The records of 108 patients admitted to a medical intensive care unit (ICU) for syncope during a two-year period were reviewed. Explicit criteria were used to classify patients by presumed etiologic diagnosis. Thirty-six percent of the cases of syncope were due to cardiovascular disease, 17% were due to noncardiovascular disease, and 47% were unexplained at hospital discharges. Seventy-two percent of presumed etiologic diagnoses were based on information available at the time of patient admission. The remainder were based on ICU monitoring and additional diagnostic tests. Patients were prospectively studied after hospital discharge. The one-year mortality was 19% in the cardiovascular group, 6% in the noncardiovascular group, and 6% among patients whose syncope remained unexplained. Age-standardized comparisons between the unexplained syncope group, the US population, and other ICU patients suggest that patients with syncope unexplained at hospital discharge do not have an increased risk of death during the subsequent year.