Near-infrared spectroscopy (NIRS) is a noninvasive technique evaluating microvascular function. The aim of this study was to assess the reproducibility of NIRS parameters during reactive hyperemia induced by a 5 min brachial artery occlusion. Twenty-four healthy young males (mean 34 +/- 8 years old) had two microvascular function evaluations by NIRS over a 7 to 30-day period (mean 16 +/- 10 days). Intra-subject and inter-observer reproducibility were evaluated with intraclass correlation coefficient (ICC), coefficient of variation (CV), and standard error of measurement (SEM%) for every parameter. Mean NIRS parameters did not differ between both evaluations. Reproducibility was greatest for muscle oxygen consumption (ICC: 0.84; CV: 6.51%; SEM: 7.11%), time to basal O(2)Hb (ICC: 0.63, CV: 20.04%, SEM 27.22%), time to maximal O(2)Hb (ICC: 0.71; CV: 15.61%; SEM: 19.27%), peak of O(2)Hb (ICC: 0.63, CV: 6.68%, SEM 8.53%), time to maximal tHb (ICC: 0.73, CV: 19,61%, SEM 24.56%) and area under the O(2)Hb and tHb curves (ICC: 0.68, CV: 16.15%, SEM 22.93% and ICC: 0.62, CV: 18.59%, SEM 26.64%, respectively). Moreover, inter-observer reproducibility ranged from excellent to perfect (ICC from 0.85 to 1.00) for every parameter. NIRS parameters during reactive hyperemia are highly reproducible which enables their repeated measurement to study microvascular function in healthy subjects.