Smoking and risk of fractures requiring hospitalization in haemodialysis patients: a nationwide cohort study in Japan

Abstract

ABSTRACT Background Smoking is a well-established lifestyle risk factor for fractures in the general population, but evidence specific to haemodialysis populations is lacking. This nationwide cohort study used data from the Japanese Society for Dialysis Therapy Renal Data Registry to examine the association between smoking status and fractures. Methods A total of 154 077 patients (64.2% men; mean age 68 years) aged ≥20 years undergoing maintenance haemodialysis thrice weekly at the end of 2016 were followed for 1 year. Among these, 19 004 (12.3%) patients were current smokers. Standardized incidence rate ratios of hospitalization due to any fracture were calculated and analysed by multivariable logistic regression analysis, controlling for potential confounders. Propensity score (PS) matching and subgroup analyses were also performed as sensitivity analyses. Results During the 1-year follow-up period, 3337 fractures requiring hospitalization (1201 hip, 479 vertebral and 1657 other fractures) occurred in 3291 patients. The age- and sex-standardized incidence ratio for current smokers was 1.24 (95% confidence interval 1.12–1.39) relative to non-smokers. Standardized incidence ratios were similar after stratification by age group and sex. Multivariable logistic regression analysis revealed a 1.25-fold higher risk of any fracture requiring hospitalization (95% confidence interval 1.10–1.42) in current smokers relative to non-smokers in a fully adjusted model. PS matching showed similar results (odds ratio 1.25; 95% confidence interval 1.05–1.48). Subgroup analyses for baseline covariates revealed no significant interactions. Conclusions Smoking is a significant risk factor for any fracture requiring hospitalization in haemodialysis patients.