ABSTRACT

Purpose:

We aimed to determine the potential and accuracy of multidetector computed tomography (MDCT) coronary angiography with an eight-row scanner in the evaluation of coronary artery disease, and find the best reconstruction percen-tiles for motion free evaluation of each coronary artery according to the phase of the cardiac cycle.

Materials and Methods:

Fifty-one patients, 19 females and 32 males, aged between 39 and 80 years (mean, 55.6 ± 9.6 years) with suspected coronary artery disease were included in the study between December 2003 and July 2005. Examinations were performed using an MDCT scanner with eight detector rows (Light Speed, General Electric, Wisconsin, USA) after the intravenous injection of 120-140 cc of 400 mg I/ml non-ionic contrast media (Iomeron, Bracco, Italy) at 5 ml/sec. Reconstruct-ed images were created from the source images using retrospective ECG gating between the 30th and 80th percentiles of the cardiac cycle. The best phases of the cardiac cycle that allowed optimal imaging of the main coronary arteries were investi-gated. The lengths of the main coronary arteries that could be evaluated free from artifacts were measured. Findings of MDCT examination were compared with catheter angiography in 14 patients.

Results:

Stenoses exceeding 50% of the luminal diameter were found with MDCT in seven of 14 patients. In four patients findings of both methods were concordant (true positives). However, catheter angiography was normal in three patients with significant stenosis on MDCT (false positives). Both methods were negative for significant stenosis in the remaining seven cases (true negatives). On a per-patient basis, sensitivity of MDCT was 100%, but the specificity was 70%. There was a medium degree of correlation between both methods (kappa coefficient 0.571). On a per-lesion basis, stenosis of 50% and higher was found on MDCT in 13 segments. Seven of these were confirmed on conventional angiography. There were six false positive and four false negative results. On a per-lesion basis, the sensitivity of MDCT was 63% and the specificity was 53%. It was possible to image the left main artery in all reconstruction sets without artifacts. The left anterior descending artery was best seen in the 70th percentile of the cardiac cycle. The right coronary artery was best seen in the 50th percentile and the left circumflex artery was best seen in the 40th percentile. The average lengths of the main coronary arteries that could be evaluated were found to be 11.14 ± 2.81 mm for the left main artery, 108 ± 18.99 mm for the left anterior descend-ing artery, 68.02 ± 17.26 mm for the left circumflex artery and 105.45 ± 25.68 mm for the right coronary artery. Ratios of the artefact free lengths were 84.06 ± 16.74% for the left anterior descending artery, 75.31 ± 20.48% for the right coronary artery and 68.02 ± 17.26% for the left circumflex artery.

Conclusion:

Coronary angiography performed using MDCT systems with eight detector rows do not have the potential of replacing conventional coronary angiography. They are not satisfactory in complete and artefact-free imaging of the coro-nary arteries because of low temporal resolution. Therefore they cannot reliably detect and exclude significant coronary stenosis.