A Preoperative Assessment of Significant Coronary Stenosis Based on a Semiquantitative Analysis of Coronary Artery Calcification on Noncontrast Computed Tomography in Aortic Stenosis Patients Undergoing Aortic Valve Replacement

Invasive coronary angiography (ICA) is the recommended assessment for coronary artery disease in patients undergoing elective aortic valve replacement (AVR). Noncontrast computed tomography (CT) is useful for evaluating lung lesions and calcifications at the cannulation site of the ascending aorta. The purpose of this study was to evaluate the role of noncontrast CT in the visual assessment of coronary artery calcification (CAC) in patients undergoing AVR. We retrospectively identified patients with significant aortic stenosis (AS) who were referred for AVR between January 2006 and December 2013. Among these, we included 386 patients (53.6% males, 69.2 ± 8.4 years) who underwent both noncontrast CT and ICA. Significant coronary artery stenosis (CAS) in the ICA was defined as luminal stenosis ≥70%. The 4 main coronary arteries were visually assessed on noncontrast CT and were scored based on the Weston score as follows: 0, no visually detected calcium; 1, a single high-density pixel detected; 3, calcium was dense enough to create a blooming artifact; and 2, calcium in between 1 and 3. Four groups were reclassified by the sum of the Weston scores from each vessel, as follows: noncalcification (0); mild calcification (1-4); moderate calcification (5-8); and severe calcification (9-12). Receiver-operating characteristic (ROC) analysis was generated to identify the cutoff Weston score values for predicting significant CAS. Diagnostic estimates were calculated based on these cutoffs. In the ICA analysis, 62 of the 386 patients (16.1%) had significant CAS. All patients were divided into 4 groups. The noncalcification group had 97 subjects (Weston score 0), the mild degree group had 100 (2.6 ± 1.0), the moderate calcification group had 114 (6.6 ± 1.1), and the severe calcification group had 75 (10.7 ± 1.1). The prevalence of significant CAS in the noncalcification, mild, moderate, and severe groups was 1% (1/97), 5% (5/100), 24% (27/114), and 39% (29/75), respectively. The group with CAS had significantly more CAC than the group without CAS (8.37 ± 2.93 vs 4.01 ± 3.75, P < 0.001). The cutoff value (by Weston score) for predicting significant CAS is ≥5 (sensitivity 90.3%, specificity 59.0%, positive predictive value 29.6%, and negative predictive value 97%). The degree of CAC detected on noncontrast CT can help to predict significant CAS in AS patients who are referred for AVR. For the clinicians, the visual assessment of CAC on noncontrast CT was easy and useful for estimating CAS. Therefore, ICA should be recommended to selective patients based on patients' CAC and Weston scores during the preoperative evaluation for elective AVR.