BACKGROUND: Intraplaque T2* values help to identify symptomatic carotid plaques and correlate with intraplaque iron deposits in plaque progression. However, intracranial T2* mapping in vivo at 3T MRI is challenging due to limited resolution and signal-to-noise ratio. This study aimed to quantitatively measure T2* value of middle cerebral artery (MCA) atherosclerotic plaques using 7T MRI and to assess its correlation with cerebrovascular symptoms. METHODS: Phantom studies were performed to evaluate the accuracy of T2* mapping obtained with the proposed sequence by comparison with the ground truth acquisition. In the in vivo study, intraplaque T2* values obtained from multi-echo T2* mapping and plaque characteristics from T1-weighted 3D sampling perfection with application-optimized contrast using different flip angle evolutions sequence on 7T MRI were analyzed and compared between patients with symptomatic and asymptomatic MCA plaques. Multivariate logistic regression was used to determine the odds ratios of T2* values and plaque characteristics in discriminating symptomatic from asymptomatic plaques. Diagnostic performance was evaluated using area under the curve (AUC) values. Correlation analyses were performed between T2* values and intraplaque hemorrhage (IPH). RESULTS: Phantom T2* measurements using the proposed sequence showed excellent agreement with the ground truth sequence (ICC=0.998, p<0.001), with a mean percentage error of 3.97 ± 3.11%. The clinical cohort of this prospective cross-sectional study included 39 symptomatic patients with MCA plaques and 21 age-, sex-, and stenosis degree-matched asymptomatic patients. Scan-rescan reproducibility of T2* mapping was excellent (p < 0.001). Symptomatic plaques had significantly lower T2* values than asymptomatic plaques (22.24±5.31 vs. 30.24±7.00 ms, p<0.001). In multivariate analysis, intraplaque T2* values (OR: 0.162, 95% CI: 0.053-0.497, p=0.001) and normalized wall index (NWI) (OR: 2.150, 95% CI: 1.041-4.443, p=0.039) were independently associated with symptomatic plaques. The optimal combination of T2* values and NWI showed the best diagnostic performance (AUC=0.861, 95% CI:0.747-0.937), with 94.9% sensitivity and 66.7% specificity. T2* values were negatively correlated with and IPH (r=-0.290, p=0.027) after age- and sex- adjustments. DISCUSSION: The feasibility of intracranial T2*mapping in vivo on 7T MRI has been proven, indicating its potential as a sensitive tool for characterizing intracranial symptomatic plaques.
Bai et al. (Thu,) studied this question.