The Effect Of Collateral Circulation MR Sign On The Prognosis Of AIS

Part 1 Collateral status and its prognostic impaction on large artery atherosclerotic and cardioembolic strokes: A preliminary study based on MR angiographyObjective: Both the etiology and collateral circulation of acute ischemic stroke(AIS)affect the clinical prognosis of patients.We tested the hypothesis that patients with large artery atherosclerotic(LAA)stroke would had better collateral status compared to cardioembolic(CE)stroke patients.Additionally,we attempted to probe whether or not this kind difference of collateral status would influence the evaluation of clinical outcome of AIS.Methods: We identified all consecutive anterior circulation AIS patients in one-year period.Patients were enrolled if they were ascribed to LAA or CE etiology according to the Trial of Org 10172 in acute Stroke Treatment(TOSAT)criteria.Collateral status was rated on pre-treatment MR angiography(MRA)according to the Tan score.Clinical outcome was assessed with 90-day modified Rankin scale(m RS)and defined as favorable(m RS 0-2)and unfavorable(m RS>2).We compared the difference of the etiology and MRI factors between patients in good(Tan 2-3)and poor(Tan0-1)collateral status.Logistic regression analysis was performed to assess the factors which associated with good collateral circulation.Then we developed two clinical outcome prediction models,including clinical + conventional MRI model,and combination(clinical + conventional MRI factors + Tan scores)model.Receiver operating characteristic(ROC)curve analysis was performed to measure the diagnostic efficiency of the two models in predicting favorable clinical outcome.Results: Of 116 patients with anterior circulation AIS,68 patients with good collateral status and 48 patients with poor collateral status were retrospectively enrolled.Patients with good collateral status were more common in LAA type in etiology(33.33% versus 66.67%,P=0.004),smaller lesions volume(26.82 ml versus 10.15 ml,P=0.001),higher ASPECTS(4.5versus 6,P=0.004),lower admission National Institutes of Health Stroke Scale(NIHSS)(8 versus 4,P=0.019),and lower m RS(3 versus 1,P<0.001),than the patients in poor collateral status.Logistic regression analysis showed that stroke etiology of LAA(LAA versus CE,OR=2.970,95%CI: 1.241-7.106,P=0.014)and higher ASPECTS(OR=1.265,95%CI:1.049-1.525,P=0.014)were the independent predictors of good collateral status.The area under ROC(AUC)values of predicting favorable clinical outcome were following: 0.855 and 0.877 for whole cohort with clinical + conventional MRI and combination models separately,0.855 and 0.865 for LAA stroke patients with clinical +conventional MRI and combination models separately,0.859 and 0.910 for CE stroke patients with clinical + conventional MRI and combination models separately.Conclusions: Better collateral status was found in LAA stroke than in CE stroke.However,more pronounced impaction of good collateral status on favorable outcome in CE stroke patients.Our findings may assist with refining the prognostic prediction of collateral status on AIS patients,based on etiology of stroke.Part 2 FLAIR vascular hyperintensity and its prognostic impact: A study based on etiologyObjective: FLAIR vascular hypersignal sign(FVH)is a common imaging sign of acute ischemic stroke(AIS),but its manifestation and outcome value are still controversial in different AIS etiology.We attempted to explore whether there is a difference in FVH signs between large atherosclerotic(LAA)and cardioembolic(CE)stroke patients,and to further analyze whether or not the prognostic value of FVH in LAA and CE stroke patients would be consistent.Methods: We analyzed all consecutive anterior circulation AIS patients in one-year period.Patients were enrolled if they were ascribed to LAA or CE etiology according to the Trial of Org 10172 in acute Stroke Treatment(TOSAT)criteria.We judged the positive and negative of FVH,and further rated it as grade 0-7 according to FVH-SPECTS method.Clinical outcome was assessed with 90-day modified Rankin scale(m RS)and defined as favorable(m RS 0-2)and unfavorable(m RS>2).We compared the difference of the etiology and other image characteristics between patients in FVH positive and FVH negative.Logistic regression analysis was performed to assess the independent factors which associated with FVH positivity.Then we used Spearman correlation analysis,scatter plot and multivariate logistic regression to evaluate the prognostic value of FVH grade in total stroke patients,LAA type and CE type stroke patients.Results: Of 116 AIS patients,there were 94 patients with FVH positive and 22 patients with FVH negative.Patients with FVH positive were less hyperlipidemia(27.3% versus 8.5%),larger lesions volume(6.26 ml versus18.85ml),higher admission National Institutes of Health Stroke Scale(NIHSS)(2.5 versus 6),higher m RS(2 versus 1)and poor outcome(22.7% versus48.9%)than the patients in FVH negative group.Additionally,the patients of FVH positive(76.2% versus 93.8%)were more common in CE type AIS than LAA(P < 0.031～0.001).Logistic regression analysis showed that stroke TOAST etiology of CE(CE versus LAA,OR=5.983,95%CI:1.186-30.179,P=0.030)and higher admission NIHSS(OR=1.265,95%CI:1.049-1.525,P=0.014)were the independent predictors of FVH positive.Spearman correlation analysis showed that FVH grade was positively correlated with m RS in the total cases(r=0.260,P=0.005)and LAA type AIS(r=0.296,P=0.006),and there was no correlation between FVH grade and m RS in CE type AIS(P=0.509).Conclusions: Higher rate of positive FVH was found in CE stroke than LAA stroke in the cases of AIS without recanalization.The correlation between FVH grade and clinical prognosis varied with different etiological types of AIS.In total cases and LAA type AIS,higher FVH grade was an independent predictor of poor outcome,while in CE type AIS,FVH grade was not related to prognosis.