Interobserver variations of plaque severity score and segment stenosis score in coronary arteries using 64 slice multidetector computed tomography: A substudy of the ACCURACY trial

Abstract

Assessing the severity of coronary plaque for the risk stratification and management of coronary artery disease is important. Multidetector computed tomography has been shown to be a useful tool to measure coronary plaque; however, interreader variability is a concern. We measured interobserver variations of plaque severity score (PSS) and segment stenosis score (SSS) as measured by the total plaque severity score (TPS) and total segment stenosis score (TSS). Cardiac CT scans (n = 221) of the ACCURACY trial were interpreted by 3 different readers blinded to patient characteristics. PSS (mild, 1; moderate, 2; and severe, 3) and SSS (stenosis 1%-29%, 1; 30%-49%, 2; 50%-69%, 3; and ≥70%, 4) were calculated with the 15-segment American Heart Association model. TPS and TSS were determined by summing the segments for each interpreter. TPS and TSS were compared for correlation and variation among any 2 of the 3 readers. A highly significant correlation was observed among any 2 of the 3 readers for both TPS and TSS. For TPS, the r = 0.91, 0.93, 0.94 (P < 0.001) for A vs B, B vs C, A vs C, respectively, and for TSS, r = 0.91, 0.92, 0.93 (P < 0.001) for A vs B, B vs C, A vs C, respectively. On Bland Altman plot, the mean difference between the scores of any 2 readers was 3.33 ± 3.93, 1.65 ± 2.88, and 1.68 ± 2.92 for TPS and 4.19 ± 4.73, 2.54 ± 4.02, and 1.65 ± 3.18 for TSS. Semiquantitative measures of coronary plaque burden, including the TPS and TSS, can be determined with a high degree of interobserver agreement, suggesting their potential role as tools to aid in the assessment of coronary heart disease.