Prospectively ECG-triggered coronary CT angiography images are acquired during a window in middiastole. Additional surrounding x-ray beam on time, or padding, can be variably set, and the increased padding results in additional available phases for analysis. The purpose of this study was to assess the effect of padding duration on image interpretability and its incident effect on radiation dose. We prospectively evaluated imaging of 886 patients undergoing consecutive prospectively ECG-triggered coronary CT angiographic examinations at three centers and compared the findings in patients stratified by padding duration. We assessed the effect of padding duration on image interpretability and radiation dose. The mean patient age was 56 +/- 12 years, and 58% of the patients were men. The median heart rate was 55 beats/min (interquartile range, 50-61 beats/min). Padding duration was 0, 1-99, and 100-150 milliseconds for 268, 482, and 136 patients, respectively, with no difference in image interpretability rate between groups (per patient, 98.8%, 97.3%, and 97.1%; per artery, 99.2%, 99.2%, and 99.1%). The groups differed in median radiation dose (2.3 mSv [interquartile range, 1.5-3.2 mSv]; 3.8 mSv [interquartile range, 2.3-4.7 mSv]; 5.5 mSv [interquartile range, 3.8-6.1 mSv]; p < 0.001). Independent of patient and scan parameters, increased padding was associated with greater radiation dose (45% increase per 100-millisecond increase in padding, p < 0.001). In a large multicenter study of coronary CT angiography of patients with excellent heart rate control, the use of minimal padding was associated with a substantial reduction in radiation dose with preserved image interpretability. Use of no or reduced padding should be considered in dose-reduction strategies.