Value Of 3D-TOF-MRA Combined With T-ASL On Collateral Circulation In Acute Ischemic Stroke Patients

Purpose: To investigate the effect of anatomical morphology of circle of Willis()upon the primary collateral blood flow.Materials and Methods: One hundred and nine healthy subjects underwent conventional MR imaging,3D-TOF-MRA and territorial ASL(t-ASL)acquisition on a 3T MRI scanner.According to the t-ASL results,the anterior circulation can be classified into three types for each case,including balanced blood flow,unbalanced left-to-right blood flow and unbalanced right-to-left blood flow.The posterior circulation can also be classified into three types for each side,including balanced blood flow,one side posterior circulation is partially supplied by the internal carotid artery group(unbalanced anterior-to-posterior blood flow),and one side posterior circulation was completely supplied by the internal carotid artery group(unbalanced anterior-to-posterior blood flow).The diameters(in cm)of anterior communicating artery(Acom A),A1 segment in three t-ASL anterior circulation groups,posterior communicating artery(Pcom A),and P1 segment in three t-ASL posterior circulation groups were measured.The paired t-test was used to assess the diameter difference between bilateral A1 segments in each anterior circulation group,and the diameter difference between Pcom A and P1 segment in each posterior circulation group.One-way ANOVA test was used to assess the difference A1right-A1 left and Acom A across three anterior circulation groups,and the difference in P1-Pcom A across three posterior circulation groups.ROC curve was used to calculate the threshold of the differenceand percentage difference between the bilateral A1 segments in the unbalanced blood flow groups.The Mantel-Haenszel chi-square test was used to analyze whether the relationship between theanatomical morphology of Pcom A and P1 and the source of occipital blood supply was related.Results: In a total of 109 cases of t-ASL anterior circulation,there was no significant difference found between bilateral A1 segments in balanced blood flow group(0.19±0.30,0.20±0.20,P=0.759),the diameter of left A1 segment was larger than right A1 segment in unbalanced left-to-right blood flow group(0.24±0.24,0.16±0.04,P<0.001)and the diameter of right A1 segment was larger than left A1 segment in unbalanced right-to-left blood flow group(0.23±0.04,0.14±0.02,P<0.001).The diameter of the dominant side A1 was significantly larger than that of contralateral side in unbalanced blood flow groups(0.24±0.04,0.15±0.04,P<0.001).A1right-A1 left were significantly different across three anterior circulation groups(0.01±0.02,-0.09±0.06,0.13±0.08,P<0.001),while Acom A showed no significant difference across three groups(0.11±0.03,0.13±0.02,0.11±0.04,P=0.222).When the primary blood flow was unbalanced,the threshold for the diameter difference between the A1 segments on both sides is 0.035 cm,and the threshold for the percentage difference is 18.5%.In a total of 98sides/49 cases of t-ASL posterior circulation,in balanced blood flow group,Pcom A was always absent,and the diameters of P1 segment and Pcom A were significantly different(0.20±0.03,0.01±0.03,P<0.001).In one side posterior circulation is partially supplied by the internal carotid artery group(unbalanced anterior-to-posterior blood flow),there was no significant difference found between P1 segment and Pcom A(0.17±0.04,0.16±0.05,P=0.229).In one side posterior circulation is completely supplied by the internal carotid artery group(unbalanced anterior-to-posterior blood flow),the diameter of Pcom A was larger than ipsilateral A1 segment(0.19±0.03,0.04±0.05,P < 0.001).In addition,P1-Pcom A were significantly different across three posterior circulation groups(0.19±0.05,0.02±0.06, -0.15±0.07,P<0.001).There was a certain correlation between the anatomical morphology of Pcom A and P1 segment and the source of occipital blood supply,the blood source to occipital lobe from ipsilateral internal carotid artery increased with the diameter of Pcom A(compared with ipsilateral A1),which has a significant correlation(x2=86.476,P<0.001,R=0.921,P<0.001).Conclusion: Anatomical morphology of has important influence on primary collateral blood flow.A1 segment plays the most vital role in anterior circulation blood flow.In posterior circulation,P1 segment governs primary blood flow with Pcom A together.Purpose: To compare combined t-ASL and 3D-TOF-MRA against multiphase CTA(m CTA)in assessment of collateral circulation in acute ischemic stroke(AIS)patients and explore the relationship between collateral circulation and clinical prognosis.Materials and Methods: Fifty-two AIS patients underwent conventional MR imaging,t-ASL and 3D-TOF-MRA acquisition and twenty-six patients also had a multiphase CTA scan.Chi-square test(Fisher’s accurate test)was used to compare the consistency between t-ASL combined 3D-TOF-MRA and multiphase CTA,kappa cofficient was calculated.Modified Rankin Scale(m RS)scores were recorded of all patients 1 month after discharge.A binary regression model was used to analyze the relationship between the establishment of collaterals and clinical outcome.Results: In 15 patients with arterial stenosis or occlusion,evaluation of secondary collaterals by 3D-TOF-MRA and t-ASL agreed well with multiphase CTA(Kappa=0.667,P < 0.022).Secondary collateral circulation was significantly associated with a good outcome in 19 AIS patients with acute arterial stenosis or occlusion(m RS scores of 0-2 1 month after discharge,OR: 0.006;95%CI: 0-0.689;P=0.035),In a multivariable ordinal logistic regression model,controlling for the confounding factors,Logistic regression shows that secondary collateral circulation was not an independent predictor of good prognosis for AIS patients at 1 month after discharge(OR: 0.018;95%CI: 0-1.295;P=0.063).Conclusion: 3D-TOF-MRA combined with t-ASL can evaluate collaterals in AIS patients,and it is in good agreement with m CTA.In AIS patients with acute arterial stenosis and occlusion,the presence of secondary collaterals is a factor that promotes good clinical outcomes,but it can’t independently predict the prognosis at 1 month after discharge.