Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease (Jun 2021)

Mapping Cerebrovascular Reactivity Impairment in Patients With Symptomatic Unilateral Carotid Artery Disease

  • Martina Sebök,
  • Christiaan Hendrik Bas van Niftrik,
  • Sebastian Winklhofer,
  • Susanne Wegener,
  • Giuseppe Esposito,
  • Christoph Stippich,
  • Andreas Luft,
  • Luca Regli,
  • Jorn Fierstra

DOI
https://doi.org/10.1161/JAHA.121.020792
Journal volume & issue
Vol. 10, no. 12

Abstract

Read online

Background Comprehensive hemodynamic impairment mapping using blood oxygenation‐level dependent (BOLD) cerebrovascular reactivity (CVR) can be used to identify hemodynamically relevant symptomatic unilateral carotid artery disease. Methods and Results This prospective cohort study was conducted between February 2015 and July 2020 at the Clinical Neuroscience Center of the University Hospital Zurich, Zurich, Switzerland. One hundred two patients with newly diagnosed symptomatic unilateral internal carotid artery (ICA) occlusion or with 70% to 99% ICA stenosis were included. An age‐matched healthy cohort of 12 subjects underwent an identical BOLD functional magnetic resonance imaging examination. Using BOLD functional magnetic resonance imaging with a standardized CO2 stimulus, CVR impairment was evaluated. Moreover, embolic versus hemodynamic ischemic patterns were evaluated on diffusion‐weighted imaging. Sixty‐seven patients had unilateral ICA occlusion and 35 patients unilateral 70% to 99% ICA stenosis. Patients with ICA occlusion exhibited lower whole‐brain and ipsilateral hemisphere mean BOLD‐CVR values as compared with healthy subjects (0.12±0.08 versus 0.19±0.04, P=0.004 and 0.09±0.09 versus 0.18±0.04, P<0.001) and ICA stenosis cohort (0.12±0.08 versus 0.16±0.05, P=0.01 and 0.09±0.09 versus 0.15±0.05, P=0.01); however, only 40 (58%) patients of the cohort showed significant BOLD‐CVR impairment. Conversely, there was no difference in mean BOLD‐CVR values between healthy patients and patients with ICA stenosis, although 5 (14%) patients with ICA stenosis showed a significant BOLD‐CVR impairment. No significant BOLD‐CVR difference was discernible between patients with hemodynamic ischemic infarcts versus those with embolic infarct distribution (0.11±0.08 versus 0.13±0.06, P=0.12). Conclusions Comprehensive BOLD‐CVR mapping allows for identification of hemodynamically relevant symptomatic unilateral carotid artery stenosis or occlusion.

Keywords