Non-Invasive Assessment of Carotid Perivascular Adipose Tissue Phenotyping and Calcific Burden Using MRI Fat Fraction and R2* Mappings: Validation Against CTA

J Magn Reson Imaging. 2025 Jun 29. doi: 10.1002/jmri.70007. Online ahead of print.

Abstract

Background: Perivascular adipose tissue (PVAT) attenuation elevation, as measured by computed tomography angiography (CTA), indicates localized vascular inflammation. Dixon-based fat fraction (FF) and R2* mappings, established for assessing carotid high-risk plaque components, demonstrate potential for quantifying PVAT phenotypic changes and calcification burden. However, studies validating FF/R2* mappings against CTA attenuation metrics are lacking.

Purpose: To assess the correlation between FF/R2* mappings and CTA measurements in evaluating carotid PVAT characteristics and calcification burden.

Study type: Prospective validation study.

Population/subjects: Twenty five patients with ultrasonographically-confirmed carotid atherosclerotic plaque (62.3 ± 11.3 years) having matched carotid CTA/MRI and five healthy volunteers (36.2 ± 11.9 years).

Field strength/sequence: 3T, 3D multi-echo Dixon sequence for co-registered FF/R2* mapping.

Assessment: Volunteers were scanned twice on the same day (with removal from scanner between scans) to assess scan-rescan reproducibility of FF mapping. PVAT FF values and calcification burden measured on Dixon-derived FF and R2* maps, respectively, were compared to spatially matched CTA Hounsfield units (HU) and calcification volumetry in the atherosclerosis cohort.

Statistical tests: Intraclass correlation coefficient (ICC), Pearson correlation coefficients (R), and Bland-Altman analysis. A p-value < 0.05 was considered significant.

Results: Dixon imaging demonstrated high scan-rescan reproducibility for PVAT FF quantification (ICC = 0.982). In patients, PVAT FF exhibited a strong inverse correlation with CTA HU values (R = -0.868). R2* mapping achieved near-perfect agreement with CTA in calcification volumetry across 40 calcified carotids (R = 0.991).

Conclusion: FF/R2* mappings show excellent agreement with CTA for PVAT phenotyping and calcific burden quantification, while potentially enabling simultaneous assessment of vascular inflammation and high-risk plaque components.

Evidence level: 2.

Technical efficacy: Stage 2.

Keywords: adipose tissue; atherosclerotic; carotid arteries; plaque.