Intrathecal decompression versus epidural decompression in the treatment of severe spinal cord injury in rat model: a randomized, controlled preclinical research

Background: In the setting of severe spinal cord injury (SCI), there is no markedly efficacious clinical therapeutic regimen to improve neurological function. After epidural decompression, as is shown in animal models, the swollen cord against non-elastic dura and elevation of intrathecal pressure may be the main causes of aggravated neurologic function. We performed an intrathecal decompression by longitudinal durotomy to evaluate the neuroprotective effect after severe SCI by comparing with epidural decompression. Methods: Eighty-four adult male Sprague-Dawley rats were assigned to three groups: sham group (group S), epidural decompression (group C), and intrathecal decompression group (group D). A weight-drop model was performed at T9. The Basso-Beattie-Bresnahan (BBB) score was used to evaluate neurological function. Animals were sacrificed at corresponding time points, and we performed pathohistological examinations including HE staining and immunohistochemical staining (IHC) of glial fibrillary acidic protein (GFAP), neurocan, and ED1 at the epicenter of injured cords. Finally, the lesions were quantitatively analyzed by SPSS 22.0. Results: The mortality rates were, respectively, 5.55 % (2/36) and 13.9 % (5/36) in groups C and D, and there was no significant difference between groups C and D (P = 0.214). Compared with epidural decompression, intrathecal decompression could obviously improve BBB scores after SCI. HE staining indicated that more white matter was spared, and fewer vacuoles and less axon degradation were observed. The expression peak of GFAP, neurocan, and ED1 occurred at an earlier time and was down-regulated in group D compared to group C. Conclusions: Our findings based on rat SCI model suggest that intrathecal decompression by longitudinal durotomy can prompt recovery of neurological function, and this neuroprotective mechanism may be related to the down-regulation of GFAP, neurocan, and ED1.