Non-invasive diagnosis of coronary artery disease with 16-slice computed tomography.

BACKGROUND: The gold standard for the diagnosis of coronary artery disease (CAD) is catheter angiography. However, catheter angiography is invasive and may not always be followed by interventional therapy. We report our results with the use of multislice computed tomography (MSCT) as a non-invasive diagnostic tool for CAD. METHODS: Thirty-one patients (26 with chronic stable angina, 5 with coronary anomalies) underwent 16-slice MSCT and catheter angiography. Vessels < 1.5 mm in diameter were excluded. The ability of MSCT to detect obstructive CAD (stenosis > or =50% of the diameter) was evaluated in pre-defined vessel segments. The association of calcium score with obstructive CAD, and the effect of heart rate on distal vessel visibility were also studied. RESULTS: There were 29 men and 2 women (age range: 36-80 years; mean [SD]: 53 [11] years). Of the 403 vessel segments, 391 were > 1.5 mm in diameter and 321 were interpretable on both modalities. The non-interpretability rates were 7% (26/391) for catheter angiography and 14% (54/ 391) for MSCT, with distal location (64%; 34), motion artifacts (29%; 16) and calcification (7%; 4) being chiefly responsible in case of the latter technique. The sensitivity, specificity, positive and negative predictive values of MSCT were 85% (95% confidence interval [CI]: 73-93), 94% (95% CI: 90-96), 76% (95% CI: 64-85) and 96% (95% CI: 93-98), respectively. MSCT correctly classified patients with no, single-, double- and triple-vessel disease in 87% of cases. One patient was incorrectly excluded on MSCT; catheter angiography showed 50%-70% stenosis in this case. Patients with obstructive CAD had a higher Agatston score equivalent (p=0.03). There was no significant effect of heart rate on distal segment visibility. MSCT correctly identified all coronary anomalies. CONCLUSION: MSCT has a good potential for the detection of coronary stenosis, and may be most useful for excluding CAD (due to its high negative predictive value). It accurately delineates coronary anomalies.