Abstract Body: Introduction: Ischemic stroke is a leading cause of death and disability worldwide, but there has been limited success in translating putative treatments from preclinical trials to patients. The Stroke Preclinical Assessment Network (SPAN) is a large-scale multicenter trial (six sites plus a coordinating center).
Methods: SPAN implemented a multiparametric MRI protocol with minimal human input to assess tissue outcomes after mouse endovascular middle cerebral artery occlusion (MCAO). Imaging from 1766 mice was used in this study. The sample size includes three comorbidity models (Young mice, Aging mice, and diet-induced hyperglycemia/ obesity mice.
Results: Infarct volumes were variable across the network but right-skewed at all sites. Striatal and cortical infarcts were more common than thalamic and hippocampal infarcts (present in 75%, 65%, 20%, and 22% of mice, respectively). Total lesion was more associated with striatal and cortical infarct volumes (R=0.94 and 0.97). Infarct volumes were strongly associated with midline shift on days 2 (R=0.74) and 30 (R=-0.80), reflecting ischemic swelling and encephalomalacia, respectively. Factor analysis identified the underpinnings of the covariance among MRI variables (Figure 1). We found 13 MRI readouts linked to the injury severity after MCAO. These included the total, striatal, and cortical infarct volumes, indices of ischemic edema on day 2, and indices of tissue loss on day 30. Five variables were linked to infarction involving the posterior cerebral artery. Another set of five variables was linked to ventricular volume, suggesting an independent contributing factor, such as age. Average R2_rate in the contralateral tissue was lower in diet-induced obesity and aged mice compared with normal young mice, suggesting higher brain water content. Within the infarct, the R2_rate was significantly lower only in diet-induced obese mice versus the other groups.
Conclusion: Our data revealed critical insight into the stroke model regarding lesion distribution, swelling, and atrophy. Our findings also shed light on previously underappreciated biological associations among these variables.