Effect of different types of exercise on bone mineral density in postmenopausal women: A systematic review and meta-analysis

Postmenopausal women face a high risk of bone loss. Exercise has a beneficial effect to prevent bone deterioration in elderly women. This study was conducted to investigate the effect of different types of exercise on bone mineral density at the lumbar spine and femoral neck in postmenopausal women. This review followed the guidelines recommended by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and was registered in advance in the International prospective register of systematic reviews (PROSPERO) (ID: CRD42018095097). Literature searches with no language restriction were conducted through PubMed, Scopus, Web of Science, Cochrane, Science Direct, Eric, ProQuest, and Primo up to March 1 2019 using related MeSH terms and key words. Controlled trials with >= 6 months intervention duration among postmenopausal women that comprised at least one exercise and one control group were included in this study. Meta-analyses were performed using random-effects models and effect sizes calculated using the standardised mean difference (SMD). Seven categories of exercise were classified and used in sub-group analyses. P-values < 0.05 were considered statistically significant. Seventy-five studies were included, assessing the bone mineral density of 5300 (range of mean age: 50-79) post-menopausal women who performed exercise or control protocols. Interventions ranged from six to 30 months and the most common exercise frequency was three sessions per week. The majority of studies employed multicomponent exercise interventions. Results of the meta-analyses showed a significant effect (SMD = 0.37, 95%-CI: 0.25-0.50) for lumbar spine and femoral neck (SMD = 0.33, 0.23-0.43) (all P < 0.0001). Sub-group analysis revealed large heterogeneity and no significant differences in exercise categories at the lumbar spine (P = 0.36) and femoral neck (P = 0.43). However, weight-bearing exercise + jumping + resistance training exhibited the largest effect (SMD = 0.71, 0.33-1.10) at the lumbar spine. Moreover, non-weight-bearing combined + resistance exercise revealed the largest effect (SMD = 0.68, 0.16-1.19) at the femoral neck. Contrary to our expectation, jumping alone indicated a greater decrease in bone mineral density compared to the control group at both the lumbar spine (-0.07, -0.46-0.32) and femoral neck (-0.12, -0.62-0.37). In conclusion, the current study provided a positive small effect of exercise on bone mineral density at the lumbar spine and femoral neck. Findings from sub-group analyses suggest larger benefits in multicomponent interventions that include weight-bearing and resistance training exercises compared to them as a single mode of exercise. Unequal numbers of studies were found for each category (e.g. jumping: two trials; dynamic resistance training: 15 trials), affecting the comparison between them. Therefore, findings should be interpreted with caution; they are not yet generalizable to specific types of exercise.