A Study on the Evaluation of a Risk Score of Osteonecrosis of the Femoral Head Based on Survival Analysis

Abstract

Background: This study aimed at developing a risk score predicting the probability of developing an osteonecrosis of the femoral head (ONFH) in patients with femoral neck fracture within 3 years after closed reduction internal fixation and evaluating its performance, clinical utility, and internal validity.

Methods: A retrospective follow-up study of 378 newly injured femoral neck fracture patients treated with 3 partially threaded parallel cannulated screws in 3 hospitals in Shanghai from March 2013 to January 2017 was performed. The patients were divided into development (n = 284) and validation (n = 94) cohorts. The risk score was constructed by Cox regression analysis in a form of nomogram. The performance and clinical utility were illustrated by box plots, calibration plots, and decision curve analysis.

Results: Eighty-three of 378 patients had developed ONFH within 3 years. Garden alignment index, time to surgery, preoperative displace, impaction, and postoperative malposition were used as predictors to construct the risk score in a form of nomogram. In the development and validation cohort, the concordance index was 0.96 and 0.94, respectively; the discrimination slope was 0.51 and 0.47, respectively. In both cohorts, the calibration slopes and intercepts were 1 and 0, respectively. The risk score was clinically useful between the risk threshold of 0% and 88%. The performance and utility in the validation data illustrated good repeatability.

Conclusion: The risk score had satisfactory discrimination and calibration performance and demonstrated clinical utility with good internal validity. It managed to distinguish high-risk groups for post-traumatic ONFH.

Keywords: clinical utility; femoral neck fracture; internal fixation; nomogram; post-traumatic osteonecrosis of femoral head; risk score.