Author: Cecelia Brewington, Connor Edwards Miles All Leska-Kent, Kyrillos Grace, Katherine Hyde, Sabee Molloi, Shiva Mostafavi, Liqiang Ren, Alireza Shojazadeh, Justin Truong 👨🔬
Affiliation: Department of Radiology, UT Southwestern Medical Center, University of California, Irvine, University of Texas Southwestern 🌍
Purpose: To retrospectively assess image quality in coronary CT angiography (CCTA) using a patient-specific post-trigger delay (PTD) protocol, developed using half of the injection time plus a dispersion delay constant, and compare it to a standard clinical protocol in a human dataset.
Methods: Forty patients suspected of lung lesions underwent dynamic CT perfusion, with imaging every 2 seconds for 28 seconds after receiving 40-50 mL of Isovue 370 and 20 mL of saline at 5 mL/s. Enhancement curves from the aortic root were fitted with a gamma variate function to determine true peak time, coronary artery enhancement (HU), and contrast-to-noise ratios (CNR) in the left anterior descending, right coronary, and left circumflex arteries as reference values. Those curves were then used to simulate enhancement (HU) and CNR for both the standard protocol with a fixed 5-second PTD scan and the individualized PTD protocol. The individualized protocol predicts contrast peak time using half the injection time plus a fixed 2.3-second dispersion delay.
Results: The predicted peak time from the individualized PTD method was within -0.80±0.85 s of the true peak time, with enhancement and CNR across all arteries within 9.8% and 8.1% of reference values, respectively. Regression analysis showed a strong correlation between predicted and reference values (Pearson’s correlation=0.92, R²=0.70, RMSE=60.03 HU for enhancement; Spearman’s correlation=0.93, R²=0.79, RMSE=1.62 for CNR). In contrast, for the standard protocol, acquisition time was within -2.11±0.82 s of the true peak time, with coronary enhancement and CNR 37% and 27% lower than the reference values. A poor correlation was found between standard and reference values (Pearson’s correlation=0.71, R²=-0.54, RMSE=146.97 HU for enhancement; Spearman’s correlation=0.73, R²=-0.28, RMSE=4.18 for CNR).
Conclusion: The individualized PTD CCTA protocol accurately predicts the time to peak of coronary artery enhancement and provides superior image quality compared to the standard fixed PTD technique.