A prospective randomized trial comparing image quality, study interpretability, and radiation dose of narrow acquisition window with widened acquisition window protocols in prospectively ECG-triggered coronary computed tomography angiography

BackgroundProspectively triggered coronary computed tomography angiography (CTA) is commonly performed with a widened acquisition window to provide flexibility in image reconstruction.ObjectiveWe conducted a randomized controlled trial to determine whether the use of a narrow acquisition window in prospectively triggered coronary CTA would allow lower radiation dose while preserving image quality and interpretability.MethodsProspective 2-center 2- platform randomized trial that evaluated 205 consecutive patients 96 with widened acquisition (WA) and 109 narrow acquisition (NA) referred for coronary CTA in sinus rhythm and heart rate <65 beats/min. Patients scanned with WA had phases reconstructed at 5% intervals, and each phase was assigned an individual study ID. Images were reviewed with individual phase reconstructions interpreted randomly by 2 level 3 readers with a third for consensus. Images were evaluated with a 5-point Likert scale on a per-vessel basis (best score on any phase). Scores were then dichotomized into diagnostic (score 3-5) compared with nondiagnostic (score 1-2). Readers also reported obstructive coronary artery disease on a per-patient basis. Agreement for the diagnosis of obstructive disease and per-artery interpretability was performed. Signal and noise measurements were also performed.ResultsNo difference in demographics between groups (P = NS). The signal-to-noise ratio was comparable 12.99 ± 3.4 NA and 12.53 ± 4.13 for the WA (P = 0.45). The median effective dose was 1.78 mSv for NA compared with 3.26 mSv for WA (P < 0.001). Image quality, diagnostic interpretability, interreader agreement, and downstream testing were not significantly different between the 2 groups (P= NS for all).ConclusionsCoronary CTA with NA resulted in a 47% lower radiation dose without significant difference in study interpretability or image quality or increased downstream resource use or testing.