Hip Preservation or Total Hip Arthroplasty? A Retrospective Case–Control Study of Factors Influencing Arthroplasty Decision‐Making for Patients with Osteonecrosis of the Femoral Head in China

OBJECTIVE: At present, there is no consensus or guidance on indications for osteonecrosis of the femoral head (ONFH) patients to receive hip arthroplasty (THA) treatment. This study aims to explore the factors that influence the decision‐making for THA in patients with ONFH, and to provide references for clinical decision for ONFH patients to be indicated for THA or hip preservation. METHODS: This retrospective case–control study involved data for ONFH patients from July 2016 to October 2021 from the China Osteonecrosis of the Femoral Head Database (CONFHD). The patients with ONFH, and unilateral hip affected at the first visit were divided into THA group and non‐THA group according to if they had undergone THA treatment. The differences between the two groups of patients in terms of gender, age at the time of consultation, body mass index (BMI), etiology, onset side, association research circulation osseous (ARCO) stage, hip joint function, visual analog scale (VAS), etc. were analyzed. Multivariate binomial logistic regression analysis was then applied to evaluate the risk factors of ONFH patients who underwent THA during the first visit. RESULTS: A total of 640 patients were recruited for analysis, including 209 cases from the THA group and 431 cases from the non‐THA group. The results of univariate analysis showed that the two groups of patients were significantly different in the following six indicators: age (59 vs. 46, Z = −9.58, p < 0.001), duration of disease (78 vs. 17, Z = –16.14, p < 0.001), gender composition (χ(2) = 8.09, p = 0.004), disease etiology (χ(2) = 33.04, p < 0.001), ARCO stage (χ(2) = 334.86, p < 0.001), flexion of hip joint (χ(2) = 172.33, p < 0.001). However, the comparison between the two groups on VAS (Z = –0.82, p = 0.41), BMI (Z = –1.35, p = 0.18), and onset side (χ(2) = 1.53, p = 0.22) did not obviously differ. The results regression analysis showed that the age at the time of consultation, duration of disease, ARCO stage, and the hip joint function affected the decision making if the patients should undergo THA. The results of receiver operating characteristic curve (ROC) analysis showed that aforementioned indicators were satisfactory in predicting whether patients with ONFH would be treated with THA. The regression model using the above four indicators as comprehensive indicators has satisfactory performance in predicting whether to perform THA, and the area under the curve (AUC) is 93.94%. CONCLUSION: These factors such as age, duration of disease, ARCO stage, and hip flexion function should be considered comprehensively before making decisions to perform THA or not in our clinical practice.