Analysis Of Risk Factors For Hemorrhage Transformation And The Establishment And Test Of The Predictive Model After Mechanical Embolization In Acute Large Vessel Occlusion

Objective:Discuss the risk factors and prediction model of hemorrhage transformation after acute large vascular occlusion and test the prediction modelMethods:Selected 110 patients with acute large vascular thrombosis after mechanical thrombosis between February 2017 and November 2019,divided into bleeding group and non-bleeding group,and counted two groups of general data,preoperative blood sugar,closed blood vessels Alberat stroke project early CT score(ASPECT),lateral branch circulation,whether bridged,preoperative NIHSS score,arterial puncture to vascular re-pass time(PTR),morbidity to vascular re-pass time(OTR),and the number of in-operative hydrants,hydration method,application in tiloffi class,m TICI grading,oozing agent oozing agent,surgery time,surgical intervention(de-bone valve decompression),postoperative 24 h National Institutes of Health stroke scale(NIHSS)score.The collected data are analyzed by single-factor and multi-factor logistic regression,independent risk factors and risk coefficients are found,prediction equations are made,AND ROC analysis is carried out on independent risk factors and prediction equations,sensitivity is tested and their optimal critical values are calculated,and prediction heat maps are made.Results:The results of single-factor analysis showed that there were statistical differences between 2 groups of indicators such as alcohol consumption,antiplate plateplates,blocked blood vessels,oozing agent,OTR value,PTR value,ASPECT score,postoperative 24 h NIHSS score,(P <0.1),bridging therapy(P-0.773)does not increase the risk of hemorrhage conversion after mechanical embolism,multi-factor logistic regression analysis shows:arterial astrology in the brain,ASPECT score,PTR for acute cerebral infarction Independent influence factors of intracranial bleeding after dead mechanical embolism,in which arterial cysts in the brain(OR=24.130,95% CI:1.360-428.055,P=1 0.030),PTR(OR=0.555,95% CI:1.006 to 1.042,P=0.008)are risk factors.ASPECT score are protective factors,and whether there is bleeding in the skull after acute cerebral infarction mechanical embolism mainly depends on the ASPECT score(standardized regression coefficient is-0.935),followed by whether arterial thrombosis in the brain occurs(standardized regression coefficient is 0.881),and finally the size of the PTR value(standardized regression factor.The standardized regression coefficient is 0.556),and the predicted equation is: Logit P= 0.024* PTR,+3.183* MCA(closure)-1.065*ASPECT score+ 0.680.If Logit P >0.187,there is an increase in intracranial bleeding,and if Logit P≤0.187,intracranial bleeding is not easy to occur after hemorrhage transformation.The area under the ROC curve of the forecast model is 0.924(95% CI: 0.858 to 0.966,P<0.001),the Youden index is 0.7875,the sensitivity is 100.00%,and the specificity is 78.75%.The diagnostic value of the predictive model was higher than that of THEPD score(AUC=0.821,P=0.0024),arterial clement in the brain(AUC=0.562,P<0.001),and PTR(AUC=0.775,P=0.001).ASPECT score≤6.PTR>72minutes.Closed blood vessels are arterial tracts in the brain,which are prone to acute mechanical thrombosis of large blood vessels and intracranial hemorrhage conversion.In order to facilitate the clinical application of predictive models,predictive models can be converted into heat maps to quickly judge patient prognosis.Conclusion:Middle cerebral artery occlusion,ASPECT score and PTR are independent influencing factors of intracranial hemorrhage after mechanical thrombectomy for acute cerebral infarction.And the predicted equation is: Logit P= 0.024* PTR,+3.183* MCA(closure)-1.065*ASPECT score+0.680.