To assess the impact of different scanning protocols on radiation dose and image quality for pediatric coronary computed tomography.

100 children underwent coronary computed tomography after arterial switch operation from November 2012 to March 2014. Scans were done using two different scanner models without difference in scanning and image reconstruction parameters: Light speed VCT and Discovery HD750, 64-slice from GE Healthcare. Two consecutive changes in scanning protocols were performed: 1) the use of adaptive statistical iterative reconstruction (ASIR) instead of filtered back projection (FBP) for image reconstruction; 2) the optimization of scan acquisition parameters (current and tube voltage reduction). Premedication (beta-blocker) was used for all children to obtain heart rate<80 BPM. Effective dose (ED) was calculated with the dose-length product method with a conversion factor adjusted for patient age. Image quality was evaluated by the referent physician. Scans were classified as “excellent”, “good” or “with significant artifacts”.

Patients were divided in three age groups: 0–4, 5–7 and 8–18years. After adjustment for scan settings, median ED decreased by 28% (3.9mSv, IQR 2.8–4.2), 40% (0.9mSv, IQR 0.6–2.6) and 65% (0.7mSv, IQR 0.5–0.9) for 0–4, 5–7 and 8–18years age groups (P<0.05), respectively. The prospective protocol (PULSE) was used in 40% of children. The reduction in radiation dose was not associated with reduction in diagnostic image quality.

Coronary CT can be obtained at very low radiation doses in pediatric patients using ASIR and prospective ECG-triggered acquisition with optimized scan parameters.