Risk factors for mechanical failure of intertrochanteric fractures after fixation with proximal femoral nail antirotation (PFNA II): a study in a Southeast Asian population

Introduction Intertrochanteric hip fractures pose a significant health problem. The proximal femur nail anti-rotation (PFNA IIDePuySynthes) is the most commonly used implant for intramedullary nailing of unstable intertrochanteric fractures at our institution. We aim to identify the risk factors predisposing to mechanical failure of intertrochanteric hip fractures fixation with PFNAII in our Southeast Asian population. Materials and methods We retrospectively reviewed 295 consecutive patients who underwent PFNA fixation for intertrochanteric fractures of the proximal femur between January 2014 and June 2018 at our institution. Exclusion criteria included a follow-up period of less than 4 months and patients with polytrauma. 204 patients were eligible for analysis, of which 22 had mechanical failure of PFNAII. We compared these patients with respect to their demographics, medical co-morbidities, quality of reduction according to the Baumgartner scale, calcar restoration, blade position according to Cleveland Zones, the stability of fracture according to OTA/AO classification, neck-shaft-angle, tip-apex distance, as well as neck of femur bone mineral density T-score. Patients were followed up for a minimum of 4 months and until fracture union or complication had occurred. Logistic regression analysis was performed to determine the odds ratio for mechanical failure for selected variables. Results Good quality of reduction reduced risk of mechanical failure whilst posterior blade position in lateral hip X-ray predicted it, with odds ratios of 0.147 (95% confidence interval, 0.030-0.733; p = 0.019) and12.12 (95% confidence interval, 1.583-92.825; p = 0.016) respectively. On univariate analysis, the mechanical failure group were older, had poorer calcar restoration, more unstable fracture patterns, more varus neck-shaft angle, and trochanteric starting points that were lateral to or on the tip of the greater trochanter. However, these were not significant in multivariate analysis. Tip apex distance, the severity of osteoporosis, presence of diabetes, chronic kidney disease and serum vitamin D levels were not significant predictors of failure. Conclusions To avoid mechanical failure, one should aim to achieve a good quality of reduction and centre blade position on lateral hip X-ray. Tip apex distance did not predict mechanical failure in our study.