Central obesity and risks of pre- and postmenopausal breast cancer: a dose-response meta-analysis of prospective studies.

Epidemiologic evidence has shown inconsistent findings regarding the relationships between abdominal fatness, as measured by waist circumferences (WC) or waist-to-hip ratio (WHR), and risks of pre- and postmenopausal breast cancer (BC). A dose-response meta-analysis of prospective studies was conducted to address these issues. Potentially eligible studies were identified by searching PubMed and EMBASE databases, and by carefully reviewing the bibliographies of retrieved publications and related reviews. The summary relative risks (RRs) with 95% confidence intervals (CIs) were calculated using a random-effects model. When the most fully adjusted RRs were combined, both WC (14 studies, RR  = 1.06, 95% CI: 1.04-1.09, I  = 29.9%) and WHR (15 studies, RR  = 1.07, 95% CI: 1.01-1.14, I  = 52.9%) were significantly positively associated with postmenopausal BC, but neither WC (eight studies, RR  = 1.05, 95% CI: 0.99-1.10, I  = 0%) nor WHR (11 studies, RR  = 1.07, 95% CI: 0.95-1.21, I  = 59.7%) were associated with premenopausal BC. The WHR-postmenopausal BC association lost statistical significance after correcting publication bias (RR  = 1.06, 95% CI: 0.99-1.13). When considering BMI-adjusted RRs, WC was associated with both pre- (five studies, RR  = 1.09, 95% CI: 1.02-1.16, I  = 0%) and postmenopausal BC (seven studies, RR  = 1.05, 95% CI: 1.02-1.08, I  = 6.3%), whereas WHR was not associated with either pre- (seven studies, RR  = 1.12, 95% CI: 0.94-1.34, I  = 70.9%) or postmenopausal BC (eight studies, RR  = 1.05, 95% CI: 0.98-1.13, I  = 57.3%). Among non-current (former or never) users of hormone replacement therapy, the summary RR of postmenopausal BC associated with WC was 1.08 (95% CI: 1.03-1.05, I  = 69.2%, seven studies; BMI-adjusted RR = 1.05, 95% CI: 1.02-1.09, I  = 22.8%, four studies). This meta-analysis indicates that central obesity measured by WC, but not by WHR, is associated with modestly increased risks of both pre- and postmenopausal BC independent of general obesity.