oa Valve Calcification on Computed Tomography Can Estimate Aortic Stenosis Severity

Use of coronary CT angiography for screening of coronary artery disease is being advocated for the general population. Meanwhile, CT is increasingly used in preoperative planning for transcatheter aortic valve replacement (TAVR). Although aortic stenosis (AS) severity can be evaluated by measurement of incidentally found aortic valve calcification on CT, it has not been validated by pathologic specimens and can be confounded by calcification of adjacent structures as well as motion artefact. Micro-computed tomography (microCT) provides ultra-high resolution imaging of small structures to yield excellent estimation of tissue calcification. In this study, excised aortic valves from patients with confirmed AS were used to determine if the amount of calcium on microCT correlated with severity of aortic stenosis. Thirty-five aortic valves excised during surgical valve replacement underwent micro-CT imaging with resolution of 76μm in the axial direction. Amount of calcium was determined by absolute and proportional values of calcium volume. Correlation of calcium volume and preoperative mean aortic valve gradient (MAVG), peak transaortic velocity (Vmax), and aortic valve area (AVA) on echocardiography was evaluated. For the patients who had a preoperative CT scan with acceptable image quality, the amount of valvular calcification was also measured by a well-experienced radiologist using modified Agatston algorithm. Mean amount of calcium across all valves was 603.2±398.5mm3, while mean ratio of calcium volume to total valve volume was 0.36±0.16. Mean aortic valve gradient correlated positively with both calcium volume and ratio (r=0.72, p<0.001). Vmax also positively correlated with calcium volume and ratio (r=0.69 and 0.76 respectively, p<0.001). A logarithmic curvilinear model was best fit to the correlation. Calcium volume of 480mm3 showed sensitivity and specificity of 0.76 and 0.83, respectively for severe AS diagnosis, while calcium ratio of 0.37 yielded sensitivity and specificity of 0.82 and 0.94, respectively. Calcium volume and its proportion to total valve volume were found to be good predictive parameters for severe AS when estimated radiologically. Calcium volume quantification may be a complimentary measure for AS severity evaluation in situations where aortic valve calcification is found incidentally on CT as well as in preoperative assessment of aortic valves for TAVR.