Abstract
We have previously studied the radiographic outcome of femoral neck fracture osteosynthesis with either two hook-pins or a four-flanged nail performed by a small group of surgeons with special interest in the methods. In 138 femoral neck fractures a backwards stepwise logistic regression analysis was used to study the significance of preoperative fracture-related factors, intraoperative factors and the osteosynthesis. The development of non-union/re displacement and segmental collapse of the femoral head was influenced by fracture displacement (P 0.001) and method of osteosynthesis (P 0.007). The postoperative scintimetric ratio was influenced by the method of osteosynthesis (P 0.0003), fracture displacement (P 0.004) and by the presence of a posterior fragment (P 0.03).
Reduction of the fracture and positioning of the osteosynthesis were to a large extent within the accepted limits. This may explain why the previously well documented negative effects of malpositioning of the osteosynthesis and inferior reduction were not demonstrated to influence the rate of healing-complications. We conclude that neither patient age, sex nor preoperative fracture variables, with the exception of the extent of fracture displacement, can be used to predict radiographic healing-complications in femoral neck fractures.