Abstract 16417: Carotid Plaque Component Reproducibility at 3.0 Tesla Mri in Patients with Overt Atherosclerosis

Abstract

Background: Carotid MRI has recently shown to provide compositional information about carotid atherosclerosis. Currently available reproducibility data is performed using an in-plane resolution of approximately 0.50 x 0.50 mm² and different segmentation methods. Reproducibility data for carotid plaque components at 3.0 Tesla MRI in patients with overt atherosclerotic plaques is lacking.

Methods: We compared the reproducibility and accuracy, as well as the within- and between reader reproducibility of high and routinely used spatial resolution scans in patients with advanced atherosclerotic carotid artery disease. In 31 patient two high (0.25 x 0.25 mm²; HR) and two routinely used (0.50 x 0.50 mm²; LR) spatial resolution carotid 3.0 Tesla MRI scans were performed within one month. A fully blinded and an open segmentation method were used to quantify the lipid rich necrotic core, calcified and loose matrix plaque area at both resolutions.

Results: We found excellent intraclass correlation coefficients (all ICCs > 0.85) between scans and rescans for the lipid rich necrotic core, calcified and loose matrix plaque area with the open segmentation method. We observed a significant lower signal-to-noise (HR 17.8 ± 8.1 vs. LR 52.2 ± 13.4; p <0.001) and contrast-to-noise ratio (HR 15.7 ± 7.2 vs. LR 48.0 ± 13.1; (P <0.001) for the HR compared to the LR measurements. Despite the reduced SNR, we found significantly lower loose matrix measurements for he HR compared to the LR measurements (P<0.05), suggesting an increased measurement accuracy due to a reduced partial volume effect for the HR measurements. No significant differences were observed between the within-reader and between-reader reproducibility for HR versus LR measurements.

Conclusions: In patients with advancedcarotid atherosclerosis, 3.0 Tesla MRI component analyses using an open segmentation method has an excellent reproducibility. Increasing the in-plane resolution improves the measurement accuracy of loose matrix plaque area measurements. Clinical trials investigating the efficacy of novel compounds on specific plaque components can use Carotid 3.0 Tesla MRI as a surrogate marker in patients with overt atherosclerosis.