Purpose: The purpose of this study is to provide an additional tool to determine the stability of AO 31A2 pertrochanteric hip fractures. This study is based on the lateral hip radiograph, which has been ignored in the current debate over stability.

Method: One-hundred and thirty-one patients were identified through medical records with a diagnosis of pertrochanteric hip fracture treated with sliding hip screw from 2003–2008. Thirty-nine patients had AO 31A2 hip fractures, cross-table lateral injury films, intra-operative fluoroscopy and follow-up films. Only 23 had follow-up films beyond discharge. The landmarks of interest were angulation and translation between the femoral shaft and neck on cross-table lateral injury films. The neck was defined in three ways: the anterior cortex, two key points in the anterior cortex and the neck bisector. The most consistent measure was used. Translation of the neck was measured as a percentage of the shaft diameter. Measurements were taken by two blinded researchers with different levels of experience. Film sequence was randomized. The primary outcome was shortening of the sliding hip screw greater than one centimetre. This is the exact midpoint between 0.61 centimetres, which is not associated with reduced patient mobility, and 1.34 centimetres which is associated with reduced patient mobility, as described by Muller-Farber. The hip screw was measured from its tip to the point it enters the barrel. The diameter of the hip screw was known and provided scale. The measurement from intra-operative fluoroscopy films with the leg in traction, represented zero shortening.

Results: The average follow up was 190 days. Using the neck bisector to measure angulation was most consistent (95% of measurements available versus 89% and 88% with other methods). More than 30° angulation and/or 30% translation on the lateral predicted shortening greater than one centimetre with 91% specificity and 33% sensitivity. The average shortening in this group was 1.6 centimeters, which is greater than shortening associated with reduced patient mobility (1.34 centimeters). Agreement between two researchers was 91% and considered “substantial” (kappa 0.71) as per Landis and Koch criteria.

Conclusion: This is a highly specific and reproducible tool to detect a subset of AO 31A2 hip fractures which acquire unwanted collapse if treated with a sliding hip screw. This information adds clarity to the debate over stability of some AO 31A2 fracture cases, at no additional cost for the surgeon and facility. The “30/30 rule” (30° angulation and 30% translation) should not be used in isolation due to low sensitivity. Other factors may affect shortening, such as the degree of comminution and the antero-posterior film should still be considered.