2D Cine Phase-Contrast MRI for Volume Flow Evaluation of the Brain-Supplying Circulation in Moyamoya Disease

Transcription

1 MRI in Moyamoya Disease Neuroradiology Original Research A C D E M N E U T R Y L I A M C A I G O F I N G K. Wolfgang Neff 1 Peter Horn 2 Peter Schmiedek 2 Christoph Düber 1 Dietmar J. Dinter 1 Neff KW, Horn P, Schmiedek P, Düber C, Dinter DJ Keywords: cardiovascular disease, cine MRI, hemodynamics, MR angiography, neuroimaging DOI:1.2214/AJR Received February 9, 25; accepted after revision April 25, Department of Clinical Radiology, University of Heidelberg, Universitätsklinikum Mannheim, Theodor-Kutzer Ufer 1 3, Mannheim, Germany. Address correspondence to K. W. Neff (wolfgang.neff@rad.ma.uni-heidelberg.de). 2 Department of Neurosurgery, University of Heidelberg, Universitätsklinikum Mannheim, Mannheim, Germany. WEB This is a Web exclusive article. AJR 26; 187:W17 W X/6/1871 W17 American Roentgen Ray Society 2D Cine Phase-Contrast MRI for Volume Flow Evaluation of the Brain-Supplying Circulation in Moyamoya Disease OBJECTIVE. The purpose of this study was to evaluate and quantify hemodynamic compromise in patients with moyamoya disease by measuring blood volume flow in the brain-supplying arteries. SUBJECTS AND METHODS. Thirty-five patients with angiographically proven moyamoya disease (31 women, 4 men; mean age, 39.4 ± 12.2 years; range, years; adult moyamoya disease) and 15 age-matched healthy controls were examined prospectively with 2D cine phase-contrast MRI. Blood volume flow was measured in both common carotid arteries (CCAs), both internal carotid arteries (ICAs), and the basilar artery. The diagnosis of moyamoya disease was based on results of selective intraarterial digital subtraction angiography. RESULTS. Blood volume flow of the brain-supplying arteries in age-matched controls was ± 47.9 ml/min for the CCA, ± 25.3 ml/min for the ICA, and ± 13.2 ml/min for the basilar artery. Patients with bilateral moyamoya disease had decreased mean blood flow in the CCA (39.4 ± 89.9 ml/min) and ICA (117.9 ± 64. ml/min) and increased blood volume flow in the basilar artery (433.7 ± ml/min). CONCLUSION. Moyamoya disease causes a significant decrease in carotid artery circulation, particularly ICA blood volume flow, with a compensatory increase in blood flow in the basilar artery to nearly 2.5 times normal basilar artery blood flow. 2D cine phase-contrast MRI with measurement of blood volume flow in the brain-supplying arteries is useful in the initial evaluation of moyamoya disease and in continuing assessment of hemodynamics in patients with this disease. oyamoya disease is a specific, M chronic, and rare cerebrovascular disease of unknown origin. It is predominantly seen in Japan and was first reported by Japanese surgeons in 1957; however, the occurrence of this disease in other ethnic groups is being reported [1 4] with increasing frequency. The disease is generally characterized by stenosis and even occlusion of the terminal portions of the internal carotid arteries (ICAs) and the middle and anterior cerebral arteries and by the development of an abnormal vascular network with parenchymal, leptomeningeal, and transdural collateral vessels that supply the ischemic basal ganglia and brain [5, 6]. Chronic ischemia leads to extensive development of collateral vessels that results in the characteristic moyamoya ( puff of smoke in Japanese) appearance at the level of the basal ganglia. These collaterals involve the thalamoperforate, and anterior and posterior choroidal, lenticulostriate arteries, and transdural external to internal carotid anastomoses develop from the middle meningeal, internal maxillary, and other branches of the external carotid artery [2, 7]. The main feature of moyamoya disease is progressive occlusion of the intracranial ICAs and the proximal portions of the anterior and middle cerebral arteries [7]. In the form of moyamoya disease seen in Japan, the clinical features are transient ischemic attacks and cerebral infarction in children and a preponderance of hemorrhagic stroke in adults. In whites, ischemic rather than hemorrhagic stroke predominates in adult cases [8]. Although the disease is distributed among all age groups, there are two peaks, one in early childhood and another at approximately 4 years old. The highest peak occurs among children younger than 1 years old [9]. Among adults, moyamoya disease is included in the so-called moyamoya syndrome or phenomenon, which has a wide differential diagnosis. The presence of moyamoya disease is presumed only by exclusion of other causes of angiographic signs of moyamoya phenomenon, such as atherosclerosis, collagen vascu- AJR:187, July 26 W17

2 A C Fig year-old woman with bilateral moyamoya disease. A and B, Right internal carotid artery (ICA) arteriogram in frontal (A) and lateral (B) projections shows severely stenosed ICA and thus middle cerebral and anterior cerebral arteries are highly stenosed. Marked moyamoya vessels at level of basal ganglia are evident. Peripheral branches of middle and anterior cerebral arteries are filled via collateral vessels. C and D, Left ICA arteriogram in frontal (C) and lateral (D) projections reveals distal ICA stenosis, middle cerebral artery occlusion, and anterior cerebral artery stenosis. Basal cerebral moyamoya vessels are evident. Peripheral branches of left middle and anterior cerebral arteries are delineated by collateral vessels. (Fig. 1 continues on next page) B D lar disease, infection, other inflammatory disorders, and dissection. Ischemic symptoms usually are caused by hemodynamic perfusion failure rather than thromboembolism. MR flow quantification with phase-contrast techniques has proved accurate in both in W18 AJR:187, July 26

3 MRI in Moyamoya Disease vitro and in vivo evaluation of blood flow velocity and volumetric flow rate [1 12]. This noninvasive technique has been used to obtain information about hemodynamic compromise in patients with ICA stenosis or occlusion [13 16] and even in patients with bilateral ICA occlusion. Except for a few studies in which cerebral diffusion and perfusion were evaluated and regional cerebral blood flow and regional cerebral blood volume were measured in moyamoya disease [6, 17, 18], to our knowledge the quantitative changes in cerebropetal blood flow in large brain-supplying arteries (macrocirculation) in patients with moyamoya disease have not been described. The aims of our study were to determine the presence of and quantify blood volume flow of the brain-supplying circulation (common carotid artery [CCA], internal carotid artery [ICA], and basilar artery) in patients with moyamoya disease. In a prospective study, we measured blood volume flow in both CCAs, both ICAs, and the basilar artery in patients with angiographically proven moyamoya disease to evaluate intraindividual and interindividual hemodynamic compromise. We compared the findings with those for age-matched healthy controls. E Fig. 1 (continued) 31-year-old woman with bilateral moyamoya disease. E and F, Left vertebral arteriogram in frontal (E) and lateral (F) projections shows collateralization from posterior to anterior circulation with filling of middle and anterior cerebral artery peripheral branches via developed leptomeningeal collaterals. Subjects and Methods Study Subjects From March 1997 to June 22, 35 consecutively evaluated patients with moyamoya disease confirmed by cerebral selective digital subtraction angiography participated in a prospective study of MR flow quantification. All patients had a history of transient ischemic attacks or cerebral infarction and were referred to the department of neurosurgery for evaluation for cerebral revascularization surgery. The 31 women and four men had an age range of years (mean, 39.4 ± 12.2 years), and all patients were considered to have adult moyamoya disease. The patients were screened for other causes of angiographic presentation of moyamoya phenomenon, including diabetes, severe hypertension, heavy smoking, or a combination of these factors as well as high risk of intracranial atherosclerosis and evidence of collagen disease or inflammatory disease. If none of these possible underlying diseases was found, the diagnosis of moyamoya disease was established. None of the patients had collateral flow through the ophthalmic artery or leptomeningeal vessels, according to findings on conventional angiography. Fifteen age-matched healthy volunteers who had no history of ischemic neurologic deficits and had normal MRI and MR angiography findings served as controls for blood volume flow measurements. Angiography All 35 patients underwent selective cerebral angiography of the cerebropetal vessels by intraarterial digital subtraction angiography. This examination included selective bilateral internal and external carotid artery arteriography and unilateral or bilateral vertebral artery arteriography through a standard femoral artery approach (Integra system, Philips Medical Systems). Vessels were shown in at least two projections, frontal and lateral (Fig. 1). Angiography was performed no more than 1 week before the MR examination. MR Flow Quantification All MRI and blood volume flow measurements were obtained with a 1.5-T MR unit (Magnetom Vision, Siemens Medical Solutions) with a circular polarized head coil and a Helmholtz neck coil. The MRI protocols were identical for patients and control subjects. T1-weighted scout images were acquired (TR/TE, 545/15; slice thickness, 4 mm) for anatomic reference information. 2D cine phase-contrast MRI was used for quantitative assessment of volumetric flow rate in the CCAs, ICAs, and basilar artery (Fig. 2). An ECG-triggered fast low-angle shot gradient-echo sequence with the following sequence parameters was used: 28/5; flip angle, 3 ; field of view, 22 mm; matrix, ; number of acquisitions, 1. The total examination time was 15 2 minutes. F AJR:187, July 26 W19

4 A C Fig year-old woman with bilateral moyamoya disease depicted in Figure 1. Single sections of 2D cine phase-contrast MR images show levels at which measurements were obtained for determination of blood volume flow. A and B, Anatomic reference image (A) and single 2D cine phase-contrast image (B) show both common carotid arteries (arrows) 2 3 mm proximal to carotid artery bifurcation. C and D, Anatomic reference image (C) and single 2D cine phase-contrast image (D) show both internal carotid arteries (arrows) at at level of C3 segment. (Fig. 2 continues on next page) B D To avoid aliasing, velocity encoding was set between 4 and 25 cm/s, depending on blood flow velocity in the vessel in question (Fig. 3) [1, 12 14]. Depending on the patient s heart rate, single 2D velocity-encoded phase images were acquired per cardiac cycle to obtain a time resolution of 28 milliseconds. Blood flow was always measured perpendicular to the course of the arteries being studied. Sections were positioned 2 3 mm proximal to the W11 AJR:187, July 26

5 MRI in Moyamoya Disease carotid artery bifurcation for the CCA measurement and through the C3 segment of the ICA for ICA blood flow. In the basilar artery, flow was determined between the origins of the anterior inferior cerebellar artery and the superior cerebellar artery. All flow data were obtained by integration of defined regions of interest (ROIs) nearly matching the lumen of the vessel being investigated. For assessment of maximal peak systolic velocity, an ROI was defined in the area of the maximal intraluminal signal intensity, usually in the middle of the vessel lumen. For the volumetric flow and velocity quantification program, the area of the vessel lumen was carefully evaluated as the ROI at the peak systolic phase image. This ROI was used for the entire series of 2D velocity-encoded phase images, providing values that represented the average volumetric flow rates within the vessels [12 15]. Total brain blood supply was calculated as the sum of blood volume flow measurements in both ICAs and the basilar artery. Values of volumetric flow rates were calculated as mean ± SD in milliliters per minute. Statistical Analysis After testing of all MR flow variables for normal distribution with the Kolmogorov Smirnov one-sample test, the data were statistically analyzed by paired Student s t test for normally distributed data samples. E Fig. 2 (continued) 31-year-old woman with bilateral moyamoya disease depicted in Figure 1. Single sections of 2D cine phase-contrast MR images show levels at which measurements were obtained for determination of blood volume flow. E and F, Anatomic reference image (E) and single 2D cine phase-contrast image (F) show upper portion of basilar artery (arrows) between origins of anterior inferior cerebellar artery and superior cerebellar artery. Results Blood volume flow in both CCAs, both ICAs, and the basilar artery was quantified in the 35 patients with moyamoya disease and 15 age-matched healthy subjects serving as controls. In five of the 35 patients, unilateral moyamoya disease, so-called probable moyamoya disease, was diagnosed with digital subtraction angiography. The other 3 patients had angiographically proven typical bilateral moyamoya disease. Healthy Controls In the 15 age-matched controls, the following blood volume flow values were determined: CCA, ± 47.9 ml/min (range, ml/min); ICA, ± 25.3 ml/min (range, ml/min), and basilar artery, ± 13.2 ml/min (range, ml/min). In control subjects, no significant differences in blood volume flow were found between the left-side and the right-side carotid artery circulation. Moyamoya Patients Blood volume flow in the CCAs of patients with bilateral moyamoya disease (39.4 ± 89.9 ml/min; range, ml/min) was significantly (p <.1) lower than in control subjects (435.6 ± 47.9 ml/min). The 6 ICAs of the 3 patients with bilateral moyamoya disease had a mean blood volume flow of ± 64. ml/min (range, ml/min), and mean blood flow was significantly (p <.1) lower than in controls (254.1 ± 25.3 ml/min). Mean basilar artery blood volume flow (433.7 ± ml/min; range, ml/min) in the patients with bilateral moyamoya disease was much greater than in controls (173.3 ± 13.2 ml/min), representing an approximately 2.5- fold higher basilar artery blood flow (p <.1). No significant differences (p >.5) in left- and right-sided ICA and CCA blood volume flow were found in patients with bilateral moyamoya disease (left ICA, ± 66.8 ml/min; right ICA, 12.4 ± 62. ml/min; left CCA, 35.8 ± 86.3 ml/min; right CCA, ± 94.7 ml/min) (Fig. 4). Unilateral probable moyamoya disease was found in five patients. Three of these five patients had right-sided and two patients left-sided probable moyamoya disease. In all five patients, mean blood volume flow in the ICA (89.5 ± 54. ml/min; range, ml/min; p <.1) and F AJR:187, July 26 W111

6 A 672 C B D Fig year-old woman with bilateral moyamoya disease. Blood flow velocity in systolic diastolic modulation obtained by 2D cine phase-contrast MR measurements for patient in Figures 1 and 2. With integration over vessel diameter with respect to baseline correction, blood volume flow for each vessel was obtained. A and B, Both common carotid arteries. C and D, Both internal carotid arteries. (Fig. 3 continues on next page) W112 AJR:187, July 26

7 MRI in Moyamoya Disease Fig. 3 (continued) 31- year-old woman with bilateral moyamoya disease. Blood flow velocity in systolic diastolic modulation obtained by 2D cine phase-contrast MR measurements for patient in Figures 1 and 2. With integration over vessel diameter with respect to baseline correction, blood volume flow for each vessel was obtained. E, Basilar artery. Collateralization via basilar artery circulation in moyamoya disease that results in basilar artery blood flow approximately 25% of normal basilar artery blood flow corresponds to extensive increase in basilar artery blood flow velocity, which is approximately 13 cm/s in the CCA (314.9 ± 64.6 ml/min; range, ml/min; p <.1) on the side affected by moyamoya disease was significantly lower than in the arteries of controls (ICA, ± 25.3 ml/min; CCA, ± 47.9 ml/min). Blood volume flow in the contralateral ICA (38.6 ± 21.9 ml/min; range, ml/min; p <.1) and CCA (554.2 ± 75.1 ml/min; range, ml/min; p <.1) was significantly greater than in controls and the diseased carotid arteries. However, in the basilar arteries of these five patients, blood volume flow (32.2 ± 47.6 ml/min; range, ml/min; p <.1) was significantly greater than in controls (173.3 ± 13.2 ml/min) (Fig. 5). Discussion Moyamoya disease occurs in all ethnic groups, but is rare outside Japan and the Far East [19, 2]. Although many studies have been undertaken over a long period, the cause of moyamoya disease remains unclear [9]. MRI is sensitive in detection of the structural cerebrovascular abnormalities of moyamoya disease. The findings suggesting this diagnosis include multiple infarctions, absent or diminished vascular flow voids in the circle of Willis, and prominent flow voids at the level of the basal ganglia. MR angiography and MRI are complementary methods that show morphologic features and the vessels involved [21 23]. In this study, we found that blood volume flow decreases markedly, a mean of more than 5%, from normal in the ICA circulation of patients with bilateral moyamoya disease. Correspondingly, blood volume flow in the basilar artery increases dramatically, approximately 25% of normal flow, in these patients (Fig. 4). In patients with unilateral, so-called probable moyamoya disease, blood volume flow in the ICAs affected by moyamoya disease was significantly lower than normal and comparable with the results for patients with bilateral disease. Blood volume flow in the contralateral ICA and the basilar artery, however, was significantly greater (Fig. 5), with a less dramatic increase in basilar artery blood volume flow compared with flow associated with bilateral moyamoya disease and normal flow. MR phase-contrast flow quantification is a verifiable and reliable technique for in vivo measurement of cerebropetal blood flow [13 15]. The procedure is completely noninvasive and safe because contrast material, ionizing radiation, and radionuclides are not used. Disadvantages and limitations of 2D E cine phase-contrast MR flow quantification are its limited availability and potential vulnerability to patient movement during the examination [14]. In our study, all measurements were obtained with respect to a baseline correction [13 15]. The velocity encoding selected for phase-contrast flow measurements should be slightly higher than the highest expected peak velocity in the artery being examined. In accordance with the findings of a study by Vanninen et al. [14], selected velocity encoding in our study ranged from 4 to 25 cm/s for the CCA, ICA, and basilar artery measurements. Another source of error was the possibility of artifacts caused by turbulence just distal to the stenosis. In this study, the ICA flow measurement ROI was placed far away from the narrowed or nearly occluded vessel segment. Another potential error in phase-contrast MR flow measurement is that the number of pixels within the vessel lumen is too small to provide accurate results. Tang et al. [24] found that at least 16 pixels must cover the vessel lumen to yield measurement accuracy within 1%. With the given in-plane resolution of our technique, the expected accuracy was much higher. In patients with symptoms of moyamoya disease, cerebral hemodynamic compromise can be measured with noninvasive methods for quantifying regional cerebral blood flow and blood volume. PET, stable xenon-enhanced CT, and MRI can be used for measuring regional cerebral perfusion and for evaluating cerebrovascular reserve capacity [25]. MRI and MR angiography are reliable methods for visualizing primary signs (occlusion of circle of Willis and collateral formation, including moyamoya vessels) and secondary signs (cerebral infarction, white matter lesions, atrophy, and hemorrhage) and postoperative results. Identification of primary findings is nearly as good as that with conventional cerebral angiography. In a comparative study, Saeki et al. [21] found that findings on MR angiography lead to overestimation of stenosis and underestimation of moyamoya vessels. Those authors found a compatible rate of 85% between MR angiography and digital subtraction angiography in seven patients and concluded that MR angiography is a useful method for follow-up and has the possibility of replacing conventional cerebral angiography for initial diagnosis. The average rate of cerebral blood flow through the healthy adult brain is approximately ml/min [14, 26] measured with the nitrous oxide method. Several cross-sectional and longi- AJR:187, July 26 W113

8 tudinal studies [14] of nitrous oxide, 133 Xe, or 15 O PET have shown a decline in cerebral blood flow with advancing age. In an early study, Marks et al. [27] measured cerebral blood flow with the cine phase-contrast MR technique as a sum of the volume flow rates in both ICAs and the basilar artery. In 24 volunteers with normal neurologic findings, these authors found a mean of 777 ml/min in women and 885 ml/min in men. Buijs et al. [28] found a mean total cerebral blood flow of 616 ± 143 ml/min with a significant yearly decrease with age of 4.8 ml/min. In their study, mean total cerebral blood flow ranged from 748 ml/min to 474 ml/min in healthy volunteers and 8 89 years old without sex differences. The relative contributions of the right and left ICAs and the basilar artery were 41%, 4%, and 19% with no significant changes according to age. BVF (ml/min) MM ICA L ** ** CON MM CON ICA L ICA R ICA R Fig. 4 Blood volume flow (BVF) values (mean ± standard error of the mean [SEM], median) of 3 patients with bilateral idiopathic moyamoya disease in comparison with values for healthy subjects serving as controls. Symptomatic (moyamoya) internal carotid arteries have lower blood volume flow than controls. Moyamoya patients had basilar artery blood volume flow approximately 2.5-fold greater than that of controls. All results were highly significant. Bars indicate blood volume flow within arteries studied. **Statistically significant difference between groups (p <.5). MM = patients with bilateral moyamoya disease, CON = controls, ICA L = left internal carotid artery, ICA R = right internal carotid artery, BA = basilar artery. In our study, we found a mean cerebropetal blood volume flow of ± 38.1 ml/min in the age-matched controls and ± ml/min in patients with moyamoya disease for the sum of both ICAs and the basilar artery. In differentiating volume flow rates in these vessels, we found the blood volume flow rate for bilateral moyamoya disease for both ICAs decreased to 236 ml/min (sum of left- and right-sided ICA) and that basilar artery volume flow increased to as much as 434 ml/min. Healthy volunteers had a volume flow of 494 ml/min in both ICAs with a basilar artery volume flow rate of approximately 17 ml/min. These data show a significant shift of blood volume flow from decreased flow rates in the ICAs to increased flow rates in the basilar artery in moyamoya disease, to reach nearly normal total cerebropetal blood flow in the major brain-supplying arteries. MM BA ** CON BA BVF (ml/min) In conclusion, the main hemodynamic result of phase-contrast MR flow quantification in the large brain-supplying arteries of patients with moyamoya disease compared with age-matched controls was asssssss flow shift from the ICAs to the basilar artery. Mean ICA blood flow in moyamoya disease decreases to less than 5% of normal ICA blood flow, and basilar artery blood flow increases in moyamoya disease to approximately 25% of normal basilar artery blood flow. These findings are for cerebral ischemia in the anterior circulation, which is more common in moyamoya disease than ischemia in the posterior circulation [6]. The use of the 2D cine phase-contrast MRI technique with absolute flow quantification gives diagnostic information about hemodynamic compromise in major brain-supplying arteries in moyamoya disease (CCA, ICA, basilar artery). Direct PMM ICA S ** ** PMM ICA C CON ICA PMM BA CON BA Fig. 5 Blood volume flow (BVF) values (mean ± standard error of the mean [SEM], median) of five patients with probable unilateral moyamoya disease in comparison with values for healthy subjects serving as controls. Symptomatic (moyamoya) internal carotid arteries (ICA) have significantly lower blood volume flow than control arteries and contralateral ICAs. Blood volume flow was greater in contralateral (normal) ICA and basilar artery than in controls. All results were highly significant. Bars indicate blood volume flow within arteries studied. **Statistically significant difference between groups (p <.5). PMM = probable (unilateral) moyamoya disease, CON = controls, ICA S = symptomatic (moyamoya) internal carotid artery, ICA C = contralateral (normal) internal carotid artery, ICA = internal carotid artery, BA = basilar artery. ** W114 AJR:187, July 26

9 MRI in Moyamoya Disease parenchymal perfusion measurement, which only gives information about cell viability, is the pinnacle of regional cerebral blood flow techniques. Measurement and quantification of the blood supply to the brain do not preclude but supplement cerebral angiography in the diagnostic evaluation of symptomatic moyamoya disease and in decision making regarding cerebral revascularization surgery. References 1. Nishimoto A, Takeuchi S. Abnormal cerebrovascular network related to the internal carotid arteries. J Neurosurg 1968; 29: Suzuki J, Takaku A. Cerebrovascular moyamoya disease: disease showing abnormal net-like vessels in base of brain. Arch Neurol 1969; 2: Taveras JM. Multiple progressive intracranial arterial occlusions: a syndrome of children and young adults. AJR 1969; 16: Pecker J, Simon J, Guy G, Herry JF. Nishimoto s disease: significance of its angiographic appearances. Neuroradiology 1973; 5: Yamada I, Himeno Y, Suzuki S, Matsushima Y. Posterior circulation in moyamoya disease: angiographic study. Radiology 1995; 197: Yamada I, Himeno Y, Nagaoke T, et al. Moyamoya disease: evaluation with diffusion-weighted and perfusion echo-planar MR imaging. Radiology 1999; 212: Farrugia M, Howlett DC, Saks AM. Moyamoya disease. Postgrad Med J 1997; 73: Chiu D, Shedden P, Bratina P, Grotta JC. Clinical features of moyamoya disease in the United States. Stroke 1998; 29: Fukui M, Kono S, Sueishi K, Ikezaki K. Moyamoya disease. Neuropathology 2; 2[suppl]: Pelc LR, Pelc NJ, Rayhill SC, et al. Arterial and venous blood flow: noninvasive quantification with MR imaging. Radiology 1992; 185: Spitzer CE, Pelc NJ, Lee JN, Evans AJ, Sostman HD, Riederer SJ. Rapid MR imaging of blood flow with a phase-sensitive, limited flip-angle, gradient recalled pulse sequence: preliminary experience. Radiology 199; 176: Vanninen R, Koivisto K, Tulla H, Manninen H, Partanen K. Hemodynamic effects of carotid endarterectomy by magnetic resonance flow quantification. Stroke 1995; 26: van Everdingen KJ, Klijn CJM, Kappelle LJ, Mali WPTM, van der Grond J. MRA flow quantification in patients with a symptomatic internal carotid artery occlusion. Stroke 1997; 28: Vanninen RL, Manninen HI, Partanen KPL, Vainio PA, Soimakallio S. Carotid artery stenosis: clinical efficacy of MR phase-contrast flow quantification as an adjunct to MR angiography. Radiology 1995; 194: Neff KW, Horn P, Dinter D, Vajkoczy P, Schmiedek P, Düber C. Extracranial intracranial arterial bypass surgery improves total brain blood supply in selected symptomatic patients with unilateral internal carotid artery occlusion and insufficient collateralization. Neuroradiology 24; 46: Neff KW, Schwarz U, Motsch L, Steinke W, Schwartz A, Düber C. Long-term hemodynamic compromise in internal carotid artery dissection: quantitative blood volume flow evaluation using 2D cine phase-contrast MR imaging. Rofo 24; 176: Harada M, Miyoshi H, Masaaki U, et al. Neuronal impairment of adult moyamoya disease detected by quantified proton MRS and comparison with cerebral perfusion by SPECT with Tc-99m HM-PAO: a trial of clinical quantification of metabolites. J Magn Reson Imaging 1999; 1: Calamante F, Ganesan V, Kirkham FJ, et al. MR perfusion imaging in moyamoya syndrome: potential implications for clinical evaluation of occlusive cerebrovascular disease. Stroke 21; 32: Hoffman HJ. Moyamoya disease and syndrome. Clin Neurol Neurosurg 1997; 99[suppl 2]: Numaguchi Y, Gonzalez CF, Davis PC, et al. Moyamoya disease in the United States. Clin Neurol Neurosurg 1997; 99[suppl 2]: Saeki N, Silva MN, Kubota M, et al. Comparative performance of magnetic resonance angiography and conventional angiography in moyamoya disease. J Clin Neurosci 2; 7: Yamada I, Nakagawa T, Matsushima Y, Shibuya H. High-resolution turbo magnetic resonance angiography for diagnosis of moyamoya disease. Stroke 21; 32: Yoon HK, Shin HJ, Lee M, Byun HS, Na DG, Han BK. MR angiography of moyamoya disease before and after encephaloduroarteriosynangiosis. AJR 2; 174: Tang C, Blatter DD, Parker DL. Accuracy of phasecontrast flow measurements in the presence of partial-volume effects. J Magn Reson Imaging 1993; 3: Derdeyn CP, Grubb RL, Powers WJ. Cerebral hemodynamic impairment: methods of measurement and association with stroke risk. Neurology 1999; 53: Ganong WF. Cerebral blood flow. In: Review of medical physiology, 8th ed. Norwalk, CT: Appleton & Lange, 1977: Marks MP, Pelc NJ, Ross MR, Enzmann DR. Determination of cerebral blood flow with a phasecontrast cine MR imaging technique: evaluation of normal subjects and patients with arteriovenous malformations. Radiology 1992; 182: Buijs PC, Krabbe-Hartkamp MJ, Bakker CJG, et al. Effects of age on cerebral blood flow: measurement with ungated two-dimensional phasecontrast MR angiography in 25 adults. Radiology 1998; 29: AJR:187, July 26 W115