Granulocyte colony-stimulating factor effects on neurological and motor function in animals with spinal cord injury: a systematic review and meta-analysis

BACKGROUND: Spinal cord injury (SCI) is a severe neurological injury for which no effective treatment exists. Granulocyte colony-stimulating factor (G-CSF) is used to treat autologous bone marrow transplantation, chemotherapy-induced granulocytopenia, Acquired Immune Deficiency Syndrome (AIDS), etc. Recent research has revealed the potential application of G-CSF on neuroprotective effectiveness. In central nervous system diseases, G-CSF can be used to alleviate neuronal injury. OBJECTIVE: To investigate the effects of G-CSF on Basso, Beattie, and Bresnahan (BBB) scale score, inclined plane test, electrophysiologic exam, quantitative analysis of TUNEL-positive cells, and quantitative analysis of glial fibrillary acidic protein (GFAP) immunostaining images in animal models of SCI. METHODS: We searched PubMed, Web of Science, and Embase databases for all articles on G-CSF intervention with animal models of SCI reported before November 2022. A total of 20 studies met the inclusion criteria. RESULTS: Results revealed that G-CSF intervention could improve the BBB scale score in both groups at 3, 7, 14, 28, and 35 days [at 35 days, weighted mean differences (WMD) = 2.4, 95% CI: 1.92–2.87, p < 0.00001, I(2) = 69%]; inclined plane test score; electrophysiologic exam; quantitative analysis of TUNEL-positive cell numbers; quantitative analysis of GFAP immunostaining images in animal models of SCI. Subgroup analysis revealed that treatment with normal saline, phosphate-buffered saline, and no treatment resulted in significantly different neurological function effectiveness compared to the G-CSF therapy. SD rats and Wistar rats with SCI resulted in significant neurological function effectiveness. C57BL/6 mice showed no difference in the final effect. The T9–T10 or T10 segment injury model and the T8–T9 or T9 segment injury model resulted in significant neurological function effectiveness. The BBB score data showed no clear funnel plot asymmetry. We found no bias in the analysis result (Egger’s test, p = 0.42). In our network meta-analysis, the SUCRA ranking showed that 15 mg/kg-20 mg/kg was an optimal dose for long-term efficacy. CONCLUSION: Our meta-analysis suggests that G-CSF therapy may enhance the recovery of motor activity and have a specific neuroprotective effect in SCI animal models. Systematic review registration: PROSPERO, identifier: CRD42023388315.