Room: Track 1
Purpose: optimize the signal-to-noise (SNR) and contrast-to-noise (CNR) ratios of prospective CT pulmonary angiography (CTPA) using an optimal timing protocol, as compared to a standard fixed-delay protocol, with retrospective CTPA as the reference standard.
Methods: total of 24 swine (48.5 ± 14.3 kg) underwent contrast-enhanced dynamic CT acquisition over 20-30 seconds to capture the pulmonary enhancement curves. Multiple contrast injections were made under different cardiac outputs (1.4 - 5.1 L/min), resulting in a total of 154 pulmonary arterial input functions (AIF). The volume scan with maximal enhancement was selected as the reference CT pulmonary angiogram (group A). Two prospective CTPA simulations were then performed on the AIF data using two different timing protocols: one had a fixed acquisition delay of 5 seconds (group B), while the other had an optimal acquisition delay (group C), following bolus-tracking. Specifically, the optimal acquisition delay was calculated using one-half of the contrast injection duration. The objective (SNR and CNR) and subjective image quality of each CTPA were then evaluated. Mean attenuation differences between the three groups were compared using Student’s t-test and linear regression. The SNR, CNR, and subjective evaluation scores were also compared by Wilcoxon-Mann-Whitney test.
Results: CTPA mean attenuation values for group A, group B and group C were 870.1 ± 242.5, 758.7 ± 246.7, and 829.2 ± 236.8 HU respectively (p <.001 between any two groups). The SNR and CNR were both higher in group C than group B (C vs. B: SNR of 62.4 vs. 56.9, p = 0.131; CNR of 56.4 vs. 50.9, p = 0.086), which was in general with the subjective image quality assessment.
Conclusion: optimal timing protocol improves CTPA image quality through improved acquisition timing at the peak contrast enhancement, resulting in higher contrast attenuation of the pulmonary arteries.