The Effect of Circle of Willis Morphology on Retinal Blood Flow in Patients with Carotid Stenosis Measured by Optical Coherence Tomography Angiography.

The Circle of Willis (CoW) is the main collateral system, and its morphological variants are more common in patients who have severe carotid artery stenosis. Earlier data suggest that optical coherence tomography angiography (OCTA) may help to assess the changes in cerebral vascular perfusion by imaging the retinal blood flow. In this single-center prospective clinical study, patients scheduled for carotid endarterectomy (CEA) underwent preoperative computed tomography angiography (CTA) of the extra- and intracranial cerebral circulation. OCTA imaging was performed one week before surgery and postoperatively one month later. The patients were divided into two subgroups based on CTA evaluation of CoW: compromised CoW or non-compromised CoW (containing hypoplastic and normal segments). The effect of the patient's age, OCTA scan quality (SQ), CoW morphology, laterality, and surgery on superficial capillary vessel density (VD) in the macula were assessed in multivariable regression models using linear mixed models. We found that VD significantly decreased with aging (-0.12%; 95%CI: -0.07--0.15; p < 0.001) and was significantly higher in patients with non-compromised CoW morphology (by 0.87% 95%CI (0.26-1.50); p = 0.005). After CEA, retinal blood flow significantly improved by 0.71% (95%CI: 0.18-1.25; p = 0.01). These results suggest that in the case of carotid artery occlusion, patients with non-compromised CoW have more preserved ocular blood flow than subjects with compromised CoW due to remodeling of the intra-orbital blood flow. Measuring the retinal blood flow might be used as a relevant and sensitive indicator of collateral cerebrovascular circulation.

Multidetector CT angiography of the Circle of Willis: association of its variants with carotid artery disease and brain ischemia.

PURPOSE: (1) to estimate the prevalence of Circle of Willis (CoW) variants in patients undergoing carotid endarterectomy, (2) to correlate these variants to controls and (3) cerebral ischemia depicted by computed tomography (CT). MATERIALS AND METHODS: After Institutional Review Board approval, data of 544 carotid endarterectomy patients (331 males, mean age 69±8 years) and 196 controls (117 males, mean age 66±11 years) who underwent brain CT and carotid CT angiography (CTA) were retrospectively analysed. Two observers independently classified each CoW segment as normal, hypoplastic (diameter <0.8 mm) or non-visualized. Four groups of CoW variants based on the number of hypoplastic/non-visualized segments were correlated with clinical data (ANOVA, χ2 and multivariate logistic regression analysis). Intra- and inter-observer agreement was estimated using Cohen κ statistics. RESULTS: High prevalence of CoW variants (97%) and compromised CoW (81%) was observed in the study group and significant difference was found in the distribution of CoW variants compared to controls (p<0.001), internal carotid artery (ICA) stenosis being the only independent predictor of CoW morphology (p<0.001). Significant correlation was found between CoW configuration and brain ischemia in the study group (p=0.002). ICA stenosis of ≥90% was associated to higher rate of ipsilateral A1 hypoplasia/non-visualization (p<0.001). Intra- and inter-observer agreement was from substantial to almost perfect (Cohen κ=0.75-1.0). CONCLUSION: Highly variable CoW morphology was demonstrated in patients undergoing endarterectomy compared to controls. Likely compromised CoW in relation to cerebral ischemia was observed in a large cohort of carotid endarterectomy subjects. KEY POINTS: • CoW variant distribution significantly differed in the study and control groups (p<0.001). • ICA stenosis was the only independent predictor of CoW morphology (p<0.001). • Severely compromised CoW configuration showed significant association with brain ischemia (p=0.002).

Incomplete circle of Willis is associated with a higher incidence of neurologic events during carotid eversion endarterectomy without shunting.

OBJECTIVE: A complete circle of Willis (CoW) is considered an important collateral network to maintain blood flow during cross-clamping in carotid endarterectomy (CEA). The aim of this study was to evaluate the impact of an incomplete CoW with isolated middle cerebral artery (iMCA) on immediate neurologic events (INEs) after CEA. METHODS: We prospectively collected the clinical data and outcomes of 902 patients who underwent CEA under general anesthesia between 2013 and 2015. All patients had preoperative computed tomography angiography of the extracranial and intracranial cerebral circulation. Indications were asymptomatic (52%) and symptomatic (48%) carotid artery disease. Patients who had CEA with shunt (n = 35) and those with inadequate intracranial imaging to assess CoW were excluded (n = 322) only. Computed tomography angiography images were reviewed retrospectively and independently by two vascular radiologists who were blinded for treatment outcomes. Imaging assessment included the vertebral and carotid circulation and each segment of the CoW, which was classified as normal, hypoplastic (diameter < 0.8 mm) or absent. The ipsilateral MCA was considered isolated if there was an absence of the anterior and posterior communicating branches from the contralateral carotid or posterior circulations. INE was defined as any transient ischemic attack (TIA) and stroke diagnosed immediately after the procedure. RESULTS: Of the 545 included patients (331 males; mean age, 69 ± 8 years), 12 (2.2%) had a stroke in the postoperative period. There were 20 INEs (8 strokes and 12 TIAs). A complete CoW was rare; it was only detected in 19 patients (3.5%) and an iMCA was found in 34 patients (6.3%). When at least one collateral circulation was complete (in 330 patients), we observed only four INEs (1.2%). Of the 34 patients with an iMCA, 8 (24%) had INE (6 TIAs and 2 strokes). Overall, iMCA was an independent predictor of INEs (odds ratio, 11.12; 95% confidence interval, 3.57-35.87; P < .001). With logistic regression, the model included hypertension, smoking, diabetes, hyperlipidemia, carotid clamping time (minutes), contralateral significant internal carotid artery stenosis of greater than 90%, ipsilateral significant internal carotid artery stenosis of greater than 90%, preoperative symptoms in 6 months, and iMCA; above iMCA only symptomatic patients had significant risk (odds ratio, 3.34; 95% confidence interval, 1.19-9.73; P = .02), whereas all other parameters were not significant. CONCLUSIONS: An iMCA carries more than a 10-fold higher the risk of INEs after CEA with cross-clamping without shunt protection. In these patients, routine shunting is recommended to prevent INEs.