To assess the accuracy of computed tomography (CT) performed for planning transcatheter aortic valve replacement (TAVR) for diagnosing obstructive coronary artery disease (CAD).

With institutional review board approval, waiver of informed consent and in HIPAA compliance we retrospectively analyzed data of 100 consecutive TAVR candidates (61 male, mean age 79.6±9.9 years) who underwent both TAVR planning CT and CCA. At both modalities, the presence of coronary artery stenosis was assessed. Additionally, all coronary bypass grafts were rated as patent or occluded. Using CCA as the reference standard, we calculated the accuracy of CT for lesion detection on a per-vessel and per-patient basis. We further analyzed the accuracy of...

In a per-vessel/per patient analysis, CT had 94.4/98.6 % sensitivity and 68.4/55.6 % specificity for the detection of >50% stenosis in the native coronary arteries. Negative and positive predictive values were 94.7/93.8 % and 67.0/85.7 %, respectively. Per-patient sensitivity for >70% stenosis was 100.0%. All 12 vessels on which percutaneous coronary intervention was performed were correctly identified as demonstrating >50% stenosis on CT. There was agreement between CT and CCA regarding graft patency in 114/115 grafts identified on CCA.

TAVR planning CT has high sensitivity and negative predictive value in excluding obstructive CAD. An additional pre-procedural CCA study may not be required in TAVR candidates with a CT negative for obstructive CAD.