Introduction Arterial calcification can limit the visualisation of vessel lumen at CT coronary angiography (CTCA). Dual energy CT (DECT) using two x-ray spectra of differing energy, either from distinct sources or using novel detectors and a single source, offers novel approaches to this problem. It allows the generation of images depicting objects as if they have been subjected to a specific photon energy (keV) rather than a polychromatic beam (virtual monochromatic images). With improved material identification, based on its attenuation coefficient at each keV and transformation into a linear combination of the two basis materials (material decomposition), it may also be possible to subtract materials from each other, such as calcium from iodine.

Single source DECT, using two energy spectra from a single anode almost simultaneously using rapid tube voltage switching, requires novel detector technology capable of distinguishing signals 0.25 ms apart without artefact from detector artefact. To date, this has not been evaluated in patients with coronary calcification. This feasibility study examined potential benefits and limitations of virtual monochromatic and material decomposition images for assessing calcified coronary arteries, and their potential diagnostic accuracy compared to invasive angiography.

Methods The study was approved by the National Research Ethics Service. Patients gave informed, written consent. Thirty patients undergoing invasive angiography on clinical grounds, with evidence of coronary calcification, underwent CTCA with a single-source DECT scanner. The results of each test were assessed by experienced, independent, blinded readers and compared in per-segment, per-vessel and per-patient analyses.

Results 403 segments in 86 vessels were analysed. The median Agatston score was 964. The accuracy of virtual monochromatic imaging is outlined in Table 1. Overall the accuracy for the identification of moderate and severe stenosis was 0.88 and 0.88 on a per-segment basis, 0.84 and 0.86 per vessel, and 0.93 and 0.97 per-patient. The weighted kappa score between invasive and CT angiography was 0.71 suggesting good agreement.

The per-segment sensitivity, specificity, PPV and NPV (and 95% confidence intervals) of the material decomposition images were 0.67 (0.57–0.76), 0.82 (0.77–0.86), 0.54 (0.45–0.63) and 0.88 (0.84–0.92) respectively for moderate stenosis, and 0.70 (0.57–0.80), 0.79 (0.75–0.83), 0.40 (0.31–0.49) and 0.93 (0.89–0.96) respectively for severe stenosis. Overall accuracy was 0.78 for both moderate and severe stenosis. Calcium subtraction was highly inconsistent, mainly due to image noise with resultant misidentification of calcium and excessive subtraction.

Conclusions The study suggests that single source DECT is feasible in patients with severe coronary calcification, and virtual monochromatic imaging may improve accuracy compared to conventional CT. A larger study comparing standard and DECT is merited.