Patients with diabetes mellitus have a high morbidity and mortality from acute myocardial infarction, the reason for which is not fully understood. The relationship between congestive heart failure symptoms, left ventricular ejection fraction, and long-term mortality was examined in 578 hospital survivors of acute myocardial infarction, 47 of whom had Type 2 (non-insulin-dependent) diabetes mellitus. None of the patients were treated with insulin. The prevalence of congestive heart failure during hospitalization was similar in patients with and without diabetes, although mean diuretic dose was higher in the former patients. Left and right ventricular ejection fraction was measured with radionuclide ventriculography in the second week after acute myocardial infarction. At discharge from the coronary care unit, patients with and without diabetes had similar left ventricular ejection fraction (with diabetes: median 46% vs without diabetes: median 43%; p = 0.89). Median right ventricular ejection fraction (62%) was within normal limits in both groups and did not differ statistically. Survival data were obtained for all patients. The 5-year mortality was increased in patients with diabetes compared with non-diabetic patients independent of left ventricular ejection fraction. Univariate analysis showed that the cumulative 5-year mortality rate was 53% in the group with diabetes compared with 43% in the non-diabetic group (p = 0.007). Using multivariate regression analysis presence of diabetes was found to have a significant association with long-term mortality after myocardial infarction, that was independent of age, history of hypertension, congestive heart failure symptoms during hospitalization or of either left or right ventricular ejection fractions at discharge. We conclude that the excess mortality in patients with non-insulin-dependent diabetes mellitus is not explained by available risk markers after myocardial infarction. Even though left ventricular ejection fraction and serum creatinine did not differ significantly, the apparent higher dose of Frusemide in patients with than without non-insulin-dependent diabetes mellitus might indicate that heart failure, it present, is more severe in patients with than in those without diabetes. The importance of diastolic dysfunction in this context needs to be determined.