By Michael Crawford, MD, Editor
Professor of Medicine, Chief of Clinical Cardiology, University of California, San Francisco
Dr. Crawford reports no financial relationships relevant to this field of study.
SYNOPSIS: A meta-analysis of available studies of coronary CT angiography compared to invasive angiography demonstrates that coronary CT angiography is a reasonable substitute for invasive angiography in patients with low- to intermediate-risk for coronary artery disease and without aortic stenosis undergoing valve surgery.
SOURCES: Opolski MP, Staruch AD, Jakubczyk M, et al. CT angiography for the detection of coronary artery stenoses in patients referred for cardiac valve surgery: Systematic review and meta-analysis. JACC Cardiovasc Imaging 2016;9:1059-1070.
Dewey M, Schlattmann P. Investigating patients for CAD before cardiac valve surgery: Is CT angiography enough? JACC Cardiovasc Imaging 2016;9:1071-1073.
Small single-center studies have suggested that coronary computed tomography angiography (CCTA) may be a suitable replacement for invasive coronary angiography (ICA) to detect coronary artery disease (CAD) in patients scheduled for valve surgery. Thus, an international group of investigators performed a meta-analysis of English-language studies published in peer-reviewed journals up to July 2015 that used ≥ 16 slice CCTA to evaluate adult patients for heart valve surgery and had absolute numbers of patients from which the accuracy of the technique could be calculated. Patients referred for transcatheter aortic valve replacement were excluded. A total of 17 studies involving 1,153 patients were included in the analysis. All studies used a 50% diameter narrowing cutoff for significant CAD. In the per patient analysis, 313 of 1,107 patients (28%) had at least one significant stenosis by ICA. In these patients, the sensitivity of CCTA was 93%, specificity was 89%, positive likelihood ratio (LR) was 8.44, negative LR was 0.07, and the diagnostic odds ratio was 113. Studies that excluded patients with aortic stenosis when pooled had better specificity (96%) and positive LR (21). Also, 64 slice CT scanners had better specificity (90%) and positive LR (9.5). In addition, male sex was the only covariate with a negative effect on sensitivity. The authors concluded that CCTA is a reliable alternative to ICA with excellent sensitivity and negative LR for the detection of significant CAD in patients undergoing heart valve surgery. The use of CCTA in aortic stenosis reduces its specificity.
It makes sense to evaluate patients undergoing valve surgery for CAD. Stress testing is not a good option since it is not 100% accurate and its accuracy has not been tested in valve disease patients who need surgery. Also, there are several reasons to believe that it may be less accurate in such patients. Consequently, ICA is recommended for these patients if they have angina or objective evidence of myocardial ischemia; have reduced left ventricular function; are men > 40 years of age; or women who are either postmenopausal or premenopausal with risk factors of CAD (Level of evidence = C). With the exception of aortic stenosis patients, most valve disease patients don’t have significant CAD. Thus, exposing them to the risk, cost, and inconvenience of ICA doesn’t make much sense. These considerations make CCTA seem attractive, but there are no robust studies on which to base firm recommendations. Consequently, this meta-analysis is worth considering.
Population studies employing CCTA have shown that it can accurately exclude CAD in low-to intermediate-risk subjects. CCTA in these population studies has a high negative predictive value (> 95%) and a low negative likelihood ratio (< 0.1). This meta-analysis shows similar results in presurgical valve disease patients with a negative likelihood ratio of (0.07). Thus, CCTA in valve disease patients is excellent for excluding the presence of significant CAD with only a 3% false negative rate. In the meta-analysis, a negative CCTA would mean about two-thirds of patients could avoid ICA. The positive likelihood ratio of CCTA for identifying CAD (8.4) is more modest and in older aortic stenosis patients is even lower (7.4). Of course, aortic stenosis patients have more risk factors for CAD. The specificity of CCTA for CAD in aortic stenosis patients also is lower (87%). If CCTA is positive for CAD, ICA should be performed to confirm the results and decide if coronary artery bypass also must be performed. Thus, CCTA could serve as a gatekeeper for ICA in presurgical valve disease patients.
CCTA has other uses in valve disease. It can determine the amount or location of valve calcifications; valve area can be measured; and aortic arch atherosclerosis and coronary anomalies can be detected. Currently, the radiation and contrast agent exposure are higher than with ICA, but this is improving with better equipment. Also, in the meta-analysis, the positive LR was not good using scanners with < 64 detectors. In addition, not all patients can undergo CCTA because of arrhythmias, contrast allergy, chronic kidney disease, and hemodynamic instability. This causes a selection bias in CCTA studies.
In conclusion, until better data are available, CCTA makes sense in low- to moderate-risk patients undergoing valve surgery who do not have aortic stenosis. Those with a high risk of CAD or aortic stenosis should undergo ICA.