Age-stratified association between hyperpolypharmacy (≥ 10 medications) and mortality in patients undergoing maintenance hemodialysis: a 3-year single-center study

Background Patients undergoing maintenance hemodialysis frequently experience polypharmacy because of multiple comorbidities. Although polypharmacy has been linked to increased mortality in the general elderly population, its impact on survival may differ by age among dialysis patients. Methods We conducted a single-center retrospective cohort study of 513 outpatient maintenance hemodialysis patients. The number of prescribed oral medications per patient per day was assessed at baseline and compared among non-elderly (< 65 years), early elderly (65–74 years), and late elderly (≥ 75 years) groups. Patients were classified into non-hyperpolypharmacy (< 10 medications) and hyperpolypharmacy (≥ 10 medications) groups. The 3-year all-cause mortality was evaluated using Kaplan–Meier analysis and Cox proportional hazards models. Results The median (Q1–Q3) number of medications (types/patient/day) was 9.7 (7.3–12.3) in non-elderly patients, 10.0 (8.3–12.7) in early elderly patients, and 10.7 (8.3–12.7) in late elderly patients, with no significant differences among the groups. No significant survival difference was observed between the non-hyperpolypharmacy and hyperpolypharmacy groups in the overall cohort or in the early elderly group. In contrast, hyperpolypharmacy was associated with poorer survival in the non-elderly group, whereas it was marginally associated with better survival in the late elderly group. After adjustment, the association in the non-elderly group was attenuated, while that in the late elderly group became statistically significant. A formal test for interaction between hyperpolypharmacy and age group was statistically significant ( P for interaction = 0.023). Conclusions Hyperpolypharmacy may show age-dependent associations with survival in patients undergoing maintenance hemodialysis, potentially being associated with poorer survival in non-elderly patients and with an apparently more favorable survival pattern in late elderly patients. These findings may reflect selection bias and highlights the potential importance of age-specific interpretation of polypharmacy in dialysis patients.