A PERITONEAL DIALYSIS ACCESS QUALITY IMPROVEMENT INITIATIVE: A SINGLE-CENTER EXPERIENCE

Background: Little evidence exists regarding optimal peritoneal dialysis (PD) access insertion pathways, benchmarking for patency targets, and definitions of access dysfunction. Methods: This quality improvement (QI) project evaluated patients with PD catheters inserted at a single center in Toronto, Canada, following: establishment of PD catheter insertion protocols, a PD access coordinator, PD access operator training, and outcomes reporting. We define primary vs secondary PD catheter dysfunction by presentation before/after initial home PD treatment. We report catheter dysfunction rates, interventions restoring PD catheter patency (interventional radiology [IR] vs advanced laparoscopic [AL]) (embedded vs non-embedded) between 2012 and 2017. Results: A total of 297 first PD catheters were inserted between January 2012 and December 2017. Interventional radiology PD catheters (n = 94) were placed in older patients with greater comorbidities and less prior abdominal surgery than AL-placed catheters. Indications for IR insertion included need for urgent dialysis given resource availability (36.2% [n = 34]) and prohibitive surgical risk (26.6% [n = 25]). Interventional radiology-inserted catheters had overall (primary and secondary) dysfunction rates of 17%. Non-embedded AL catheters had 16.1% overall dysfunction. Embedded AL-inserted PD catheters had a 24.6% overall dysfunction rate. Among all dysfunctional catheters, IR manipulation was successful in 31% (n = 11), and surgical revision was necessary in all unsuccessful cases with either lysis of adhesions or omentopexy to establish patency. Conclusion: Our PD catheter QI initiative involved tracking, outcome reporting, defining PD catheter dysfunction and PD access insertion pathway development, yielding important insights into opportunities for program improvement. Multicenter research initiatives are needed to further improve PD access dysfunction definitions and to establish the best benchmarks for these metrics.