Comparison of dual-energy iodine and standard subtraction methods for myocardial extracellular volume quantification using cardiac computed tomography

Abstract

Background: Computed tomography (CT)-derived extracellular volume fraction (ECV) quantifies myocardial fibrosis noninvasively, comparable to cardiovascular magnetic resonance. The conventional subtraction method (ECV-conv) requires dedicated pre- and post-contrast acquisitions, while the dual-energy spectral method (ECV-spec) uses only the post-contrast phase. We aimed to compare these methods in patients undergoing CT planning for transcatheter aortic valve replacement (TAVR).

Methods: We prospectively included patients undergoing CT-TAVR evaluation on a dual-energy dual-layer detector scanner. Baseline pre-contrast and equilibrium delay phase (at 5 ​min) prospectively ECG-triggered acquisitions were co-registered for ECV-conv calculation. Equilibrium phase spectral iodine maps were derived for ECV-spec measurement and compared with ECV-con maps using a mid-septal and a global mid-ventricular region of interest.

Results: Overall, 78 patients (53 ​% female, mean age 77 years) were analyzed. There was a minor overestimation in global measurement by ECV-spec (31.8 ​± ​4.6 ​% vs 30.3 ​± ​5.3 ​%, p ​= ​0.023. However, there were no significant differences between both methods for the mid-septal measurement (ECV-conv ​= ​30.4 ​± ​5.3 ​% vs ECV-spec ​= ​31.0 ​± ​5.6 ​%, p ​= ​0.196).). Both methods demonstrated comparable 95 ​% limits of agreement, with a strong correlation for mid-septal (r ​= ​0.75, p ​< ​0.0001) and a moderate correlation for global measurements (r ​= ​0.51, p ​< ​0.0001).

Conclusions: Myocardial CT-ECV estimation using dual-energy spectral maps was comparable to the conventional subtraction method without the need of a pre-contrast acquisition. This approach may help reduce imaging and processing times, as well as minimize radiation exposure.

Keywords: Aortic stenosis; Dual-energy CT; Myocardial extracellular volume; Spectral CT; TAVR.