创伤性急性硬膜下血肿患者神经功能预后模型的构建与验证OA

Objective To analyze the influencing factors of neurological prognosis in patients with traumatic acute subdural hematoma(ASDH),and to construct and validate a prognostic prediction model.Methods The clinical data of 351 ASDH patients hospitalized at the Affiliated Lianyungang Hospital of Xuzhou Medical University from March 2021 to March 2023 were retrospectively analyzed.The collected patient data were divided into training and validation sets in a 7∶3 ratio.The training set was used for predictive model construction,and the validation set was used for model validation.Multivariate Logistic regression analysis was used to screen for influencing factors on the neurological prognosis of ASDH patients.Based on these factors,a nomogram model was constructed to predict the probability of a good neurological prognosis in ASDH patients.The model efficacy was evaluated using receiver operating characteristic(ROC)curves,Hosmer-Lemeshow tests,and decision curve analysis(DCA),and validated on the validation set.Results Multivariate analysis revealed that age,Glasgow coma scale(GCS)score,hemoglobin,long bone fracture,subarachnoid hemorrhage,initial cranial CT hematoma volume,craniotomy treatment,and progressive hemorrhagic injury were identified as independent predictors in this study(P<0.05).The predictive model constructed from the training set yielded an area under the ROC curve(AUC)of 0.93,with sensitivity,specificity,and accuracy of 0.91,0.81,and 0.88,respectively.On the validation set,the model achieved an AUC of 0.90,with sensitivity,specificity,and accuracy of 0.89,0.78,and 0.85,respectively.Calibration curves demonstrated good model fit,and decision curve analysis indicated high clinical utility within the threshold probability range of 0.20～0.80.Conclusions Age,GCS score,hemoglobin level,initial cranial CT hematoma volume,subarachnoid hemorrhage,long bone fracture,surgical intervention,and progressive hemorrhagic injury are independent predictors of neurological outcome in patients with ASDH.A nomogram model constructed based on these predictors demonstrates good predictive performance.