Comminuted femoral-shaft fractures: treatment by roller traction, cerclage wires and an intramedullary nail, or an interlocking intramedullary nail.

In a retrospective study, we attempted to assess progress in the treatment of comminuted fractures of the femoral shaft at Parkland Memorial Hospital from 1978 to 1983. Seventy-nine comminuted femoral-shaft fractures were available for follow-up: thirty-two treated by roller traction, twenty-three treated by cerclage wires and an intramedullary nail, and twenty-four treated by an interlocking intramedullary nail. Using the classification of Winquist and Hansen, Grade-III and IV comminuted fractures accounted for 69 per cent of those treated by roller traction, 68 per cent of those treated by nailing and cerclage wires, and 96 per cent of those treated by an interlocking nail. The frequency of multiple injuries was 38 per cent in the patients treated by roller traction, 39 per cent in those treated by nailing and cerclage wires, and 58 per cent in those treated by an interlocking nail. The average hospitalization times were thirty-one days for roller traction, sixteen days for cerclage wires and an intramedullary nail, and 19.5 days for an interlocking nail. The average length of follow-up was 132 weeks after roller traction, 115 weeks after cerclage wiring and an intramedullary nail, and sixty weeks after insertion of an interlocking nail. All fractures were followed until after union; the average times to union were 18.4 weeks after roller traction, thirty-four weeks after open reduction and intramedullary nailing with cerclage wires, and 13.8 weeks after insertion of an interlocking nail. For the purposes of this study, treatment was assumed to have failed if a change in treatment was necessary, an unplanned reoperation was performed, femoral shortening exceeded 2.5 centimeters, angulation was more than 15 degrees, non-union or a deep infection developed, motion of the knee was less than 70 degrees of flexion, or a refracture occurred. By these criteria, the frequency of failure after roller traction was 66 per cent (secondary to malalignment and shortening); after insertion of an intramedullary nail with cerclage wires, 39 per cent (secondary to unplanned surgery, non-union, shortening, and infection); and after use of an interlocking nail, 4 per cent (secondary to shortening). Currently, at our institution, an interlocking intramedullary nail is the treatment of choice for comminuted femoral-shaft fractures because it encourages early union with maintenance of length and alignment and the results are predictable.