Type A Aortic Dissection Postoperative Neurological Complications Risk Factors Analysis And Risk Model Building

Aortic dissection is a cardiovascular disease that has a high mortality and divided into Stanford type A and B according to the dissection involved scope.Type A aortic dissection affects the whole process of the aorta.It is a extensive disease that has difficult surgical technique and involves many organs and complications.Studies have shown that type A aortic dissection postoperative neurological complications are more common,whose mechanismin the aortic dissection is not completely clear.And the neurological complications are characterized by complicated and variable condition,hardly forecasting,resulting in catastrophic consequences and significantly associated with patient mortality.Clinical researches have shown that operation time,stroke history,sex,age,and so on,are related to postoperative neurological complications of type A aortic dissection.But it is controversial and needs further research.Based on these risk factors,the prognostic risk model has not been established.The aim of this study is to investigate the neurological complications of type A aortic dissection,analyze the risk factors and construct the risk model.The specific research contents are as follows:Firstly,the aortic dissection database system was established.The clinical data of patients with aortic dissection were classified using access and entered into the database,which involved preoperative information,including age,gender and other basic information and preoperative biochemical indicators,past history,intraoperative information,including surgical procedures,transit time,blocking time,deep stop time,ventilator use of time,and postoperative data,including postoperative complications and postoperative biochemical indicatorsSecondly,it had a risk factors analysis of type A aortic dissection neurological complications.Univariate and multivariate logistic regression analysis were performed using SPSS software to obtain independent risk factors for neurological complications.Finally,the classification tree risk model was built.Based on the above analysis results,the risk model was constructed using CHAID algorithm and then evaluated using the Risk statistics and the area under the ROC curve.The following conclusions were obtained.First of all,the independent risk factors were surgical history,cerebral infarction history,CABG,artificial graft bypass,age,ventilator use time,time from oncome to surgery,D2 DBIL level(P <0.05)using multivariate data analysis method.Second of all,based on the multivariate analysis results,seven risk factors were selected by CHAID algorithm.The most important variables were ventilator use time,followed by cerebral infarction history,surgery history,acute aortic dissection,CABG,age and D2 DBIL.Ventilator use time greater than 128 min patients with the highest risk,followed by the history of cerebral infarction,surgical history,acute aortic dissection,CABG surgery,older than 60 years of age,direct bilirubin high levels of patients.Last but not the least,the classification tree model was constructed and cross-validation was performed using medical data.The results showed the Risk value is 0.095 and the area under the ROC curve is 0.843,which is statistically significant compared to 0.5,indicating that the model fitting effect is perfect.In conclusion,the independent risk factors for patients with neurological complications after type A aortic dissection were ventilator use time,acute aortic dissection,age,cerebral infarction history,CABG,surgery history,D2 DBIL level,and the classification tree model established by these seven risk variables had a good effect.The model showed that preoperative active anti-infective therapy,reasonable control of ventilator use time,rational use of surgical methods,focusing on acute dissection and elderly patients,pay close attention to direct bilirubin levels remain at normal levels could reduce the incidence of neurological complications.