Moyamoya disease (MMD) is characterized by progressive stenosis of the intracranial arteries with unknown etiology.^1,2 It most commonly affects the terminal portion of the internal carotid artery to the proximal portion of the anterior and the middle cerebral arteries, causing luminal narrowing and compensatory development of fine collateral networks called moyamoya vessels. MMD is typically found in children and young adults, but asymptomatic cases are sometimes found in the elderly. Intracranial atherosclerotic disease (ICAD) can also cause luminal narrowing of the intracranial arteries, most commonly in elderly patients but on rare occasions in young adults. Therefore, it would be helpful to differentiate between MMD and ICAD by imaging.

A few recent MR studies have shown that the decrease of the outer diameters of the affected arteries (i.e. vascular constriction) is characteristic in MMD, while the outer diameters are preserved in ICAD (Fig. 1).^3,4 However, these studies did not use black blood techniques.

Motion-sensitized driven-equilibrium (MSDE) preparation is a recently introduced technique to black blood imaging. T1-weighted images with MSDE preparation are expected to highlight vessel walls with intermediate to high signals and thus to clearly depict the inner and the outer margins of vessel walls. To our knowledge, however, the utility of vessel wall imaging with a MSDE technique has not been elucidated in the application to MMD. The purpose of this proof-of-concept study was to assess the diagnostic feasibility of vessel wall imaging with a MSDE technique in differentiation between MMD from ICAD.

Patient demographics

The institutional review board at our hospital approved this study. This study enrolled seven patients with MMD (4 male and 3 female patients, aged 31 to 58 years) and seven patients with atherosclerotic stenosis of the internal carotid artery or the middle cerebral artery (4 male and 3 female patients, aged 56 to 71 years). All of the patients underwent brain MRI and MRA, including 3D T1-weighted variable refocusing flip angle fast spin echo (FSE) imaging (a research version of CUBE T1) with MSDE preparation.

Methods for MR imaging

A 3-T clinical scanner (GE Signa HDxt) was used. We used a standard MSDE preparation scheme (90°_x-180°_y-90°_-x).⁵ Imaging parameters for CUBE T1 with MSDE preparation were as follows: motion sensitization gradient = 7 s/mm² in 3-axis composite value, TR/TE = 500/minimum ms, FOV = 22 cm, slice thickness/interslice spacing = 0.8/0.4 mm, using a trans-axial plane, matrix = 256 x 256 (512 x 512 after zero-fill interpolation).

Analyses

Anonymized, MSDE-prepared CUBE T1 images alone were presented to a radiologist who did not know the correct diagnosis. Imaging features of the affected arteries were assessed as follows: concentric or eccentric luminal narrowing, localized or non-localized (diffuse or multiple) luminal narrowing. Then, imaging diagnosis was chosen between MMD and ICAD. The percentage of correct imaging diagnosis was assessed with Fisher’s permutation test. Statistical significance was set at p = 0.05.

Excellent blood signal suppression was recognized and outer vessel margins were satisfactorily depicted on MSDE-prepared CUBE T1 images (Fig. 2). All of ICAD patients had eccentric luminal narrowing, while five out of the seven MMD patients showed concentric luminal narrowing. All of MMD patients showed non-localized (diffuse) luminal narrowing, while ICAD showed localized luminal narrowing in five and non-localized (at multiple locations) narrowing in two patients. Differentiation between MMD and ICAD was performed correctly in all patients by viewing through MSDE-prepared CUBE T1 images alone (p = 0.0006).

The results of our study suggest that MSDE-prepared CUBE T1 images can achieve excellent black blood capability and that it can depict the inner and the outer margins of vessel walls at the same time. Our method enabled correct differentiation between our MMD and ICAD cohorts. Therefore, addition of MSDE-prepared 3D T1-weighted variable refocusing flip angle FSE imaging may be helpful to differentiate between MMD and ICAD when intracranial arterial stenosis is found on time-of-flight MRA.

3D FSE T1-weighted images contain T1 contrast in vessel walls where lipid-rich atherosclerotic plaques and intraplaque hemorrhages typically show high signals.⁶ This could be another advantage of our method. The previous study used 3D constructive interference in steady-state (CISS) MR imaging to outline vessel outer margins.³ Image contrast between vessel walls and cerebrospinal fluid is excellent in CISS imaging, but CISS images likely lose fine tissue contrast within vessel walls.

MSDE-prepared 3D T1-weighted variable refocusing flip angle FSE MR imaging would be feasible for vessel wall imaging in MMD. We believe that this could be a helpful technique to differentiate MMD from ICAD.