Diffusion and perfusion magnetic resonance imaging studies to evaluate a noncompetitive N-methyl-D-aspartate antagonist and reperfusion in experimental stroke in rats.

Abstract

Diffusion magnetic resonance imaging (MRI) can quantitatively detect focal ischemic injury within minutes of onset, and perfusion MRI can evaluate the brain's microcirculation. N-Methyl-D-aspartate (NMDA) antagonists and reperfusion can reduce lesion size in stroke models. We used diffusion and perfusion MRI to evaluate the in vivo effects of a noncompetitive NMDA antagonist, CNS 1102, in a temporary ischemia model. Sixteen Sprague-Dawley rats underwent suture occlusion of the middle cerebral artery. Fifteen minutes after occlusion, animals were randomly assigned to treatment with CNS 1102 (n = 10) or placebo (n = 6), receiving a bolus of 1.13 mg/kg at that time and an infusion of 0.785 mg.kg-1.h-1 for the next 165 minutes. The placebo group received a saline bolus and infusion. Diffusion MRI studies by a spin-echo technique were initiated 30 minutes after occlusion and repeated every 30 minutes for the next 3 hours. Perfusion MRI studies were obtained using echo-planar imaging after injection of superparamagnetic iron oxide particles, immediately before and 15 minutes after withdrawal of the occluder at 3 hours after middle cerebral artery occlusion. At 24 hours, the animals were clinically evaluated (scale of 0 to 5) and electively killed, and the brain was stained with triphenyltetrazolium chloride to evaluate infarct size. Diffusion imaging demonstrated markedly reduced ischemic lesion area in the CNS 1102 group during occlusion--10.5 +/- 7.3% (mean +/- SEM) of the ischemic hemisphere (optic chiasm slice) at 30 minutes after occlusion versus 50.0 +/- 2.7% of the hemisphere in controls (P < .02). With reperfusion after 3 hours of temporary ischemia, diffusion imaging documented an additional 29% reduction of the ischemic lesion area in the CNS 1102-treated group (P < .01) compared with the prereperfusion ischemic lesion area, with no change in the placebo group. During occlusion, perfusion imaging demonstrated a relative signal intensity decline of 31.5 +/- 7.7% in controls and 83.4 +/- 7.6% in the CNS 1102 group (P < .005), indicating better perfusion in the latter group. After removal of the occluder, perfusion improved in both groups and was not significantly different. Post mortem infarct volume was 53.8 +/- 20.0 mm3 in the CNS 1102 group and 216.8 +/- 16.1 mm3 in the controls (P < .0001). Clinical outcome at 24 hours was 1.1 +/- 0.4 in the CNS 1102 group and 4.0 +/- 0.5 (scale of 0 to 5) in the controls (P < .005). This study demonstrates that CNS 1102 reduces early postischemic injury as documented by diffusion MRI and improves perfusion as documented by perfusion MRI and that reperfusion confers additional reduction of ischemic lesion size.