How to assess non-calcified plaque in CT angiography: delineation methods affect diagnostic accuracy of low-attenuation plaque by CT for lipid-core plaque in histology

Abstract

To compare the accuracy of two plaque delineation methods for coronary computed tomographic angiography (CTA) to identify lipid-core plaque (LCP) using histology as the reference standard. Five ex vivo hearts were analysed by CTA and histology. LCP was defined by histology as fibroatheroma with core diameter/circumference >200 μm/>60° and cap thickness <450 μm. In CTA, plaque was manually delineated either as the difference between the inner and outer vessel walls (Method A) or as a direct tracing of plaque (Method B). Low-attenuation plaque was defined as an area with <90 Hounsfield units. Of 446 co-registered cross-sections, 55 (12%) contained LCP. In CTA, low-attenuation plaque area was larger as assessed with Method A compared with Method B (difference: 120 ± 60%). Although low-attenuation plaque was associated with the presence of LCP, the delineation Method B yielded higher diagnostic accuracy than Method A [area under the curve (AUC): 0.831 vs. 0.780, respectively, P = 0.005]. After excluding 'normal' cross-sections by CTA (n = 117), AUC for detecting LCP became similar between both methods (0.767 vs. 0.729, P = 0.07, respectively). Low-attenuation plaque in CTA is a diagnostic tool for LCP but prone to error if plaque is defined as the area between the inner and outer vessel walls and normal cross-sections are included in the assessment.