firing rate. Nevertheless, it is unknown if repeated explosive isometric contractions induce mainly central or peripheral fatigue. Thus, we aimed to verify the extent of central and peripheral fatigue induced by repetitive purely explosive contractions of knee extensors. 
Methods. 30 volunteers performed 100 fast and brief (< 200 ms in duration) explosive isometric knee extensions interspersed by 3 s of rest. High-density surface electromyography (HDsEMG) was recorded from the vastus medialis and lateralis muscles. During the first (PRE) and last (POST) 10 contractions, we calculated the rate of force development (RFD) over the first 50, 100, and 150 ms of contractions. We also evaluated the contractile and EMG responses of electrically evoked single and octet stimuli (eight stimuli, 300 Hz). Neural efficacy was calculated as the ratio between octet-evoked and voluntary force during the first 50 ms of contraction. Student’s t-test was used to evaluate the difference between PRE vs. POST. This study was approved by the Ethical Advisory Committee and performed in accordance with the Helsinki Declaration. 
Results. Voluntary RFD decreased by 12±8% at 50ms and 6±5% at 100ms (all p values < 0.01). Root mean square EMG (normalized to M-wave amplitude) decreased by around 20±10% in all time intervals from 0 to 150 ms (all p values <0.001); conduction velocity (normalized to M-wave) decreased by 2.5±1.5% at 100 and 150 ms (p<0.05). Neural efficacy also declined by 15±8% (p<0.01). Some metrics related to peripheral function did not show fatigue but even increased: the RFD assessed at 0-50 ms through the electrically evoked octet increased by 3±2% (p<0.01).Similarly, the M-wave amplitude and conduction velocity increased by 7±5% (p<0.05). 
Conclusions. The neural efficacy was impaired following 100 purely explosive contractions. However, the peripheral mechanisms did not show evidence of fatigue. Therefore, the impairment in explosive capacity was mainly driven by central fatigue.