Abstract
PURPOSE: The evidence from recent epidemiological studies investigating the relationship between bone mineral density (BMD) and the risk of breast cancer (BC) remains inconsistent. MATERIALS AND METHODS: The PubMed, EMBASE, and Web of Science databases were comprehensively searched by two independent authors to identify related cohort studies from the inception of the databases through January 31, 2018. Similarly, two researchers separately extracted the data from the selected studies, and any differences were resolved by discussion. Summarized relative risks (RRs) and 95% CIs were summarized via inverse variance weighted random-effects meta-analysis. Heterogeneity among studies was assessed with the I (2) statistic. RESULTS: Ten studies with 1,522 BC patients among 81,902 participants were included in this meta-analysis. Compared to the participants with the lowest BMD at the lumbar spine, those with the highest BMD had a significantly lower RR for BC (RR =0.75; 95% CI =0.60-0.93; I (2)=23.0%). In the subgroup analyses, although the directions of the results were consistent with those of the main findings, not all showed statistical significance. We failed to detect an association between BMD at the femoral neck or total hip and the risk of BC (RR =0.94; 95% CI =0.66-1.33; I (2)=72.5%). Furthermore, the results of the dose-response analysis did not show a significant association between BMD at the lumbar spine, femoral neck, or total hip and the risk of BC. Funnel plot and statistical analyses showed no evidence of publication bias. CONCLUSION: There is no relationship between BMD and the risk of BC. More prospective cohort studies are warranted to further investigate this issue.