The recent introduction of new diagnostic criteria for acute myocardial infarction (AMI), with troponin measurement, has increased the number of patients admitted with this diagnosis. To evaluate the epidemiologic and prognostic implications of the new diagnostic criteria for AMI. This was a retrospective study of 586 patients admitted for acute coronary syndrome (ACS) to the coronary care unit of our hospital, between 2002 and 2003. Data were collected from RECIMA, the Madeira Ischemic Heart Disease Registry. The population was analyzed following two different definitions of ACS: 1 - old criteria (Group I): AMI with ST elevation (typical symptoms or ECG with ST-segment elevation and raised CK-MB >2x), AMI without ST elevation (typical symptoms or ECG without ST elevation and raised CK-MB >2x) and unstable angina (UA) (symptoms or ECG indicative of ischemia, with normal CK-MB, regardless of troponin status); 2 - new criteria (Group II): AMI with ST elevation (typical symptoms or ECG with segment ST elevation and raised CK-MB >2x or troponin), AMI without ST elevation (typical symptoms or ECG without ST-segment elevation and raised CK-MB >2x or troponin) and UA (symptoms or ECG indicative of ischemia, with normal enzymes). We evaluated whether this change in criteria had any influence on in-hospital mortality. The new criteria significantly (by 11.9 %) increased the total number of patients admitted with AMI. This was due to an increase in AMI without ST elevation (p < 0.001) and a decrease in patients with UA (p < 0.001), with no changes in AMI with ST elevation. In-hospital mortality was lower in patients with AMI diagnosed by the new criteria and in those with UA. The overall increase in AMI resulting from the new diagnostic classification was accompanied by a decrease, although not statistically significant, of in-hospital mortality, probably due to the lower risk of the population analyzed.