The Pauwels angle has been used widely, however an accurate evaluation of this angle is difficult because of deformity of the affected lower extremity. Therefore we designed a new measurement of the orientation of femoral neck fracture and applied this in a retrospective study to assess: (1) its reproducibility, (2) its advantages compared with the Pauwels angle, (3) its relationship with the short-term prognosis treated with three cannulated compression screws.

This new measurement is reproducible and has some reference meaning for the treatment of femoral neck fractures.

Two hundred and twenty-eight patients with femoral neck fractures treated with three cannulated compression screws were retrospectively analyzed. The VN angle, which was the angle between the fracture line and the vertical of the neck axis, and the Pauwels angle were measured respectively. The method of ICC was performed to assess the reproducibility of the two angles, and the absolute value of difference in pre-operative and post-operative radiographs was used to evaluate the uniformity of the two angles. These fractures were divided into four groups according to VN angle (VN<0° (n=92), 0°≤VN<10° (n=82), 10°≤VN<15° (n=26), VN≥15° (n=28)), and the short-term (within 6 months) fixation results of radiographs in these fractures were evaluated.

The ICC of the VN angle and the Pauwels angle in pre-operative radiographs were 0.937 (95% confidence interval (CI): 0.922–0.950) and 0.942 respectively (95% CI: 0.914–0.970), indicating both angles had a good inter-rater reproducibility. However, there was a great difference between the Pauwels angle in pre-operative and post-operative radiographs (P=0.037), the absolute difference was 10.66±6.47 (range: 1.72–38.48), while no statistical difference for the VN angle (P=0.084) and the absolute difference was 2.20±1.63 (range: 0.05–7.56). The overall fixation failure rate which was defined as screw loosening, varus collapse, obvious fracture displacement or femoral neck shortening was 11.84%, and the mean failure rates according to VN angles were respectively 0%, 3.24% (95% CI: 1.64–4.84), 22.69% (95% CI: 16.43–28.96), 65.45% (95% CI: 59.36–71.53). The mean failure rates of fractures according to post-operative Pauwels angle (<30°, 30–50°, >50°) were respectively 0%, 1.46% (95% CI: 1.42–1.50) and 36.24% (95% CI: 34.93–37.54).

The VN angle has a good inter-rater reproducibility, a higher reliability than the Pauwels angle and is closely related to the short-term prognosis of femoral neck fractures treated with cannulated compression screws.

Level IV, retrospective diagnostic study.