Divergent responses of adiponectin and leptin to exercise in women with overweight or obesity: a systematic review and meta-regression.

Objective: We examined how the volume of exercise relates to circulating adiponectin and leptin in women with overweight or obesity, asking whether each hormone responds in proportion to the amount of exercise accumulated or instead changes only once a sufficient stimulus has been reached. Methods: We searched five databases (Web of Science, PubMed, Cochrane Library, MEDLINE, and Embase) for randomized controlled trials (RCTs) appearing through January 2026 (PROSPERO: CRD420261438448) and pooled standardized mean differences (SMDs) under random-effects models. Weekly exercise volume (minutes per week) was modelled as a continuous moderator in univariate meta-regression, and moderate-intensity continuous training (MICT) was compared with high-intensity interval training (HIIT). Results: Thirty-four RCTs (N = 1,987) were analyzed. Exercise raised adiponectin (SMD = 0.60; P < 0.05) and lowered leptin (SMD = -0.70; P < 0.01), but neither response scaled with the weekly volume of exercise. In univariate meta-regression the slope for weekly volume was non-significant for both adiponectin (slope = -0.0039, P = 0.331) and leptin (slope = 0.0031, P = 0.394), consistent with a threshold effect rather than a gradual, volume-dependent change. Where the hormones diverged was in intensity rather than volume: the pooled adiponectin effect was numerically larger for MICT than for HIIT, although the subgroup difference was not statistically significant, whereas the leptin reduction was similar across intensities. Because energy intake and expenditure were not reported consistently, the contribution of energy balance to the leptin response could not be tested directly. Conclusion: Adiponectin and leptin both improve with exercise, but neither response scales with the weekly volume of training. Adiponectin rose more with moderate-intensity continuous training (MICT) than with HIIT, whereas the leptin reduction was similar across intensities; in both cases the change appeared once an adequate stimulus was delivered rather than accumulating with extra weekly minutes, with energy balance a likely but untested contributor to the leptin response. Prescription can therefore be tailored to the target: moderate-intensity continuous aerobic training when the aim is to raise adiponectin, and either MICT or shorter HIIT when the aim is to lower leptin, provided that energy balance is also managed. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/, identifier CRD420261438448.