Association between BMI variability and risk of fracture among Korean men and women: a population based study.

Abstract

The bone-health-related outcome, such as fractures due to BMI fluctuation, has been understudied within Asian populations. In this large-scale, population-based cohort study in Korea, we aimed to investigate the relationship between variability in body mass index (BMI) and the risk of fractures. The study included 166,932 subjects aged ≥ 50 years from the National Health Insurance Service-Health Screening Cohort. The BMI variability value from three follow-up examinations during 2002-2007 was categorized into quartiles. The hazard ratios (HRs) with 95% confidence intervals (CIs) for the effects of BMI variability on the risk of admission from hip, spine, and upper extremity fractures during 2008-2015 were evaluated using a Cox proportional hazards regression analysis. Compared to those in the lowest BMI variability (1st quartile), men in the highest BMI variability (4th quartile) showed an increased risk of spinal fractures (aHR 1.21, 95% CI 1.07-1.36) with a significant linear trend (P for trend = 0.021). Compared to those in the lowest BMI variability (1st quartile), women in the highest BMI variability (4th quartile) showed an increased risk of hip and spinal fractures (aHR 1.35, 95% CI 1.05-1.69; aHR 1.16, 95% CI 1.05-1.28) with significant linear trends (P for trend = 0.021; P for trend = 0.003, respectively). There was no association between BMI variability and incidents of upper extremity fractures for men or women. Association between BMI variability and increased fracture risk depended on sex and fracture types. BMI maintenance, instead of high BMI fluctuation, may be beneficial in terms of lowering the overall fracture risk for Korean adults over 50 years old.