BACKGROUND: In patients referred for aortic valve replacement (AVR) a pre-surgical assessment of coronary artery disease is mandatory to determine the possible need for additional coronary artery bypass grafting. The diagnostic accuracy of coronary computed tomography angiography (coronary CTA) was evaluated in patients with aortic valve stenosis referred for surgical AVR. METHODS: Between March 2008 and March 2010 a total of 181 consecutive patients were included. All patients underwent pre-surgical coronary CTA (64- or 320-detector CT scanner) and invasive coronary angiography (ICA). The analyses were performed blinded to each other. RESULTS: The mean±SD age of the included patients was 71±9years and 59% were male. The prevalence of significant coronary artery stenosis >70% by ICA was 36%. Average heart rate during coronary CTA was 65±16bpm. In a patient based analysis 94% of the patients (171/181) were considered fully evaluable. Coronary CTA had a sensitivity of 68%, a specificity of 91%, a positive predictive value of 81%, and a negative predictive value of 83%. Advanced age, obstructive lung disease, NYHA function class III/IV, and high Agatston score were found to be significantly associated with disagreement between ICA and coronary CTA in univariate analysis. CONCLUSION: In patients with aortic valve stenosis referred for surgical AVR the diagnostic accuracy of coronary CTA to identify significant coronary artery disease is moderate. Coronary CTA may be used successfully in a subset of patients with low age, no chronic obstructive lung disease, NYHA function class