BACKGROUND: Among patients hospitalized with a suspected acute coronary syndrome, a minority will eventually develop a confirmed acute myocardial infarction (AMI). In the remaining patients, coronary artery disease is the underlying cause in a large proportion. HYPOTHESIS: The aim of the study was to determine risk indicators for death and the mode of death during 5 years of follow-up among patients hospitalized and surviving hospitalization, who presented with initially suspected AMI, but in whom infarction was not confirmed. METHODS: Consecutive patients who fulfilled the above criteria and were discharged from Sahlgrenska Hospital alive during 1986 and 1987 were followed for 5 years. RESULTS: In all, 1,227 patients, of whom 396 (34%) died during the 5 years of follow-up, fulfilled the criteria. The following factors appeared to be independent risk indicators for death: age (p < 0.001); male gender (p < 0.001); a history of either current smoking (p < 0.001), congestive heart failure (p < 0.01), or myocardial infarction (p < 0.05); congestive heart failure during hospital stay (p < 0.01); and prescription of digitalis at discharge (p < 0.05). Among patients who died, only 63% were judged to have been dying a cardiac death. CONCLUSION: Among patients hospitalized with suspected acute coronary syndrome and discharged from hospital without a confirmed AMI, one third had died during the 5 years of follow-up. Risk indicators for death were related to age, male gender, history of current smoking, congestive heart failure or previous AMI, congestive heart failure in hospital, and digitalis medication at discharge.