重型颅脑损伤患者死亡风险预测模型建立

Establishment of death risk prediction model for patients with severe craniocerebral injury

目的构建重型颅脑损伤死亡预测模型,为早期判断预后提供可行依据。方法回顾性分析2012年1月至2017年12月广西医科大学第一附属医院收治的共计190例重型颅脑创伤患者的临床资料。收集入院时相关危险因素并记录出院时的生存情况,Logistic回归建立死亡预测模型,计算ROC曲线下面积并通过拟合优度检验预测模型性能。结果190例重度颅脑损伤患者住院期间有91例死亡,发生率为40%。单因素Logistic回归分析显示,年龄、气管插管、瞳孔大小、枕叶损伤、脑疝、蛛网膜下腔出血、格拉斯哥昏迷评分、体温、血小板、血肌酐、APACHEⅡ为重型颅脑损伤的死亡危险因素。多因素Logistic回归分析得出死亡预测概率公式为:Logit(P)=-2.053-1.736×(瞳孔不等大)-3.088×(瞳孔散大)+1.364×(枕叶损伤)+1.663×(脑疝)+1.112×(蛛网膜下腔出血)+0.150×(APACHEⅡ评分)。ROC曲线分析显示,瞳孔异常(不等大、散大)、枕叶损伤、蛛网膜下腔出血、脑疝及APACHEⅡ评分均能单独预测死亡概率,ROC曲线下面积分别为0.636、0.595、0.611、0.599及0.621。本风险模型预测重度颅脑损伤死亡发生的ROC曲线下面积为0.860,其敏感度和特异度88.60%及81.60%。结论本风险预测模型可用于评估重度颅脑损伤患者的预后。

Objective To construct the death prediction model for severe craniocerebral injury and provide a feasible basis for early prognosis.Methods A retrospective analysis of 190 patients with severe craniocerebral injury admitted to the First Affiliated Hospital of Guangxi Medical University from January 2012 to December 2017 was performed.Relevant risk factors at admission and record survival were collected at discharge.Logistic regression was used to establish a death prediction model.The performance of the model was predicted by fitting goodness test and calculating the area under the ROC curve.Results Of the 190 patients with severe head injury,91 died during hospitalization,with an incidence rate of 40%.Univariate Logistic regression analysis showed that age,tracheal intubation,pupil size,occipital lobe injury,cerebral hernia,subarachnoid hemorrhage,Glasgow Coma Scale,body temperature,platelet count,serum creatinine,and APACHEII were risk factors for death of severe head injury.Multivariate Logistic regression analysis showed that the formula for predicting the probability of death was:Logit(P)=-2.053-1.736×(different pupil)-3.088×(dilated pupil)+1.364×(occipital lobe injury)+1.663×(brain Hernia)+1.112×(subarachnoid hemorrhage)+0.150×(APACHEII score).ROC curve analysis showed that pupil abnormality(large and scattered),occipital lobe damage,subarachnoid hemorrhage,cerebral hernia,and APACHE II score could predict the probability of death alone,with AUC of 0.636,0.595,0.611,and 0.621 respectively.The AUC of death prediction for severe head injury was 0.860,and its sensitivity and specificity were 88.60%and 81.60%.Conclusion This risk prediction model can be used to evaluate the prognosis of patients with with severe craniocerebral injury.