Superb Microvascular Imaging Ultrasound for Cervical Carotid Artery Stenosis for Prediction of the Development of Microembolic Signals on Transcranial Doppler during Carotid Exposure in Endarterectomy

Introduction: During exposure of the carotid arteries, embolism from the surgical site is recognized as a primary cause of neurological deficits or new cerebral ischemic lesions following carotid endarterectomy (CEA), and associations have been reported between histological neovascularization in the carotid plaque and both plaque vulnerability and the development of artery-to-artery embolism. Superb microvascular imaging (SMI) enables accurate visualization of neovessels in the carotid plaque without the use of intravenous contrast. This study aimed to determine whether preoperative SMI ultrasound for cervical carotid artery stenosis predicts the development of microembolic signals (MES) on transcranial Doppler (TCD) during exposure of the carotid arteries in CEA. Methods: Preoperative cervical carotid artery SMI ultrasound followed by CEA under TCD monitoring of MES in the ipsilateral middle cerebral artery was conducted in 70 patients previously diagnosed with internal carotid artery stenosis (defined as ≥70%). First, observers visually identified intraplaque microvascular flow (IMVF) signals as moving enhancements located near the surface of the carotid plaque within the plaque on SMI ultrasonograms. Next, regions of interest (ROI) were manually placed at the identified IMVF signals (or at arbitrary places within the plaque when no IMVF signals were identified within the carotid plaque) and the carotid lumen, and time-intensity curves of the IMVF signal and lumen ROI were generated. Ten heartbeat cycles of both time-intensity curves were segmented into each heartbeat cycle based on gated electrocardiogram findings and averaged with respect to the IMVF signal and lumen ROI. The difference between the maximum and minimum intensities (ID) was calculated based on the averaged IMVF signal (ID IMVF ) and lumen (ID l ) curves. Finally, the ratio of ID IMVF to ID l was calculated. Results: MES during exposure of the carotid arteries were detected in 17 patients (24%). The incidence of identification of IMVF signals was significantly greater in patients with MES (94%) than in those without (57%; p = 0.0067). The ID IMVF /ID l ratio was significantly greater in patients with MES (0.108 ± 0.120) than in those without (0.017 ± 0.042; p < 0.0001). The specificity and positive predictive value for the ID IMVF /ID l ratio for prediction of the development of MES were significantly higher than those for the identification of IMVF signals. Logistic regression analysis r