Meta-analysis of the diagnostic accuracy of computed tomography angiography compared with invasive coronary angiography in preoperative cardiac surgery planning: a focus on valve surgery patients.

OBJECTIVE: To investigate the diagnostic performance of coronary CT angiography (CCTA) for assessing significant coronary artery disease (CAD) in patients referred for surgical aortic valve replacement or transcatheter aortic valve implantation (TAVI)\transcatheter aortic valve replacement (TAVR), with invasive coronary angiography (ICA) as the reference standard. METHODS: We performed a meta-analysis of 28 studies to compare CCTA with ICA for preoperative coronary evaluation. Studies were stratified into two subgroups: the first consisting of those which included only patients undergoing valve surgery (n=19) and the second including TAVI or mixed (TAVI and surgical) populations (n=9). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were recorded or determined, and a summary diagnostic performance was obtained by a random effects model. Pooled forest plots and summary receiver operating characteristic curves were also analysed. RESULTS: The overall sensitivity of CCTA to diagnose significant CAD varied between 18 studies, ranging from 85% to 94%; the pooled sensitivity over all 28 studies was 91% (95% CI 88% to 93%) and the specificity was 88% (95% CI 84% to 91%). The pooled PPV was 78% (95% CI 72% to 83%), while the NPV was 95% (95% CI 93% to 97%). The diagnostic performance of the study was 89.8%. CONCLUSIONS: CCTA is a trustworthy, non-invasive diagnostic option to rule out significant CAD in patients undergoing valve surgery. Its high specificity in surgical candidates favours its use as a 'gatekeeper' to ICA with a potential reduction in unnecessary invasive surgery.