Cardiac ImagingE3055. Routine Qualitative Evaluation of Coronary Calcium: Is It Reproducible?
Pedersen C, Schuleri K, Teytelboym O. Mercy Catholic Medical Center, Darby, PA
Address correspondence to C. Pedersen (firstname.lastname@example.org)
Objective: The purpose of this study is to evaluate the reproducibility of qualitative coronary calcium burden assessment on CT scans of the chest with and without contrast in a real-life clinical practice setting.
Materials and Methods: This study is a retrospective database review of 100 patients with two unenhanced chest CT scans and 100 patients with two IV contrast-enhanced chest CT scans, from November 15, 2015, to November 15, 2016, who were randomly chosen from our institution PACS. Each study was reported by one of 12 board-certified radiologists with 1–34 years of clinical experience who had completed training in a subspecialty and had completed a one-time review of a published qualitative calcium assessment scale. All measurements were done on PACS images. For coronary calcifications, we created a grading system: 0, none; 1, mild/minimal; 2, moderate; 3, heavy/marked/severe/dense/diffuse/extensive/multiple; and 4, status postcoronary artery bypass grafting or coronary stent placement. Weighted kappa was calculated using commercially available software.
Results: Grade 4 classifications (n = 9) were not included in the study. Three patients were excluded due to unclassifiable wording; 188 patients were included in the final analysis. The average time between the two studies was 98 days (range,1–323 days). The paired t test showed no statistical differences between the groups. Both the unenhanced group and contrast-enhanced group showed good correlation with weighted kappa of 0.68 (95% CI, 0.57–0.79) for the unenhanced group, 0.77 (95% CI, 0.67–0.87) for the contrast-enhanced group, and 0.72 (95% CI, 0.64–0.79) for both groups combined. The standard errors were 0.06, 0.05, and 0.04, respectively. The weighted kappa values were interpreted according to Altman.
Conclusion: Reproducibility of qualitative coronary calcium burden assessment on chest CT scans with and without a contrast agent is adequate to permit its use in real-life clinical practice settings. Further training and standardization of terms is indicated to accurately describe changes in calcification. This study investigates reproducibility in standardized reporting for qualitative coronary calcium scoring, which should be considered when reporting coronary calcification burden.