Neurological outcome of patients with aneurysmal

Transcription

1 Early Perfusion Computerized Tomography Imaging as a Radiographic Surrogate for Delayed Cerebral Ischemia and Functional Outcome After Subarachnoid Hemorrhage Nima Etminan, MD*; Kerim Beseoglu, MD*; Hi-Jae Heiroth, MD; Bernd Turowski, MD; Hans Jakob Steiger, MD; Daniel Hänggi, MD Background and Purpose To date, there is no immediate radiographic surrogate to quantify primary cerebral injury to identify patients at risk for delayed cerebral ischemia and poor clinical outcome after aneurysmal subarachnoid hemorrhage. Therefore, we investigated the relation of early cerebral perfusion computerized tomography and clot volume with radiological events of delayed cerebral ischemia and clinical outcome in patients with aneurysmal subarachnoid hemorrhage. Methods Data from 2 cohorts of patients (51 in main, 28 patients in control cohort) with aneurysmal subarachnoid hemorrhage, receiving computerized tomography and perfusion-computerized tomography scanning <12 hours after ictus, were included. A risk group model for functional outcome was developed on the basis of early mean transit time (MTT) and volumetric blood clot measurements. The relation of the risk group model with subsequent MTT, angiographic vasospasm, new cerebral infarction, and functional outcome was analyzed. Actual and predicted functional outcomes based on the risk group model were compared in the control cohort. Results The risk group model correlated significantly with subsequent MTT measurements, cerebral infarction, and functional outcome. Odds for poor outcome were significantly higher in case of concomitant increase of early MTT and clot volumes, as opposed to exclusive early MTT or clot volume increase. For patients in the high- or low-risk groups, neurological outcome in the control cohort correlated significantly with predicted outcomes. Conclusions Assessment of early cerebral perfusion and intracranial blood clot may serve as a radiographic surrogate for delayed cerebral ischemia and functional outcome in patients with aneurysmal subarachnoid hemorrhage using risk group modeling. (Stroke. 2013;44: ) Key Words: CT perfusion delayed cerebral ischemia early brain injury early perfusion computed tomography functional outcome microcirculation subarachnoid hemorrhage vasospasm Neurological outcome of patients with aneurysmal subarachnoid hemorrhage (asah) is predominantly associated with the extent of primary cerebral injury and the incidence of delayed cerebral ischemia (DCI) as well as cerebral infarction. 1,2 Ideally, the medical management of patients with asah would be adapted to a prognostic surrogate, related to the individual patient and the extent of primary cerebral injury. During the past decades, the main grading scales for assessment of primary cerebral injury and prediction of outcome in patients with asah were the clinical World Federation of Neurosurgical Societies (WFNS) grading scale and the radiological Fisher score, based on computerized tomography (CT). 3,4 These historical grading scales, however, may be too insensitive as a surrogate for DCI and neurological outcome, especially because of interobserver variability and improved CT imaging quality. 5 Nevertheless, in clinical practice, an early radiological surrogate for primary cerebral injury to predict the subsequent extent of DCI and functional outcome has not yet been established. Data from animal and clinical studies strongly suggest that primary brain injury after asah is mainly reflected in early microvascular perfusion alterations, despite physiological cerebral perfusion or intracranial pressures. 6,7 After experimental SAH, these microcirculatory alterations were reported to exacerbate with increasing amounts of subarachnoid blood clot. Assuming, Received September 5, 2012; accepted February 22, From the Department of Neurosurgery (N.E., K.B. H.H, H.-J.S., D.H.) and Institute for Diagnostic and Interventional Radiology (B.T.), Medical Faculty, Heinrich-Heine-University, Düsseldorf, Germany. *Drs Etminan and Beseoglu contributed equally to this article. The online-only Data Supplement is available with this article at /-/DC1. Correspondence to Daniel Hänggi, MD, Department of Neurosurgery, Medical Faculty, Heinrich-Heine-University, Moorenstraße 5, Düsseldorf, Germany. daniel.haenggi@uni-duesseldorf.de 2013 American Heart Association, Inc. Stroke is available at DOI: /STROKEAHA

2 Etminan et al Early PCT and SAH 1261 then, that these alterations may be radiologically detected using early perfusion CT scanning and intracranial blood clot measurements, we investigated a potential association with radiographic events for DCI and clinical outcome in patients with asah. Methods Patient Population Out of a total of 153 patients admitted to our department between April 2010 and October 2011 because of asah, the present retrospective study analyzed data from a (main) cohort of 51 patients meeting the following selection criteria: (1) asah present on initial nonenhanced CT, applicable for volumetric measurements; (2) perfusion-ct (PCT) scanning time <12 hours after SAH ictus; and (3) no early angiographic vasospasm on admission demonstrated by digital subtraction angiography. Patients with unknown onset of SAH as well as patients with insufficient cerebral perfusion pressures (<65 mm Hg) attributable to intractable primary increase in intracranial pressures on admission were excluded from this analysis. A control cohort of 28 patients out of a total of 82 patients admitted to our department because of asah between November 2011 and September 2012, with identical selection criteria, was used to test the validity of the below described risk group model. In addition to consent from the patient or next of kin for data analysis, scientific use of the clinical data were approved by the local ethics committee of the Medical Faculty of the Heinrich-Heine University, Düsseldorf, Germany (study ID: 3573). The management of patients in the presented cohorts was in accordance with current SAH guidelines. 8 Patients with a Glasgow Coma Score <12 received an external ventricular drain for intracranial pressure control. In case of moderate-to-severe angiographic vasospasm, medical management consisted of induced hypertension and euvolemia. In case of refractory angiographic vasospasm, rescue therapy was initiated using endovascular intra-arterial infusion of nimodipine or balloon angioplasty (for severe proximal vasospasm refractory to intra-arterial nimodipine application). SAH Screening Protocol and PCT Methods All patients admitted to our department are subjected to a standardized PCT screening protocol. In addition to digital subtraction angiography on admission and on day 7 after SAH ictus, PCT measurements were routinely performed immediately on admission, 6 to 12 hours after aneurysm treatment, on days 3 to 4 as well as 9 to 11 after SAH ictus or in case of secondary neurological deterioration, that is, occurrence of clinical features of DCI. 9 PCT data were calculated using singular-value decomposition with an acquisition time of 35 seconds. The automated image analysis integrated a 1 cm cortical layer with 180 overlapping regions of interests and corresponding values, using 360 cortical banding analysis (Figure 1). After generation of perfusion maps, PCT data, such as mean transit time (MTT), time to peak of the residue function (T max ), cerebral blood flow, and cerebral blood volume were calculated using the STROKETOOL-CT software (Version 2.5, DigitalImageSolutions, Frechen, Germany). 10 Only MTT values were used for correlation with radiological events of DCI and poor outcome. The individual 180 (90 per hemisphere) MTT values of each PCT scan were pooled and expressed as mean values. Definitions Intracranial clot volume was calculated based on all pixels indicative for acute hemorrhage (60 90 Hounsfield units), on initial CT scan during admission. 11 Thus, the full subarachnoid clot as well as its intracerebral, intraventricular, or even subdural extension was included. Clot volume was measured using 3-dimensional volume rendering (OsiriX Version 3.7.1, Pixmeo SARL, Bernex, Switzerland) based on the initial CT scan on admission. Clot volumes for basal (2.5 mm slice thickness) and cranial slices (5 mm slice thickness) were added and expressed in ml. Early MTT was defined as MTT scanning time measured within 12 hours after asah ictus. Subsequent or follow-up MTT was defined as the pooled MTT data between day 1 and day 14 from each individual patient, excluding the early MTT measurements. Although absolute MTT threshold values may differ from previous studies, a 1.5- fold prolongation of reference MTT values from healthy patients was considered suggestive for DCI; this corresponded, according to our institutional hard- and software calibration, to an absolute MTT threshold of 4.2 seconds (normal range, seconds). 12 Angiographic vasospasm was defined as narrowing of the arterial diameter of >30% from baseline. New cerebral infarction was diagnosed and defined as the presence of cerebral infarction on CT scan within 6 weeks after SAH, or on the latest CT made before death within 6 weeks and resulting from DCI when not attributable to other causes, such as aneurysm treatment or intraparenchymal hematoma. 9 Functional outcome, as measured by Glasgow Outcome Score (GOS), was assessed at discharge and after 6 weeks and dichotomized in poor (GOS, 1 3) and good neurological outcomes (GOS, 4 5). Assessment of radiographic and neurological outcome measures was performed by investigators (N.E., H.H.) blinded to clot volume and CT perfusion measurements. Statistical Analysis A logistic regression model with stepwise backward exclusion of the following variables was used to determine their influence on dichotomized functional outcome: patient age, sex, initial WFNS grade, Fisher score, early MTT, and clot volume. Predictive values of early MTT and clot volume for functional outcome were analyzed using receiver operator characteristics analysis. The area under the curve was calculated to determine the cutoff values based on highest sensitivity and specificity for outcome prediction. Based on these cutoff values for early MTT and clot volume, patients were categorized into 3 different risk groups: (1) low risk with low-early MTT values and low-clot volumes (2) intermediate risk with low-early MTT and highclot volumes or high-early MTT values and low-clot volumes, and (3) high risk with high-early MTT and high-clot volumes. The correlation of this risk group allocation with subsequent MTT values, incidence of angiographic vasospasm, and new cerebral infarction, as well as functional outcome at discharge and at 6 weeks, were calculated. Odds ratios, likelihood ratios, sensitivity, and specificity for new cerebral infarction, and poor outcome after 6 weeks (poor GOS) were calculated for the different risk groups. Additionally, the initial WFNS and Fisher grading scores were correlated with incidence of cerebral infarction and functional outcome at 6 weeks. Finally, outcome prediction was tested in the control cohort using the below described risk group model derived from the main cohort, by correlating actual clinical outcomes and projected outcomes, based on early MTT and clot volume risk group allocation. To evaluate the quality of the risk group model, a logistic regression analysis with the patient data from the control cohort was done. Spearman correlation for ordinal variables and Pearson product-moment correlation were used for metric or continuous variables. According to scale levels either Student t test or nonparametric tests were used to determine differences between groups. Significance was accepted at a level of P All statistic analysis was performed using SPSS (Ulead Technologies, Chicago, IL). Results All 51 patients received CT and PCT scanning <12 hours (mean 6±3.5 hours) after SAH ictus as well as before treatment of the ruptured aneurysm. Mean patient age was 54.8 ± 12.0 years (mean±sd), and mean intracranial clot volume was 48.3±44.8 ml in the cohort (Table 1). In addition to the initial WFNS grade, the logistic regression model identified both intracranial clot volume and early MTT as significant and independent contributors to predict dichotomized outcome (Nagelkerke R²=0.842). The receiver operator characteristics curves designed to determine the cutoff values for early MTT and clot volume with the highest sensitivity/specificity analysis for functional outcome prediction were used to define the following risk groups: (1) low risk with early MTT <4.2

3 1262 Stroke May 2013 Figure 1. Exemplary illustration of the perfusion-computerized tomography (PCT) algorithm for delayed cerebral ischemia (DCI) monitoring in a patient with subarachnoid hemorrhage (SAH). A, The axial PCT scanning level was set through the medial cella of the lateral ventricles, parallel to the orbito-meatal plain, thus representing 6 supratentorial vessel territories. Within the cortical band (green), every 2 a region of interest was defined in a clockwise manner. B, Semiquantitative perfusion maps were generated for mean transit time (MTT), time to peak of the residue function (Tmax), relative cerebral blood flow (rcbf), and cerebral blood volume (rcbv). C, All 180 values for each perfusion VALUE were integrated. D, Mean values from each individual PCT measurement were used to monitor for DCI during the observational period until day 11 after SAH ictus. ACA indicates anterior cerebral artery; ABZ, anterior border zone; MCA, middle cerebral artery; PBZ, posterior border zone; and PCA, posterior cerebral artery. *Uncalibrated volume data. seconds and clot volume <50 ml (n=27), (2) intermediate risk with early MTT <4.2 seconds and clot volume >50 ml or early MTT >4.2 seconds and clot volume <50 ml (n=13), and (3) high risk with early MTT >4.2 seconds and clot volume >50 ml (n=11). Risk Group Analyses There was a strong correlation of risk group allocation, derived from early PCT and clot volume measurements, with radiographic and neurological outcome events, as compared with the WFNS and Fisher grading scores in the cohort (Figure 2). Furthermore, there was a significant difference in neurological outcome, as measured by GOS after 6 weeks, between the 3 risk groups (Figure 3). Additionally, patients within the highrisk group demonstrated significantly greater odds to have cerebral infarctions and poor outcome at 6 weeks (Table 2). Interestingly, there was no difference in the incidence of angiographic vasospasm between the risk groups (P=0.733). As illustrated in the Table in the online-only Data Supplement, projected and actual outcomes of patients in the low- or high-risk group correlated significantly (R2=0.584; P<0.001) within the control cohort. Because clinical outcome in patients classified into the intermediate risk group (n=8) was not clearly associated with good or poor functional outcome according to the risk group model, outcome for these patients was defined as unknown. The logistic regression model for the control cohort confirmed the overall validity of the risk group model (Nagelkerke R2=0.654; P<0.0001). Discussion The main findings of this retrospective analysis in patients with asah are the following: (1) the risk group model correlated significantly with subsequent MTT measurements, cerebral infarction, and functional outcome, (2) odds for poor outcome were significantly higher in case of concomitant increase of early MTT and clot volumes, as opposed to exclusive early

4 Etminan et al Early PCT and SAH 1263 Table 1. Summary of Initial Neurological (WFNS) and Radiological (Fisher) Grade as well as Incidence of Radiographic Features for DCI (Angiographic VSP, New Cerebral Infarction) and Functional Outcome (GOS at Discharge and at 6 wk) N (%) WFNS I III 24 (47) IV V 27 (53) Fisher 0 II 8 (15.7) III IV 43 (84.3) Aneurysm location Anterior 39 (76.5) Posterior 12 (23.5) Aneurysm treatment Surgical 25 (49.0) Endovascular 24 (47.1) None 2* (3.9) Angiographic VSP No 18 (35.3) Yes 33 (64.7) New cerebral infarction No 36 (70.6) Yes 15 (29.4) GOS d/c Poor (1 3) 31* (60.8) Good (4 5) 20 (39.2) GOS 6 weeks Poor (1 3) 18* (35.3) Good (4 5) 33 (64.7) Angio VSP indicates angiographic vasospasm; DCI, delayed cerebral ischemia; GOS, Glasgow Outcome Score; GOS d/c, GOS at discharge; and WFNS, World Federation of Neurosurgical Societies. *Two patients with severe subarachnoid hemorrhage died (1 from acute cardiac failure, 1 from aneurysm rerupture) within 48 hours after admission before aneurysm treatment. MTT or clot volume increase, and (3) for patients in the highor low-risk groups, neurological outcome in the control cohort correlated significantly with predicted outcomes. Perfusion CT during the course of asah is increasingly thought to provide a more comprehensive insight on pathological macrocirculatory and microcirculatory alterations, as opposed to sole angiographic assessment of cerebral arteries Nevertheless, the role of early cerebral perfusion imaging for prediction of the subsequent extent of secondary cerebral injury, thus neurological outcome after SAH remains unclear. In general, previous studies within this context partially support our findings but (1) used heterogeneous definition of early PCT scanning, hours after SAH ictus, (2) solely investigated poor grade SAH patients, WFNS IV and V, (3) reported on data derived from smaller patient cohorts and, importantly, (4) did not include clot volume measurements in early radiographic risk prediction. 7,16 19 It is known that (1) 10% to 20% of patients may present with early angiographic vasospasm during 24 hours after SAH ictus and (2) poor grade SAH is strongly associated with poor outcome, per se. 2,20 However, such studies should particularly address the early and more sensitive identification of those patients with SAH, with initial good clinical grades, who are at increased risk for subsequent DCI and poor outcome. Nevertheless, the largest study on 75 patients, with PCT scanning within 72 hours after SAH ictus, reported an association with the subsequent incidence of DCI but not with cerebral infarction or neurological outcome. 18 Despite the prolonged PCT imaging time-span of up to 72 hours, these data partially underscore our findings, as exclusive early MTT prolongation or high-clot volumes (intermediate risk group) seemed to have less severe impact on outcome. Thus, our study is the first to describe the additive impact of concomitant early PCT alterations and intracranial clot volumes on radiological events of DCI and poor neurological outcome during the course of SAH based on risk group allocation. There are different explanations for our results. Most likely, patients with early MTT prolongation and increased clot volumes (high risk group) in the absence of large vessel spasm have early microcirculatory disturbances, which additively results in an increased predisposition to secondary ischemic damage and poor functional outcome during the course of SAH. Early and delayed microcirculatory dysfunction after experimental or aneurysmal SAH is increasingly thought to be associated with the incidence of DCI, in the absence of macrovascular spasm or changes in cerebral perfusion pressure. 6,21 One possible pathomechanism for this phenomenon might be endothelin-1 mediated vasoconstriction of microvessels. 22 Alternatively, endothelial dysfunction may also impose an increased vulnerability for DCI in patients early after SAH. 22,23 Furthermore, the increased incidence of DCI after concomitant early perfusion impairment and increased subarachnoid blood clot (high-risk group) in our study may reflect an exacerbation of SAH-related endothelial dysfunction. 24 Finally, our results reflect recent studies on the dissociation of angiographic vasospasm and poor neurological outcome and support the assumption that cerebral infarction might reflect the ultimate end point of different, proischemic pathomechanisms (eg, microthromboembolism, microvascular spasm, cortical spreading ischemia, impaired autoregulation). 2,25,26 We acknowledge several limitations of our study. First, CT perfusion screening may only be applied in patients with normal renal function. In the present study, 3 out of the 153 patients in the main cohort, as well as 1 out of 82 patients in the control cohort, were not included to standard PCT imaging because of renal insufficiency. Furthermore, aside from the radiation exposure, ideally, diffusion-weighted MRI would be more sensitive than CT scanning for detection of ischemic lesions after asah. 27 However, aside from its impractical use for intensive care unit patients, the nonlinear relation of contrast-agent concentration and signal intensity in MR diffusion/perfusion may result in intraindividual and interindividual data bias, thus not permitting the establishment of MRI screening protocols or thresholds. Second, because of its high sensitivity/specificity, we solely focused on MTT as the main PCT parameter in our analysis. Nevertheless, the

5 1264 Stroke May 2013 Figure 2. Correlation analysis of risk group allocation (low, intermediate, high), based on early mean transit time (MTT) and clot volumes, with radiological events for secondary brain injury (angiographic vasospasm, follow-up MTT measurements, and new cerebral infarction) as well as neurological outcome (Glasgow Outcome Score at discharge and after 6 weeks). Initial World Federation of Neurosurgical Societies (WFNS) and Fisher grading scores on admission were also correlated with cerebral infarction and functional outcome at weeks. Angio VSP indicates angiographic vasospasm; f/u MTT, follow-up MTT, GOS d/c, Glasgow Outcome Score at discharge; and SAH, subarachnoid hemorrhage. **P< presented absolute MTT cutoff values may differ from other studies and might not be directly attributable to other PCT scanners or algorithms. Considering cutoff values at 1.5-fold prolongation of reference MTT, our results are well in line with those from other studies. 12 Third, we excluded patients with early macrovascular spasm and primary intractable increase of intracranial pressure, thus eliminating patients with primary insufficient cerebral perfusion pressure from our analysis. However, as in previous studies, we defined our selection criteria to further elucidate the sequence of primary brain injury, early microcirculatory alterations, DCI, and functional outcome. 7,21,28 Fourth, our softwarebased calculation of intracranial blood volumes did not permit further differentiation or subanalysis of subarachnoid, intraventricular or intraparenchymal blood clot components. However, the benefit from Hounsfield-unit derived blood clot definition as opposed to manual inclusion and exclusion of different blood clot components is less interobserver bias. Ultimately, the role of risk prediction in patients with exclusive increase of early MTT or blood clot remains to be elucidated. In summary, risk group allocation, derived from early MTT and intracranial blood clot measurements correlated significantly with subsequent MTT measurements, incidence of new cerebral infarction, and poor functional outcome. Our data suggest that acute PCT disturbances may reflect early microcirculatory impairment, which is significantly amplified with concomitant increase in clot volumes, predisposing patients for subsequent ischemic damage. Therefore, early MTT and intracranial blood clot might serve for assessment of primary Figure 3. Dichotomized (good vs poor) functional outcome GOS at 6 weeks between the different risk groups (low, intermediate, high). GOS 1 to 3 was defined as poor, GOS 4 to 5 was defined as good outcome. GOS indicates Glasgow Outcome Score; and MTT, mean transit time.

6 Etminan et al Early PCT and SAH 1265 Table 2. OR and 95% CI, Likelihood Ratios, Sensitivity, and Specificity of Intermediate and High-Risk Groups to Have New Cerebral Infarction and Poor Outcome After 6 Weeks (Poor GOS) in Relation to the Low-Risk Group as Well New CerebraI Infarction Poor GOS at 6 Weeks Low-risk group 1 1 Intermediate risk group OR ( ) OR ( ) Positive likelihood ratio Negative likelihood ratio Sensitivity % Sensitivity % Specificity % Specificity % High-risk group OR ( ) OR ( ) Positive likelihood ratio Positive likelihood ratio Sensitivity 81.82% Sensitivity % Specificity 85.00% Specificity 82.50% WFNS grade OR ( ) OR ( ) Positive likelihood ratio Positive likelihood ratio Sensitivity 40.74% Sensitivity 48.14% Specificity 83.33% Specificity 79.16% Fisher grade OR ( ) OR ( ) Positive likelihood ratio Positive likelihood ratio Sensitivity 39.28% Sensitivity 57.14% Specificity 82.61% Specificity 91.30% CI indicates confidence interval; GOS, Glasgow Outcome Score; OR, odds ratios; and WFNS, World Federation of Neurosurgical Societies. cerebral injury for prediction of the extent of DCI and poor neurological outcome. Disclosures Dr Etminan received grant support from the Physicians Services Incorporated Foundation. Dr Hänggi is a scientific advisor for Edge Therapeutics, Inc. The other authors have no conflicts to report. References 1. Fergusen S, Macdonald RL. Predictors of cerebral infarction in patients with aneurysmal subarachnoid hemorrhage. Neurosurgery. 2007;60: ; discussion Vergouwen MD, Ilodigwe D, Macdonald RL. Cerebral infarction after subarachnoid hemorrhage contributes to poor outcome by vasospasmdependent and -independent effects. Stroke. 2011;42: Report of world federation of neurological surgeons committee on a universal subarachnoid hemorrhage grading scale. J Neurosurg. 1988;68: Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery. 1980;6: Rosen DS, Macdonald RL. Subarachnoid hemorrhage grading scales: a systematic review. Neurocrit Care. 2005;2: Friedrich B, Müller F, Feiler S, Schöller K, Plesnila N. Experimental subarachnoid hemorrhage causes early and long-lasting microarterial constriction and microthrombosis: an in-vivo microscopy study. J Cereb Blood Flow Metab. 2012;32: Schubert GA, Seiz M, Hegewald AA, Manville J, Thomé C. Acute hypoperfusion immediately after subarachnoid hemorrhage: a xenon contrastenhanced CT study. J Neurotrauma. 2009;26: Connolly ES Jr, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, et al; American Heart Association Stroke Council; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; Council on Cardiovascular Surgery and Anesthesia; Council on Clinical Cardiology. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2012;43: Vergouwen MD, Vermeulen M, van Gijn J, Rinkel GJ, Wijdicks EF, Muizelaar JP, et al. Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke. 2010;41: Wittsack HJ, Wohlschläger AM, Ritzl EK, Kleiser R, Cohnen M, Seitz RJ, et al. CT-perfusion imaging of the human brain: advanced deconvolution analysis using circulant singular value decomposition. Comput Med Imaging Graph. 2008;32: Hänggi D, Liersch J, Turowski B, Yong M, Steiger HJ. The effect of lumboventricular lavage and simultaneous low-frequency head-motion therapy after severe subarachnoid hemorrhage: results of a single center prospective Phase II trial. J Neurosurg. 2008;108: Wintermark M, Ko NU, Smith WS, Liu S, Higashida RT, Dillon WP. Vasospasm after subarachnoid hemorrhage: utility of perfusion CT and CT angiography on diagnosis and management. AJNR Am J Neuroradiol. 2006;27: Chaudhary SR, Ko N, Dillon WP, Yu MB, Liu S, Criqui GI, et al. Prospective evaluation of multidetector-row CT angiography for the diagnosis of vasospasm following subarachnoid hemorrhage: a comparison with digital subtraction angiography. Cerebrovasc Dis. 2008;25: Sanelli PC, Ugorec I, Johnson CE, Tan J, Segal AZ, Fink M, et al. Using quantitative CT perfusion for evaluation of delayed cerebral ischemia following aneurysmal subarachnoid hemorrhage. AJNR Am J Neuroradiol. 2011;32: Washington CW, Zipfel GJ; Participants in the International Multidisciplinary Consensus Conference on the Critical Care Management of Subarachnoid Hemorrhage. Detection and monitoring of vasospasm and delayed cerebral ischemia: a review and assessment of the literature. Neurocrit Care. 2011;15: Kamp MA, Heiroth HJ, Beseoglu K, Turowski B, Steiger HJ, Hänggi D. Early CT perfusion measurement after aneurysmal subarachnoid hemorrhage: a screening method to predict outcome? Acta Neurochir Suppl. 2012;114: Lagares A, Cicuendez M, Ramos A, Salvador E, Alen JF, Kaen A, et al. Acute perfusion changes after spontaneous SAH: a perfusion CT study. Acta Neurochir (Wien). 2012;154: ; discussion Sanelli PC, Jou A, Gold R, Reichman M, Greenberg E, John M, et al. Using CT perfusion during the early baseline period in aneurysmal

7 1266 Stroke May 2013 subarachnoid hemorrhage to assess for development of vasospasm. Neuroradiology. 2011;53: Tsuang FY, Chen JY, Lee CW, Li CH, Lee JE, Lai DM, et al. Risk profile of patients with poor-grade aneurysmal subarachnoid hemorrhage using early perfusion computed tomography. World Neurosurg. 2012;78: Baldwin ME, Macdonald RL, Huo D, Novakovic RL, Goldenberg FD, Frank JI, et al. Early vasospasm on admission angiography in patients with aneurysmal subarachnoid hemorrhage is a predictor for in-hospital complications and poor outcome. Stroke. 2004;35: Uhl E, Lehmberg J, Steiger HJ, Messmer K. Intraoperative detection of early microvasospasm in patients with subarachnoid hemorrhage by using orthogonal polarization spectral imaging. Neurosurgery. 2003;52: ; discussion Park KW, Metais C, Dai HB, Comunale ME, Sellke FW. Microvascular endothelial dysfunction and its mechanism in a rat model of subarachnoid hemorrhage. Anesth Analg. 2001;92: Hatake K, Wakabayashi I, Kakishita E, Hishida S. Impairment of endothelium-dependent relaxation in human basilar artery after subarachnoid hemorrhage. Stroke. 1992;23: ; discussion Nagatani K, Masciopinto JE, Letarte PB, Haworth RA, Duff TA. The effect of hemoglobin and its metabolites on energy metabolism in cultured cerebrovascular smooth-muscle cells. J Neurosurg. 1995;82: Etminan N, Vergouwen MD, Ilodigwe D, Macdonald RL. Effect of pharmaceutical treatment on vasospasm, delayed cerebral ischemia, and clinical outcome in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab. 2011;31: Vergouwen MD, Etminan N, Ilodigwe D, Macdonald RL. Lower incidence of cerebral infarction correlates with improved functional outcome after aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab. 2011;31: Ibrahim GM, Weidauer S, Vatter H, Raabe A, Macdonald RL. Attributing hypodensities on CT to angiographic vasospasm is not sensitive and unreliable. Stroke. 2012;43: Ohkuma H, Manabe H, Tanaka M, Suzuki S. Impact of cerebral microcirculatory changes on cerebral blood flow during cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke. 2000;31: