A three-dimensional measurement based on CT for the posterior tilt with ideal inter-and intra-observer reliability in non-displaced femoral neck fractures

Posterior tilt is associated with prognosis of non-displaced femoral neck fractures (FNFs). Knowledge of their association is critical and informs surgeons whether to choose internal fixation or arthroplasty in treatment of non-displaced FNFs. This study aimed to design a novel three-dimensional (3D) posterior tilt measurement and evaluate the intra- and inter-observer variability compared to two-dimensional (2D) measurement proposed by Palm. We hypothesized that 3D measurement would be more accurate and realistic with higher reliability. To test the hypothesis, three observers measured the posterior tilt on the radiographs of 50 non-displaced FNFs, twice with both methods. Intra- and inter-observer reliability for each measurement method used were determined. The measured angle was divided into two categories, at the cut-off of 20° for clinical practice simulation. Intra- and inter-observer reliability were identified for clinical effectiveness. The results indicated that inter- and intra-observer reliability for 3D measurement and its classification was almost perfect with an intraclass coefficient of 0.995 (0.994) and a kappa value of 0.927(0.947), respectively. Conversely, a substantial inter- and intra-observer reliability for the 2D measurement was obtained with an interclass coefficient of 0.764 as well as an intraclass coefficient of 0.773. The clinical validity for 2D measurement showed slight inter-reliability and moderate intra-reliability with a kappa value of 0.192 and 0.587, respectively. Hence, the novel 3D measurement appears to be more reliable with a strong inter- and intra-observer reliability measurement. Further clinical studies are needed to carry out to validate this hypothesis.