The circle of Willis predicts the antihypertensive effects of carotid artery stenting.

Abstract

OBJECTIVE Carotid artery stenting (CAS) has antihypertensive effects, but the durability and degree of this response remain variable. The authors propose that this clinical variability is a function of the presence or absence of a complete circle of Willis (COW). Incomplete COWs perfuse through a higher-resistance pial collateral pathway, and therefore patients may require a higher mean arterial pressure (MAP). Carotid artery revascularization in these patients would reduce the end-organ collateral demand that has been hypothesized to drive the MAP response. METHODS Using a retrospective, nonrandomized within-subject case-control design, the authors compared the postoperative effects of CAS in patients with and without a complete COW by using changes in MAP and antihypertensive medication as end points. They recorded MAP and antihypertensive medications 3 months prior to surgery, preoperatively, immediately postoperatively, and at the 3-month follow-up. RESULTS Data were collected from 64 consecutive patients undergoing CAS. Patients without a complete COW (25%) were more likely to demonstrate a decrease in BP response to stenting (i.e., a drop in MAP of 10 mm Hg and/or a reduction or cessation of BP medications at 3 months postoperatively). Of the patients in the incomplete COW cohort, 75% had this outcome, whereas of those in the complete COW cohort, only 41% had it (p < 0.041). These findings remained statistically significant in a logistic regression analysis for possible confounders (p < 0.024). A receiver operating curve analysis of preoperative data indicated that a MAP > 96.3 mm Hg was 55.5% sensitive and 57.4% specific for predicting a complete COW and that patients with a MAP > 96.3 mm Hg were more likely to demonstrate a good MAP decrease following CAS (p < 0.0092). CONCLUSIONS CAS is associated with a significant decrease in MAP and/or a reduction/cessation in BP medications in patients in whom a complete COW is absent.