In a prospective randomized trial of prophylactic therapy in fat embolism syndrome (FES) 64 patients with femoral and/or tibial shaft fractures uncomplicated by other significant injuries were treated with hypertonic glucose (A), methylprednisolone (B), and placebo (C). Clinical findings and several laboratory parameters were used to establish the diagnosis. Two of the 23 patients in the placebo group (C), 3 of the 21 patients in the hypertonic dextrose (A) group and none of the 20 patients in the methylprednisolone (B) group developed clinical fat embolism syndrome. All patients with fat embolism syndrome were hypoxemic (pO2 less than 65 mm Hg) and exhibited central nervous system symptoms. Four out of 5 were hypocalcemic and 3 patients had thrombocytopenia. There was a high incidence of subclinical hypoxemia or subclinical fat embolism syndrome (29% in A Group, 15% in B Group and 39% in C Group). A comparison between groups of the mean pO2 for each patient demonstrated a statistically significant difference between the methylprednisolone group and the control group (p less than 0.025) as was the comparison of proportions of pO2 less than 70 mm Hg (methylprednisolone versus control p less than 0.01, glucose versus control p less than 0.03). Methylprednisolone given prophylactically may reduce the incidence of fat embolism syndrome and can reduce the degree of hypoxemia associated with long bone fractures of the lower extremmity.