P1444CHANGES IN TREATMENTS AND OUTCOMES OF PATIENTS INITIATING PERITONEAL DIALYSIS DURING THE LAST 10 YEARS: DATA FROM SWEDISH RENAL REGISTRY

Background and Aims The recent years have witnessed several therapeutic improvements regarding both peritoneal dialysis (PD) treatment itself and the prevention and management of complications. We evaluated trends in implementation of therapeutic improvements and their relationship with survival and other relevant clinical outcomes in PD patients from Sweden. Method Using the Swedish Renal Registry (SRR), clinical data from all patients initiating PD in Sweden between 2006 and 2015 were linked to other national healthcare registries to obtain information on vital status, hospitalization and diagnoses during in and out-patient care, along with information on all dispensed drugs, from inclusion to the end of 2017. We first evaluated trends in the use of PD-related treatments and of selected key medications within 2-year blocks. We then evaluated the incidence of death, major cardiovascular events (MACE), technique failure, transplantation and peritonitis episodes within one and two years from PD initiation using standardized incidence rates via logistic regression to account for differences in patient characteristics over time. In subsequent Cox regression models, we studied whether adjustment for implementation of these therapeutic changes modified observed trends in clinical outcomes. Finally, patient survival was compared against the age-sex matched general Swedish population using standardized incidence ratios (SIR). Results 3,312 patients (mean age 63±15 years; 34% women) initiated PD in Sweden during the study period. Across 2-year time blocks, there was no difference over time in the distribution of age, sex or comorbidities with the exception of an increased proportion of patients with chronic pulmonary disease. The proportion of patients undergoing automated peritoneal dialysis and using icodextrin increased over time, while mean standard Kt/V and weekly creatinine clearance did not change. The use of non-calcium phosphate binders, cinacalcet and calcium-channel blockers increased, and the use of angiotensin-converting enzyme inhibitors /angiotensin receptor blockers, erythropoietin, iron, and calcium supplements showed decreasing trends. After standardization for differences in demography and comorbidities, the one-year incidence of peritonitis decreased by 13% and the rate of kidney transplantation increased by 52% in 2014/15 compared to 2006/07. The rates of death, MACE or technique failure did not show differences over time. Similar results were observed with 2-years of follow up and when applying Cox models; Per 2-year period, the risk of peritonitis within one year decreased by 8% (HR 0.92, 95% CI 0.87-0.98) regardless of age, sex, comorbidity and the abovementioned therapeutic changes. The risk of death for PD patients was in 2006/2007 4 times higher [SIR 4.08 (3.15–5.29)] and in 2014/15 3.7 times higher [SIR 3.65 (2.70–4.94)] than the general population, corresponding to a reduction of SIR by 11%. Conclusion During the last 10 years, there has been a gradual implementation of new and established evidence-based treatments in patients starting PD in Sweden. Although survival rates are improving compared to the general population, there were no clear differences in the rates of technique failure, MACE or death across PD patients over time. We observed, however, consistent improvements in peritonitis frequency and access to transplantation.