A Preoperative Mortality Risk Assessment Model For Stanford Type A Acute Aortic Dissection

BACKGROUND:Acute aortic dissection type A is a life-threatening disease required emergency surgery during acute phase.It requires emergency treatment or emergency surgery in the acute phase.If it is not treated in time or undergoes surgery,50%of patients die within 24 hours from the onset.The hourly mortality rate increases by approximately 1%-2%.Different clinical manifestations,laboratory tests,and imaging features of patients with acute aortic dissection type A are the risk factors of preoperative mortality.At present,many preoperative risk assessment models for cardiac surgery and some risk prediction models related to aortic dissection type A had been developed.However,these models had problems that were not suitable for patients with acute aortic dissection type A,limited samples,and improper selection of variables.Therefore,their predictive performance were questionable.Compared with foreign literature reports,the diagnosis and treatment of aortic dissection in my country has the characteristics of diversified clinical manifestations,younger age of onset,high mortality during hospitalization and before surgery,and more complications in clinical treatment.Therefore,there is an urgent need for a preoperative death risk scoring system for Chinese patients predicts the risk of early death for the emergency rooms of various institutions.METHODS:A total of 673 Chinese patients with acute aortic dissection type A who were admitted to our hospital were retrospectively included.All patients were unable to receive surgically treatment within 3 days from the onset of disease.The patients included were divided into the survivor and deceased groups,and the endpoint event was preoperative death.Collected patient information,including general information,medical history,physical examination,laboratory examination,imaging examination,etc.Built a patient database.Using random sampling method,673 patients were divided into 75%and 25%two groups,and a developmental dataset of 505 cases and a validation dataset of 168 cases were established respectively.Normally distributed continuous variables were compared using by a two-tailed t-test.Non-normally distributed continuous variables were compared using the Mann-Whitney U test.Categorical variables were analyzed using by χ2 or Fisher’s exact test as needed.P<0.05 was considered statistically significant.The stepwise multivariate analysis was performed for determining the variables that were independently associated with preoperative mortality.RESULTS:Among the 673 patients,527 patients survived(78.31%)and 146 patients died(21.69%).The developmental dataset had 505 patients,calibration by Hosmer Lemeshow was significant(χ2=3.260,df=8,P=0.917)and discrimination by area under ROC curve was 0.8448(95%CI,0.8007-0.8888).The validation dataset had 168 patients,calibration was significant(χ2=5.500,df=8,P=0.703)and the area under the ROC curve was 0.8086(95%Cl,0.7291-0.8881).The following independent variables increased preoperative mortality:age(OR=1.0085,P=0.510),abrupt chest pain(OR=3.534,P<0.001),lactic in arterial blood gas≥3 mmol/L(OR=3.636,P<0.001),inotropic support(OR=8.615,P<0.001),electrocardiographic myocardial ischemia(OR=3.300,P=0.001),innominate artery involvement(OR=1.625,P=0.104),right common carotid artery involvement(OR=3.487,P=0.001),superior mesenteric artery involvement(OR=2.651,P=0.001),false lumen/true lumen of ascending aorta≥0.75(OR=2.221,P=0.007).Our data suggest that a simple and effective preoperative death risk assessment model has been established.CONCLUSIONS:Using a simple and effective risk assessment model can help clinicians quickly identify high-risk patients and make appropriate medical decisions.