"Median paralyzing dose" and "multiple regression analysis", a new viewpoint to the research method of spinal cord...

Spinal cord impact is a mature method for building models of spinal cord injury (SCI). However, a common problem is that the degree of elicited paralysis may not be identical even though animals receive the same impact. We hypothesize that this difference may be caused by the difference in the secondary injury mechanism of SCI and there might be an impact dosage named "median paralyzing dose (PD₅₀)", similar as the "median lethal dose (LD₅₀)" in pharmacology. In addition, since SCI is a result of multiple mechanisms, we hypothesize that it is more suitable to employ multiple regression analysis to analyze the related factors for complete paraplegia. So the present study aimed to calculate the existence of PD₅₀ and analyze the related factors of SCI-induced complete paralysis using logistic regression under the PD₅₀ which represents identical primary injury.

Rat models of SCI were built using the weight-drop method under PD₅₀. PD₅₀ was calculated by Karber's method. Rats were allocated into two groups according to whether they developed complete or incomplete paralysis 2 weeks after injury. Cavity and spared tissues in the two groups were compared. Neuronal preservation, microglia/macrophage reaction, T-lymphocyte infiltration, astrocyte activation and neuronal apoptotic were compared by immunohistochemistry. The logistic regression model was constructed and significant related factors of complete paralysis were selected.

Of the two groups, the cavity in the injured spinal cord of the complete-paralysis rats was significantly larger and the spared white matter volume (SWMV%) was obviously smaller. Whereas, the spared grey matter volume was not different between groups. Macrophage reaction, T-lymphocyte infiltration and neuronal apoptosis were significantly more severe in the complete-paralysis rats. Astrocyte activation and neuronal preservation showed no difference between groups. Logistic regression analysis showed that cavity volume, SWMV%, microglia/macrophage reaction and neuronal apoptosis were significantly correlated with SCI-induced complete paralysis.

As a non-mainstream method, it is feasible to analyze the secondary factors of SCI-induced complete paralysis using multiple regression analysis in the condition of identical primary injury (PD₅₀). SWMV% and microglia/macrophage reaction are important factors that contribute to complete paralysis at the early phase of severe SCI.