Effect of scanner type on the reproducibility of extracoronary measures of calcification: the multi-ethnic study of atherosclerosis

Abstract

Rationale and objectives: Cardiac computed tomography (CT) has been used extensively to measure coronary artery calcification. However, extracoronary calcifications, such as aortic valve calcification (AVC), may have independent clinical significance as well. The ability to track calcification is dependent on the reproducibility of the original measurement, and the variability of extracoronary calcification measurements still is unknown. Accurate quantification of calcification of the aortic valve, mitral annulus (MAC), and thoracic aortic (TAC) may be possible by using cardiac CT.

Methods: A total of 1,729 randomly chosen participants (ages 45-84, 53% female, 28% African-American, 36% Caucasian, 11% Chinese, 25% Hispanic) of the Multi-Ethnic Study of Atherosclerosis underwent dual scanning by electron beam CT (EBT) or multidetector CT (MDCT) to assess coronary and extra-coronary calcifications. Two calcium measurement methods--Agatston score (AS) and volume score (VS)--were measured for each scan. Concordance for calcium positivity was assessed among all scans. Mean absolute and relative differences between calcium measures on scans 1 and 2, excluding cases for which both scans had a measure of zero, was modeled by using linear regression to compare variability between scanner types. A repeated measures analysis of variance test was used to compare variability across calcium measures, with mean percentage absolute difference as the outcome measure.

Results: Concordances for the presence of calcium between duplicate scans were high and similar for both EBT and MDCT. Concordance was high for all three extracoronary measures, with a kappa statistic of kappa = 0.94-0.96. For all three extracoronary sites, Bland-Altman plots demonstrated excellent agreement, with almost all measures falling within the boundaries of the 95% confidence limits of reproducibility. AVC interscan variability was approximately 8% for both AS and VS, with improved variability for EBT as compared with MDCT. Mitral annular calcification demonstrated slightly lower variability than AVC for both scanner types (approximately 6%), with no significant differences between MDCT and EBT. Of the three extracoronary sites, TAC had the highest variability (10%), with MDCT variability slightly lower than EBT variability (9.3 vs. 10.2%, respectively, P = NS). Agatson and volume scores for each of the three extracoronary sites were similar.

Conclusions: Overall rescan measurement variabilities for extracoronary calcification are low and should not be an impediment to the use of this test for studying progression of extracoronary calcification over time.