Prediction Model Of Zelan Mixture Combined With Sodium Hyaluronate Intra-articular Injection In The Treatment Of KOA

Objective: To explore the influencing factors of short-term efficacy of "Zelan mixture oral combined with intra-articular injection of sodium hyaluronate" in the treatment of knee osteoarthritis,and establish a prediction model to provide reference for the decision-making of non-surgical treatment plan for knee osteoarthritis patients.Methods:From November 2021 to December 2022,the patients who received "Zelan mixture oral combined with sodium hyaluronate intra-articular injection" for knee osteoarthritis in our hospital were collected,and the baseline information,clinical and imaging data of the patients were collected for follow-up.After 5 weeks of treatment,the patients were followed up by outpatient service or telephone at the 6th week to understand the improvement of the patients after treatment.According to the improvement of WOMAC score before and after treatment,the patients were divided into excellent group and poor group.The factors with statistical differences between the two groups were screened for logistic regression to build a prediction model.The model was validated,and the accuracy of the model was analyzed by Bootstrap internal validation method.The patient working curve(ROC curve),decision curve(AUC),Hosmer-Lemeshow test(HL test)were used to analyze the differentiation and calibration of the model,and DCA curve was used to evaluate the clinical applicability of the model.The Nomogram nomogram is made using R language,and the electronic calculator is developed using the winform framework of c # to display the prediction model.Results:Of the 122 cases finally included in the analysis,74(60.7%)were in the excellent effect group and 48(39.3%)in the poor effect group.Traditional univariate analysis showed that there were significant differences in BMI,course of disease,KL grade,Lysholm,age,and WOMAC function between the excellent group and the poor group(P<0.05);Univariate logistic regression analysis showed that there were significant differences in BMI,course of disease,KL grade,Lysholm,age,and WOMAC function partial scores between the excellent and poor effect groups(P<0.05).Multivariate logistic regression showed that BMI,course of disease,Lysholm,and WOMAC function were independent risk factors for efficacy(P<0.05).The prediction model was constructed: P=1/[1+exp Logit(P)],Logit(P)=-11.174+0.419 × BMI+0.187 × Course of disease-0.172 × Lysholm+0.315 ×WOMAC function.Model AIC value=104.292.The VIF value of variance expansion factor is less than 10,indicating that the four variables are independent of each other and there is no collinearity.In the validation of the model,the ROC curve shows that the AUC value of the area under the model curve is 0.889,(95% CI: 0.831-0.946),with good discrimination.Hosmer-Lemeshow test P=0.840 > 0.05,indicating that the prediction model has good calibration ability.DCA curve shows that with the increase of high risk probability threshold,the intervention curve is always higher than the reference line,and the model has good clinical applicability.Conclusions:1.BMI,course of disease,Lysholm and WOMAC function scores were independent risk factors for the efficacy of "Zelan mixture oral combined with sodium hyaluronate intra-articular injection" in the treatment of KOA.2.The prediction model has good discrimination,accuracy and clinical applicability.