The Preserved Thickness Ratio of the Femoral Head Contributes to the Collapse Predictor of Osteonecrosis.

objectivesThe collapse of femoral head is a serious symptom of osteonecrosis of the femoral head (ONFH), resulting in hip pain and deformity. However, it is hardly possible to reestablish the femoral head nonoperatively once the collapse happens. Predicting femoral head collapse is of great value for the prognosis of ONFH. This study aimed to develop a new method to quantify the preserved thickness of femoral head and to assess its diagnostic contribution in predicting femoral head collapse on plain radiographs.

methodsThis was a single-center retrospective study. A total of 101 hips (85 patients) with ARCO stage II from January 2008 to December 2016 were included in this study. The preserved thickness was measured on standard anteroposterior (AP) and frog-leg (FL) radiographs. The anteroposterior view's preserved thickness ratio (APTR) and the frog-leg view's preserved thickness ratio (FPTR) were calculated to show the preserved thickness ratio of the femoral head anteriorly and laterally. Univariate and multivariate logistic regression was performed to determine the risk factors for collapse. Sensitivity, specificity, and cut-off values for APTR and FPTR were determined by the receiver operating characteristic (ROC) curve analysis. Kaplan-Meier (K-M) analysis was applied to determine femoral head survival in ONFH patients.

resultsThe mean age of the 27 females and 58 males was 38.93 years old. The mean follow-up time was 74.62 (36-124) months in the non-collapse group and 18.66 (3-82) months in the collapse group. Femoral head collapse was observed in 62 hips during the follow-up period. Logistic regression analysis and ROC results showed that APTR <24.79% and FPTR <10.62% were significantly correlated with femoral head collapse. The Kaplan-Meier survival curve suggested that the overall survival rate of APTR ≥24.79% was 68.2% at 5 and 10 years and FPTR ≥10.62% was 71.63% at 5 and 10 years. At the last follow-up, 26 hips had collapse on the anterior side of the femoral head, 12 hips occurred on the lateral side, and 24 hips happened to collapse on both anterior and lateral sides.

conclusionFemoral head collapse predominantly occurred anteriorly rather than laterally in ONFH patients. The measurements of APTR and FPTR have noticeable implications for the prediction of femoral head collapse, and contribute to the selection of treatment options for ONFH patients with types B and C1 according to the JIC classification.