Objectives To evaluate the effects of mean heart rate (HR) and heart rate variation (HRV) on image quality and diagnostic accuracy of prospectively ECG-triggered sequential dual-source coronary CT angiography (CCTA) in patients with atrial fibrillation (AF). Methods Eighty-five patients (49 women, 36 men; mean age 62.1 +/- 9.5 years) with persistent AF underwent prospectively ECG-triggered sequential second-generation dual-source CCTA. Tube current and voltage were adjusted according to body mass index (BMI) and iterative reconstruction was used. Image quality of coronary segments (four-point scale) and presence of significant stenosis (>50%) were evaluated. Diagnostic accuracy was analyzed in 30 of the 85 patients who underwent additional invasive coronary angiography (ICA). Results Only 8 of 1102 (0.7%) segments demonstrated poor image quality. No significant impact on image quality was found for mean HR (94.9 +/- 21.8 bpm; r = 0.034, p = 0.758; F = 0.413, p = 0.663) or HRV (67.5 +/- 22.8 bpm; r = 0.097, p = 0.377; F = 0.111, p = 0.895). On per-segment analysis, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 89.7% (26/29), 99.4% (355/357), 92.9% (26/28), and 99.2% (355/358), respectively, with excellent correlation (kappa = 0.91) with ICA. Mean effective dose was 3.3 +/- 1.0 mSv. Conclusions Prospectively ECG-triggered sequential dual-source CCTA provides diagnostic image quality and good diagnostic accuracy for detection of coronary stenosis in AF patients without significant influence by HR or HRV.