Multimodal CT Based Study Of The Predicting Factors Of The Hemorrhagic Transformation And Prognosis In Acute Ischemic Stroke After Alteplase Thrombolysis Treatment

Objective:This is a multimodal CT based study to looking for imaging quantitative prediction indicators of the hemorrhagic transformation and prognosis in acute ischemic stroke after intravenous thrombolysis treatment,to provide important guidance for the majority of patients with AIS,decrease the occurrence of adverse events,improve the prognosis.Methods:A total of 266 AIS patients who received thrombolysis treatment with rt-pa in the green channel of emergency stroke treatment in the first affiliated hospital of Soochow university from January 2017 to March 2019 were continuously recruited.All of them were examined by multi-mode CT(NCCT+CTA+CTP)and re-examined by head CT 24 hours after thrombolysis.Quantitative imaging data such as CTP ischemia-hypoperfusion volume,infarct core,ischemia-penumbra volume and mismatching of each patient were calculated automatically by MIStar software,and ASPECT score and rLMC score were recorded.Meanwhile,demographic data,clinical baseline data and relevant laboratory data were recorded.According to the presence or absence of hemorrhagic transformation,the patients were divided into two groups:those with HT and those without HT,There were significant differences in imaging data,demographic data,clinical baseline data and related laboratory data between the two groups,to analyze the relationship between the parameters of multi-mode CT images and bleeding subtype transformation,further using the multi-factor logistic regression analysis of bleeding into the correlation of factors,ROC curve analysis was used to predict the threshold values of clinical and imaging factors for bleeding.Patients were divided into two groups,group of good prognosis and group of bad prognosis,based on the mRS score recorded 90 days after thrombolysis treatment.The same comparison between the two groups of imaging data,demographic data,clinical and related laboratory baseline data for statistical differences,further using the multi-factor logistic regression analysis of the clinical prognosis independent correlation factors,and using ROC curve analysis predicts poor prognosis related factors of the boundary value.Results:1.Case data analysisA total of 266 AIS patients who received thrombolysis treatment with rt-pa were included,with an average age of 66 years old(66.47±12.97 years),107 female(40.2%),215 cases(80.8%)without hemorrhagic transformation,and 51 cases(19.2%)with hemorrhagic transformation.170 cases(63.9%)with good prognosis,and 96 cases(36.1%)with poor prognosis were included.2.Single factor and multivariate logistic regression analysis and ROC curve analysisof hemorrhage transformationFinally 266 patients who received venous rt-PA in the treatment of patients with acute ischemic stroke,HT group and none HT group single factor analysis,the results showed that bleeding into groups more age,baseline blood glucose high,high baseline NIHSS,longer prothrombin time,high c-reactive protein,high triglycerides,low platelet count,low rLMC score,the ASPECT score,low core hypoperfusion,ischemia and infarction volume is big,and patients with atrial fibrillation,a past medical history of antithrombotic;However,hemorrhagic transformation was not correlated with gender,blood pressure,OTT,previous hypertension,diabetes,stroke,smoking,uric acid,total cholesterol,etc.Multivariate logistic regression analysis showed that:Atrial fibrillation(OR=5.26,95%CI=1.541-17.951,P=0.008),platelet count(OR=0.988,95%CI=0.978-0.998,P=0.019),and ASPECT score(OR=0.502,95%CI=0.342-0.737,P<0.001),Volume of ischemia-hypoperfusion(OR=0.984,95%ci=0.969-1,P=0.043)were correlated with hemorrhage transformation after intravenous thrombolysis.Patients with atrial fibrillation had a significantly increased incidence of hemorrhage transformation after intravenous thrombolysis,while patients with small ischemia-hypoperfusion volume,high platelet count and high ASPECT score have relatively low incidence of bleeding transformation.In ROC analysis,platelet count with a cut-off value of 199 could better predict hemorrhage transformation(AUC=0.821),displaying a sensitivity of 72.6%and specificity of 76.5%.Baseline ASPECT score was the best candidate for predicting hemorrhage transformation when the cut-off value was 8(AUC=0.602),showing a sensitivity of 43.3%and specificity of 78.4%.AF was a strong predictor of hemorrhage transformation,with a sensitivity of 70.6%and specificity of 76.6%.Ischemia and hypoperfusion volume(DT>3.0s)with a threshold of 24 could well predict hemorrhage transformation,with a sensitivity of 88.2%and specificity of 55.6%.3.Multimode CT parameters predict the hemorrhagic transformation subtypes,as well as clinical and imaging characteristics of patients with different collateral levelsIn all the parameters in multimodal CT rLMC grade,baseline ASPECT and Infarction core volume(CBF<30%)to predict hemorrhagic transformation subtype were statistically significant(P<0.05),The rLMC score,baseline ASPECT score of parenchymal hematoma(PH)group were lower than that of hemorrhagic infarction(HI)group,and infarction core volume(CBF<30%)is higher than in the HI group.There were 222 cases with good collateral circulation level and 44 cases with poor collateral circulation level.The baseline NIHSS score of patients with poor collateral circulation level was significantly higher than that of patients with good collateral circulation level(P<0.001).The volume of ischemia and hypoperfusion(DT>3.0s)and the volume of core infarction(CBF<30%)of the patients are larger(P<0.001)in patients with poor collateral circulation.The proportion of patients with hemorrhagic transformation and poor prognosis in the group with poor collateral level was significantly higher than that in the group with good collateral level(P<0.001).4.Univariate and multivariate logistic regression analysis and ROC curve analysis of neurological function prognosisAfter 3 months of follow-up,170 patients had a good prognosis and 96(36%)had a poor prognosis.In single-factor analysis,poor prognosis was common in patients with advanced age,hyperglycemia,severe stroke,OTT extension,hyperlipidemia,high CRP,hemorrhagic transformation,previous history of atrial fibrillation and stroke,low rLMC and ASPECT scores of multi-mode CT imaging.Patients with baseline ischemia-hypoperfusion and large infarct cores tend to have a poor prognosis.Multivariate logistic regression analysis showed that age(OR=1.052,95%CI=1.006-1.100,P=0.025),baseline blood glucose(OR=1.179,95%CI=1.003-1.385,P=0.046)and baseline NIHSS(OR=1.217,95%CI=1.116-1.327,P=0.000)were significantly correlated with poor prognosis of neurological function after 3 months.Patients with advanced age,hyperglycemia and severe stroke had a poor prognosis.In ROC curve analysis,the baseline NIHSS cut-off value of 12 can be better distinguish good prognosis from poor prognosis(AUC=0.855),showing a sensitivity of 74.7%and specificity of 86.9%.The best prognostic assessment was when the age threshold was 70(AUC=0.686),showing a sensitivity of 66.3%and specificity of 65.5%.With baseline blood glucose threshold of 7.07,good prognosis and poor prognosis could be distinguished(AUC=0.616),showing sensitivity of 57.9%and specificity of 66.7%.Conclusions:1.Baseline ASPECT score and ischemia-hypoperfusion volume(DT>3.0s)in multimode CT images are closely related to hemorrhage transformation after alteplase thrombolysis treatment in AIS(The blood supply area of the middle cerebral artery)patients.Baseline ASPECT score ?8 and ischemia-hypoperfusion volume?24ml can be used as imaging predictors of hemorrhage transformation after intravenous thrombolysis.Patients with platelet count?199(109/L)are prone to hemorrhage transformation after received intravenous thrombolysis treatment.Atrial fibrillation is a risk factor for hemorrhagic transformation after intravenous thrombolysis.2.The hemorrhagic transformation rate and poor prognosis rate of patients with poor collateral circulation in acute ischemic stroke were significantly higher than those with good collateral circulationbut,but rLMC score was not qualified to be a predictor of hemorrhagic transformation and prognosis.3.Baseline NIHSS,age and baseline blood glucose were independently correlated with clinical prognosis for 3 months after intravenous thrombolysis in AIS patients;The prognosis of patients with baseline NIHSS?12 points,age?70 years old,baseline blood glucose?7.07mmol/1 was relatively poor.