The efficacy of corticosteroid treatment in the prophylaxis of the fat embolism syndrome was evaluated in a prospective, randomized, double-blind study of high-risk patients with long-bone fractures. Using a set of objective diagnostic criteria, we saw a significant difference in the incidence of the syndrome between corticosteroid- (0 of 21) and placebo-treated patients (9 of 41) (p less than 0.05). There were no complications related to corticosteroid treatment. No routine laboratory or physical findings reliably predicted the development of the fat embolism syndrome except petechial rash, which occurred only in 5 placebo-treated patients who developed the syndrome. Complement activation was present in all patients studied who had the syndrome (5 of 27) but also in many patients who did not meet our diagnostic criteria, suggesting a multifactorial cause. These data support the prophylactic value of corticosteroid treatment in patients at high risk for the fat embolism syndrome, particularly if several unfavorable predictors are present.