Plasma sodium and mortality risk in patients with myocardial infarction and a low LVEF.

Hyponatremia is a known prognostic factor for mortality in patients with heart failure but has not been extensively studied in patients with myocardial infarction (MI). This study was, therefore, designed to evaluate whether plasma sodium and hyponatremia (< 135 mM) are associated with mortality risk in patients with MI. In retrospective analyses using data from the Trandolapril Cardiac Evaluation (TRACE) study--a randomized, double-blind, placebo-controlled trial of trandolapril in 1749 patients with MI and left ventricular ejection fraction (LVEF) ≤ 35%--associations between plasma sodium or hyponatremia and more than 15-year mortality risk were evaluated in multivariate Cox proportional hazard models including traditional clinical confounders before and after additional adjustment for renal function, use of diuretics or both. During the extended follow-up time, 1462 patients died. Both hyponatremia [Hazard ratio: 1·30 (95% CI: 1·13-1·50), P < 0·001] and plasma sodium [Hazard Ratio(pro mM increase in P-Na): 0·98 (95% CI: 0·96-0·99), P = 0·004] were associated with mortality risk, and the adjusted parameter estimates were not affected by additional adjustment for renal function, use of diuretics or both. Hyponatremia and plasma concentrations of sodium are associated with long-term mortality risk in patients with MI complicated by left ventricular systolic dysfunction. Importantly, these associations are independent of renal function and use of diuretics. Whether the associations between plasma sodium or hyponatremia and long-term mortality risk reflect a causation or merely the severity of the underlying cardiac disease remains to be clarified.