Aseptic femoral nonunion treated with exchange locked nailing with intramedullary augmentation cancellous bone graft

BACKGROUND: Closed reamed locked intramedullary nailing has been the treatment of choice for most of femoral shaft fractures. A high union rate with a low complication rate is generally predictable. For an aseptic femoral shaft nonunion with a prior inserted intramedullary nail, exchange nailing is one of favored surgical techniques for treatment. However, a greatly varied success rate of 72–100% has been reported. To improve the success rate of exchange femur nailing, a modified bone grafting technique was developed. The purpose of this retrospective study intended to evaluate outcomes of such a revised technique. METHODS: From July 2011 to March 2019, 48 consecutive adult patients (average, 38 years; range, 19–67 years) with aseptic femoral shaft nonunions after intramedullary nailing treatment were studied. All femoral shaft fractures were initially caused by traffic accidents, which were treated by a closed or open intramedullary nailing technique at various hospitals. The current revision treatment was performed after an average of 2.2 years (range 1.1–6.2 years) from initial injuries. In the surgery, the prior nail was removed and the marrow cavity was reamed widely (at least 2 mm as possible). Sufficient cancellous bone grafts harvested on the trochanteric marrow wall from the inside were placed in the marrow cavity of the junction of nonunion fragments. A new 1-mm smaller size locked intramedullary nail was inserted. Whether the dynamic or static mode of nails were used mainly depended on the nonunion level. Postoperatively, protected weight bearing with crutches was allowed for all patients. RESULTS: Forty-one patients were followed for an average of 2.8 years (85.4%; range, 1.9–4.5 years) and all fractures healed. The union rate was 100% (41/41, p < 0.001) with a union time of an average of 3.4 months (range, 2.5–5.0 months). There were no complications of deep infection, nonunions, malunions, implant failures or an avulsed trochanter tip fracture. The satisfactory knee function improved from 73.2% (30/41) preoperatively to 92.7% (38/41) at the latest follow-up (p = 0.019). CONCLUSIONS: The described modified bone grafting technique may effectively improve a union rate of exchange femur nailing while the surgical procedure is not complicated. It may therefore be used concomitantly in all aseptic femoral shaft nonunions when exchange nailing is performed.