Exercise dose and modality for blood pressure reduction in predominantly postmenopausal women: a dose-response network meta-analysis

Abstract

Background Blood pressure rises during the menopausal transition and postmenopause, yet exercise guidance seldom specifies the type and dose most effective for lowering it. We examined the dose-response relationship between exercise and blood pressure in trials enrolling peri- or postmenopausal women, while recognizing that the available evidence was predominantly derived from postmenopausal populations. Methods We searched PubMed, Web of Science, Embase, and the Cochrane CENTRAL register from inception to March 2026 for randomized controlled trials of exercise interventions lasting at least 4 weeks in peri- or postmenopausal women. To limit dose misclassification, weekly dose was derived only from trials that reported exercise frequency, intensity, and session duration, expressed as METs-min/week and adjusted for attendance where this was reported. A Bayesian model-based dose-response network meta-analysis was fitted, with risk of bias assessed using RoB 2 and certainty of evidence using CINeMA. Results Seventy-four trials involving 3,573 women (mean age, 59.5 years) were included; 70 exclusively enrolled postmenopausal women and four included perimenopausal women. Among eight candidate dose-response functions, the quadratic model yielded the lowest DIC and was retained as the primary model. Under this model, the largest predicted reductions occurred at 1,400 METs-min/week for SBP (−6.70 mmHg, 95% credible interval −9.18 to −4.29) and 1,500 METs-min/week for DBP (−4.20 mmHg, −6.52 to −1.74). The apparent attenuation at higher doses was model-dependent. At the highest observed AQE dose of 800 METs-min/week, aquatic exercise produced the largest model-estimated point reductions for both outcomes (systolic −11.63 mmHg; diastolic −7.73 mmHg); however, these estimates were based on only three AQE intervention arms and the ranking should be considered exploratory. Meta-regression found no significant moderation by age, baseline blood pressure, or intervention length. Certainty of evidence ranged from very low to low. Conclusion Among trial populations composed predominantly of postmenopausal women, exercise was associated with a non-linear, dose-related reduction in blood pressure. The largest model-estimated reductions occurred near 1,400–1,500 METs-min/week. Aquatic exercise appeared promising, but the small number of contributing trials and the generally low certainty of evidence preclude definitive modality ranking. Evidence remains insufficient to draw stage-specific conclusions for perimenopausal women. Systematic review registration https://www.crd.york.ac.uk/prospero/, identifier [CRD420261354599].