Dietary sodium intake and outcomes: a secondary analysis from SODIUM - HF

Abstract

Abstract Background Sodium restriction is common advice provided to patients with heart failure, however, evidence from the SODIUM-HF trial and a subsequent meta-analysis failed to show a beneficial effect of dietary sodium restriction on outcomes. Purpose This secondary analysis of SODIUM HF is aimed to assess association between baseline dietary sodium intake, the change in dietary sodium at 6 months and a composite of cardiovascular hospitalization, emergency department visit, and all-cause death at 12 and 24 months. Methods Multivariable Cox proportional hazard regression model was used to assess association of dietary sodium level measured at randomization with primary and secondary endpoints. Adjusted Hazard ratios and 95% confidence intervals were presented. A responder analysis also explored the achieved dietary sodium at 6 months with subsequent risk for clinical outcomes. Results 792 participants were included. At baseline, sodium intake was ≤ 1500 mg/day in 19.9% (n = 158), >1500-3000 mg/day in 56.5% (n = 448) and >3000 mg/day in 23.4% (n= 186) of the participants. Women, older patients, and those with lower glomerular filtration rate and higher ejection fraction were more often represented in the group with sodium intake ≤1500 mg/day; conversely, diabetes, prior smoking, use of mineralocorticoid receptor antagonists, and ICD were observed in the groups of higher sodium intake levels (>3000 mg/day)(Table 1). Cox proportional hazard regression analysis showed no association between sodium level (per doubling) at baseline with clinical outcomes (all associations p=ns), however, adjusted models exhibited a trend towards a decreased risk of CV hospitalizations at 2 years. In the responder analysis, a trend towards a decreased adjusted risk for composite outcome and CV hospitalizations at 1 year in those achieving a sodium level <1500 mg at 6 months was observed (Table 2). Adjusted Cox proportional hazard regression analysis showed no association for 1- and 2-year outcomes per 500 mg increase in absolute difference in sodium level between baseline and 6 months. Conclusion In SODIUM HF, there was no association between sodium level at baseline with clinical outcomes at any time point but there is a trend towards a decreased risk of CV hospitalizations at 2 years and in patients achieving <1500 mg / day by 6 months, a trend towards fewer subsequent clinical events was present. This finding should be explored via a prospective clinical trial.Baseline characteristics of populationOutcomes