Quantitative assessment of atherosclerotic plaques on (18)F-FDG PET/MRI: comparison with a PET/CT hybrid system

Abstract

Purpose: PET with (18)F-FDG has the potential to assess vascular macrophage metabolism. (18)F-FDG is most often used in combination with contrast-enhanced CT to localize increased metabolism to specific arterial lesions. Novel (18)F-FDG PET/MRI hybrid imaging shows high potential for the combined evaluation of atherosclerotic plaques, due to the superior morphological conspicuity of plaque lesions. The purpose of this study was to evaluate the reliability and accuracy of (18)F-FDG PET/MRI uptake quantification compared to PET/CT as a reference standard in patients with carotid atherosclerotic plaques.

Methods: The study group comprised 34 consecutive oncological patients with carotid plaques who underwent both PET/CT and PET/MRI with (18)F-FDG on the same day. The presence of atherosclerotic plaques was confirmed by 3 T MRI scans. Maximum standardized uptake values (SUVmax) for carotid plaque lesions and the average SUV of the blood pool within the adjacent internal jugular vein were determined and target-to-blood ratios (TBRs, plaque to blood pool) were calculated.

Results: Atherosclerotic lesions with maximum colocalized focal FDG uptake were assessed in each patient. SUVmax values of carotid plaque lesions were significantly lower on PET/MRI than on PET/CT (2.3 ± 0.6 vs. 3.1 ± 0.6; P < 0.01), but were significantly correlated between PET/CT and PET/MRI (Spearman's r = 0.67, P < 0.01). In contrast, TBRmax values of plaque lesions were similar on PET/MRI and on PET/CT (2.2 ± 0.3 vs. 2.2 ± 0.3; P = 0.4), and again were significantly correlated between PET/MRI and PET/CT (Spearman's r = 0.73, P < 0.01). Considering the increasing trend in SUVmax and TBRmax values from early to delayed imaging time-points on PET/CT and PET/MRI, respectively, with continuous clearance of radioactivity from the blood, a slight underestimation of TBRmax values may also be expected with PET/MRI compared with PET/CT.

Conclusion: SUVmax and TBRmax values are widely accepted reference parameters for estimation of the radioactivity of atherosclerotic plaques on PET/CT. However, due to a systematic underestimation of SUVmax and TBRmax with PET/MRI, the optimal cut-off values indicating the presence of inflamed plaque tissue need to be newly defined for PET/MRI.

Keywords: 18F-FDG; Atherosclerosis; Carotid plaque; Inflammation; PET/CT; PET/MRI.

Fig. 1

Fused PET/CT (a–f) and PET/MRI (g–l) images of the right carotid artery in a 52-year-old patient with a head and neck tumour (blue circles ROIs of a plaque at the carotid bifurcation). The right carotid artery shows focal pathological ¹⁸F-FDG uptake on the PET images with CT-based attenuation correction. Colocalized focal uptake of ¹⁸F-FDG is seen on the PET image with MR attenuation correction

Fig. 2

a, b Linear regression analysis of SUV (a) and TBR (b) for 34 carotid plaques obtained on ¹⁸F-FDG PET/MRI and ¹⁸F-FDG PET/CT. c, d Bland Altman analysis of the agreement between the two systems (SUV and TBR on PET/CT minus SUV and TBR on PET/MRI, with a lower bias of −0.07 for TBR_mean compared with a bias of 0.8 for SUV_mean)

Fig. 3

Mean SUV_max (a) and TBR_max (b) in each individual patient (n = 34) on ¹⁸F-FDG PET/CT and PET/MRI. In all patients, SUV_max values were higher on PET/CT, and in a majority of patients (n = 20), TBR_max values were higher on PET/MR

Fig. 4

Regression analysis between amount of radioactivity and circulation time (minutes). Both SUV_max and TBR_max of carotid plaque, on both PET/CT and PET/MRI, showed an upward trend from early to late imaging time-points with continued decreasing activity as a result of blood clearance