慢性化脓性中耳炎手术预后的预测模型构建及初步评价OA

Objective To report a prediction model for surgical prognosis in chronic suppurative otitis media(CSOM)and its clinical application value.Methods Univariate and multivariate logistic regression analyses were employed to identify factors affecting surgical prognosis in CSOM,and a nomogram model was constructed based using identified independent predictors,based on data from 236 CSOM patients admitted to a specific treatment group at the Chinese PLA General Hospital from January 2024 to August 2025.The cases were allocated to a training set(n=141)and a prospective validation set(n=95).Using the training set,variables were first screened via univariate analysis and then incorporated into multivariate regression to identify independent predictors for nomogram construction.The model's discrimination and calibration were validated using the receiver operating characteristic(ROC)curve,calibration curve,Hosmer-Lemeshow test,and clinical decision curve analysis,with validation performed on the validation set.Candidate predictors were determined through literature review and by incorporating core elements from the middle ear risk index(MERI)and the ossiculoplasty outcome parameter staging index(OOPSI).These included history of smoking,preoperative dry ear status,contralateral ear pathology,tympanic membrane perforation size,tympanic mucosal lesions,presence of granulation tissue or effusion,Eustachian tube function,ossicular chain status,and revision surgery.Surgical success was defined as an intact tympanic membrane postoperatively with an air-bone gap≤20 dB HL.Results Multivariate analysis identified preoperative dry ear status,Eustachian tube function and ossicular chain status as independent predictors of surgical outcomes(P<0.05).The area under curve of the nomogram model was 0.856(95%CI:0.792～0.919)in the training set and 0.859(95%CI:0.782～0.936)in the validation set.The calibration curve showed high consistency between predicted and observed probabilities,and the Hosmer-Lemeshow test indicated good model fit(training set P=0.90,validation set P=0.25).Conclusions This study successfully developed and validated a nomogram model based on three key preoperative indicators.The model can effectively predict postoperative hearing outcomes in CSOM patients,providing an intuitive quantitative tool for clinical identification of high-risk patients,formulation of personalized treatment plans,and preoperative communication.