Photon-counting CT-derived extracellular volume in acute myocarditis: Comparison with cardiac MRI

Abstract

Purpose

The purpose of this study was to evaluate the feasibility and the accuracy of myocardial late iodine enhancement for extracellular volume (ECV) quantification using dual-source photon-counting detector computed tomography (PCD-CT) in patients with suspected acute myocarditis by comparison with cardiac MRI.

Materials and methods

Patients with clinical suspicion of myocarditis who were referred for coronary CT angiography (CCTA) to exclude coronary artery disease were included in this retrospective study. All patients underwent CCTA examination using a first-generation PCD-CT, which included slate iodine enhancement images. ECV was calculated from the iodine ratio of the myocardium to the blood pool on late iodine enhancement PCD-CT images. A comprehensive cardiac MRI protocol was used as the reference method to confirm myocarditis according to the Lake Louise 2018 criteria. All subjects underwent CCTA using PCD-CT and cardiac MRI within 24 h. The mean dose-length product of late enhancement PCD-CT scanning was calculated. Correlations between ECV PCD-CT (endocardial, epicardial, midcardial, and global), cardiac MRI-LGE, and right and left ventricular ejection fractions were assessed using Pearson correlation test. ECV values derived from PCD-CT and those from cardiac MRI were compared using Bland Altman plots and linear regression analysis. Areas under the receiver operating characteristic curves (AUCs) were used to determine the optimal thresholds of ECV-PCD-CT and ECV-MRI for differentiating patients with myocarditis from those not meeting the Lake Louise criteria.

Results

Thirty-two patients were included. There were 19 men and 13 women with a mean age of 35.9 ± 15.0 (standard deviation [SD]) years; age range: 21–51). The mean dose-length product of late enhancement PCD-CT scanning was 96 ± 32 (SD) mGy.cm. No significant differences in mean global ECV were found between ECV calculated with the PCD-CT (29.4 ± 4.5 [SD] %) and that calculated with cardiac MRI (30.0 ± 4.1 [SD] %) (P = 0.69). ECV-CT was greater in patients with cardiac MRI-confirmed myocarditis (31.65 ± 3.6 [SD] %) by comparison with those with normal findings (25.6 ± 3.2 [SD] %) (P < 0.01). ECV-CT strongly correlated with LGE mass (r = 0.82) and showed strong segmental correlation with ECV-MRI (basal: r = 0.95; mid-ventricular: r = 0.91). An ECV-CT threshold of 26.9 % yielded an AUC of 0.95 (95 % CI: 0.84–1.00) for the diagnosis of myocarditis.

Conclusion

Calculation of ECV using iodine maps derived from late iodine enhancement cardiac PCD-CT images is both feasible and accurate at low radiation doses. PCD-CT appears as a promising non-invasive imaging modality for the diagnostic and prognostic assessment of acute myocarditis in the setting of chest pain.