Comparative Study Of Multiphase CTA And CTP In The Evaluation Of Collateral Circulation Status And Outcome Prediction For Acute Ischemic Stroke Due To Anterior Circulation Large Vessel Occlusion

Objective1.To investigate the clinical and practical value of mCTA-ColorViz score based on FastStroke software and traditional mCTA-MIP score in the evaluation of collateral circulation in patients with anterior circulation LVO-AIS.2.To investigate the correlation between HIR derived from CTP,mCTA-ColorViz score and mCTA-MIP collateral score and its clinical practical value in the evaluation of collateral circulation in patients with anterior circulation LVO-AIS.Simultaneously,to assess the correlation between HIR,mCTA-ColorViz score,mCTA-MIP collateral score with functional outcome and their clinical and practical value in functional outcome prediction of patients with anterior circulation LVO-AIS.Materials and Methods1.SubjectsConsecutive patients who had acute neurological deficit were confirmed as having AIS with ’ one-stop-shop’ multimodal CT examination(including NCCT,CTP and multiphase CTA)within 24 h after onset(Time from last known well,TFLKW),were admitted to our single-institution stroke center from January 2019 to July 2021,and were retrospectively reviewed from our prospective database for AIS.Patients with a final clinical and imaging diagnosis of anterior circulation LVO-AIS were included in this study according to inclusion/exclusion criteria.2.MethodsThe mCTA angiograms were reconstructed at a workstation(AW 4.7,GE Healthcare,Milwaukee,Wisconsin)with the GE FastStroke workflow and multi-plane axial,coronal,and sagittal reconstructed images,and axial maximum density projection(MIP)were obtained for all three phases.The collateral status were evaluated by two neuroradiologists using the mCTA-ColorViz collateral score and the traditional mCTA-MIP collateral score and compared with the gold standard DSA.The original CTP images were automatically processed using the SHUKUN AI perfusion software(SK-CTPDoc,StrokePro V2.4,SHUKUN Technology,Beijing,China),and six parameter maps of tMIP,CBF,CBV,MTT,TTP,and Tmax were obtained rapidly.The infarct core(rCBF<30%),ischemic area volume(Tmax>6s),mismatch ratio and mismatch volume(MMV)or ischemic penumbra(IP)volume were calculated automatically.Further,the volumes of Tmax>4s,Tmax>6s,Tmax>8s,and Tmax>10s and HIR(Tmax>10s/Tmax>6s)were also calculated automatically.The correlations between the HIR and the mCTA-ColorViz collateral score and the mCTA-MIP collateral score and the DSA collateral score were calculated using the Spearman correlation.Receiver operating characteristic(ROC)curve analysis was conducted to evaluate the diagnostic performance of the HIR and the mCTA-ColorViz collateral score and the mCTA-MIP collateral score in predicting favorable DSA collateral score.The Youden index was used to identify the optimal threshold from the ROC curve for predicting favorable DSA collateral score.The area under the curve(AUC)of each predictor was compared using the Delong test.The mCTA collateral score evaluated by each of the two raters were tested for interrater reliability using linear-weighted Kappa analysis.Patients were dichotomized into favorable collateral score(4-5 points)group and unfavorable collateral score(0-3 points)group based on the mCTA-ColorViz collateral score,and the univariate analysis was conducted to compare initial stroke severity assessed using NIHSS score,infarct core volume,Tmax volumes,and perfusion mismatch ratio between two groups using the Mann-Whitney U test.The multivariate logistic regression were used to analyze the relationship between the clinal and imaging characteristic variables and the collateral status,and the influencing factors of independent prediction of favorable collateral status.Patients were then dichotomized into favorable functional outcome(mRS≤2)group and unfavorable functional outcome(mRS>2)group based on 90-d mRS and were analyzed and compared respectly in general population group,reperfusion treatment arm and conservative medical treatment arm.The correlation between the imaging characteristic variables(mCTA-ColorViz collateral score,ASPECTS score,perfusion parameters including HIR,follow-up infarct volume,etc.)and functional outcome(90-d mRS)among the above three study groups were calculated using the Spearman correlation.ROC curve analysis was conducted to evaluate the diagnostic performance of mCTA-ColorViz collateral score,mCTA-MIP collateral score,HIR,various perfusion parameters,and follow-up infarct volume in predicting favorable functional outcome.The Youden index was used to identify the optimal threshold from the ROC curve for predicting favorable functional outcome.The area under the curve(AUC)of each predictor was compared using the Delong test.The multivariate logistic regression were used to analyze the relationship between the imaging characteristic variables(mCTA-ColorViz collateral score,ASPECTS score,perfusion parameters including HIR,follow-up infarct volume,etc.)and functional outcome among the above three study groups,and the influencing factors of independent prediction of favorable functional outcome.3.Results1.Finally,for analysis,235 eligible patients(median age,66 years(interquartile range[IQR],55.5,73 years);178 males)were identified.Reperfusion therapy was performed in 118 of 235 patients(50.2%),among whom 31underwent IVT,73 underwent EVT,and 14 underwent both.A total of 127 of 235(54%)patients had favorable functional outcomes.In the reperfusion group,the proportion of achieved favorable functional outcome(0-1 points)(51 cases(43.2%)vs.24 cases(20.5%))and the proportion of independent functional outcome(0-2 points)(70 cases(59.3%)vs.58 cases(49.6%))were significantly higher than those in the conservative medical treatment group.2.The median mCTA-ColorViz collateral score was 3([IQR],3,4)and the consistency among raters were excellent(k=0.913,95%CI:0.855-0.935;P<0.05).Overall,40.4%(95/235)patients had favorable collateral scores(4-5 points),43.8%(103/235)had intermediate(3 points)and 15.8%(26/285)poor collateral scores(0-2 points).Patients with favorable collateral scores were more likely had a lower baseline NIHSS score and a higher baseline ASPECTS and a larger mismatch ratio,and along with a smaller infarct core volume and a smaller ischemic volume(Tmax>4s、Tmax>6s、Tmax>8s、Tmax>10s)and a lower HIR and smaller FIV than those with unfavorable collateral scores.Moreover,patients with favorable collateral scores were more likely had a lower 90d mRS than those with unfavorable collateral scores.Multivariable logistic regression analysis was shown that the baseline NIHSS score and IC volume and HIR were independently associated with a favorable collateral score.3.There was a significant correlation between HIR and the mCTA-ColorViz collateral scores and the mCTA-MIP collateral score and the DSA collateral score.ROC analysis revealed that the optimal cut-off point of the mCTA-ColorViz collateral scores and the mCTA-MIP collateral scores to predict a favorable DSA collateral score was 4(>3).The area under the curve(AUC)of the HIR to predict a favorable DSA collateral score was 0.972(95%CI:0.912-0.995;p<0.001),which was more higher than the mCTA-ColorViz collateral score(AUC,0.941;95%CI:0.868-0.980;p<0.001)and the mCTA-MIP collateral score(AUC,0.914;95%CI:0.835-0.964;p<0.001),also the Youden index and sensitivity and specificity were more higher for the HIR than for the mCTA collateral scores.The consistency among different raters by using the mCTA-ColorViz collateral score was excellent,which needed less time and was more time-saving.4.There was a significant positive correlation between HIR and functional outcome(90-d mRS),and the correlation coefficient between HIR and functional outcome was significantly higher than the correlation coefficient between mCTA collateral score and functional outcome.ROC analysis revealed that the optimal cut-off point of the HIR to predict a favorable functional outcome was 0.3(≤0.3).The area under the curve(AUC)of the HIR to predict a favorable functional outcome was 0.968(95%CI:0.937-0.987;p<0.001),which was more higher than the mCTA-ColorViz collateral score and mCTA-MIP collateral score and DSA collateral score,also the Youden index and sensitivity and specificity was more higher for the HIR than for the others.Simultaneously,the area under the curve(AUC)and Youden index and sensitivity and specificity of the HIR to predict a favorable functional outcome were the highest than other perfusion parameters(such as IC(rCBF<30%)volume,Tmax>4s,Tmax>6s,Tmax>8s,Tmax>10s,mismatch ratio,IP volume,etc.)and clinical parameters(baseline NIHSS score)and other imaging parameters(ASPECTS score,follow-up infarct volume(FIV)).At univariate logistic regression analysis,the HIR was significantly associated with a favorable outcome not only in total cohort but also in reperfusion treatment arm and supportive medical treatment arm.Multivariable logistic regression analysis showed that the HIR was independently associated with a favorable outcome not only in total cohort but also in reperfusion treatment arm and in the EVT arm and in the supportive medical treatment arm.The mCTA-ColorViz collateral score and mCTA-MIP collateral score were not significantly associated with functional outcome after adjustment both in total cohort and in reperfusion treatment arm and in supportive medical treatment arm.4.Conclusions1.The mCTA-ColorViz collateral score is a novel,simple,rapid,accurate,efficient and visual method for collateral circulation assessment that has a strong correlation with DSA collateral score and mCTA-MIP collateral score,which can be used as a new collateral circulation scoring method to assess collateral staus in patients with anterior circulation LVO-AIS.2.The mCTA-ColorViz collateral score is more accurate and reliable compared with the traditional mCTA-MIP collateral score.It can effectively improve the consistency among readers,shorten the scoring time,reduce the difficulty of scoring,and has obvious advantages.It can replace the application of mCTA-MIP collateral score in the evaluation of collateral circulation in patients with anterior circulation LVO-AIS.3.HIR derived from CTP is a fast,objective,accurate and reliable quantitative parameter that can accurately assess collateral staus.It is superior to mCTA-ColorViz collateral score and mCTA-MIP collateral score in the assessment of collateral circulation in patients with anterior circulation LVO-AIS.4.HIR is a very practical imaging parameter that can accurately and reliably predict the outcome of LVO-AIS patients.As a simple,rapid,automatic and quantitative evaluation tool,it is better understand than the subjective mCTA collateral score,and can be applied for determining patient eligibility before EVT.