The influence of the hips position on measurements of posterior tilt in a valgus-impacted femoral neck fracture.

INTRODUCTION: Lateral radiographs are important for the evaluation of Garden I and II femoral neck fractures. These fractures appear undisplaced in the anteroposterior view, but posterior tilt of the femoral head may still be present in the lateral view. The influence of posterior tilt is, however, debated, which could be caused by the use of non-standardized cross-table radiographs in the conflicting reports. The aim of this bone-model study was therefore to evaluate the influence of the hips position on measurements of posterior tilt.

MATERIALS AND METHODS: We generated models of a Garden I-II femoral neck fracture and the non-injured contralateral femur from CT reconstructions with a 3D-printer. Lateral radiographs of the models were obtained in positions ranging from 90° internal rotation (IR) to 90° external rotation (ER) and 0-80° flexion. Two investigators independently measured posterior tilt on 27 lateral radiographs obtained from each model.

RESULTS: The inter-tester ICC was 0.91 (0.84-0.94), standard error of measurement (SEM) 2.6, and minimal detectable change (MDC) 7.2. The median (range) posterior tilt for the fracture model was 21.9° (5.0, 33.8) and 23.6° (2.2, 28.6) for observer 1 and observer 2, respectively. The corresponding posterior tilt within the range of 10° IR to 40° ER and 0 to 30° flexion of the fracture model, was 27.3° (24.0, 33.8) and 26.3° (24.8, 28.6).

DISCUSSION: The range of posterior tilt measurements for positions of the proximal femur restricted from 10° IR to 40° ER and 0 to 30° flexion, was above the MDC for observer 1, and below the MDC for observer 2. These findings indicate that rotation and flexion affect measurements of posterior tilt, but the influence may be negligible for positions of the injured extremity that are clinically relevant during cross-table lateral radiographs. A larger study that accounts for variations in anatomy and fracture displacement is required to confirm these findings.