Background
Coronary artery disease (CAD) is frequently seen in patients suffering from severe aortic valve stenosis (AS), as both pathologies share the same pathophysiology. In a transcatheter aortic valve implantation (TAVI) work-up, patients beneficiate from both computed tomography angiography (CTA) and invasive coronary angiography (ICA). Some studies evaluated the performance of CTA to diagnose CAD among patients undergoing TAVI and showed interesting results1-4. Nevertheless, data remain scarce and this diagnostic method is not validated in this population. In this context, we thought to evaluate the diagnostic performance of CTA to diagnose CAD among patients selected for TAVI.
Methods
A total of 199 patients that had a TAVI in the Lausanne University Hospital between the 1st of June 2013 and the 31st of December 2017 were retrospectively included. Exclusion criteria were coronary artery bypass graft (CABG) prior to CTA and unavailable CTA images. Finally, 127 patients were included. Two independent radiologists – blinded for ICA report – were asked to read the CTA of these patients and to indicate the presence of ≥50% and ≥70% stenosis in the 4 main coronary vessels. Their evaluation was then compared with ICA reports and analyses were performed at vessel and patient levels.
Results
A total of 342 vessels were analyzable. Based on ICA, significant CAD (at least 1 ≥50% stenosis) was present in 49 (38.6%) patients. Severe CAD (≥70% stenosis) was found in 29 (22.8%) patients. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of CTA to diagnose significant CAD were 81.1%, 87.9%, 44.8%, 97.5% and 87.1% at vessel level using the cut-off of 50% and 42.8%, 97.8%, 56.3%, 96.3% and 94.4% for severe CAD, using the cut-off of 70%. At patient level, sensitivity, specificity, positive and negative predictive values were respectively 84.6%, 64.6%, 56.4% and 88.6% for significant CAD.
Conclusion
Pre-TAVI CTA shows good performance to rule out significant and severe CAD and could be used as a gatekeeper for ICA. Positive findings on CTA should be confirmed with ICA given the low positive predictive value.