Optimal post-contrast timing for photon-counting CT-based myocardial extracellular volume quantification in a swine model: Cardiac magnetic resonance as the reference.

RATIONALE AND OBJECTIVES

Advances in computed tomography (CT) have enabled myocardial extracellular volume (ECV) quantification, but the optimal post-contrast timing for CT-derived ECV remains undefined. In this preclinical study using a swine model, we aimed to determine the optimal post-contrast acquisition time for photon-counting CT (PCCT)-derived ECV using cardiac magnetic resonance (CMR) as the reference standard.

MATERIALS AND METHODS

Twenty-three CMR and PCCT scans were performed on infarcted swine, and nine scans on healthy controls. PCCT scans were acquired at 3-, 5-, 7-, 9-, 12-, and 15-minutes after contrast administration. Pearson correlation coefficients (r) and intraclass correlation coefficients (ICCs) were calculated between PCCT- and CMR-derived ECV across all myocardial segments, infarcted and normal myocardium. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of PCCT-ECV in differentiating LGE-positive from LGE-negative myocardial segments. Subgroup analyses were conducted based on heart rate (HR).

RESULTS

Across all segments, r ranged from 0.79 to 0.91 and ICCs from 0.88 to 0.95 and the area under the ROC curve (AUC) values ranged from 0.799 to 0.847 (all p < 0.001). Peak correlation and agreement occurred at 5 min (r = 0.91; ICC = 0.95, 95 % CI: 0.92-0.96), with similarly high values at 3 and 7 min (r = 0.90; ICC = 0.93 and 0.94, respectively), followed by a gradual decline. At elevated HRs (>84 bpm), PCCT still maintained robust agreement with CMR (ICC = 0.78-0.92) and satisfactory diagnostic performance (AUC values ranged from 0.677 to 0.777).

CONCLUSION

PCCT enables accurate and consistent ECV quantification with strong concordance to CMR from 3 to 7 min post-contrast in swine models, providing a wide temporal window for clinical scanning. This was further confirmed by ROC analysis, and the method demonstrated robustness across varying heart rates.