Association of Sex With All-Cause and Cause-Specific Peritoneal Dialysis Discontinuation.

Abstract

RATIONALE & OBJECTIVE Little is known about the association between sex and specific causes of peritoneal dialysis (PD) discontinuation. This study sought to assess the association of sex with all-cause and cause-specific PD discontinuation and explore the factors mediating these relationships. STUDY DESIGN Retrospective cohort study. SETTING & PARTICIPANTS All patients with kidney failure who started PD between 2005-2019 in Australia. EXPOSURE Sex. OUTCOMES All-cause (transfer to hemodialysis for ≥30 days or death) PD discontinuation, PD discontinuation related to inadequate dialysis, and PD discontinuation related to infection. ANALYTICAL APPROACH Adjusted cause-specific proportional hazards regression models were used to assess the association of sex with all-cause and cause-specific PD discontinuation. Counterfactual mediation analysis was conducted to explore potential mediators (sociodemographic status, geographical remoteness, cardiovascular disease, diabetes, history of peritonitis, late referral, smoking status, and BMI) of these associations. Sensitivity analyses using the Fine and Gray method were implemented to address the competing risks of death, kidney transplantation, and other causes of PD discontinuation. RESULTS 6001 out of 9748 incident patients experienced PD-discontinuation from any cause (2098 died, 793 inadequate dialysis-related, 1442 infection-related, 1668 other cause), with a median follow-up of 1.47 years (IQR 0.67-2.73). Males were more likely to experience PD discontinuation from any cause (HR 95% CI) (1.09, 1.03-1.14, p=0.002) and inadequate dialysis (1.71, 1.47-1.99, p<0.001), but not from infection (0.95, 0.85-1.05, p=0.3). The mediation analyses found that 76.9% of the total effect of sex on all-cause PD discontinuation was explained by mediators, including cardiovascular disease, smoking status, and diabetes, whereas less than 10% of the total effect of sex on PD discontinuation from inadequate dialysis was explained by mediators. LIMITATIONS Residual and unmeasured confounders, such as biological differences, behavioral patterns, hospitalizations, frailty, and severity of comorbidities. CONCLUSION Males were more likely than females to experience PD discontinuation from any cause and from inadequate dialysis. This relationship was mediated by multiple morbidities for PD discontinuation from any cause but not for PD discontinuation from inadequate dialysis. These findings may inform future studies evaluating biological and sociodemographic factors that may contribute to these observed sex differences.