T2*-Weighted Magnetic Resonance Imaging with Hyperoxia in Acute Ischemic Stroke Krishna A. Dani, MBChB, 1 Celestine Santosh, MD, 2 David Brennan, PhD, 2 Christopher McCabe, PhD, 3 William M. Holmes, PhD, 3 Barrie Condon, PhD, 2 Donald M. Hadley, PhD, 2 I. Mhairi Macrae, PhD, 3 Martin Shaw, MSci, 4 and Keith W. Muir, MD 1 Objective: We describe the first clinical application of transient hyperoxia (“oxygen challenge”) during T2*- weighted magnetic resonance imaging (MRI), to detect differences in vascular deoxyhemoglobin between tissue compartments following stroke. Methods: Subjects with acute ischemic stroke were scanned with T2*-weighted MRI and oxygen challenge. For regions defined as infarct core (diffusion-weighted imaging lesion) and presumed penumbra (perfusion-diffusion mismatch [threshold T max 4 seconds], or regions exhibiting diffusion lesion expansion at day 3), T2*-weighted signal intensity–time curves corresponding to the duration of oxygen challenge were generated. From these, the area under the curve, gradient of incline of the signal increase, time to maximum signal, and percentage signal change after oxygen challenge were measured. Results: We identified 25 subjects with stroke lesions 1ml. Eighteen subjects with good quality T2*-weighted signal intensity–time curves in the contralateral hemisphere were analyzed. Curves from the diffusion lesion had a smaller area under the curve, percentage signal change, and gradient of incline, and longer time to maximum signal ( p 0.05, n 17) compared to normal tissue, which consistently showed signal increase during oxygen challenge. Curves in the presumed penumbral regions (n 8) showed varied morphology, but at hyperacute time points (8 hours) showed a tendency to greater percentage signal change. Interpretation: Differences in T2*-weighted signal intensity–time curves during oxygen challenge in brain regions with different pathophysiological states after stroke are likely to reflect differences in deoxyhemoglobin concen- tration, and therefore differences in metabolic activity. Despite its underlying complexities, this technique offers a possible novel mode of metabolic imaging in acute stroke. ANN NEUROL 2010;68:37– 47 T he ischemic penumbra is a region of potentially viable tissue after acute ischemic stroke, which is eventually either recruited into infarct core or salvaged by reperfu- sion. 1 Because penumbral evolution varies widely among individuals, methods of imaging the penumbra have po- tential clinical importance. For example, they may be used for individual case selection for reperfusion therapies beyond current time windows, 2 or avoidance of thrombo- lytic therapy where reperfusion has already occurred. Gold standard penumbral imaging with 15 O positron emission tomography (PET) to measure both metabolic activity and perfusion 3 is impractical in routine Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.22032 Received Jun 9, 2009, and in revised form Feb 9, 2010. Accepted for publication Feb 16, 2010. Address correspondence to Dr Muir, Division of Clinical Neurosciences, University of Glasgow, Institute of Neurological Sciences, Southern General Hospital, Glasgow G51 4TF, Scotland, UK. E-mail: k.muir@clinmed.gla.ac.uk From the 1 Division of Clinical Neurosciences, University of Glasgow, Institute of Neurological Sciences, Southern General Hospital; 2 Department of Neuroradiology, Institute of Neurological Sciences, Southern General Hospital; 3 Glasgow Experimental MRI Centre, Division of Clinical Neurosciences, University of Glasgow, Garscube Estate; and 4 Department of Clinical Physics, Institute of Neurological Sciences, Southern General Hospital, Glasgow, UK. Additional Supporting Information can be found in the online version of this article. ORIGINAL ARTICLE © 2010 American Neurological Association 37