Advanced MRI of calcified carotid artery plaques: pathological validation of accuracy.

Abstract

Objective: Carotid artery atherosclerotic plaques undergo various pathological changes during disease progression, and their characterization relies on MRI, CT, and ultrasound imaging. The diagnosis of calcified lesions using only MRI is challenging because of irregular low-signal areas across all sequences. Conventionally, a diagnosis is made by combining multiple modalities such as contrast-enhanced CT (CECT) and carotid ultrasonography. Field echo resembling CT using restricted echo-spacing (FRACTURE) is a new MRI sequence with enhanced bone contrast that uses a 3D gradient echo pulse sequence. The aim of this study was to evaluate the ability of FRACTURE to detect carotid plaque calcification by comparing its images with pathological findings from carotid endarterectomy (CEA). The study also compared the accuracy of conventional CECT and MRI plaque imaging with that of FRACTURE and assessed its accuracy in stenosis rate evaluation using fusion imaging with MR angiography (MRA).

Methods: This retrospective analysis included 21 consecutive patients (average age 74 years, 16 males) who underwent FRACTURE, CECT, and MRA (including time-of-flight and black-blood techniques) as preoperative examinations for CEA. Pathological specimens from extracted plaques were analyzed, and the calcified areas were measured and statistically compared across imaging modalities. Digital subtraction angiography (DSA) was used as the reference standard for stenosis rate evaluation.

Results: The sensitivity and specificity of FRACTURE in detecting calcified lesions in pathological specimens were comparable with those of CECT (sensitivity 91.7% vs 95.8%, specificity 96.6% vs 93.1%) and superior to those of MRA (sensitivity 75.0%, specificity 93.1%). Furthermore, for lesions < 2 mm2, significant differences in the measured calcified areas were observed between pathological specimens and both CECT and MRA, whereas no significant difference was found with FRACTURE, suggesting its superiority for small lesions. In stenosis rate assessment, no significant differences were found among FRACTURE/MRA, CECT, and DSA in overall stenosis measurements. However, in cases in which calcified lesions directly faced the vascular lumen, CECT tended to overestimate stenosis because of beam-hardening artifacts, while FRACTURE/MRA showed a higher agreement with DSA, indicating greater accuracy.

Conclusions: FRACTURE is a useful MRI sequence with higher sensitivity and specificity than those of MRA and accuracy comparable with that of CECT for calcified lesion detection. Additionally, it may provide superior detection of small, calcified lesions and more accurate stenosis assessment in cases with calcified plaques.