Abstract
INTRODUCTION: Dialysis withdrawal represents an increasingly common cause of death in patients receiving kidney replacement therapy internationally. Prognostic information about stopping dialysis guides clinicians counseling patients and families regarding end-of-life care. However, few studies examine prognostication after withdrawal. We aimed to determine median survival time after withdrawal of dialysis, and to determine which patient and dialysis-related factors are significantly associated with prognosis. METHODS: This retrospective cohort study used registry data. We included all adult patients from the Western Renal Services who were receiving peritoneal dialysis (PD) or hemodialysis prior to death, whose cause of death was documented as "withdrawal from dialysis" and whose date of death was between January 1, 2016 and June 30, 2022. Demographic, clinical, and biochemical data was extracted. The primary outcome was time-to-death, defined as days from last dialysis session to date of death. RESULTS: Median survival time from last dialysis to death for the PD group (n = 53) was 4 days (interquartile range [IQR]: 3-10 days), not significantly different from the hemodialysis group which was 6 days (IQR: 2-11 days, P = 0.72). For PD, the only variable significantly associated with survival time was reason for withdrawing (P = 0.01). Median survival time was significantly longer for patients withdrawing for psychosocial reasons compared to those withdrawing for other reasons (P = 0.002). For hemodialysis (n = 186), variables significantly associated with survival time from last dialysis to death was reason for withdrawing (P = 0.001), urine production at the time of withdrawal (P = 0.005), serum sodium (P = 0.02) and smoking status (P = 0.009). CONCLUSION: Median survival time was longer for withdrawals for psychosocial reasons compared to medical reasons. The data presented could inform withdrawal discussions regarding prognostication and end-of-life planning with patients and family.