BACKGROUND: From 1989 to 1997, 1507 fractures of the shaft of the femur were treated with intramedullary nailing at The R Adams Cowley Shock Trauma Center. Fifty-nine (4 percent) of those fractures were treated with early external fixation followed by planned conversion to intramedullary nail fixation. This two-stage stabilization protocol was selected for patients who were critically ill and poor candidates for an immediate intramedullary procedure or who required expedient femoral fixation followed by repair of an ipsilateral vascular injury. The purpose of the current investigation was to determine whether this protocol is an appropriate alternative for the management of fractures of the femur in patients who are poor candidates for immediate intramedullary nailing. METHODS: Fifty-four multiply injured patients with a total of fifty-nine fractures of the shaft of the femur treated with external fixation followed by planned conversion to intramedullary nail fixation were evaluated in a retrospective review to gather demographic, injury, management, and fracture-healing data for analysis. RESULTS: The average Injury Severity Score for the fifty-four patients was 29 (range, 13 to 43); the average Glasgow Coma Scale score was 11 (range, 3 to 15). Most patients (forty-four) had additional orthopaedic injuries (average, three; range, zero to eight), and associated injuries such as severe brain injury, solid-organ rupture, chest trauma, and aortic tears were common. Forty fractures were closed, and nineteen fractures were open. According to the system of Gustilo and Anderson, three of the open fractures were type II, eight were type IIIA, and eight were type IIIC. Intramedullary nailing was delayed secondary to medical instability in forty-six patients and secondary to vascular injury in eight. All fractures of the shaft of the femur were stabilized with a unilateral external fixator within the first twenty-four hours after the injury; the average duration of the procedure was thirty minutes. The duration of external fixation averaged seven days (range, one to forty-nine days) before the fixation with the static interlocked intramedullary nail. Forty-nine of the nailing procedures were antegrade, and ten were retrograde. For fifty-five of the fifty-nine fractures, the external fixation was converted to intramedullary nail fixation in a one-stage procedure. The other four fractures were associated with draining pin sites, and skeletal traction to allow pin-site healing was used for an average of ten days (range, eight to fifteen days) after fixator removal and before intramedullary nailing. Follow-up averaged twelve months (range, six to eighty-seven months). Of the fifty-eight fractures available for follow-up until union, fifty-six (97 percent) healed within six months. There were three major complications: one patient died from a pulmonary embolism before union, one patient had a refractory infected nonunion, and one patient had a nonunion with nail failure, which was successfully treated with retrograde exchange nailing. The infection rate was 1.7 percent. Four other patients required a minor reoperation: two were managed with manipulation under anesthesia because of knee stiffness, and two underwent derotation and relocking of the nail because of rotational malalignment. The rate of unplanned reoperations was 11 percent. The average range of motion of the knee was 107 degrees (range, 60 to 140 degrees). CONCLUSIONS: We concluded that immediate external fixation followed by early closed intramedullary nailing is a safe treatment method for fractures of the shaft of the femur in selected multiply injured patients.