BACKGROUND The effect of recent advances in critical care and the emphasis on early fracture fixation in patients with fat embolism syndrome (FES) are unknown. OBJECTIVE To better define FES in current practice by conducting a 10-year review of the experiences at our level I trauma center. METHODS The medical records of all patients in whom FES was diagnosed from July 1, 1985, to July 1, 1995, were reviewed for demographics, injury severity and pattern, diagnostic criteria, and management. METHODS A level I trauma center. RESULTS Twenty-seven patients with clinically apparent FES were identified. This resulted in an incidence of 0.9% of all patients with long-bone fractures. The mean injury severity score was 9.5 (range, 4-22). The diagnosis of FES was made by clinical criteria, including hypoxia, 26 patients (96%); mental status changes, 16 patients (59%); petechiae, 9 patients (33%); temperature higher than 39 degrees C, 19 patients (70%); tachycardia (heart rate > 120 beats per minute), 25 patients (93%); thrombocytopenia (platelet count < 150 x 10(9)/L), 10 patients (37%); and unexplained anemia, 18 patients (67%). Thirteen patients (48%) had multiple long-bone fractures, and 14 patients (52%) had a single long-bone fracture. Seven patients (26%) had open fractures, 15 (56%) had closed fractures, and the remaining 5 (18%) had both. Of the total fracture population, the distribution was 81% closed, 15% open, and 4% both. Management included ventilatory support for 12 (44%) of the patients; early operative fixation was emphasized, and 74% of the fractures were stabilized within 24 hours of injury. This was comparable with 76% of the total fracture population. There were 2 deaths, for a mortality of 7%. CONCLUSIONS (1) Fat embolism syndrome remains a diagnosis of exclusion and is based on clinical criteria. (2) Clinically apparent FES is unusual but may be masked by associated injuries in more severely injured patients. (3) No association could be identified between FES and a specific fracture pattern or location. (4) Early intramedullary fixation does not increase the incidence or severity of FES. (5) While FES seems to have a direct effect on survival, the management of FES remains primarily supportive.