Communication of inwardly projecting neovessels with the lumen contributes to symptomatic intraplaque hemorrhage in carotid artery stenosis

Abstract

OBJECT: Recent studies have demonstrated that plaque morphology can contribute to identification of patients at high risk of carotid artery atherosclerosis as well as the degree of stenosis in those with carotid atherosclerosis. Neovascularization of carotid plaques is associated with plaque vulnerability. However, the mechanism of neovascularization in intraplaque hemorrhage (IPH) and its clinical contribution remain undetermined. In this study, the authors aimed to clarify the characteristics of neovessel appearance with a focus on inwardly projecting neovessels, which are reportedly important in plaque advancement. METHODS: Consecutive patients with moderate to severe carotid atherosclerosis who underwent carotid endarterectomy were prospectively analyzed from 2010 to 2014. The neovessel appearance was categorized into 3 groups based on intraoperative indocyanine green (ICG) videoangiography: early appearance of neovessels from the endothelium (NVe), late appearance of neovessels from the vasa vasorum (NVv), and no appearance of vessels. Each neovessel pattern was evaluated with respect to clinical, radiological, and pathological findings including IPH, neovascularization, hemosiderin spots, and inflammation. RESULTS: Of 57 patients, 13 exhibited NVe, 33 exhibited NVv, and 11 exhibited no neovessels. Overall, the interobserver and intraobserver reproducibilities of neovessel appearance were substantial for ICG videoangiography (κA = 0.76) and at 7 days postoperatively (K = 0.76). There were no significant differences in baseline characteristics among the 3 groups, with the exception of a higher percentage of symptomatic presentations in patients with NVe (artery-to-artery embolic infarction in 61.5% and transient ischemic attack in 23.1%). Moreover, patients with NVe exhibited larger infarctions than did those with NVv (9675.0 ± 5601.9 mm3 vs 2306.6 ± 856.9 mm3, respectively; p = 0.04). Pathologically, patients with NVe had more severe IPH (47.2 ± 8.3 mm2 vs 19.8 ± 5.2 mm2, respectively; p < 0.01), hemosiderin spots (0.5 ± 0.2 mm2 vs 0.2 ± 0.1 mm2, respectively; p = 0.04), neovessels (0.4 ± 0.7 mm2 vs 0.1 ± 0.4 mm2, respectively; p = 0.11), and inflammation (1.0 ± 1.1 mm2 vs 0.6 ± 0.9 mm2, respectively; p = 0.26) around the endothelium than did patients with NVv, and all of these parameters were correlated with hyperintensity on time-of-flight MRI. However, the neovessel and inflammation differences were nonsignificant. Interestingly, inflammation was significantly correlated with neovessel formation (r = 0.43, p = 0.0008), hemosiderin spots (r = 0.62, p < 0.0001), and IPH (r = 0.349, p = 0.0097), suggesting that inflammation may be a key factor in plaque vulnerability. CONCLUSIONS: Communication of inwardly projecting neovessels with the lumen and inflammation synergistically contribute to IPH and symptomatic presentations in patients with carotid stenosis and are more specific than the vasa vasorum. This condition could be a new therapeutic target, and regression of luminal neovessel sprouting and inflammation may help to prevent IPH development and a symptomatic presentation.