Hyponatremia in the early phase of acute myocardial infarction (AMI) is a well-known predictor of poor prognosis. However, little is known about the clinical implication of sodium levels at discharge in hospital survivors after AMI. The study included 1290 consecutive patients (64 ± 12 years; 877 men) who survived the index hospitalization after AMI. We determined the 12-month mortality rates of these patients. Patients who died during the 12-month follow-up had lower sodium levels at discharge than those who had survived (137 ± 6 vs. 139 ± 4 mmol/L; P < 0.014). Hyponatremia at discharge, defined as a serum sodium level ≤135 mmol/L, was present in 210 patients (16.3 %). In the Cox-proportional hazard model, hyponatremia at discharge (hazard ratio, 2.264; 95 % confidence interval, 1.119–4.579; P = 0.023) was an independent predictor of 12-month mortality. Moreover, hyponatremia at discharge had an incremental prognostic value over conventional risk factors (χ ² = 7, P = 0.007), and conventional risk factors and log N-terminal Pro-B-type natriuretic peptide combined (χ ² = 5, P = 0.021). In the subgroup analysis, the 12-month mortality of patients with hyponatremia at discharge was significantly higher than in those without, irrespective of age, Killip class, left ventricular ejection fraction, percutaneous coronary intervention at index hospitalization, and prescription of diuretics at discharge. Hyponatremia at discharge is an independent predictor of 12-month mortality in hospital survivors after AMI.