基于生物力学参数的髋关节置换术后翻修风险预测模型研究OA

Objective To construct a risk prediction model for revision after hip arthroplasty,providing a reference for early identification of high-risk revision populations in clinical practice.Methods Forty patients who underwent revision hip ar-throplasty in our hospital from June 2024 to June 2025 were selected as the study group,and 80 patients who were still using the prosthesis 5 years after surgery during the same period were selected as the control group.We collected general preoperative infor-mation,postoperative biomechanical parameters(imaging parameters:acetabular cup abduction angle,anteversion angle,etc.;func-tional parameters:hip joint flexion/abduction range of motion,etc.),and clinical indicators from two groups of patients.After data cleaning,we divided the modeling set and validation set into a 7∶3 ratio.LASSO regression was used to screen key predictive fac-tors,and a prediction model was constructed using a random forest model.The performance of the prediction model was evaluated based on AUC,sensitivity,specificity,accuracy,and the Hosmer-Lemeshow test.Subgroup analysis was also carried out.Results There was no statistically significant difference in preoperative general information between the two groups(P＞0.05);There were statistically significant differences(P＜0.05)in 9 indicators including Harris score,WOMAC score,joint range of motion flexion/abduction,and peak axial load after surgery.Subgroup analysis showed that there was no statistically significant difference(P＞0.05)in the number of revision cases with different BMI,gender,prosthesis fixation method,and surgical approach.LASSO regres-sion identified five key indicators:complications,dynamic stability center of gravity shift,peak axial load,joint range of motion ab-duction,and joint range of motion flexion.Hosmer-Lemeshow test x2=6.82 and P=0.559;The AUC,sensitivity,specificity,and accuracy of the random forest model on the validation set are 0.785,0.746,0.723,and 0.731,respectively.The Hosmer-Lemeshow test x2=7.53 and P=0.489,with joint range of motion flexion being the most important predictive feature of the model.Conclu-sion LASSO regression and random forest model based on biomechanical parameters can preliminarily predict the revision risk af-ter hip replacement surgery,but further verification is needed,which can provide reference for individualized clinical diagnosis and treatment.