Abstract
BACKGROUND: It is important to consider biological sex as a variable that might influence exercise adaptation in order to optimize exercise prescription for men and women. OBJECTIVE: The aim of this study was to quantify the impact of biological sex on maximal oxygen uptake ([Formula: see text]O(2max)) and performance outcomes after high-intensity interval training (HIIT). METHODS: A systematic search and review was conducted by two independent reviewers up to 8 September 2022 using MEDLINE, SPORTDiscus, and Sports Medicine & Education Index in ProQuest. Trials including healthy adults were included if they presented data for or compared male and female [Formula: see text]O(2max) or performance outcomes in response to HIIT. Performance outcomes included measures of exercise performance and concurrently measured physiological adaptations. Where appropriate, a random-effects, pre-post meta-analysis was undertaken. Data were sub-grouped for men and women, baseline training level, mean age, intervention type, and intervention length. Heterogeneity was assessed using Chi(2), Cochran's Q, and Higgins I(2) and sensitivity analyses, where required. Study quality was assessed using the Newcastle-Ottawa Scale and publication bias was assessed through visual inspection of funnel plots. RESULTS: Thirty-three references from 28 trials were included in the review (n = 965; 462 women and 503 men). Meta-analyses included 19 studies for [Formula: see text]O(2max), eight for peak power output from [Formula: see text]O(2max) testing (PPO), and five for threshold power (power(AT)). Meta-analyses revealed similar increases in [Formula: see text]O(2max) in women (g = 0.57; 95% CI 0.44-0.69) and men (g = 0.57; 95% CI 0.42-0.72), and power(AT) in women (g = 0.38; 95% CI 0.13-0.64) and men (g = 0.38; 95% CI 0.11-0.64). Raw mean differences for change in [Formula: see text]O(2max) were Δ 0.32 L·min(-1) and 3.50 mL·kg(-1)·min(-1) in men, versus Δ 0.20 L·min(-1) and 3.34 mL·kg(-1)·min(-1) for women. No significant sex differences were present for the primary analysis of any outcome. After sub-grouping, significant differences were present for PPO where the effect size was higher for well-trained women (g = 0.37) compared with well-trained men (g = 0.17), and for [Formula: see text]O(2max) where interventions with a duration of 4 weeks or less had significantly smaller effect sizes compared with those longer than 4 weeks (p < 0.001). Unweighted mean percentage change in [Formula: see text]O(2max), PPO, and power(AT) across studies was 11.16 ± 7.39%, 11.16 ± 5.99%, and 8.07 ± 6.55% for women, and 10.90 ± 5.75%, 8.22 ± 5.09%, and 7.09 ± 7.17% for men, respectively. Significant heterogeneity was present for both [Formula: see text]O(2max) and PPO (I(2), range: 62.06-78.80%). Sub-grouping by baseline training status and intervention length decreased heterogeneity in most groups. A qualitative synthesis of other outcomes indicated similar improvements in fitness and performance for men and women with some evidence suggesting differences in the mechanisms of adaptation. LIMITATIONS AND RISK OF BIAS: Publication bias is unlikely to have significantly influenced results for [Formula: see text]O(2max) or power(AT), but the meta-analysis of PPO could have benefitted from additional study data to strengthen results. The overlap in age categories and sensitivity of the analysis limits the accuracy of the results of the sub-grouping by age. CONCLUSIONS: Findings indicated no sex-specific differences for any fitness or performance outcomes. Baseline training status and intervention length accounted for most variability in outcomes. PROSPERO registration number: CRD42021272615.