Long-term clinical outcomes of peritoneal dialysis patients: single center experience from Korea.

Of a large body of literature reporting clinical outcomes for patients maintained on peritoneal dialysis (PD), most publications have focused on relatively short-term results. Few reports have focused on long-term survival in PD patients. Here, we present our experience with long-term patient outcomes and further analyses of the trends in demographics and clinical outcomes of 2301 end-stage renal disease (ESRD) patients treated with continuous ambulatory PD (CAPD) during a 25-year period (1981 - 2005) at our institute. Outcomes were analyzed for 1656 patients, excluding those younger than 15 years of age at initiation of CAPD, those having less than 3 months' follow-up, or those who had been on hemodialysis or who received a kidney graft before starting CAPD. In the study patients, technique survival at 5 and 10 years was 71.9% and 48.1% respectively. Patient survival was 69.8% and 51.8%. Mean age at the start of PD (50.4 +/- 13.9 years vs. 44.2 +/- 13.9 years, p < 0.01), ESRD incidence as a result of diabetic nephropathy (30.5% vs. 19.5%, p < 0.01), and incidence of cardiovascular comorbidities (26.6% vs. 20.5%, p < 0.01) were all significantly greater in patients who started PD during the second half of the study period (1993 - 2005) as compared with the first half (1981 - 1992). A multivariate analysis adjusting for these changes in demographics and comorbid conditions revealed that PD therapy starting in 1993 - 2005 was associated with a significant reduction in technique failure [hazard ratio (HR): 0.65; p < 0.01] and mortality (HR: 0.68; p < 0.01) as compared with the earlier period. However, in subgroup analyses, technique survival was not observed to be significantly improved in patients with diabetes. In summary, technique and patient survival have significantly improved despite increases in patient age, cardiovascular comorbidity, and ESRD caused by diabetes. Although diabetes, older age, and cardiovascular comorbidities are not factors that are easily modifiable to improve PD outcomes, results at our institution are encouraging in an era of declining PD utilization.