100 Diagnostic accuracy of high definition computed tomographic coronary angiography compared to invasive coronary angiography in the assessment of patients with high pre-test probability of or established coronary artery disease

Background 64-slice Computed Tomographic Coronary Angiography (CTCA) has high negative predictive value in assessment of patients with low-risk of coronary disease (CAD). However, it has a low specificity in assessment of patients with high pre-test probability of or established CAD, due to limited spatial resolution and blooming artefact from coronary artery calcium. Recently published NICE guidelines recommend the use of conventional CTCA for the assessment of patients with low pre-test probability of CAD, but not for patients with high pre-test probability of CAD or those with calcified coronary atheromatous disease (Agatston Calcium score >400). High-definition CT (HDCT) combined with the use of iterative reconstruction (ASIR), aims to address the shortcoming of conventional CT technology by improving spatial resolution and reducing calcium blooming artefact, without increasing ionising radiation exposure. Methods Patients with high pre-test probability of and established CAD, were prospectively enrolled in our HD-CTCA accuracy trial. We present the interim results of our 50 consecutive patients who underwent HD-CTCA following invasive coronary angiography (ICA) for the assessment of coronary disease. HD-CTCA was conducted on all patients within 30 days of ICA. Anonymised ICA and HD-CTCA studies were evaluated separately and results compared with ICA as the reference standard. Results All HD-CTCA studies were acquired using prospective gating, 100 kV tube voltage and optimum radiation dose reduction strategies and images were reconstructed using 50% ASIR. The male: female ratio was 37:13 and the median (IQR) age, BMI and Agatston Calcium Score of patients at the time of scanning were 67.5 (60–76.5) years, 26.5 (24.4–28.6) kg/m 2 and 708 (293–1615) respectively. The median (IQR) radiation dose was 151 (131–275) mGy.cm, representing effective doses of 4.2 (3.7–7.7) mSv using a cardiac specific conversion factor (0.028/cm). All coronary segments visualised on ICA were demonstrated on HD-CTCA. Of the 726 coronary segments evaluated on HD-CTCA, 96.4% were of excellent, 2.8% moderate and 0.8% poor diagnostic quality. Compared to ICA, HD-CTCA had sensitivity and specificity of 97% (95% CI 81% to 100%) and 95% (95% CI 72% to 99%) on a per-patient basis and sensitivity and specificity of 94% (95% CI 87% to 980%) and 98% (95% CI 97% to 99%) on a per-coronary segment basis respectively. Conclusion Our interim results demonstrate that HD-CTCA has excellent accuracy compared to ICA in the assessment of patients with high pre-test probability of CAD or with established CAD and can be performed within acceptable radiation dose limits.