The FRAX ® as a predictor of mortality in Japanese incident hemodialysis patients: an observational, follow-up study.

Abstract

The World Health Organization Fracture Risk Assessment Tool (FRAX(®)) was recently developed to estimate the 10-year absolute risk of osteoporotic fracture among the general population. However, the evidence for its use in chronic kidney disease patients has been lacking, and the association between the FRAX(®) and mortality is unknown. Therefore, a hospital-based, prospective, cohort study was conducted to evaluate the predictive ability of the FRAX(®) for mortality in hemodialysis patients. A total of 252 patients who had been started on maintenance hemodialysis, 171 men and 81 women, with a mean age of 67 ± 14 years, was studied. The endpoint was defined as all-cause death. The Cox proportional hazards model was used to calculate hazard ratios and 95 % confidence intervals. During the mean follow-up period of 3.4 ± 2.7 years, 61 deaths occurred. The median (interquartile range) of the FRAX(®) for major osteoporotic fracture was 6.9 (4.6-12.0) % in men and 19.0 (7.6-33.0) % in women. Cumulative survival rates at 5 years after starting dialysis, with the FRAX(®) levels above and below the median, were 51.9 and 87.9 %, respectively, in men and 67.4 and 83.7 %, respectively, in women. Overall, in men, the multivariate Cox regression analyses revealed that the log-transformed FRAX(®) remained an independent predictor of death after adjusting by confounding variables. However, in women, the significant association between the FRAX(®) value and the outcome was eliminated if age was put into these models. Among Japanese hemodialysis patients, the FRAX(®) seems to be useful for predicting death, especially in men.