Spinal cord injury (SCI) is a devastating disease with no clear molecular mechanisms or effective treatments. Murine models of SCI have been widely used to explore its pathogenesis.
In this study, a comprehensive bioinformatic analysis using GEO datasets was performed to evaluate the characteristics of different SCI models.
We found that the contusion model was similar to the compression model, with inflammation and apoptosis significantly enriched, while more complex biological processes existed in hemisection and transection model. Inflammatory markers can be used as a primary evaluation index of SCI models not only in the acute and subacute phases, but also in the chronic phase. In the meantime, apoptosis markers are more suitable for evaluating mouse SCI models while inflammatory markers are more suitable for rat SCI models. In addition, SCI models with different ages, genders, injury positions, and injury levels were also analyzed.
Our findings indicate that SCI is a heterogeneous disease and play an instructive role in model selecting.