OBJECTIVE The aim of this study was to evaluate the diagnostic accuracy (sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV)) of 64-slice multidetector computed tomography (MDCT) compared with quantitative coronary angiography (QCA) for detection of coronary artery disease (CAD). METHODS Sixty-nine patients participating in a study of coronary in-stent restenosis were investigated. After a 64-slice MDCT scan patients were evaluated by QCA. The coronary arteries were divided into 15 segments and stenosis was graded for each segment by both methods. The diagnostic accuracy of 64-slice MDCT was evaluated using the QCA as the gold standard. RESULTS Among the 69 patients included in the study 13 (19%) were female and 56 male. The mean age was 63 (SD 10) years. The following risk factors were present: high blood pressure 67%, elevated blood cholesterol 54%, diabetes 12% and family history of CAD 71%. Current smokers were 22% and previous smokers were 48%. Altogether 663 segments were examined. Of those 221 (33%) segments were excluded; 103 because of stents, 48 because of heavy calcification, 41 because of motion artifacts and 29 because the segments were less than 1.5 mm in diameter. The mean time between MDCT and QCA was 6.3 (SD 12.1) days. The sensitivity of 64-slice MDCT for diagnosing significant stenosis (>or= 50% according to QCA) was 20%, the specificity was 94%, PPV was 16%, NPV was 95% and the accuracy was 89%. CONCLUSIONS High NPV and specificity indicates that MDCT is useful for accurately excluding significant CAD but the low sensitivity and low PPV indicate that the method is not accurate in diagnosing coronary artery stenosis of 50% or more according to QCA.