Predictive model for postoperative secondary acute cerebral infarction in patients with severe traumatic brain injury

Zhang Ruijuan; Cao Qian; Jin Tietuo; Yan Xiuyou; Dong Liang

doi:10.11850/j.issn.1009-122X.2024.07.002

Abstract：Objective To develop and validate a predictive model for postoperative secondary acute cerebral infarction in patients with severe traumatic brain injury(TBI).Methods Clinical data of 150 patients with severe TBI were analyzed retrospectively.Based on the occurrence of postoperative secondary acute cerebral infarction,patients were divided into an acute cerebral infarction group(n=52)and a control group(n=98).Clinical characteristics were compared between the two groups,and risk factors for postoperative secondary acute cerebral infarction in patients with severe TBI were analyzed.Based on the identified risk factors,a predictive model was constructed and internally validated.Results Statistically significant differences were observed between the two groups in terms of brain hernia,hypotension,diabetes,serum creatinine levels,and disseminated intravascular coagulation(DIC)scores(all P<0.05).Multivariate logistic regression analysis revealed that brain hernia,hypotension,diabetes,and a DIC score≥5 were independent risk factors for postoperative secondary acute cerebral infarction in patients with severe TBI(all P<0.05).Using R4.0.3 statistical software,the dataset was randomly divided into a training set(n=105)and a validation set(n=45).The area under the curve(AUC)was 0.790(95%CI:0.694-0.887)for the training set and 0.753(95%CI:0.599-0.907)for the validation set.The Hosmer-Lemeshow goodness-of-fit test for the model in the validation set yielded χ²=13.322,P=0.101,indicating good predictive value and reliability of the model.Conclusions The predictive model developed in this study has potential value in identifying high-risk patients with severe TBI who are prone to develop postoperative secondary acute cerebral infarction.