Correlation between MDCTA and Carotid Plaque Histological Heterogeneity: A Pilot Study

• Published in: European journal of vascular and endovascular surgery 2018-07, Vol.56 (1), p.7-14
• Main Authors: Veraldi, Gian Franco, Nocini, Pier Francesco, Eccher, Albino, Fenzi, Alberto, Sboarina, Andrea, Mezzetto, Luca
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Area Under Curve; Asymptomatic Diseases; Biopsy; Carotid Arteries - diagnostic imaging; Carotid Arteries - pathology; Carotid atherosclerotic plaque; Carotid Stenosis - complications; Carotid Stenosis - diagnostic imaging; Carotid Stenosis - pathology; Computed Tomography Angiography; Female; Histological heterogeneity; Humans; Male; Middle Aged; Multidetector Computed Tomography; Multidetector computed tomography angiography; Observer Variation; Pilot Projects; Plaque, Atherosclerotic; Predictive Value of Tests; Radiographic Image Interpretation, Computer-Assisted; Reproducibility of Results; Risk Assessment; Risk Factors; ROC Curve; Severity of Illness Index; Software; Stroke - etiology
• ISSN: 1078-5884; 1532-2165
• DOI: 10.1016/j.ejvs.2018.04.001

The aim of this pilot study was to identify multidetector computed tomography angiography (MDCTA) features that may help identify carotid atherosclerotic plaques (CAPs) with severe histological heterogeneity. Thirty-one CAPs (9 symptomatic) were evaluated histologically using a semiquantitative scale. The CAPs were assessed for the presence of ulceration, lipids, fibrosis, thrombotic deposits, haemorrhage, neovascularisation, and inflammation. A CAP presenting at least five of these histological features was defined as a severe heterogeneous plaque (Group A); in all other cases it was defined as a mild to moderate heterogeneous plaque (Group B). The non-calcified plaque tissue was segmented in pre-operative MDCTA. Median and mean intensity and percentages of soft tissue voxels with a value smaller than or equal to certain thresholds (from 20 HU to 200 HU with a constant distance of 20 HU) were calculated. Comparison of intensity measurements was analysed by Mann–Whitney U test and receiver operating characteristic (ROC) analysis. In order to assess the method reliability, values showing better performance were compared using the Wilcoxon signed rank test and k-Cohen test according to ROC analysis. According to histological analysis 18 CAPs were classified as belonging to Group B and 13 to Group A. The percentages of soft tissue with density ≤ 40 (TH_40), 60 (TH_60), 80 (TH_80), and 100 HU (TH_100) were statistically significantly greater in plaques of Group A (respectively p = .016, p = .002, p = .001, p = .007). The mean (p = .025) and median (p = .014) intensity were statistically significantly lower in the plaques of Group A. TH_60 and TH_80 showed the greatest accuracy (0.81) with similar performance in term of AUC and sensitivity/specificity (TH_60: 0.82, 0.62, 0.94; TH_80: 0.83, 0.69, 0.89, respectively, for AUC, sensitivity and specificity). Reliability analysis showed good repeatability and reproducibility of these measurements. The findings have demonstrated lower density of the non-calcified tissue in the plaques of Group A with higher TH_60 and TH_80 soft tissue percentages with respect to CAPs of Group B.