Comparison of coronary computed tomography angiography using 64- and 320-row computed tomography scanners in diagnosing coronary artery disease

Background: Coronary computed tomography angiography (CTA) is a noninvasive tool used to evaluate coronary artery disease. Our study aims to compare the diagnostic performance of the two commonly used commercialized CT systems, 64- and 320-row CT, in evaluating significant coronary artery stenosis, which is defined as a reduction of luminal diameter by ≧50%. Materials and Methods: Two radiologists reviewed the CT images of coronary arteries in our hospital that were performed from 2006 to 2014. We enrolled 259 patients aged from 37 to 87 years. All individuals had their cardiac catheterization performed within a month. We used the results of angiographically measured stenosis as the standard. We applied the two-sample z-test to compare the diagnostic performance of the two commonly used CT modalities. Results: A total of 131 patients were evaluated with a 64-row CT scanner, and 128 patients, with a 320-row CT scanner. The amount of contrast medium used in the 64-row CT scanner was significantly higher than that used in the 320-row CT scanner (90.16 ± 4.45 mL vs. 60.52 ± 2.45 mL, respectively; P < 0.001). The comparison was performed based on significant stenosis of the coronary artery diagnosed by conventional coronary angiography. The overall sensitivity, specificity, positive predictive value, negative predictive value, and accuracy did not show statistical difference between the diagnostic performances of the 64- and 320-row CT systems. However, the effective radiation dose of 64-row was significantly higher than that of 320-row (14.25 ± 3.49 mSv vs. 6.26 ± 3.83 mSv, respectively; P < 0.001). Conclusions: Both 64-and 320-row CTA systems have a diagnostic performance in detecting significant coronary artery stenosis without significant difference. However, the 320-row CT is superior to 64-row CT in less radiation dose exposure and administrating less amount of contrast medium.