High spatial resolution intracranial vessel wall imaging and blood oxygenation level-dependent (BOLD) measures of hypercapnia-induced cerebrovascular reactivity (CVR) are being applied more frequently to evaluate structural and functional changes in patients with intracranial stenosis at risk for stroke. However, no study to date has assessed the relationship between intracranial vessel plaque and wall thickening and its impact on tissue-level function. Specifically, it is unclear whether altered CVR in regions of intracranial stenosis represent delayed blood arrival time, reduced reactivity, or delayed reactivity time, all of which may represent different indicators of disease severity and endothelial function, and may require distinct treatments. The purpose of this work is (1) to apply a novel intracranial vessel wall imaging protocol in a group of patients with intracranial stenosis secondary to atherosclerosis and moyamoya disease (non-atherosclerotic stenosis), and (2) to separately quantify reactivity time and reactivity magnitude in flow territories perfused by vessels with vessel wall disease.Experiment. All volunteers (n=36) provided informed, written consent. Inclusion criteria were stroke or TIA within 30 days, intracranial stenosis > 50%, and extracranial stenosis < 70%. Vessel wall imaging was performed on a 3T whole-body system (Philips) in the coronal plane using a custom 3D turbo spin echo (TSE) proton density-weighted sequence with long TSE readout (blood water nulling) and anti-driven equilibrium module (CSF water nulling). Imaging parameters: FOV=200mm x 166mm, spatial resolution=0.6 x 0.5 x 0.5 mm3, TR/TE=1500/38.5ms, TSE factor=56. BOLD imaging (TR/TE 2000/35 ms) was performed using a three-minute hypercapnic stimulus (5% CO2) interleaved with three-minutes normocapnic room air and repeated twice. Analysis. The time of maximum correlation of BOLD data with the applied stimulus, the CVR time, was calculated using a novel time regression technique1 and the maximum z-statistic was recorded as a qualitative metric of CVR. Means and standard deviations of the CVR time and maximum CVR were calculated within each of the ASPECTS2 territories (right and left A1, A2, M1-M6, P1, and P2) as well as the total vascular territories including the right and left anterior cerebral artery (ACA), middle cerebral artery (MCA), and posterior cerebral artery territories (PCA). A board certified radiologist performed a blinded review of the vessel wall imaging for lesion determination, defined as concentric or eccentric vessel wall thickening or increased signal on proton density imaging. The intracranial internal carotid arteries, A1, M1, and P1 segments were evaluated for each patient. The CVR time and maximum CVR values were compared (two-tailed Student’s t-test) between patients with and without a proximal vessel wall lesion for each vascular territory and for all territories combined. 36 patients were included in the study, 16 with atherosclerosis and 20 with moyamoya disease. Representative patient examples are shown in Figure 1. CVR time was significantly prolonged in flow territories with vs. those without a proximal vessel wall lesion for both the atherosclerosis and moyamoya patient groups (Table 1). Similar analysis for maximum CVR showed a significant decrease in the moyamoya group but not the atherosclerosis group. When each vascular territory was separately evaluated, the average CVR time was prolonged and the maximum CVR decreased in territories with a proximal vessel wall lesion compared to those without a proximal vessel wall lesion for the moyamoya group (Figure 2). The atherosclerosis group shows a similar trend, though the differences do not meet criteria for statistical significance. The PCA territory was not included in this analysis due to the small number of posterior circulation vessel wall lesions.Intracranial vessel wall imaging without intravenous contrast can depict lesions in patients with intracranial stenosis, and these lesions are shown to correlate with functional changes of cerebrovascular reactivity. Vascular territories with a proximal vessel wall lesion are shown to have a prolonged CVR time in both patient groups, though the differences are more marked in the moyamoya group than the atherosclerosis group possibly indicative of different mechanisms of endothelial dysfunction in atherosclerotic vs. non-atherosclerotic disease. The maximum CVR was shown to be significantly decreased in the presence of a vessel wall lesion in the moyamoya patients only, possibly indicating that it is CVR time, rather than commonly-assumed overall reactivity, that is impaired in many patients with atherosclerotic disease. Ongoing work focuses on assessing the effect of luminal stenosis on CVR time and maximum CVR.