High-resolution MRI of the vessel wall helps to distinguish moyamoya disease from atherosclerotic moyamoya syndrome

2021 
AIM To evaluate the value of high-resolution magnetic resonance imaging of the vessel wall (VWI) for differentiating moyamoya disease (MMD) from atherosclerotic moyamoya syndrome (AS-MMS). MATERIALS AND METHODS Twenty-one patients with MMD or AS-MMS were assessed retrospectively by two independent raters regarding and magnetic resonance angiography (MRA) stage grading score; collateral development in the lateral fissure and basal ganglia on MRA; and pattern of the thickening of the arterial wall; presence, degree, and pattern of enhancement; presence and distribution of deep tiny flow voids (DTFVs) and collateral development in the lateral fissure and basal ganglia on VWI. After univariate analysis between the two groups, logistic regression models based on imaging findings of MRA or VWI were implemented respectively, and receiver operating characteristic (ROC) curves were generated to compare the discriminatory power of the two imaging methods for diagnosis of MMD. Interrater agreement was analysed using an unweighted Cohen's κ or interclass correlation coefficient (ICC). RESULTS MMD manifested as more concentric thickening, more homogeneous enhancement, higher presence of DTFV, smaller outer-wall boundary area of stenosis or occlusion, and smaller remodelling index on VWI. After Bonferroni–Holm correction for multiple comparisons, for AS-MMS, collaterals in both the lateral fissure and basal ganglia were not usually present on either MRA or VWI. The diagnostic performance of the multivariate logistic regression model based on VWI with an accuracy of 87.1% for classification was higher than MRA. Interrater agreement was moderate or substantial for all the imaging findings. CONCLUSIONS VWI might be a useful and feasible method for differentiating MMD from AS-MMS and a prospective tool for guiding first-line treatment.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    23
    References
    0
    Citations
    NaN
    KQI
    []