Prognostic factors for mortality in middle-aged and older hemodialysis patients: a 5-year observational study.

Clinical guidelines for hemodialysis therapy have been described in an evidence-based manner with most evidence from randomized control trials or retrospective studies in which all generations of the hemodialysis patients were enrolled. Therefore, the question still remains whether these guidelines can be applied to increasing older patients. This study is an observational study, including 735 patients who received maintenance hemodialysis in April 2006. At baseline, the participants' age was 62.1 ± 12.8 years (mean ± SD). Hemodialysis duration was 103.7 ± 89.3 months. In a 5-year observation period (actual follow-up period: 1551 ± 499 days), 175 patients died. Prognostic factors were investigated by multivariate analysis with Cox proportional hazard model. Next, we stratified the patients according to their age. 363 patients were included in the middle-aged patient's category between 40 and 64 years, and 314 were involved in the older patient's category between 65 and 84 years old. As a subanalysis, significant predictors of 5-year survival were examined in the age-stratified cohort. Then, Kt/V, serum β2-microglobulin and calcium concentration were significant predictors in our entire cohort, as well as body mass index, neutrophil count, and serum sodium concentration even after adjustment for age, gender, diabetic status and hemodialysis duration. However, Kt/V, serum β2-microglobulin and calcium concentration controlled by hemodialysis prescriptions were independent risk factors especially in older patients, not in middle-aged patients. In conclusion, hemodialysis prescriptions for lowering uremic toxins and managing mineral-bone disorder are important to decrease the risk of death even in older hemodialysis patients.