• atherosclerosis
• cardiac computed tomography
• computed tomography
• scanning
• coronary artery disease
• imaging

Contemporary multidetector row computed tomography (CT) scanners permit the acquisition of non-invasive coronary angiograms with good diagnostic accuracy for the detection of coronary stenosis. Early studies have demonstrated that, in addition to the depiction of luminal narrowing, CT also allows visualisation of coronary atherosclerotic plaque.1 2 However, given the small dimensions of coronary atherosclerotic lesions, further characterisation of coronary atherosclerosis and the exact quantification of plaque volumes has remained challenging. Correlation of plaque volumes determined by CT and intravascular ultrasound (IVUS) has been reported early.1 2 However, margins of agreement have been wide because the accuracy of CT plaque volume quantification has been hampered by insufficient spatial resolution, calcium blooming artefacts, the presence of motion artefacts and image noise. Nevertheless, as CT technology has matured and image acquisition protocols have been improved, newer studies have shown stepwise improvements in our ability to quantify coronary plaque burden.1–3

In this issue of Heart, Schepis et al4 report the results of their study comparing the quantification of non-calcified coronary plaque volumes by a very recent CT scanner generation, dual-source computed tomography (DSCT) (see page 610). The authors examined high-quality CT coronary angiograms of 70 subjects and measured plaque volumes in 100 non-calcified plaques. Corresponding IVUS images were used as the gold standard for the measurements of plaque volumes. The agreement between DSCT and IVUS was good (correlation r=0.89, mean difference −1±34 mm³). However, the limits of agreement were still relatively wide (−67 to +65 mm³). Studies with previous CT scanner generations have demonstrated that the size of non-calcified plaques was typically underestimated, and calcified or partly calcified plaques were overestimated.1–3 5 The results of the study by Schepis et al4 suggest that the next generation of scanners with better temporal resolution may provide improved quantification …