INTRODUCTION: Repeated-sprint training in hypoxia (RSH) is a popular and effective method to improve physical performance (Brocherie et al., 2017). However, it requires hypoxic chambers, tents or hypoxic simulators. RSH induced by voluntary hypoventilation at low lung volume (RSH-VHL) is a recent method which can provide a low-cost alternative to RSH (Woorons et al., 2019). This method requires athletes to perform repeated-sprint efforts while holding their breath at the end of expiration, provoking systemic hypoxia as well as hypercapnia. The aim of this study was therefore to review the scientific literature and perform a meta-analysis on the effects of RSH-VHL vs. similar training with normal breathing (RSN) on physical performance. METHODS: The PubMed/MEDLINE, SportDiscus®, ProQuest, and Web of Science online databases were examined for articles published up to February 2023 which evaluated changes in physical performance following RSH-VHL and RSN. The meta-analysis was conducted to determine the standardized mean difference (SMD) between the effects of RSH-VHL vs. RSN on repeated-sprint ability (RSA) related variables: best and mean performance (velocity or power output), sprint decrement score (Sdec), maximal blood lactate concentration ([Lamax]), single-sprint reference (Ref) and maximal oxygen consumption (VO2max). RESULTS: After screening, 6 studies were considered for this meta-analysis, including a total of 127 individuals (mean age 21.2 ± 7 years, 115 males). These studies included 6-10 training sessions over 2-5 weeks. The training protocols consisted in 3 ± 1 sets of 9 ± 4 repetitions of 7 ± 1 s sprint with 22 ± 3 s intra-set recovery and 6 ± 6 min inter-set rest. Average recorded oxygen saturation (SpO2) for sprints in RSH-VHL was 89.4 ± 2.7 %. Maximal (SMD = 0.37, P = 0.04; small to moderate effect) and mean performance (SMD = 0.49, P = 0.01; small to moderate effect), as well as Sdec (SMD = 0.58, P < 0.01; moderate to large effect) were significantly enhanced with RSH-VHL vs. RSN. The increase in [La]max following RSA in RSHVHL compared to RSN was not significant (SMD = 0.32, P = 0.1; small to moderate effect). No significant effect was observed for RSH-VHL Ref (SMD = -0.14, P = 0.51) and VO2max (SMD = 0.54, P=0.11). CONCLUSION: The present meta-analysis indicates that repeated-sprint training in hypoxia induced by hypoventilation at low lung volume provides putative gains in repeated-sprint maximal and mean velocity and fatigue resistance (sprint decrement score) than similar training with normal breathing. Whether this method induces similar or additional benefits compared to repeated-sprint training in hypoxia – demonstrated to induce higher hypoxic stress (Imai et al., 2022) but without hypercapnic stimulus – remains to be elucidated.