Optimal resistance training parameters for improving bone mineral density in postmenopausal women: a systematic review and meta-analysis.
Level 1 - systematic review of randomized trials
Systematic review and meta-analysis of randomized controlled trials
PubMed 40420105 · doi:10.1186/s13018-025-05890-1
What was done
A systematic review and meta-analysis of randomized controlled trials evaluated the effects of resistance training on bone mineral density (BMD) across different skeletal sites (lumbar spine, femoral neck, total hip, and trochanter) in postmenopausal women. Four databases (PubMed, Embase, Web of Science, Cochrane Library) were searched through March 2025. Two reviewers independently screened studies, extracted data, assessed risk of bias using the Cochrane Handbook, and conducted meta-analyses with RevMan 5.4 and Stata 18 across different training intensities, durations, frequencies, and session lengths.
What was found
Across 17 RCTs with 690 participants, resistance training significantly improved BMD at: - Lumbar spine (LS): SMD = 0.88 (95% CI [0.21, 1.56], P = 0.01, I² = 91%) - Femoral neck (FN): SMD = 0.89 (95% CI [0.40, 1.39], P = 0.0004, I² = 87%) - Total hip (TH): SMD = 0.30 (95% CI [0.10, 0.50], P = 0.003, I² = 25%) No significant effect was observed at the trochanter (SMD = 0.23, 95% CI [-0.01, 0.47], P = 0.06, I² = 19%). Subgroup analyses showed: - High intensity (≥ 70% 1RM) significantly improved TH and FN BMD (P < 0.05) - Frequency of 3 sessions/week significantly improved LS, FN, TH, and trochanter BMD (P < 0.05) - Durations ≥ 48 weeks significantly improved FN and TH BMD (P < 0.05) - Session durations of 40 minutes significantly improved LS BMD (P < 0.05)
Why it matters
This meta-analysis identifies specific resistance training parameters—notably high loads (≥ 70% 1RM) performed three times weekly for at least 48 weeks—that optimize bone density gains at major osteoporotic fracture sites in postmenopausal women.
Limits
High statistical heterogeneity for the lumbar spine (I² = 91%) and femoral neck (I² = 87%) pooled analyses reduces confidence in the effect sizes. Total sample size across 17 trials was relatively small (690 total subjects, averaging ~41 per study), and clinical fracture outcomes were not evaluated.