Optimal image reconstruction intervals for non-invasive coronary angiography with 64-slice CT

Abstract

The reconstruction intervals providing best image quality for non-invasive coronary angiography with 64-slice computed tomography (CT) were evaluated. Contrast-enhanced, retrospectively electrocardiography (ECG)-gated 64-slice CT coronary angiography was performed in 80 patients (47 male, 33 female; mean age 62.1+/-10.6 years). Thirteen data sets were reconstructed in 5% increments from 20 to 80% of the R-R interval. Depending on the average heart rate during scanning, patients were grouped as < 65 bpm (n = 49) and > or = 65 bpm (n = 31). Two blinded and independent readers assessed the image quality of each coronary segment with a diameter > or = 1.5 mm using the following scores: 1, no motion artifacts; 2, minor artifacts; 3, moderate artifacts; 4, severe artifacts; and 5, not evaluative. The average heart rate was 63.3 +/- 13.1 bpm (range 38-102). Acceptable image quality (scores 1-3) was achieved in 99.1% of all coronary segments (1,162/1,172; mean image quality score 1.55 +/- 0.77) in the best reconstruction interval. Best image quality was found at 60% and 65% of the R-R interval for all patients and for each heart rate subgroup, whereas motion artifacts occurred significantly more often (P < 0.01) at other reconstruction intervals. At heart rates < 65 bpm, acceptable image quality was found in all coronary segments at 60%. At heart rates > or = 65 bpm, the whole coronary artery tree could be visualized with acceptable image quality in 87% (27/31) of the patients at 60%, while ten segments in four patients were rated as non-diagnostic (scores 4-5) at any reconstruction interval. In conclusion, 64-slice CT coronary angiography provides best overall image quality in mid-diastole. At heart rates < 65 bpm, diagnostic image quality of all coronary segments can be obtained at a single reconstruction interval of 60%.