METHODS: This was an observational cohort on incident end-stage kidney disease (ESKD) patients from January 1, 2007 to December 31, 2008. The primary outcome was all-cause mortality. Patients contributed person-time from the date of ESKD diagnosis until death, transplant or end of study on December 31, 2014, whichever occurred first. An extended Cox regression model with time-varying exposure to dialysis was used to account for immortal time bias.
RESULTS: Of 3990 incident ESKD patients included, 70.2% patients initiated dialysis; 78.8% with haemodialysis (HD) while the remaining 21.2% with peritoneal dialysis (PD). Dialysis reduced hazard of death in both elderly and non-elderly patients even after controlling for comorbidities (hazard ratio [HR] 0.58, 95% confidence interval [CI] 0.50, 0.68 and HR 0.76, 95% CI 0.69, 0.85, respectively). HD was protective in both the elderly and non-elderly (HR 0.53, 95% CI 0.45, 0.63 and HR 0.71, 95% CI 0.64, 0.80, respectively). PD significantly reduced risk of death compared to no dialysis in the elderly but not in the non-elderly.
CONCLUSION: Dialysis improved survival in all incident ESKD patients. The findings suggested a larger protection offered by HD. Although improvement in survival from initiating dialysis was large, its true benefit should take overall quality of life into account. SUMMARY AT A GLANCE This observational study showed that initiation of dialysis improves the survival of end-stage kidney disease (ESKD) patients of all age groups, but the quality of life is an important aspect that has not been explored.