Background and purposeTraumatic brain injury (TBI) with brain herniation predisposes to posttraumatic cerebral infarction (PTCI), which in turn seriously affects the prognosis of patients. At present, there is a lack of effective indicators that can accurately predict the occurrence of PTCI. We aimed to find possible risk factors for the development of PTCI by comparing the preoperative and postoperative clinical data of TBI patients with brain herniation.MethodsThe clinical data of 120 patients with craniocerebral trauma and brain herniation were retrospectively analyzed. Among them, 54 patients had cerebral infarction within 3–7 days after injury. The two groups of patients were compared through univariate and multivariate logistic regression analysis, and a classification tree model and a nomogram model were constructed. Finally, receiver operating characteristic curve analysis and decision curve analysis were conducted to analyze the clinical utility of the prediction model.ResultsLogistic regression analysis showed that factors like the Glasgow Coma Scale (GCS) score (P = 0.002), subarachnoid hemorrhage (SAH) (P = 0.005), aspiration pneumonia (P < 0.001), decompressive craniectomy (P < 0.05), intracranial pressure (ICP) monitoring (P = 0.006), the shock index (SI) (P < 0.001), the mean arterial pressure (MAP) (P = 0.005), and blood glucose (GLU) (P < 0.011) appeared to show a significant statistical correlation with the occurrence of infarction (P < 0.05), while age, sex, body temperature (T), D-dimer levels, and coagulation tests were not significantly correlated with PTCI after cerebral herniation. Combined with the above factors, Classification and Regression Tree was established, and the recognition accuracy rate reached 76.67%.ConclusionsGCS score at admission, no decompressive craniectomy, no ICP monitoring, combined SAH, combined aspiration pneumonia, SI, MAP, and high GLU were risk factors for infarction, of which SI was the primary predictor of PTCI in TBI with an area under the curve of 0.775 (95% CI = 0.689–0.861). Further large-scale studies are needed to confirm these results.