The effect of exercise intensity on bone in postmenopausal women (part 2): A meta-analysis.
Level 1 - systematic review of randomized trials
Meta-analysis of randomized controlled trials
PubMed 33357834 · doi:10.1016/j.bone.2020.115697
What was done
A systematic review and meta-analysis of randomized controlled trials was conducted to determine the effects of low-, moderate-, and high-intensity exercise compared to control on DXA-derived bone mineral density (BMD) at the lumbar spine, femoral neck, and total hip in healthy postmenopausal women. Random-effects models calculated mean differences (MD). Subgroup analyses evaluated exercise modalities (resistance and impact training), concomitant bone medications, and fracture risk.
What was found
Fifty-three trials evaluating 63 interventions (19 low, 40 moderate, 4 high intensity) were included. At the lumbar spine, high-intensity exercise produced significantly greater BMD gains (MD = 0.031 g/cm², 95% CI [0.012, 0.049], p = 0.002) than moderate (MD = 0.012 g/cm², 95% CI [0.008, 0.017], p < 0.001) or low intensity (MD = 0.010 g/cm², 95% CI [0.005, 0.015], p < 0.001). At the femoral neck, low and moderate intensities were equally effective (low: MD = 0.011 g/cm², 95% CI [0.006, 0.016], p < 0.001; moderate: MD = 0.011 g/cm², 95% CI [0.007, 0.015], p < 0.001), with no effect observed for high intensity. Moderate-intensity exercise increased total hip BMD (MD = 0.008 g/cm², 95% CI [0.004, 0.012], p < 0.001), whereas low intensity did not; high-intensity data were insufficient for total hip pooling. Resistance training, especially combined with impact training, was the most effective modality. Exercise did not influence medication efficacy.
Why it matters
The findings show that load magnitude directly influences human bone adaptation, confirming that high-intensity exercise yields superior spine BMD benefits in postmenopausal women.
Limits
Only 4 high-intensity interventions were identified in the literature, leaving hip BMD analyses underpowered. Total participant n was not reported in the abstract. Risk of bias in primary studies was frequently unclear due to poor reporting, and meta-regressions were uninformative.