| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Departments of Radiology (I.Y., Y.H., H.S.) and Neurosurgery (T.N., H.A., Y.M.), Faculty of Medicine, and the Department of Neuropathology (T.K.), Medical Research Institute, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. Received June 2, 1998; revision requested August 5; final revision received November 10; accepted February 22, 1999. Address reprint requests to I.Y. (e-mail: yamada.crad@med.tmd.ac.jp).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Seventeen patients with moyamoya disease were examined prospectively with diffusion-weighted and perfusion echo-planar MR imaging and conventional angiography. The change in the effective transverse relaxation rate (
R2*) peak value, R2* peak time, and R2* integral were calculated to assess regional cerebral perfusion. The MR images were compared with angiographic images.
RESULTS: Of the 34 posterior cerebral arteries (PCAs) of the 17 patients, 14 PCAs (41%) in 11 patients showed stenosis or occlusion. The R2* peak value ratio in the cerebral hemispheres decreased significantly, and the R2* peak time ratio increased significantly, with PCA stenosis and occlusion. However, no correlation was apparent between perfusion and extent of the stenotic or occlusive lesions of the internal carotid artery bifurcation. The frequency of cerebral infarctions was significantly increased in patients with stenotic or occlusive PCA lesions. For three acute infarctions, a decrease in the apparent diffusion coefficient was significantly correlated with a decrease in the R2* peak value, an increase in the R2* peak time, and a decrease in the R2* integral.
CONCLUSION: Regional cerebral perfusion in moyamoya disease is decreased and delayed with PCA stenosis, with greater decrease and delay with PCA occlusion. Diffusion-weighted and perfusion imaging are useful for evaluating cerebral ischemia in moyamoya disease.
Index terms: Carotid arteries, MR, 17.121416, 17.12144 • Cerebral angiography, 17.1243, 17.1246, 17.1247 • Cerebral blood vessels, MR, 17.121416, 17.12144 • Cerebral blood vessels, stenosis or obstruction, 17.72134 • Moyamoya disease, 17.72134, 17.781
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Diffusion-weighted and perfusion echo-planar magnetic resonance (MR) imaging have recently been applied to evaluate occlusive cerebrovascular diseases (7–10). However, to our knowledge, few studies have described the application of such imaging in patients with moyamoya disease (11). We now present the results of a prospective study that was undertaken to determine the clinical efficacy of diffusion-weighted and perfusion echo-planar MR imaging, compared with that of conventional angiography, in the evaluation of moyamoya disease.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Imaging Examinations
All MR images were obtained with a 1.5-T superconducting system with a 25 mT/m maximal gradient capability (Magnetom Vision; Siemens, Erlangen, Germany) and a circularly polarized head coil. Initially, all patients underwent conventional MR imaging. Axial T1-weighted images were obtained with a spin-echo sequence (600/14 [repetition time msec/echo time msec]), a matrix of 192 x 256, and two signals acquired. Axial T2-weighted images were obtained with a turbo spin-echo sequence (4,000/96, echo train length of seven), a matrix of 210 x 512, and two signals acquired. All images were acquired with a field of view of 165 x 220 mm and a section thickness of 5 mm, with a 1-mm intersection gap.
Diffusion-weighted imaging.—Diffusion-weighted imaging was performed with a multisection, single-shot, spin-echo, echo-planar sequence, which included an echo time of 123 msec, a bandwidth of 1,250 Hz per pixel, a matrix of 128 x 200, a field of view of 230 x 230 mm, a section thickness of 5 mm with a 1-mm intersection gap, and one signal acquired. Fat suppression was performed by applying the frequency-selective radio-frequency pulse before imaging with the pulse sequence.
The diffusion gradient was applied along the section-select direction and had a duration of 26 msec, a separation time of 59.7 msec, and a gradient strength of 3.5 or 21.1 mT/m. The resultant values for the gradient factor b were 30 or 1,100 sec/mm2. Apparent diffusion coefficient (ADC) maps were calculated by using the following equation: ADC = ln(S0/S)/(b - b0), where S0 and S are the signal intensities of the two diffusion-weighted images, and the gradient factors b0 and b are 30 and 1,100 sec/mm2, respectively.
Perfusion imaging.—Perfusion imaging was performed with a multisection, single-shot, free induction decay, echo-planar sequence, which included an echo time of 54 msec, a bandwidth of 1,190 Hz per pixel, a matrix of 128 x 128, a field of view of 230 x 230 mm, a section thickness of 5 mm, and one signal acquired. We obtained 50 images every 1.2 seconds, with seven sections in each image. The seven sections were set to cover both cerebral hemispheres.
At the beginning of the image acquisition, a bolus of 0.1 mmol of gadodiamide (Omniscan; Daiichi Seiyaku, Tokyo, Japan) per kilogram of body weight was administered intravenously by means of manual injection over approximately 5 seconds, followed by a saline flush. Injection was performed with an 18-gauge intravenous catheter in an antecubital vein to minimize resistance to the injection.
Images based on the change in the effective transverse relaxation rate (R2*) were constructed from each time series at each time point along the first pass of the contrast agent bolus. The R2* was computed on a pixel-by-pixel basis according to the following formula: R2* = -ln(St/S0)/TE, where St is signal intensity at time t, S0 is the precontrast baseline signal intensity, and TE is the sequence echo time. R2* is proportional to the concentration of the contrast agent and can be used to estimate perfusion parameters.
On R2* images, regions of large signal change appear as areas of high signal intensity, whereas regions of small signal change appear as areas of low signal intensity. These images were processed to create maps of the R2* peak value, R2* peak time, and R2* integral, with the use of a numerical integration technique similar to those described previously (12).
Conventional angiography.—All patients also underwent cerebral angiography that included bilateral internal and external carotid arteriography and unilateral or bilateral vertebral arteriography, with the use of the transfemoral catheterization technique for both carotid and vertebral arteriography. Conventional angiograms were obtained with a digital subtraction technique. Digital subtraction images were obtained with a 0.6-mm focal spot and a 10-inch (25.4-cm) cesium iodide image-intensifying tube (KXO-80C/D and DFP-2000A; Toshiba Medical Systems, Tokyo, Japan). Images were displayed and printed with a 1,024 x 1,024 matrix. A dose of 8–10 mL of ioxaglate (Hexabrix 320; Eiken Kagaku, Tokyo, Japan) was injected for each arteriographic examination. Conventional angiography was performed within 1 month before (n = 14) or after (n = 3) MR imaging. In these three patients, the diagnosis was made with conventional angiography after MR imaging.
Image Analysis
On the basis of the angiographic findings, we classified stenotic or occlusive lesions of the ICA bifurcation, according to the extent of narrowing in the supraclinoid portion of the ICA and the proximal portions of the anterior cerebral artery and middle cerebral artery, into five different stages: stage 1, mild to moderate stenosis of the ICA bifurcation (
80% reduction in diameter); stage 2, severe stenosis of the ICA bifurcation (>80% reduction in diameter); stage 3, occlusion of either the anterior cerebral artery or middle cerebral artery; stage 4, occlusion of the ICA bifurcation, with partial retention of the anterior cerebral artery or middle cerebral artery main trunk; and stage 5, occlusion of the ICA bifurcation, with no detectable anterior cerebral arterial or middle cerebral arterial main trunk (6). Stenotic or occlusive lesions in the posterior cerebral artery (PCA) were also identified. Thus, the PCA was graded as normal, stenotic, or occluded.
We also classified the basal cerebral moyamoya vessels on the basis of their presence and appearance into four grades: none, slight, moderate, or marked. "Marked" indicated the moyamoya vessels formed a vascular network that showed high contrast and that extended above the basal ganglia, with visualization of the medullary arteries. "Moderate" meant the moyamoya vessels formed an intermediate vascular network localized in the basal ganglia, without visualization of the medullary arteries. "Slight" indicated the moyamoya vessels showed less contrast and a more orderly arrangement near only the internal carotid bifurcation, and they assumed the appearance of unusually dilated perforating arteries.
The R2* images were reviewed by two radiologists (I.Y., Y.H.) for the distribution of areas that showed decreased regional cerebral perfusion. Circular regions of interest were thus set on R2* images in the frontal, temporal, parietal, and occipital lobes and in the basal ganglia, and the mean value of three or four regions of interest was calculated in these five cerebral regions. Ratios of the R2* peak value, R2* peak time, and R2* integral were calculated by dividing the value in these regions by the mean cerebral value obtained from both cerebral hemispheres to assess the regional cerebral perfusion parameters. Values were expressed as means ± 1 SD.
Diffusion-weighted and conventional MR images were reviewed by two radiologists (I.Y., Y.H.) for cerebral infarctions. At conventional MR imaging, an infarction was diagnosed for lesions that showed an abnormal signal intensity on both T1- and T2-weighted images but not for lesions that showed an abnormal signal intensity in the deep white matter on only T2-weighted images. On diffusion-weighted MR images, an acute infarction was diagnosed for lesions that showed a high signal intensity on heavily diffusion-weighted images with a gradient factor b of 1,100 sec/mm2.
The ADC values were estimated for these lesions, and the ADC ratios were calculated by dividing the ADC value in these regions by the normal ADC value in the anatomically matched site of the contralateral cerebral hemisphere. To examine the relationships between the perfusion parameters and ADC values, we selected several regions of interest for these acute infarctions.
Imaging studies were evaluated on the basis of blinded, separate interpretations by two independent radiologists (I.Y., Y.H.). Furthermore, MR images were interpreted independently, without knowledge of the angiographic findings. In instances in which the radiologists did not fully agree, diagnosis was achieved by consensus.
Finally, a statistical analysis of the comparisons between groups was performed with the Spearman rank correlation test, unpaired Student t test, 
2 test, or Mann-Whitney U test. The statistical tests were corrected for multiple comparisons. Linear regression analysis was used to assess the relationships between the perfusion parameters and ADC values. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
R2* Images
|
|
|
|
|
|
|
|
|
|
R2* peak value ratios for the frontal, temporal, parietal, and occipital lobes decreased significantly with PCA stenosis, with a greater decrease with PCA occlusion (Spearman rank correlation test) (Figs 1, 2). In contrast, the R2* peak value ratio for the basal ganglia increased with PCA stenosis or occlusion (Spearman rank correlation test). Furthermore, the R2* peak time ratios for the frontal, temporal, parietal, and occipital lobes increased significantly with PCA stenosis, with a greater increase with PCA occlusion (Spearman rank correlation test).
|
R2* integral ratios did not significantly correlate with PCA stenosis and occlusion, with the exception of that for the occipital lobe, which increased with PCA stenosis or occlusion (Spearman rank correlation test). Thus, the presence of stenotic or occlusive PCA lesions was markedly related to both the R2* peak value and R2* peak time ratios in patients with moyamoya disease.
In contrast, on R2* images, the stage of stenotic or occlusive lesions of the ICA bifurcation did not correlate significantly with the R2* peak value, R2* peak time, or R2* integral ratios (P > .05 for all, Spearman rank correlation test). When stages 1, 2, and 3 were grouped as stenotic and stages 4 and 5 were grouped as occluded, there was no significant difference between the two groups of the ICA bifurcation lesions for the R2* peak value, R2* peak time, and R2* integral ratios (P > .05 for all, unpaired Student t test).
Collateral Vessels and R2* Images
In all 17 patients, basal cerebral moyamoya vessels were detected bilaterally in the cerebral hemispheres (Figs 1, 2). The grades of the moyamoya vessels in the 34 cerebral hemispheres were as follows: slight, 20 hemispheres; moderate, eight hemispheres; and marked, six hemispheres. Furthermore, leptomeningeal collateral vessels from the PCA to the anterior circulation were detected in 18 cerebral hemispheres (53%) (Fig 1). However, in hemispheres that had occlusion of the PCA, no leptomeningeal collateral vessels from the PCA to the anterior circulation were apparent (Figs 1, 2).
As shown in Table 2, the R2* peak value ratios for the frontal, temporal, parietal, and occipital lobes were significantly greater when leptomeningeal collateral vessels were present than when they were not (unpaired Student t test) (Figs 1, 2). The R2* peak time ratios for the frontal, temporal, parietal, and occipital lobes were significantly smaller when leptomeningeal collateral vessels were present than when they were not (unpaired Student t test). However, the R2* integral ratios did not significantly correlate with the presence of the leptomeningeal collateral vessels, with the exception of that for the occipital lobe, which was smaller in the presence of such vessels (unpaired Student t test). Thus, the presence of leptomeningeal collateral vessels was closely related to both the R2* peak value and R2* peak time ratios.
|
R2* images, the grade of the basal cerebral moyamoya vessels did not significantly correlate with the R2* peak value, R2* peak time, or R2* integral ratios, with the exception of the R2* peak time ratio for the basal ganglia, which decreased as the grade of moyamoya vessels increased (P < .01, unpaired Student t test).
Cerebral Infarctions on Diffusion-weighted and Conventional MR Images
Of the 34 cerebral hemispheres in the 17 patients, cerebral infarctions were detected in 14 hemispheres (41%) on conventional MR images (Figs 1, 2). As shown in Table 3, the frequency of cerebral infarction significantly increased with PCA stenosis, with a greater increase with PCA occlusion (Mann-Whitney U test) (Figs 1, 2). Thus, the presence of stenotic or occlusive PCA lesions was related to the occurrence of cerebral infarction in patients with moyamoya disease. However, the stage of stenotic or occlusive lesions of the ICA bifurcation did not significantly correlate with cerebral infarction.
|
|
|
|
|
|
R2* peak value ratio was postively correlated with the ADC value (r = 0.872, P < .01), and the R2* peak time ratio was negatively correlated with the ADC value (r = -0.680, P < .05). The R2* integral ratio also was positively correlated with the ADC value (r = 0.831, P < .01). Thus, the decrease in the R2* peak value ratio, the increase in the R2* peak time ratio, and the decrease in the R2* integral ratio correlated significantly with the decrease in the ADC value for acute infarctions in patients with moyamoya disease.
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
R2* peak value ratio for the cerebral hemispheres decreased significantly and the corresponding R2* peak time ratio increased significantly. Thus, the presence of stenotic or occlusive PCA lesions is significantly associated with a decrease in the R2* peak value and an increase in the R2* peak time in patients with moyamoya disease. However, no significant correlation was apparent between the stenotic or occlusive lesions of the ICA bifurcation and the R2* peak value or R2* peak time. These data indicate cerebral ischemia in individuals with moyamoya disease increases with stenotic or occlusive lesions of the posterior circulation, rather than with stenotic or occlusive lesions of the anterior circulation. To our knowledge, the role of the posterior circulation in moyamoya disease has not previously been evaluated at perfusion echo-planar MR imaging.
Furthermore, the presence of leptomeningeal collateral vessels from the PCA to the anterior circulation was shown to be significantly correlated with the R2* peak value and R2* peak time for the cerebral hemispheres. In patients with moyamoya disease in whom both ICA bifurcations are occluded, the leptomeningeal collateral vessels become a major provider of a compensatory cerebral blood supply (13,14). Thus, any decrease in the number of leptomeningeal collateral vessels would critically influence the cerebral blood flow. The reduction in the number of leptomeningeal collateral vessels may be due to the progressive stenotic or occlusive lesions of the PCA. When the PCA finally becomes occluded, the leptomeningeal collateral vessels presumably disappear, thereby resulting in severe cerebral ischemia.
In contrast, the grade of the basal cerebral moyamoya vessels in patients with moyamoya disease did not correlate with the R2* peak value or R2* peak time. Thus, although the presence of basal cerebral moyamoya vessels is a characteristic of moyamoya disease (1,2), the contribution of the basal cerebral moyamoya vessels to the cerebral blood flow appears to be much less than that of the leptomeningeal collateral vessels.
On the basis of the conventional MR images, the frequency of cerebral infarction significantly increased with PCA stenosis, with a greater increase with PCA occlusion. Thus, the presence of stenotic or occlusive PCA lesions appears to be significantly related to the occurrence of cerebral infarction in patients with moyamoya disease. However, no significant correlation was apparent between the stage of stenotic or occlusive lesions of the ICA bifurcation and cerebral infarction.
Our data demonstrate that diffusion-weighted MR images were useful for the detection of acute cerebral infarctions in patients with moyamoya disease. The acute infarctions also occurred in cerebral hemispheres with stenotic or occlusive PCA lesions. Furthermore, the decrease in the R2* peak value ratio, the increase in the R2* peak time ratio, and the decrease in the R2* integral ratio correlated significantly with the decrease in ADC value for these acute infarctions.
Results of recent studies (15–18) with animal models of ischemia have revealed an important relationship between the peak R2* ratios and ADC values. The results of these studies and our data indicate reduced and delayed perfusion is directly related to the mechanism of ADC reduction in acute infarctions due to moyamoya disease.
Diffusion-weighted and perfusion echo-planar MR imaging may affect the treatment of patients with moyamoya disease. The noninvasive assessment of cerebral ischemia with diffusion-weighted and perfusion echo-planar MR imaging may have value in following up disease progression, planning surgical treatment, and evaluating patients postoperatively.
In conclusion, regional cerebral perfusion in patients with moyamoya disease is decreased and delayed with PCA stenosis, with a greater decrease and delay with PCA occlusion. If the PCA is occluded, the number of leptomeningeal collateral vessels to the anterior circulation presumably decreases, thereby causing severe cerebral ischemia that results in an infarction. In acute infarctions, the ADC reduction correlates with decreased and delayed perfusion. Thus, diffusion-weighted and perfusion echo-planar MR imaging are useful methods for evaluating cerebral ischemia in patients with moyamoya disease.
|
Footnotes |
|---|
R2* = change in the effective transverse relaxation rate Author contributions: Guarantor of integrity of entire study, I.Y.; study concepts, I.Y.; study design, I.Y., Y.H.; definition of intellectual content, I.Y.; literature research, I.Y., T.K.; clinical studies, I.Y., Y.M.; data acquisition, I.Y., T.N.; data analysis, I.Y., H.A.; statistical analysis, I.Y., T.K.; manuscript preparation, I.Y., Y.H.; manuscript editing, I.Y.; manuscript review, I.Y., H.S.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. J. Yun, J.-E. Cheon, D. G. Na, W. S. Kim, I.-O. Kim, K.-H. Chang, K. M. Yeon, I. C. Song, and K.-C. Wang Childhood Moyamoya Disease: Quantitative Evaluation of Perfusion MR Imaging--Correlation with Clinical Outcome after Revascularization Surgery Radiology, April 1, 2009; 251(1): 216 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.H. Kang, H.S. Kim, and S.Y. Kim Quantitative Cerebrovascular Reserve Measured by Acetazolamide-Challenged Dynamic CT Perfusion in Ischemic Adult Moyamoya Disease: Initial Experience with Angiographic Correlation AJNR Am. J. Neuroradiol., September 1, 2008; 29(8): 1487 - 1493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Neff, P. Horn, P. Schmiedek, C. Duber, and D. J. Dinter 2D cine phase-contrast MRI for volume flow evaluation of the brain-supplying circulation in moyamoya disease. Am. J. Roentgenol., July 1, 2006; 187(1): W107 - W115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Togao, F. Mihara, T. Yoshiura, A. Tanaka, T. Noguchi, Y. Kuwabara, K. Kaneko, T. Matsushima, and H. Honda Cerebral Hemodynamics in Moyamoya Disease: Correlation between Perfusion-Weighted MR Imaging and Cerebral Angiography. AJNR Am. J. Neuroradiol., February 1, 2006; 27(2): 391 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kassner, X.-P. Zhu, K.-L. Li, and A. Jackson Neoangiogenesis in Association with Moyamoya Syndrome Shown by Estimation of Relative Recirculation based on Dynamic Contrast-Enhanced MR Images AJNR Am. J. Neuroradiol., May 1, 2003; 24(5): 810 - 818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-K. Lee, D. I. Kim, E.-K. Jeong, S.-Y. Kim, S. H. Kim, Y. K. In, D.-S. Kim, and J.-U. Choi Postoperative Evaluation of Moyamoya Disease with Perfusion-Weighted MR Imaging: Initial Experience AJNR Am. J. Neuroradiol., April 1, 2003; 24(4): 741 - 747. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Calamante, V. Ganesan, F.J. Kirkham, W. Jan, W.K. Chong, D.G. Gadian, and A. Connelly MR Perfusion Imaging in Moyamoya Syndrome: Potential Implications for Clinical Evaluation of Occlusive Cerebrovascular Disease Stroke, December 1, 2001; 32(12): 2810 - 2816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Yamada, T. Nakagawa, Y. Matsushima, and H. Shibuya High-Resolution Turbo Magnetic Resonance Angiography for Diagnosis of Moyamoya Disease Stroke, August 1, 2001; 32(8): 1825 - 1831. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Yoshino, I. Yamada, N. Ohbayashi, E. Honda, M. Ida, T. Kurabayashi, K. Maruyama, and T. Sasaki Salivary Glands and Lesions: Evaluation of Apparent Diffusion Coefficients with Split-Echo Diffusion-weighted MR Imaging—Initial Results Radiology, December 1, 2001; 221(3): 837 - 842. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
/54) shows areas of decreased perfusion (arrows) in the right cerebral hemisphere, which has PCA occlusion, as compared with the left cerebral hemisphere. (d) Time-
), which had PCA occlusion, as compared with the left cerebral hemisphere (•). (e) Axial T2-weighted turbo spin-echo MR image (4,000/96) shows infarctions in the right frontal (arrow) and parietal lobes. There are numerous moyamoya vessels (arrowheads) in the bilateral basal ganglia.
