In this prospective same-day comparison of 29 patients, myocardial extracellular volume (ECV) measured with PCCT showed strong agreement with cardiac MRI, supporting spectral PCCT as a quantitative tool for myocardial tissue characterization.
Dual-energy PCCT achieved correlations of r = 0.91 for global ECV and r = 0.82 for midventricular ECV versus MRI. Dual-energy PCCT slightly overestimated ECV by approximately 2%, whereas the single-energy approach underestimated ECV by approximately 3%. Despite this small systematic bias, reliability versus MRI was good to excellent, with ICCs of 0.90 globally and 0.81 at the midventricular level for dual-energy PCCT.
A major advantage of the spectral approach was that dual-energy ECV required only the late-enhancement acquisition, eliminating the true unenhanced scan needed for single-energy ECV. This reduced radiation exposure by 40% (CTDIvol 10.1 vs 16.8 mGy) while also reducing potential interscan registration errors.
Access publication
Figure 3. Images show subepicardial late enhancement in a 50-year-old woman with history of arrhythmogenic right ventricular cardiomyopathy. Short- and long-axis cardiac MRI and CT scans demonstrate patchy left ventricular midmyocardial and subepicardial involvement. Enhancement pattern may suggest arrhythmogenic right ventricular cardiomyopathy with left ventricular involvement, although remote myocarditis cannot be ruled out. (A–C) Top row: Images show (A) basal and (B) midventricular short-axis late gadolinium enhancement (LGE) MRI scans with subepicardial involvement (arrows), similarly in (C) the long-axis image. (D–F) Middle row: Images are corresponding CT extracellular volume (ECV) maps showing a similar pattern of myocardial involvement (arrows). (G–I) Bottom row: Images show (G, H) corresponding, short-axis MRI ECV maps demonstrating similar areas of involvement (arrows), and a (I) representative CT ECV American Heart Association 17-segment subepicardiac (50% to outer 10% of the myocardial wall) polar map, which shows the hematocrit (Hct) and the global ECV mean, as well as the means and SDs of ECV for each myocardial segment.
Protocol
Detail | Value |
|---|
Scanner | NAEOTOM Alpha |
Scan area | Cardiac / myocardium |
Scan mode | QuantumPlus sequential |
Acquisition mode | PCCT, full spectral data |
Scan direction | Axial |
Tube voltage [kV] | 120 |
Effective mAs [mAs] | CARE Dose4D automatic |
IQ level | 20 unenhanced; 44 CCTA; 50 late enhancement |
Dose modulation | CARE Dose4D |
CTDIvol [mGy] | Dual-energy ECV: 10.1; single-energy ECV: 16.8 |
DLP [mGy·cm] | Dual-energy ECV: 148; single-energy ECV: 248 |
Rotation time [s] | 0.25 |
Acquisition slice thickness [mm] | 0.4 |
Reconstruction slice width [mm] | 1.5 |
Reconstruction increment [mm] | 1.0 |
Reconstruction kernel | Qr40 |
Iterative reconstruction | QIR 3 |
Reconstruction energy | 65 keV |
Reconstruction comparison | Single-energy ECV vs dual-energy iodine-based ECV |
ECG gating | ECG-triggered |
ECG window | 280 ms RR interval for unenhanced and late enhancement; 30–80% RR for CCTA |
Contrast agent | Iopromide, Ultravist 370 mg I/mL |
Contrast volume | 100 mL |
Saline flush | 20 mL |
Injection rate | 5–6 mL/s |
Bolus triggering | n.a. |
Late-enhancement phase | 5 min after contrast administration |
ECV reconstruction | Dual-energy iodine maps from late-enhancement scan; automated 17-segment ECV maps |
Key Results
Parameter | Main finding |
|---|
Study population | 29 patients |
Global ECV, dual-energy PCCT vs MRI | r = 0.91; ICC = 0.90 |
Midventricular ECV | r = 0.82; ICC = 0.81 |
Dual-energy bias | ECV overestimated by 1.9% globally |
Radiation reduction | 40% with dual- vs single-energy ECV |
Clinical implication | Spectral PCCT may add quantitative myocardial tissue characterization to cardiac CT without requiring an additional unenhanced scan |
Source:
Aquino GJ, O’Doherty J, Schoepf UJ, Ellison B, Byrne J, Fink N, Zsarnoczay E, et al. Myocardial Characterization with Extracellular Volume Mapping with a First-Generation Photon-counting Detector CT with MRI Reference. Radiology. 2023;307(2). doi:10.1148/radiol.222030.
