A Criterion of Reduction Quality of Femoral Neck Fractures Based on Spatial Residual Displacements of Femoral Head.

objectiveCurrently, quality criteria for reduction after femoral neck fractures such as the Garden index are mainly based on two-dimensional x-rays. Research shows that current reduction quality criteria are no longer sufficient to meet the needs to reduce the incidence of osteonecrosis of the femoral head (ONFH). The purpose of this study is to construct a reduction quality criterion based on spatial residual displacements of 3D reconstruction and to predict the occurrence of ONFH.

methodsThe subjects were from the Tianjin Hospital Hip Fracture cohort, all of whom experienced femoral neck fractures and underwent reduction and internal fixation at Tianjin Hospital and received 12 months of follow-up minimally postoperatively. CT scans were used for 3D reconstruction, and six spatial displacement indicators were measured. Risk factors of ONFH were identified using logistic regression, and a reduction quality criterion based on spatial residual displacements was constructed. Multivariate logistic regression models were performed to estimate the effect of reduction quality by the new criterion on ONFH.

resultsData from 391 patients were included in the final analysis. Preoperative displacement of the center of the femoral head (CFH), postoperative residual displacement of the CFH, and residual rotational displacement were proved to be independent risk factors for ONFH, with OR values of 3.83 (95% CI: 1.98, 7.48), 2.62 (95% CI: 1.05, 6.49), and 5.36 (95% CI: 2.46, 11.64), respectively. The new criterion was composed of two indicators and had three grades: when residual displacement of CFH is ≤ 4.075 mm, and residual rotational displacement of femoral head is ≤ 18.595°, the quality is excellent; when residual displacement of CFH is ≤ 4.075 mm and residual rotational displacement of femoral head is > 18.595° or when residual displacement of CFH is > 4.075 mm and residual rotational displacement of femoral head is ≤ 18.595°, the quality is moderate; when residual displacement of CFH is > 4.075 mm and residual rotational displacement is > 18.595°, the quality is poor. The risk of ONFH would increase 3.99 times (95% CI: 2.35, 6.78) for each lower grade in reduction quality. The logistic regression model was well discriminated with an AUC area of 0.802 and had good calibration with a p-value of > 0.05 by the Hosmer-Lemeshow test.

conclusionA new reduction quality criterion for femoral neck fracture based on CT was constructed, and reduction quality before fixation by the new criterion was proved to be an independent predictive factor for ONFH. The logistic regression model had quite good discrimination and calibration for postoperative ONFH.