Diabetes is a risk factor for increased mortality after a myocardial infarction. Whether this applies for patients with hyperglycemia during the acute phase of a acute myocardial infarction is unclear. Therefore we determined the relation between admission plasma glucose level and mortality in a prospectively collected series of 336 consecutive AMI patients. Patients were divided in four groups based on WHO criteria for glucose levels: I: <5.6 mmol/l, II: 5.6--8.3 mmol/l, III: 8.4--11.0 mmol/l, IV: 11.1 mmol/l. The average age was 68+/-11 years with a peak CK of 1378+/-160 U/l, 34% were anterior wall AMIs and 52% were treated with thrombolysis. All patients had a long-term follow-up control at an average of 14.2 months. One year mortality rate was 19.3% and rose to 44% in patients with glucose levels >11.1 mmol/l. The mortality was higher in diabetic patients than in non-diabetic patients (40 vs. 16%; P<0.05). Multivariate analysis revealed an independent effect of glucose level on mortality. In conclusions, our study in an unselected patient population demonstrates that admission plasma glucose level independently predicts 1 year mortality even in absence of diagnosed diabetes mellitus. Further studies evaluating the effect of acute insulin intervention in reducing mortality are warranted.