Comparison of Two Contemporary Quantitative Atherosclerotic Plaque Assessment Tools for Coronary Computed Tomography Angiography: Single-Center Analysis and Multi-Center Patient Cohort Validation

• DOI: 10.3390/diagnostics14020154
• Publication Date: 2024-01-09T15:48:06Z
• Authors: Loris Weichsel; Alexander Giesen; Florian André; Matthias Renker; Stefan Baumann; Philipp Breitbart; Meinrad Beer; Pal Maurovitch-Ho...; Bálint Szilveszter; Borbála Vattay; Sebastian J. Buss; Mohamed Marwan; Andreas A. Gianno...; Sebastian Kelle; Norbert Frey; Grigorios Korosoglou
• Keywords: Computed Tomography Angiography; Coronary arteries

Background: Coronary computed tomography angiography (CCTA) provides non-invasive quantitative assessments of plaque burden and composition. The assessment components requires the use analysis software that reproducible semi-automated detection analysis. However, commercially available can vary widely in degree automation, resulting differences terms reproducibility time spent. Aim: To compare spent two CCTA tools using different algorithms for coronary volumes composition independent patient cohorts. Methods: study population included 100 patients from cohorts: 50 a single-center (Siemens Healthineers, SOMATOM Force (DSCT)) another multi-center (5 > 64 slice CT scanner types). Quantitative measurements total calcified non-calcified volume right artery (RCA), left anterior descending (LAD), circumflex (LCX) were performed on 300 coronaries by readers, (Tool #1: Siemens syngo.via Frontier Plaque Analysis Tool #2: successor prototype). In addition, was recorded with both programs. Results: cohorts 1 2 62.8 ± 10.2 70.9 11.7 years old, respectively, 10 (20.0%) 35 (70.0%) female 34 (68.0%) 20 (40.0%), had hyperlipidemia. Cohort #1, inter- intra-observer variabilities per #1 versus #2 22.8%, 22.0%, 26.0% 2.3%, 3.9%, 2.5% 19.7%, 21.4%, 22.1% 0.2%, 0.1%, 0.3%, total, noncalcified, lesions (p < 0.001 all between Tools intra-observer). remained low at 2.9%, 2.7%, 3.0% 3.8%, 3.7%, 4.0%, non-calcified, #2. For each dataset, median processing higher (459.5 s IQR = 348.0–627.0 208.5 s; 198.0–216.0) 0.001). Conclusion: (CT-guided PCI) encompassing automated support required less manual editing, more time-efficient, showed intra- inter-observer representative validation cohort.