Effects of timing of dialysis initiation on the prognosis of peritoneal dialysis patients: a retrospective cohort study

Objective To analyze the effects of dialysis therapy initiation on the prognosis of peritoneal dialysis (PD) patients. Methods PD patients who were newly catheterization and long-term followed-up in Peking University Shenzhen Hospital from January 1, 2012 to March 25, 2019 were retrospectively analyzed. According to the estimate glomerular filtration rate (eGFR) at the time of patients catheterization, the patients were divided into early-dialysis group eGFR＞5.5 ml?min-1?(1.73 m2)-1] and late-dialysis group eGFR≤5.5 ml?min-1?(1.73 m2)-1]. The endpoint events were transferred to other renal replacement therapy (such as hemodialysis, kidney transplantation) or death. Kaplan-Meier method was used to draw survival curve, and log-rank test was used to compare the difference of survival rate between the two groups. Cox proportional hazard model was used to analyze the influencing factors of all-cause death and technical death in PD patients. Results A total of 342 PD patients were enrolled in this study, and there were 165 cases and 177 cases in the early-dialysis and the late-dialysis group respectively. Compared with the early-dialysis group, the proportion of patients with diabetes and men, and the level of hemoglobin, serum calcium and CO2 binding capacity in the late-dialysis group were lower, while the incidence of hypertension, serum phosphorus, blood uric acid and blood urea nitrogen level were higher in the late-dialysis group (all P＜0.05). The median follow-up time was 33(16, 57) months. Kaplan-Meier survival analysis showed that the cumulative survival rate of late-dialysis group was significantly higher than that of early-dialysis group (Log-rank χ2=12.004, P＜0.001). After adjusting for gender, age of catheterization, body mass index (BMI), diabetes mellitus and hypertension, the risk ratio of all-cause death in the early-dialysis group was 1.950 times higher than that in the late-dialysis group (HR=1.950, 95%CI 1.019-3.730, P=0.044). Subgroup analysis showed that the timing of dialysis and the risk of end-point events were not affected by BMI, diabetes stratification and other factors (interactive P＞0.05), but there was interaction between dialysis time and catheter age (interactive P＜0.05). According to the age of catheterization, the risk of all-cause death were higher in the early dialysis group at a young age (≤48 years old) (HR=21.287, 95%CI 2.609-173.665, P=0.004). Conclusions The mortality rate of PD patients is higher in early-dialysis group, which is independent of gender, age, BMI, diabetes and hypertension. The difference is more distinct in low age group.