Effects of exercise on fracture reduction in older adults: a systematic review and meta-analysis.

UNLABELLED

In this meta-analysis, we evaluated the effect of exercise on fracture reduction in the elderly. Our results determined a significantly positive effect on overall fractures, whereas the possibility of a publication bias indicates the need for well-designed (multi-center) trials that generate enough power to focus on osteoporotic fractures.

INTRODUCTION

The preventive effect of exercise on fracture incidence has not been clearly determined yet. Thus, the purpose of this study is to evaluate the effectiveness of exercise in preventing overall and vertebral fractures in older adults by meta-analyses technique.

METHODS

This study followed the PRISMA recommendations for systematic reviews and meta-analyses. A systematic review of English articles between 1980 and March 2012 was performed. Terms used were: "exercise", "fractures", "bone", "falls", "osteoporosis", "BMD", "BMC", "bone turnover", while the search was limited to "clinical trial" and "humans". Controlled exercise trials that reported fracture number as endpoint or observation in subjects 45 years and older were included.

RESULTS

Ten controlled exercise trials that reported overall fractures and three exercise trials that reported vertebral fractures met our inclusion criteria. Overall fracture number in the exercise group was 36 (n = 754) compared with 73 fractures in the CG (n = 670) (relative risk [RR] = 0.49; 95 % confidence interval [CI], 0.31-0.76). No significant heterogeneity of trial results (p = 0.28; I (2) = 17) was determined; however, there was some evidence to suggest a publication bias. The overall RR for vertebral fracture number (0.56; 95 % CI, 0.30-1.04) (EG: 19 fractures/103 subjects vs. CG: 31 fractures/102 subjects) was borderline non-significant while the heterogeneity of trial results also cannot be ruled out.

CONCLUSION

Although there is evidence that exercise reduces overall and, to a lesser degree, vertebral fractures in the elderly, the possibility of publication bias weakens our result and demonstrates the imperative for large exercise studies with dedicated exercise protocols that focus on fractures as a primary endpoint.