Measuring Cerebral Autoregulation using Magnetic Resonance Imaging

Abstract

BACKGROUND: The mechanisms by which the brain regulates cerebral blood flow (CBF) have been extensively investigated. Significant insights have been gained into the homeostatic mechanism known as cerebral autoregulation at the large vessel level but information on tissue perfusion during hemodynamic stress has been more difficult to investigate. MRI techniques such as phase contrast angiography (PCA), dynamic contrast enhanced (DCE) and arterial spin labeling (ASL) MRI have the potential to deepen understanding of these mechanisms. This theses describes the investigation of cerebral blood flow dynamics in patients with stenotic carotid artery disease, describes the development and testing of O2, CO2 and lower body negative pressure (LBNP) as stimuli for investigating cerebral autoregulation at the macrovascular and tissue perfusion level using PCA and ASL MRI.METHODS: Subjects with occlusive unilateral carotid artery stenosis underwent imaging DCE and PCA MRI to investigate the effects of artery stenosis on CBF dynamics. A normalized measure of cerebral perfusion (nCBF) was derived using information from both DCE and PCA. Experimental protocols and feasibility studies were conducted using anesthetic circuits to deliver high concentrations of gases to subjects and the use of LBNP within a MRI scanner. Healthy volunteers were given O2, CO2 and LBNP challenges and their cerebral blood flow, grey matter perfusion and bolus arrival time (BAT) were measured and compared to baseline.RESULTS: In the carotid stenosis patients there was minimal asymmetry in middle cerebral artery (MCA) flow despite wide variability in CBF. Measures of nCBF in the anterior watershed areas showed strong correlation with CBF and demonstrated bilateral reduction in watershed blood flow. Only subjects using anesthetic delivery circuits showed CBF decreases during 100% O2 inhalation. GM perfusion increases and BAT decreases were seen during CO2 inhalation, during O2 inhalation GM perfusion was maintained but BAT increased. LBNP caused decreased aortic blood flow consistently within the MRI scanner. LBNP caused a decrease in CBF but no change in grey matter (GM) perfusion but an increase in BAT.CONCLUSIONS: Scaled DCE methods enable comparisons of perfusion metrics, which would be unreliable by conventional means. There is a perfusion reduction in anterior watershed areas in the hemisphere contralateral to the stenosis; the circulation is sharing out the flow deficit. To accurately detect CBF changes caused by O2 inhalation anesthetic circuits are required. ASL detects tissue BAT delays during inhalation of O2 suggesting autoregulatory processes are occurring to maintain perfusion, whilst during CO2 inhalation CBF and tissue perfusion increase due to uncontrolled vasodilation which results in decreased BAT. LBNP of -20 mmHg is capable of producing a normotensive hypovolemic stimulus with can be combined with MRI. -20 mmHg LBNP causes a decrease in CBF, which is compensated by autoregulation resulting in preserved GM perfusion but prolonged BAT. We have demonstrated the potential of MRI in the investigation of CBF dynamics and introduced BAT as a potential biomarker of autoregulatory function.

Date of Award	30 Nov 2011
Original language	English
Awarding Institution	The University of Manchester