Abstract
Spinal cord injury (SCI) is a cause for significant morbidity, often resulting in long-term disability. We compared outcomes after administration of granulocyte-colony stimulating factor (G-CSF) versus controls. MEDLINE, Embase, and Cochrane Library database searches yielded 222 records; six met study inclusion criteria. Fixed-effect and random-effects models were used to establish odds ratios (ORs) and mean difference (MD) with 95% confidence intervals (CIs) for each outcome. The results of the pooled analysis showed that in patients with incomplete SCI, G-CSF resulted in increased American Spinal Cord Injury Association (ASIA) motor scores at 3 months (MD = 0.57 [95% CI = 0.04, 1.10], I-2 = 63.84%, p = 0.036), 6 months (MD = 4.18 [95% CI = 0.55, 7.80], I-2 = 98.75%, p = 0.024), change in ASIA pinprick scores at 6 months (MD = 3.38 [95% CI = 1.48, 5.28], I-2 = 89.78%, p < 0.001), and increased Spinal Cord Independence Measure (SCIM) III score at 6 months (MD = 3.27 [95% CI = 1.13, 5.41], I-2 = 91.86%, p = 0.003). G-CSF resulted in more adverse events than the non-MP control groups (OR = 1.44 [95% CI = 0.38, 2.50], I-2 = 0%, p = 0.008), but fewer than the MP control groups (OR = -4.2 [95% CI = -5.72, -2.68], I-2 = 0%, p < 0.001). Systemic white blood cell count increased after administration of G-CSF in comparison to baseline (MD = 3.57 [95% CI = 2.79, 4.35], I-2 = 55.06%, p < 0.001). G-CSF did not statistically increase ASIA Impairment Scale at 3 months (MD = 0.48 [95% CI = -0.33, 1.28], I-2 = 0%, p = 0.246) or at 6 months (MD = 1.84 [95% CI = -0.10, 3.79], I-2 = 50.09%, p = 0.063). These meta-analyses of six studies suggest that G-CSF for the treatment of incomplete SCI may result in improved neurological outcomes when compared to the controls. The results are limited by a small sample size with heterogeneity between studies. More robust prospective, randomized studies are necessary to help inform the safety and efficacy of G-CSF.