Purpose of the study: The constantly increasing incidence of extracapsular fractures of the proximal femur are a public health concern. The basic therapeutic options are screw-plate fixation and proximal reconstruction with nails. The purpose of this retrospective study was to assess the mid-term results with a new osteosynthesis material, the proximal femoral nail antirotation (PFN-A^®) which has a spiral blade.

Material and methods: One hundred eight 108 PNF-A^® performed from January 2007 to July 2008 were included in a retrospective clinical and radiographic study. These series included exclusively extracapsular fractures of the proximal femur in subjects aged over 70 years. All patients were assessed with the Parker score pre- and postoperatively. Blood loos, position of the spiral blade on the AP and laterals views and operative time were analysed. We searched for complications (femoral head slide, blade protrusion, head rotation, non-union, fracture on material, and operative site infection). We searched for risk factors.

Results: One hundred eight patients (94% ASA 2 or 3) were reviewed at mean 5.3 months (±1.5). None of the patients were lost to follow-up. At revision, 19 patients had died (17.6%). The mean Parker score declined 1.4 points. All fractures healed at mean 10.4 weeks (±0.6). Six complications were noted: three operative site infections, three head slidings, one intraacetabular protrusion. No statistically significant could be identified. Nevertheless, the three cases with femoral head sliding occurred on fractures that were unstable (type 31-A2) which had a malpositioned blade.

Discussion: There appears to be a consensus on the treatment of proximal fractures of the femur: screw-plate fixation for stable fractures, centromedullary nailing for the others. Arthroplasty is a second-line solution. There are few publications on the new spiral blade of the PFN-A^®. This method spares bone stock and allows compaction of the cancellous bone, particularly adapted for osteoporotic bone: the efficacy is comparable with reference techniques with lower rates of sliding (2.%) and acetabular protrusion (< 1%).