Objective  To explore the effect of resistance training on bone mineral density (BMD) of total body and parts with high incidence of fracture of lower body (lumbar spine, total hip, trochanter and femoral neck) in elderly people over the age of 60.

  Methods  CNKI, PubMed, Elsevier and WOS databases were searched, and all randomized controlled trials of resistance training on BMD in older adults were collected. Cochrane bias risk assessment tool was used to evaluate the quality and publication bias of the included articles; ReMan5.3.5 software was used for statistical analysis and data sensitivity test.

  Results  A total of 17 articles were included into meta-analysis, 6 with low risk of methodological bias, 9 with moderate risk of methodological bias and 2 with high risk of methodological bias. The total sample size was 719, and the average age was more than 60 years old. Meta analysis results show that resistance training can improve the total body (SMD=0.21, I²=68%, P=0.38) and trochanter (SMD=0.22, I²=56%) BMD in the elderly, but there is no significance. Resistance training can significantly improve the lumbar spine (SMD=0.46, I²=70%, P=0.01), total hip (SMD=0.34, I²=16%, P=0.005) and femoral neck BMD (SMD=0.53, I²=60%, P=0.000 5) in the elderly.

  Conclusion  resistance training can improve the BMD of the elderly, and can significantly improve the lumbar spine BMD, total hip BMD and femoral neck BMD in the elderly. Appropriate resistance training can effectively protect the bones from osteoporosis.