Aims/hypothesis This post hoc analysis from the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) 2 trial reports on extended long-term outcome in relation to glucose-lowering agents in patients with myocardial infarction and type 2 diabetes. Methods Patients were randomised as follows: group 1, insulin-based treatment; group 2, insulin during hospitalisation followed by conventional glucose control; and group 3, conventional treatment. Treatment according to the above protocol lasted 2.1 years. Using the total DIGAMI 2 cohort as an epidemiological database, this study presents mortality rates in the randomised groups, and mortality and morbidity rates by glucose-lowering treatment during an extended period of follow-up (median 4.1 and max 8.1 years). Results Follow-up data were available in 1,145 of the 1,253 patients. The mortality rate was 31% (72% cardiovascular) without significant differences between treatment groups. The total number of fatal malignancies was 37, with a trend towards a higher risk in group 1. The HR for death from malignant disease, compared with group 2, was 1.77 (95% CI 0.87-3.61; p=0.11) and 3.60 (95% CI 1.24-10.50; p=0.02) compared with group 3. Insulin treatment was associated with non-fatal cardiovascular events (OR 1.89 95% CI 1.35-2.63; p=0.0002), but not with mortality (OR 1.30, 95% CI 0.93-1.81; p=0.13). Metformin was associated with a lower mortality rate (HR 0.65, 95% CI 0.47-0.90; p=0.01) and a lower risk of death from malignancies (HR 0.25, 95% CI 0.08-0.83; p=0.02). Conclusions/interpretation Patients with type 2 diabetes and myocardial infarction have a poor prognosis. Glucoselowering drugs appear to be of prognostic importance. Insulin may be associated with an increased risk of nonfatal cardiac events, while metformin seems to be protective against risk of death.