Modified pauwels' intertrochanteric osteotomy in neglected femoral neck fractures in children : A report of 10 cases followed for a minimum of 5 years

Objectives: To evaluate the role of a modified Pauwels' intertrochanteric osteotomy (MPIO) in neglected femoral neck fractures in children. Design: Prospective study with retrospective analysis. Setting: Tertiary care Postgraduate Institute of Medical Sciences. Patients: Ten children (8 males, 2 females) with an average age of 10.2 years with neglected femoral neck fractures were seen from 1990 to 1998. A femoral neck fracture was considered neglected when no proper medical treatment was instituted for at least 1 month following the fracture. Nonunion was accompanied by coxa vara and resorption of the femoral neck in 9 patients; a 10th patient had a neglected femoral neck fracture for 1 month without coxa vara. Three patients at time of presentation with Delbet Type II displaced fractures with associated nonunion and coxa vara (2 with Ratliff Type III and 1 with Type I) also had avascular necrosis using plain radiographic criteria of increased density. Intervention: Modified Pauwels' intertrochanteric osteotomy. The children were immobilized in a hip spica for 6-10 weeks postoperatively and weightbearing was started after hip spica removal. Main Outcome Measurements: Fracture healing, neck-shaft angle, avascular necrosis, and functional outcome. Results: Patients were followed for an average of 8.2 years (range 5-12 years). All patients had union of their fracture within an average of 16.6 weeks (12-20 weeks) and of the osteotomy site within 8.2 weeks (7-9 weeks). Radiologic signs of avascular necrosis disappeared completely in the 3 patients who presented with avascular necrosis. In 1 patient with a preoperatively viable femoral head, radiologic signs of Ratliff Type I avascular necrosis appeared between 60 and 98 weeks. This radiologic finding became normal again, indicating viability of the femoral head somewhere between 98 to 205 weeks of follow-up. Postoperatively, an average of 135-degree neck-shaft angle was achieved (range 125-160 degrees). The average preoperative neck-shaft angle was 104.4 degrees (range 92-120 degrees) and on the normal hip side it was 127.7 degrees (range 124-132 degrees). Significant improvement in the neck-shaft angle was seen compared with the preoperative angle (P < 0.001) and normal hip angle (P < 0.05). Coxa vara and signs of chondrolysis were not observed in any of the patients. Premature proximal femoral epiphyseal closure resulting in a 1-cm and a 1.5-cm leg-length discrepancy was seen in 2 patients as compared with their normal side. A mild Trendelenburg gait was observed in 1 patient (10%). Using Ratliff's criteria, 9 patients (90%) were graded as a good result and 1 patient (10%) was graded as a fair result. The osteotomy plate was removed in 1 patient (10%). Conclusion: An MPIO creates a biomechanical environment conducive to healing of a neglected femoral neck nonunion in a child while simultaneously correcting an associated coxa vara. The procedure also seems to have a biological role in helping restore viability to a noncollapsed femoral head with avascular necrosis.