Associations Between Serum Sodium, Peritoneal Dialysis-Associated Peritonitis, and Mortality in the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS).

RATIONALE & OBJECTIVE The clinical consequences of hyponatremia among patients receiving peritoneal dialysis (PD) are poorly understood. This study sought to evaluate the association of variations in serum sodium with peritoneal dialysis-associated peritonitis and death. STUDY DESIGN Multicenter observational cohort study. SETTINGS & PARTICIPANTS 23,707 participants in the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS) in 8 countries between 2014 and 2022 with a serum sodium available at study enrollment. PREDICTOR Serum sodium categories (categories: <135, 135-137, 138-139, 140-141, >142 mEq/L) at study enrollment. OUTCOMES Time to first peritonitis episode and all-cause mortality. ANALYTICAL APPROACH Cause-specific hazards models adjusted for demographic, comorbidity, and treatment characteristics. Secondary analyses using average serum sodium levels over time and evaluation of modification of the association between serum sodium and study outcomes by use of icodextrin, as well as patient characteristics and peritoneal dialysis (PD) modality. RESULTS Compared to a serum sodium of 140 - 141 mEq/L (n=5065), those with a sodium of <135 mEq/L (n=3601) had longer dialysis vintage and were more likely to have diabetes and use icodextrin. Across serum sodium categories, there were no differences in the adjusted peritonitis risks. Compared to individuals with a sodium of 140-141 mEq/L, those with a sodium of <135 mEq/L (adjusted hazard ratio [AHR] 1.45, 95% CI 1.29-1.63), a sodium of 135-137 mEq/L (AHR 1.26, 95% CI 1.13-1.42) and a sodium ≥142 mEq/L (AHR 1.16, 95% CI 1.03-1.30) were all associated with higher mortality. Associations between serum sodium and mortality were similar across all patient characteristic and PD modality subgroups. Peritonitis risk was not detectably different across serum sodium categories regardless of treatment with icodextrin. LIMITATIONS Lack of standardization/validation of serum sodium measures across sites; icodextrin use was limited to a subset of patients. CONCLUSIONS Variations in serum sodium were associated with death but not peritonitis risk. Future studies are needed to understand the mechanisms underpinning these associations and whether modification of serum sodium would improve outcomes among those receiving PD.