Intracranial vessel wall imaging at 7.0 tesla MRI

Abstract

Intracranial atherosclerosis is one of the main causes of ischemic stroke. Current conventional imaging techniques assessing intracranial arterial disease in vivo only visualize the vessel wall lumen instead of the pathological vessel wall itself. Therefore, not much is known about the imaging characteristics of intracranial vessel wall pathology in vivo, and distinguishing different vessel wall pathologies remains difficult. In this thesis, several 7.0 tesla MRI sequences – first with small coverage of only the circle of Willis and its major branches, and subsequently with whole-brain coverage and several image contrast weightings – were designed specifically for visualization of the intracranial arterial vessel wall. Besides imaging healthy vessel walls, they proved successful in visualizing intracranial vessel wall lesions in patients with ischemic stroke and TIA that were recruited in the Intracranial Vessel wall Imaging (IVI) study in the University Medical Center Utrecht. Preliminary results of the first 35 included IVI patients showed that 66% of these patients had one or more vessel wall lesions on MR imaging, of which only ¼ of lesions could be seen on conventional imaging; ¾ would therefore have been missed without the use of these dedicated MRI sequences. In a larger group of IVI patients, it was shown that most visualized lesions do not change over time (1 month), possibly reflecting a more generalized atherosclerotic process. Currently the main culprit lesion is thought to be the ‘vulnerable’ atherosclerotic plaque: a plaque with specific characteristics like a lipid-rich necrotic core and intraplaque hemorrhage. For the extracranial carotid artery, validation studies have determined specific in vivo image contrast characteristics for plaque characterization. However, it is not known whether MRI has the image contrast necessary to image intracranial plaque components as well. In this thesis, in an ex vivo setting ultrahigh-resolution 7.0 tesla MRI sequences were indeed able to distinguish areas of different signal intensities that spatially corresponded to plaque components within advanced atherosclerotic plaques. Although 7.0 tesla MRI has several advantages over imaging on lower field strengths – like an increased contrast-to-noise and signal-to-noise ratio, enabling faster scanning or imaging at higher spatial resolution – a major drawback are the strict safety rules regarding metallic implants, which are based on conflicting results of RF heating of implants at different field strengths. In this thesis, fundamental experiments were performed on 7.0 tesla, explaining the discrepancies of RF heating at different field strengths; further, 20 peripheral arterial stents were tested and found MR compatible for 7.0 tesla. In conclusion, 7.0 tesla MR imaging of the intracranial arterial vessel wall has shown potential in directly visualizing vessel wall pathology in patients with ischemic stroke or TIA, not only in the circle of Willis and its main branches, but also in more distally located arteries. Although not yet performed in vivo, results from ex vivo studies show that 7.0 tesla MRI also has the image contrast to identify different components within atherosclerotic plaques. Future studies should focus on larger patient groups with different intracranial arterial diseases, possibly enabling patient-based diagnosis of specific vessel wall pathologies.