Evaluation of Extracellular Volume and Coronary Artery Disease in Cardiac Amyloidosis Using Photon-Counting CT

In cardiac amyloidosis (CA) protein misfolding and consecutive storage into the extracellular myocardial compartment causes left ventricular hypertrophy and, in later stages of the disease, heart failure. The aim of this study was to compare extracellular volume (ECV) measurements obtained from photon-counting CT (PCCT) to the imaging reference cardiac magnetic resonance imaging (CMR) and to evaluate coronary artery disease (CAD) in a CA cohort.

Thirty CA patients (mean age 77.5 +/- 7.9 years) underwent clinically indicated coronary CT angiography (CCTA) for the evaluation of CAD on a first-generation PCCT including a late-phase scan for assessment of ECV. ECV in PCCT was derived using 2 different techniques: (I) a single-energy (SE) technique, based on attenuation changes between the precontrast calcium scoring scan and delayed CCTA in the equilibrium phase (II) a dual-energy (DE) technique, based on iodine density maps from the delayed scan. Both methods were compared with CMR-derived ECV. Statistical analysis included repeated-measures analysis of variance (RM-ANOVA) with Bonferroni-adjusted pairwise comparisons. Correlations between methods were assessed using Pearson's correlation coefficient, and agreement was evaluated using Bland-Altman analysis.

CMR exhibited the highest mean ECV value (42.93 ± 10.14), followed by the SE method (42.5 ± 9.1), while the DE method yielded the lowest ECV values (40.7 ± 9.2). When compared with CMR, ECV obtained via the DE method was significantly lower ( MDiff  = -2.24, P =  0.04). In contrast, no significant difference was observed between CMR and the SE method ( MDiff  = 0.43, P  = 1.00). Differences between the DE and SE methods were significant ( MDiff  = -1.82, P <  0.001). Despite these differences, all 3 methods demonstrated excellent positive correlations. The strongest correlation was observed between the DE and SE methods ( r  = 0.98, P  < 0.001), indicating high consistency in their measurements. Comparatively, the correlation between CMR and DE ( r  = 0.892, P  < 0.001) was slightly stronger than that between CMR and SE methods ( r  = 0.882, P  < 0.001). CAD was present in 29 (97.0%) CA patients with a mean Agatston score of 1086 ± 1398 (range 0-6848.5). Despite this high mean plaque burden and 14 (47.6%) patients presenting with atrial fibrillation, image quality was preserved in 29 (97.0%) patients with 17 (57.6%) of the patients having nonobstructive CAD.

Compared to the imaging reference standard CMR, ECV derived from the DE and SE methods via PCCT demonstrated excellent positive correlations with CMR. The DE method exhibited minor differences compared to CMR, which were clinically not relevant. CAD with an extensive burden of calcified plaque was highly prevalent in CA; however, 57.6% of patients presented with nonobstructive CAD. Therefore, PCCT is a valuable tool for imaging both the coronary arteries and myocardial structure in CA.