Background and Purpose: The translational roadblock has long impeded the implementation of experimental therapeutic approaches for stroke into clinical routine. Considerable interspecies differences, for example, in brain anatomy and function, render comparisons between rodents and humans tricky, especially concerning brain reorganization and recovery of function. We tested whether stroke-evoked changes in neural networks follow similar patterns in mice and patients using a systems-level perspective. Methods: We acquired resting-state functional magnetic resonance imaging data during the early poststroke phase in a sample of human patients and compared the observed network changes with data from 2 mouse stroke models, that is, photothrombosis and distal middle cerebral artery occlusion. Importantly, data were subjected to the same processing steps, allowing a direct comparison of global network changes using graph theory. Results: We found that network parameters computed for both mouse models of stroke and humans follow a similar pattern in the postacute stroke phase. Parameters indicating the global communication structure’s facilitation, such as small worldness and characteristic path length, were similarly changed in humans and mice in the first days after stroke. Additionally, small worldness correlated with concurrent motor impairment in humans. Longitudinal observation in the subacute phase revealed a negative correlation between initial small worldness and motor recovery in mice. Conclusions: We show that network measures based on resting-state functional magnetic resonance imaging data after stroke obtained in mice and humans share notable features. The observed network alterations could serve as therapeutic readout parameters for future translational studies in stroke research.

No abstract

No abstract