SODIUM GRADIENT AND ALL-CAUSE MORTALITY IN INCIDENT MAINTENANCE HEMODIALYSIS PATIENTS: RESULTS FROM THE MONITORING DIALYSIS OUTCOMES (MONDO) INITIATIVE

Introduction:

There is still uncertainty on the optimal dialysate sodium concentration (DNa) and the association between the resulting DNa-to-serum-sodium (SNa) gradient (GNa), and outcomes in hemodialysis (HD) patients. We investigated the GNa and its association with all-cause mortality in incident HD patients from an international cohort.

Methods:

Patients were required to survive the first 15 months (3 months dialysis initiation phase followed by a 12-month baseline period). All-cause death during a two years follow-up period was considered as an outcome defining an event. We included patients with at least 100 treatments during the baseline in countries contributing at least 200 patients. We built a Cox Proportional hazards model adjusted for region, age, sex, diabetes, interdialytic weight gain (IDWG), and albumin, to fit a bivariate spline function to analyze the joint association of SNa-GNa, with all-cause mortality a as contour plot.

Results:

Out of 82459 eligible patients with available data, we included 8437 chronic HD patients from 7 countries in 3 regions (2269 US, 4796 Europe, and 1372 Asia) with an overall mortality rate of 15.2 per 1000 patient-years. Mean age in the study population was 63±14 years, with 58% males, 28% diabetic patients, mean serum albumin 3.84 ±0.38g/dl, IDWGs of 2±0.9kg, mean SNa 137.9±2.58 mEq/l, mean DNa 138.8±1.7 mEq/l and mean GNa 0.94±2.99 mEq/l. At SNa levels below 138 mEq/L, a neutral to negative GNa appears to be more favorable in terms of mortality. Conversely, when SNa levels exceed 138 mEq/L, no clear association between GNa magnitude and risk of death was observed. (Figure 1 and Figure 2)

Conclusions:

Within the limitations of retrospective research our study demonstrates that there is a trend towards an increased risk of all-cause mortality at lower SNa levels. When considering both the SNa and GNa in a joint fashion, a neutral to negative gradient appears preferable at lower SNa levels, without a clear association at higher levels of SNa. While further studies are needed, our data corroborates a need for prospective trials to support clinical optimization and individualization on the most optimal DNa at different levels of SNa.  

I have no potential conflict of interest to disclose.

I did not use generative AI and AI-assisted technologies in the writing process.