#1917 Peritoneal dialysis modality and outcomes in the peritoneal dialysis outcomes and practice patterns study

Abstract Background and Aims Automated peritoneal dialysis (APD) is the most common peritoneal dialysis (PD) modality in many high-income countries, with increasing adoption in low/middle-income countries. Whether or not differences exist in outcomes between PD modality (APD vs. continuous ambulatory PD [CAPD]) remains controversial. A better understanding of this relationship could inform further APD uptake globally. We evaluated patient survival, permanent hemodialysis transfer (HDT), and peritonitis by PD modality in a contemporary cohort of patients across facilities and countries with significant variation in APD use. Method Using the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS), we identified patients prescribed either CAPD vs. APD with a dialysis vintage of 4 months or longer at study entry in Australia, Canada, Japan, New Zealand, South Korea, United Kingdom, and the United States. Thailand was excluded due to very low APD use. Cox models estimated the association between PD modality and outcomes, including patient survival, HDT, and time to first peritonitis, adjusting for case mix and facility-level factors. Subgroup analyses were prespecified. Additional analyses explored APD as the proportion APD use at the facility level with outcomes, to reduce treatment by indication bias. Results Among 17,591 included patients, 14,343 (82%) were on APD. APD use ranged from 37% in South Korea to 90% in the United States (Fig. 1A). APD patients had similar comorbidity profile but were younger, had lower urine volume, were more likely to use hypertonic glucose, and less likely to use icodextrin compared to CAPD patients. The hazard ratio (HR) for patient survival for APD vs. CAPD was 0.93 (95% CI 0.81–1.06) overall (Fig. 1B). Country-specific mortality results varied, with HR 1.08 (95% CI 0.89–1.28) in the US and HR 0.55 (95% CI 0.34–1.03) in Japan. In a subgroup analysis, APD was associated with better survival in patients with icodextrin use (HR 0.78, 95% CI 0.61–1.00). Risk of HDT was marginally higher among APD patients (HR 1.07, 95% CI 0.97–1.18), despite a lower risk of peritonitis (HR 0.87, 95% CI 0.78–0.98). In Japan and South Korea where APD use was less common, approximately one quarter of facilities reported use in fewer than 20% of patients), lower facility proportion APD use was associated with higher mortality. Conclusion Patients on APD had lower rates of peritonitis. Patient survival did not differ significantly by PD modality in the PDOPPS. However, APD patients had a marginally higher risk of HDT, possibly reflecting its use as a “rescue therapy.” Greater icodextrin use in APD patients may improve survival. Further prospective evaluations are required to explore patient reported outcome differences across PD modalities, to identify specific patients who could benefit from initial APD use or switch from CAPD to APD and to explore the impact of increasing APD use on clinical outcomes across low- and middle-income countries.