Impact of unplanned peritoneal dialysis start on patients' outcomes—A multicenter cohort study

Abstract

Background: Patients with end-stage kidney disease (ESKD) who start unplanned dialysis therapy are more likely to be treated with hemodialysis (HD) using a central venous catheter, which has been associated with a greater risk of infections and other complications, as well as with a higher long-term risk of death. Urgent-start PD is an alternative that has been suggested as an option for starting dialysis in these cases, with potentially better patient outcomes. However, the definition of urgent-start PD is not homogeneous, and no study, to our knowledge, has compared clinical outcomes among urgent start, early start, and conventional start of PD. In this study, we aimed to compare these types of initiation of dialysis therapy in terms of a composite outcome of patient survival and technique failure. Methods: This is a retrospective, multicenter, cohort study, involving data from 122 PD clinics in Brazil. We used the following: Urgent-start groups refer to patients who initiated PD within 72 h after the PD catheter insertion; early-start groups are those starting PD from 72 h to 2 weeks after the catheter insertion; and conventional-start groups are those who used the PD catheter after 2 weeks from its insertion. We analyzed the composite endpoint of all causes of patient's mortality and technique failure (within the initial 90 days of PD therapy) using the following three different statistical models: multivariate Cox, Fine and Gay competing risk, and a multilevel model. Results: We included 509 patients with valid data across 68 PD clinics. There were 38 primary outcomes, comprising 25 deaths and 13 technique failures, with a total follow-up time of 1,393.3 months. Urgent-start PD had no association with the composite endpoint in all three models. Conclusion: Unplanned PD seems to be a safe and feasible option for treatment for patients with non-dialysis ESKD in urgent need of dialysis.