Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 1 (January), 2014: pp e39-e45. e39

Transcription

1 Internal Cerebral Vein Asymmetry on Follow-up Brain Computed Tomography after Intravenous Thrombolysis in Acute Anterior Circulation Ischemic Stroke Is Associated with Poor Outcome Vijay K. Sharma, MRCP,* Leonard L. L. Yeo, MRCP,* Hock L. Teoh, MRCP,* Liang Shen, PhD, Bernard P. L. Chan, MD,* Raymond C. Seet, MRCP,* Aftab Ahmad, MRCP,* Vincent F. Chong, FRCR,x and Prakash R. Paliwal, MRCP* Background: Identifying early predictors of functional outcome after acute ischemic stroke (AIS) is important for planning rehabilitation strategies. Internal cerebral veins (ICV) drain deep parts of brain, run parallel to each other, and consistently seen on computed tomography angiography (CTA). Even minor asymmetry in their filling can be identified. We hypothesized that venous drainage would be impaired in patients with acute occlusion of internal carotid artery or middle cerebral artery. Because systemic thrombolysis can alter the vascular findings, we evaluated the relationship between ICV asymmetry on follow-up CTA and functional outcome. Methods: Consecutive AIS patients treated with intravenous thrombolysis between 2007 and 2010 were included. ICVasymmetry was assessed by 2 independent blinded stroke neurologists/neuroradiologists. Functional outcome was assessed by the modified Rankin Scale (mrs) at 3 months, dichotomized as good (0-1) and poor (2-6). Data were analyzed for predictors of functional outcome. Results: Of 2238 patients with AIS, 226 (10.1%) anterior circulation AIS patients received intravenous thrombolysis. The median age was 65 years (range 19-92), 44% were men, and median National Institutes of Health Stroke Scale (NIHSS) score was 16 points (range 4-32). Hypertension was the commonest risk factor in 173 (76.5%) patients, whereas 78 (34.5%) had atrial fibrillation. ICV asymmetry on follow-up CTA was assessed in 103 (45.5%) patients. Admission NIHSS score (odds ratio [OR] 1.07; 95% confidence interval [CI] , P 5.046), change in NIHSS score during first 24 hours (OR.737; 95% CI , P,.0001), and ICV asymmetry on follow-up CTA (OR 20.3; 95% CI , P,.0001) independently predicted poor outcome at 3 months. Conclusions: ICV asymmetry on follow-up CTA after intravenous thrombolysis is an early predictor of poor functional outcome. Key Words: Acute ischemic stroke internal cerebral vein thrombolysis CT angiography. Ó 2014 by National Stroke Association From the *Division of Neurology, Department of Medicine, National University Health System; Yong Loo Lin School of Medicine, National University of Singapore; Biostatistics Unit, Yong Loo Lin School of Medicine, National University of Singapore; and xdepartment of Diagnostic Imaging, National University Health System, Singapore. Received June 28, 2013; revision received August 8, 2013; accepted August 11, Author contributions: Study concept and design P.R.P., L.L.L.Y., B.P.L.C., and V.K.S.; acquisition of data P.R.P., L.L.L.Y., A.A., and V.K.S.; analysis and interpretation of data P.R.P., L.S., and V.K.S.; drafting of the manuscript P.R.P., L.L.L.Y., and V.K.S.; critical revision of the manuscript for important intellectual content A.A., R.C.S., B.P.L.C., V.F.C., H.L.T., and V.K.S.; statistical analysis V.K.S., L.L.L.Y., and L.S.; obtained funding none; and administrative, technical, and material support H.L.T., V.F.C., L.S., and V.K.S. Disclosure: None of the authors declare any conflicts of interest. Address correspondence to Vijay K. Sharma, MRCP, Division of Neurology, Department of Medicine, National University Health System, 1 E Kent Ridge Rd, Singapore drvijay@singnet. com.sg /$ - see front matter Ó 2014 by National Stroke Association Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 1 (January), 2014: pp e39-e45 e39

2 e40 Introduction A significant proportion of acute ischemic stroke (AIS) patients achieve good functional outcome with timely administered intravenous tissue plasminogen activator (tpa). However, the rate and extent of recovery remain variable. Considering scarce and costly resources, early identification of reliable predictors for functional outcome is important for planning rehabilitation strategies and placement after discharge from the hospital. Internal cerebral veins (ICVs) drain the deep parts of cerebral hemispheres and run backward to form the great cerebral vein. 1 Impaired arterial blood supply results in poor venous drainage, rendering the corresponding ICV less visible. Because the 2 ICVs are consistently seen on computed tomographic angiography (CTA) as running parallel and very close to each other, even minor asymmetry in their filling can be diagnosed easily on CTA (Fig 1). 2 We hypothesized that ipsilateral cerebral hypoperfusion because of acute occlusion of internal carotid artery (ICA) or middle cerebral artery (MCA) would impair the ipsilateral deep venous drainage and result in obvious asymmetric opacification of the ICVs. Therefore, ICV asymmetry might serve as a surrogate marker of inadequate cerebral perfusion and collateral circulation. Because a significant proportion of intravenous tpatreated patients achieve arterial recanalization 3 and many of them might develop sufficient collateral arteries during the early phase of AIS, 4 ICV asymmetry on the pre-tpa CTA can disappear. Therefore, ICV asymmetry on a pre-tpa CTA may not be useful for predicting the long-term outcomes. On the other hand, vascular status of the intracranial circulation is believed to acquire its near-final status in majority of AIS patients by day 2. Therefore, we aimed to evaluate whether the presence of ICV asymmetry on day 2 CTA in tpa-treated AIS patients can predict the final outcome. Subjects and Methods In this retrospective cohort study, we analyzed the data from consecutive AIS patients treated with intravenous tpa between January 2007 and July Data were entered prospectively in the AIS thrombolysis registry maintained at our tertiary care center. Because the arterial supply via the vertebrobasilar system is not drained by the ICVs, patients with posterior circulation stroke were excluded from this study. The ethics committee at our institution approved the study. Urgent noncontrast CT scan of the brain was performed for all patients suspected of having an acute stroke to exclude intracerebral hemorrhage and other mass lesions. A considerable proportion of patients considered eligible for intravenous thrombolysis at our center undergo CTA to identify the site of intracranial arterial occlusion. High-resolution brain CTAs were performed on a 64- V.K. SHARMA ET AL. Figure 1. Axial maximum intensity projection image showing the ICVs running posteriorly, parallel, and close to each other. Both ICVs show symmetrical opacification (arrows). Abbreviation: ICV, internal cerebral vein. slice multidetector helical scanner (Philips, Inc., USA) in patients without any evidence of intracranial bleeding and no contraindication for CTA (contrast medium allergy or serum creatinine levels. 110 mmol/l). CTA images were acquired with a bolus injection of ml of contrast with a bolus-tracking technique, using a threshold level of 800 Hounsfield units (HU). Scan parameters at our institution were slice thickness, 1 mm; no slice gap; field of view, 200 mm; matrix, ; mas. Coverage was from the base of skull to the vertex. The source images were reformatted into 3-mm-thick axial, coronal, and sagittal projections. Majority of the neurologically and hemodynamically stable patients without contraindications for radiocontrast underwent CTA on day 2 after intravenous tpa to assess the status of arterial patency. ICVs were assessed on the reformatted maximum intensity projection images on axial sections. We defined ICV asymmetry as a unilateral attenuation of the ICV with lower density (reduced filling) than that of the normal side. In this study, ICV asymmetry was assessed on the initial CTA and the day 2 CTA by 2 independent readers, both blinded to patient clinical data or outcome. In the event of any discrepancy, consensus was later achieved by reviewing the scans together. The absolute values for HU of individual ICVs were estimated by placing the cursor probe on the digital imaging and communications in medicine images of CTAs. We calculated the ICV asymmetry index for each CTA (HU of ICV on affected side/hu of contralateral

3 INTERNAL CEREBRAL VEIN ASYMMETRY IN ACUTE ISCHEMIC STROKE ICV). Thus, ICV asymmetry index of 1.0 defined no asymmetry. Information collected about various vascular risk factors included age, gender, hypertension, hyperlipidemia, active smoking, atrial fibrillation, and diabetes mellitus. Data were collected regarding other cardioembolic causes such as recent ischemic heart disease, poor ejection fraction, and presence of intracardiac thrombus. The time of symptom onset and time to intravenous tpa bolus were recorded in all cases. National Institutes of Health Stroke Scale (NIHSS) scores were evaluated just before tpa bolus, at 1 hour after tpa bolus and then 24 hours after the onset of stroke. AIS was classified by its etiopathogenic mechanism into various subtypes: large artery atherosclerosis, cardioembolism, small artery occlusion, stroke of other determined etiology, and stroke of undetermined cause. 5 Symptomatic intracerebral hemorrhage (sich) was defined as the presence of blood on the day 2 CT scan, accompanied by an increase in NIHSS score by 4 or more points. 6 All the patients were treated with intravenous tpa in a standard dose (0.9 mg/kg body weight, maximum 90 mg; 10% as bolus and remaining 90% as infusion over 1 hour). Informed consent was obtained from all the patients or their legally acceptable representatives. In general, we followed the inclusion and exclusion criteria of the National Institute of Neurological Disorders study for selecting candidates for thrombolysis. 7 However, patients were treated up to 270 minutes since symptom onset, and there was no upper age limit (oldest treated patient was 92 years). In addition, we excluded patients with CT evidence of extensive early ischemic change (affecting more than one third of the MCA territory). Functional outcome was assessed by modified Rankin Scale (mrs) at 3 months, dichotomized as good outcome (mrs score 0-1) and poor outcome (mrs score 2-6). Data were analyzed for the early predictors of poor functional outcome. to be significant in the univariate analysis (P,.05) to identify predictors of poor functional outcomes (mrs score 2-6) at 3 months. To maximize sensitivity, those variables with a univariable association of P less than.2 were included as candidates into a multivariable logistic regression model with backward stepwise selection procedure. Associations have been presented as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). Statistical analyses were performed using the SPSS statistical software package (version 19.0 for Windows; SPSS Inc., USA). Results Out of a total of 2238 AIS patients admitted to our tertiary care center during the study period, 240 (11%) were treated with intravenous tpa. We excluded 14 patients from the study: 12 with posterior circulation stroke, 1 that died before the day 2 CTA could be performed, and 1 who underwent decompressive hemicraniectomy for malignant MCA territory infarction before the day 2 CTA. Therefore, data from 226 patients were included in the final analysis. Various demographic characteristics and vascular risk factors in our tpa-treated patients are presented in Table 1. Briefly, the median age of our patients was 65 years (range 19-92) with men comprising 43.8%. Hypertension was the commonest vascular risk factor (in 76.5% cases), whereas atrial fibrillation was seen in 80 (34.5%) patients. Median NIHSS score at presentation was 16 (range 4-32) points. Cardioembolism (44.2%) was the commonest stroke subtype in our study population. Of the 226 patients, 144 (63.7%) underwent CTA before Table 1. Baseline characteristics of study population (n 5 226) Characteristic All (n 5 226) e41 Statistical Analysis Statistical comparisons were performed between patients with and without ICV asymmetry on the day 2 CTA scans in terms of demographic characteristics, stroke risk factors, admission NIHSS scores, time-to-treatment, stroke subtypes, change in NIHSS scores (NIHSS score on day 2 minus pre-tpa NIHSS score), site of pre-tpa intracranial occlusion on the pre-tpa CTA, sich, and the functional outcomes at 3 months. Dichotomous variables were compared with the chi-square test and continuous variables using 2-sample t test and Mann Whitney U test. Initially, univariable analyses of potential predictors (demographic characteristics, stroke risk factors, admission NIHSS score, time-to-treatment, and stroke subtypes) were performed. Multivariable analyses were performed with logistic regression of the variables found Median age, y (range) 65 (19-92) Male gender, n (%) 99 (43.8) Hypertension, n (%) 171 (76.5) Diabetes mellitus, n (%) 66 (29.2) Dyslipidemia, n (%) 129 (57.1) Atrial fibrillation, n (%) 80 (34.5) Smoker, n (%) 58 (25.7) Median NIHSS at presentation 17 (4-35) (range) Time-to-treatment, min (range) 149 (46-270) TOAST classification, n (%) Large artery atherosclerosis 68 (30.1) Cardioembolism 100 (44.2) Small artery occlusion 11 (4.9) Other determined etiology 6 (2.7) Undetermined etiology 41 (18.1) Abbreviations: NIHSS, National Institute of Health Stroke Scale; TOAST, Trial of Org in Acute Stroke Treatment.

4 e42 V.K. SHARMA ET AL. Table 2. Differences in the characteristics of patients with good (mrs score 0-1) and poor functional outcomes at 3 months (mrs score 2-6) Variable Patients with good outcomes (n 5 118) Patients with poor outcomes (n 5 108) P value Median age, y (range) 62 (35-89) 71 (19-92),.0005 Male gender, n (%) 37 (43.5) 47 (45.5).038 Prestroke mrs, median (range) 1 (0-3) 1 (0-4).827 Hypertension, n (%) 86 (72.8) 86 (79.6).196 Diabetes mellitus, n (%) 31 (26.3) 35 (32.4).292 Dyslipidemia, n (%) 69 (58.5) 60 (55.6).772 Atrial fibrillation, n (%) 35 (29.7) 45 (41.7).054 Smoker, n (%) 32 (27.1) 26 (24.1).627 Median NIHSS at presentation (range) 14 (3-35) 20 (3-30),.0005 Median change in NIHSS in 24 h, DNIHSS (range)* 10 (3-27) 4 (-4-11),.0005 Time-to-treatment, min* (range) 146 (46-260) 150 (55-270).255 TOAST classification, n (%).201 Large artery atherosclerosis 37 (31.4) 31 (28.7) Cardioembolism 46 (38.9) 54 (50) Small artery occlusion 8 (6.8) 3 (2.8) Other determined etiology 2 (1.7) 4 (3.7) Undetermined etiology 25 (21.2) 16 (41.8) Site of occlusion on pre-tpa CTA.042 Terminal ICA (%) 9 18 Proximal MCA (%) Distal MCA (%) No detectable occlusion (%) 10 8 Symptomatic intracerebral hemorrhage, n (%) 5 (4.2) 6 (5.6).425 ICV asymmetry on pre-tpa CTA, n (%) 41 (47.1) 46 (52.9).043 ICV asymmetry indexy on pre-tpa CTA, median (range).47 ( ).51 ( ).163 ICV asymmetry on follow-up day 2 CTA, n (%) 2 (8) 23 (92),.005 ICV asymmetry index on day 2 CTA.9 ( ).76 ( ).017 Abbreviations: CTA, computed tomographic angiography; HU, Hounsfield unit; ICV, internal cerebral vein; mrs, modified Rankin Scale; NIHSS, National Institute of Health Stroke Scale; TOAST, Trial of Org in Acute Stroke Treatment; tpa, tissue plasminogen activator. *Change in NIHSS scores (DNIHSS) 5 pre-tpa NIHSS 2 NIHSS score at 24 h. yicv asymmetry index 5 HU score on affected side/hu score of contralateral ICV. initiating thrombolysis and 87 of them (60.4%) were observed to have ICV asymmetry. However, there were no statistically significant differences in the functional outcomes between patients with and without ICV asymmetry on the pre-tpa CTA. On the contrary, ICV asymmetry could be assessed in 103 (45.6%) patients on their day 2 CTA scans. Importantly, these patients did not have any statistically significant differences compared with the group that did not undergo day 2 CTA. Identification of ICV asymmetry was found to have excellent inter- and intraobserver agreement.88 and.92, respectively. The median absolute HU value of ICVon the affected side was 43 (range 17-76) as compared with the normal (contralateral) ICV (median value 88; range units). Data for ICV asymmetry on pre-tpa and on day 2 CTAs were available only for 72 cases. ICV asymmetry of the pre-tpa CTA persisted in 24 (33.3%) cases and 23 of them achieved poor functional recovery at 3 months. On the contrary, ICV asymmetry on pre-tpa CTA disappeared on the day 2 CTA in 48 patients and 35 (72.9%) of them achieved good functional recovery (P,.0005). Factors associated with poor outcomes on univariable analyses were higher age, male gender, higher NIHSS score at presentation, lower change in NIHSS score during first 24 hours, proximal arterial occlusion, ICV asymmetry on admission CTA, ICV asymmetry on day 2 CTA, and ICV asymmetry index on day 2 CTA (Table 2). In addition to these factors, AF was also included in the multivariable logistic regression model. Poor functional outcome at 3 months was independently associated with higher admission NIHSS score (OR 1.073; 95% CI , P 5.046), change in NIHSS score during first 24 hours (OR.737; 95% CI , P,.0001), and ICV asymmetry on follow-up CTA (OR 20.3; 95% CI , P,.0001) (Table 3). Persistence or presence of ICV asymmetry on the follow-up day 2 CTA was not significantly associated with a risk of increased risk for sich.

5 INTERNAL CEREBRAL VEIN ASYMMETRY IN ACUTE ISCHEMIC STROKE Table 3. Predictors of poor outcome at 3 months (modified Rankin scale score 2-6 points) on multivariable analysis e43 Variable OR (95% CI) P value NIHSS (per 1-point increase) ( ).046 DNIHSS (pre-tpa NIHSS score 2 NIHSS score at 24 hours).737 ( ),.0001 Presence of ICV asymmetry on follow-up day 2 CT scan 20.3 ( ),.0001 Abbreviations: CI, confidence interval; CT, computed tomography; ICV, internal cerebral vein; NIHSS, National Institute of Health Stroke Scale; OR, odds ratio. Discussion Our study demonstrates that reduced hemispheric venous drainage, represented by ICV asymmetry in patients with acute anterior circulation ischemic stroke is associated with poor outcome. ICV asymmetry in patients with anterior circulation AIS may serve as a surrogate marker of hemispheric hypoperfusion and aid in planning various interventional strategies during the acute phase and rehabilitative strategies in the long term. To the best of our knowledge, this is the first study that has described the relationship between deep cerebral venous filling and large artery occlusion in AIS. The ICVs are paired midline vessels that arise at the foramina of Monro. They receive tributaries that drain subcortical and periventricular structures and the choroid plexus within the lateral ventricles. ICVs travel posteriorly in the roof of the third ventricle, enter the quadrigeminal plate cistern, and empty into the great cerebral vein of Galen. 8 Thus, the ICVs are responsible for a large proportion of the brain s venous drainage. After systemic thrombolysis, we observed that the persistence of ICV asymmetry on the day 2 follow-up CTA was often seen in patients with persistent occlusion of ICA or MCA and failure of development of efficient collaterals. Furthermore, presence of ICV asymmetry was significantly associated with poor clinical outcome at 3 months (Fig 2, A). Conversely, patients with resolution of the ICV asymmetry on the follow-up day 2 CTA achieved better functional outcome (Fig 2, B). Although, conventional cerebral angiography is the gold standard for assessing the patency of intracranial arteries, various leptomeningeal collaterals, and venous outflow, its applicability is limited by the invasive nature and associated risks. 9 CTA is a noninvasive and rapid tool, often employed in AIS for determining the arterial occlusion and assessing various patterns of collateral circulation in the brain Many large intracranial veins are opacified during CTA and even degrade the image quality in some cases. 13 Our study shows that this inadvertent contaminant leads to consistent visualization of the ICVs and the information from ICV asymmetry because of reduced venous drainage on the side of arterial occlusion can be used for the indirect assessment of hemispheric perfusion. The consistency and reliability of ICVs are evident from the good interobserver agreement in 192 of 224 patients (k 5.85) for their assessment. Interestingly, ICV asymmetry can be studied by other imaging modalities. Power-based transcranial colorcoded duplex sonography with low-flow settings can be used to evaluate the ICVs and quantifying the actual blood flow in the ICVs. 14 ICV asymmetry evaluated on magnetic resonance angiography has been correlated with proximal arterial occlusions and the occurrence of intracranial hemorrhagic transformation after intravenous thrombolysis. 15 We did not find any significant relationship between ICV and intracranial hemorrhage in our study cohort. Certain limitations of our study need to be acknowledged. First, we evaluated ICV asymmetry by visual inspection with variable window settings on CTA images. We believe that signal changes may be analyzed better with automated software with quantitative measurements. Although we evaluated the absolute HU values for each ICV and calculated the ICV asymmetry index on each CTA, these were not found to be statistically significant for being a predictor of functional outcome at 3 months. Perhaps, our smaller sample size accounted for these findings. Second, like other veins in the body, ICVs may differ in their configuration and tributaries, leading to inaccurate interpretation. 16 However, we did not encounter any case with developmental and significant difference in their ICVs. Moreover, any congenital morphologic asymmetry would not change within 24 hours. Third, we accept the inherent problems with the retrospective analysis as ours. Perhaps, a larger prospective study would provide stronger evidence for a relationship between ICV asymmetry and functional outcome in AIS patients after systemic thrombolysis. Fourth, information regarding ICV asymmetry was available in only 103 (45.5%) patients on their day 2 CTA scans. However, the missing data in more than half of the study population are not expected to change our findings because the patients who underwent CTA did not have any statistically significant differences as compared with the group without day 2 CTA. Similarly, persistence or disappearance of ICV asymmetry on the day 2 CTA should convey a stronger message. Although, we found a clear signal of the relationship between persistence of ICV asymmetry and poor functional outcome, we did not evaluate this relationship

6 e44 V.K. SHARMA ET AL. Figure 2. Temporal changes in the appearance of ICVs in a 56-year-old man who presented with acute left hemiparesis. NIHSS score at presentation was 26 points. Arrows show the salient findings on CTA images. CTA (A) showed a filling defect in the intracranial ICA and entire MCA. Intravenous thrombolysis was initiated at 175 minutes after symptom onset that resulted in complete recanalization of right terminal ICA and MCA (B). NIHSS was 4 points at 2 hours. Asymmetry of ICV was obvious on the pre-tpa CTA (C) that resolved completely on the follow-up scan (D). Abbreviations: CTA, computed tomographic angiography; ICV, internal cerebral vein; NIHSS, National Institute of Health Stroke Scale; tpa, tissue plasminogen activator. further because of the smaller number of cases. Finally, our study is limited to patients with Asian ethnicity that is known to have higher prevalence of intracranial stenosis 17 that might have contributed to higher observation of ICV asymmetry and potentially limit the generalizability of our findings. In conclusion, our study shows that the presence of the asymmetry of ICVs on the follow-up day 2 CTA in AIS patients treated with intravenous tpa can be used as an early predictor of poor functional outcome. A larger prospective study including various racial groups is needed to confirm this relationship. References 1. Meder JF, Chiras J, Roland J, et al. Venous territories of the brain. J Neuroradiol 1994;21: Delgado Almandoz JE, Su HS, Schaefer PW, et al. Frequency of adequate contrast opacification of the major intracranial venous structures with CT angiography in the setting of intracerebral hemorrhage: comparison of 16- and 64-section CT angiography techniques. AJNR Am J Neuroradiol 2011;32: Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke 2007;38: Shuaib A, Butcher K, Mohammad AA, et al. Collateral blood vessels in acute ischaemic stroke: a potential therapeutic target. Lancet Neurol 2011;10: Adams HP Jr, Bendixen BH, Kappelle LJ, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org in Acute Stroke Treatment. Stroke 1993; 24: Hacke W, Kaste M, Fieschi C, et al, for the ECASS Study Group. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke: The European Cooperative Acute Stroke Study (ECASS). JAMA 1995;274: The National Institute of Neurological Disorders and Stroke rt-pa Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med 1995; 333: Krayenb uhl H, Yasargil MG. Neuroradiologic diagnosis and surgical therapy of occlusions of brain vessels. Ther Umsch 1968;25: Connors JJ III, Sacks D, Furlan AJ, et al, American Academy of Neurology; American Association of Neurological Surgeons; American Society of Interventional and Therapeutic Neuroradiology; American Society of Neuroradiology; Congress of Neurological Surgeons; AANS/CNS Cerebrovascular Section; Society of Interventional Radiology; NeuroVascular Coalition Writing Group. Training, competency, and credentialing standards for diagnostic cervicocerebral

7 INTERNAL CEREBRAL VEIN ASYMMETRY IN ACUTE ISCHEMIC STROKE e45 angiography, carotid stenting, and cerebrovascular intervention: a joint statement from the American Academy of Neurology, the American Association of Neurological Surgeons, the American Society of Interventional and Therapeutic Neuroradiology, the American Society of Neuroradiology, the Congress of Neurological Surgeons, the AANS/CNS Cerebrovascular Section, and the Society of Interventional Radiology. Neurology 2005;64: Rosenthal ES, Schwamm LH, Roccatagliata L, et al. Role of recanalization in acute stroke outcome: rationale for a CT angiogram-based Benefit of Recanalization model. AJNR Am J Neuroradiol 2008;29: Tan IY, Demchuk AM, Hopyan J, et al. CT angiography clot burden score and collateral score: correlation with clinical and radiologic outcomes in acute middle cerebral artery infarct. AJNR Am J Neuroradiol 2009;30: Maas MB, Lev MH, Ay H, et al. Collateral vessels on CT angiography predict outcome in acute ischemic stroke. Stroke 2009;40: Matsumoto M, Endo Y, Sato M, et al. Acute aneurysm surgery using three dimensional CT angiography without conventional catheter angiography. Fukushima J Med Sci 2002;48: Baumgartner RW, Nirkko AC, Muri RM, et al. Transoccipital power-based color-coded duplex sonography of cerebral sinuses and veins. Stroke 1997;28: Hermier M, Nighoghossian N, Derex L, et al. Hypointense transcerebral veins at T2*-weighted MRI: a marker of hemorrhagic transformation risk in patients treated with intravenous tissue plasminogen activator. J Cereb Blood Flow Metab 2003;23: Banna M, Young JR. Normal anatomical variation and asymmetry of the Galenic venous system. Br J Radiol 1970;43: Wong KS, Ng PW, Tang A, et al. Prevalence of asymptomatic intracranial atherosclerosis in high-risk patients. Neurology 2007;68: