Construction and verification of risk prediction nomogram model for cerebral hernia formation in patients with craniocerebral injury
[Objective]To explore the risk factors of cerebral hernia formation in patients with craniocerebral injury,and to construct a risk prediction nomogram model.[Methods]The clinical data of 213 patients with craniocerebral injury admitted to our hospital from January 2019 to January 2023 were retrospectively analyzed.According to the ratio of 2:1,they were randomly divided into modeling group(n=142)and verification group(n=71).According to the presence or absence of brain herniation in patients,the modeling group was divided into cerebral hernia group(n=49)and non-cerebral hernia group(n=93).Multivariate logistic regression analysis was used to explore the risk factors of cerebral hernia formation in patients with craniocerebral injury,and a risk prediction nomogram was established based on this.Bootstrap method was used for internal verification.The receiver operating characteristic(ROC)curve was used to analyze the efficacy of nomogram in predicting the formation of cerebral hernia in patients with craniocerebral injury.The decision curve analysis(DCA)was used to verify the clinical net benefit rate of the model.[Results]The proportions of age ≥60 years old,history of hypertension,smoking,open brain injury,subarachnoid hemorrhage,cerebral contusion and laceration,hydrocephalus and hypotension at admission in the cerebral hernia group were higher than those in the non-cerebral hernia group(P＜0.05),and the Glasgow Coma Scale(GCS)score at admission,blood sodium concentration ≥135 mmol/L at admission and the proportion of decompressive craniectomy treatment were lower than those in the non-cerebral hernia group(P＜0.05).Multivariate logistic regression analysis showed that age ≥60 years,smoking history,open craniocerebral injury,subarachnoid hemorrhage,hydrocephalus and hypotension at admission were risk factors for cerebral hernia formation in patients with craniocerebral injury(P＜0.05),and GCS score at admission,serum sodium concentration at admission ≥135 mmol/L and decompressive craniectomy were protective factors for cerebral hernia formation in patients with craniocerebral injury(P＜0.05).The nomogram model constructed based on the above influencing factors was verified by Bootstrap method in the modeling group and the verification group,and the consistency indexes were 0.834 and 0.825 respectively,and the calibration curve and ideal curve fitting were good.The ROC curve results showed that the area under the curve(AUC),sensitivity and specificity of the modeling group nomogram for predicting the formation of cerebral hernia in patients with head injury were 0.864,83.67％and 89.25％respectively,while the validation group were 0.850,82.61％and 87.50％respectively.The DCA curve showed that when the patient's threshold probability was 0.01 to 0.90,the nomogram model could obtain satisfactory net returns for risk assessment.[Conclusion]Age ≥60 years old,smoking history,open craniocerebral injury,subarachnoid hemorrhage,hydrocephalus,hypotension at admission,GCS score at admission,blood sodium concentration at admission ≥135 mmol/L and decompressive craniectomy are all influencing factors for the formation of cerebral hernia in patients with craniocerebral injury.The nomogram prediction model constructed based on the above factors has good performance,which is helpful to timely screen high-risk patients with cerebral hernia formation after craniocerebral injury in clinical practice.
craniocerebral injurycerebral herniarisk factorsnomogram model
万方数据
维普
