OBJECTIVES: To provide valid clinical data of early in-hospital deaths with presumed acute myocardial infarction (AMI) who are often not included in clinical trials or registries. METHODS: From August 2004 to August 2006 all patients (age 25-84 years) dying within 24 h after hospitalization in a large tertiary care academic teaching hospital were screened regarding an underlying cardiovascular cause of death. RESULTS: After validation, 79 out of 1,352 patients remained with a final diagnosis of AMI. Sixty-six percent of these experienced prehospital cardiac arrest or shock. In 37% no resuscitation attempts were performed in-hospital, the most common reason being multimorbidity. Only 23% could be transferred to coronary angiography for revascularisation attempts. An independent panel of clinicians judged that only in one patient would another management strategy have been promising. Of interest, 33% of the deceased patients had typical or atypical chest pain the days before the lethal event. CONCLUSION: A large percentage of AMI patients who died soon after hospitalization were in critical circulatory state directly before hospitalization. In 37%, in-hospital resuscitation attempts were omitted for understandable reasons. Options for improvement in acute care in the investigated setting were not found. However, in one third of the cases earlier preventive measures might have been reasonable.