Predicting impaired cerebrovascular reactivity and risk of Hyperperfusion syndrome in carotid artery stenosis using BeamSAT magnetic resonance imaging

Abstract

Pencil-beam presaturation (BeamSAT) magnetic resonance imaging (MRI) produces selective magnetic resonance angiography (MRA) images of specific arteries, including the unilateral internal carotid artery (ICA-selective MRA) or vertebral artery (VA-selective MRA). This study aimed to evaluate the influence of collateral flow, visualized using BeamSAT MRI, on preoperative cerebral hemodynamic status and postoperative hyperperfusion syndrome (HPS). Patients undergoing carotid artery stenting or carotid endarterectomy were divided into two groups to evaluate collateral flow. Patients with neither crossflow on BeamSAT MRI nor mismatch in middle cerebral artery (MCA) signal intensity between ICA-selective and conventional MRA were classified into Group I (N = 29). Group II included all other patients comprising 19 patients. Cerebral blood flow and cerebrovascular reactivity (CVR) were assessed using single-photon emission computed tomography, and potential HPS symptoms were retrospectively assessed by chart review. Preoperative ipsilateral CVR was significantly lower in Group II (18.0% ± 20.0% vs. 48.3% ± 19.5%; P < 0.0001). Group II showed significantly impaired CVR (odds ratio 17.7, 95% confidence interval 1.82–171; P = 0.013). The partial areas under the curve of the BeamSAT logistic model (0.843) were significantly larger than those of the conventional logistic model over the range of high sensitivity (0.6–1) (P = 0.04). The incidence of postoperative HPS symptoms was significantly higher in Group II than in Group I (8/19 vs. 1/29; P = 0.001). Differences in MCA-MRA grade and crossflow between ICA-selective and conventional MRA were associated with impaired CVR. BeamSAT MRI may be a valuable and non-invasive tool for assessing cerebral hemodynamics and predicting postoperative HPS.