Cerebrospinal Fluid Inflammatory Cytokines and Biomarkers of Injury Severity in Acute Human Spinal Cord Injury

There is an urgent need for both the scientific development and clinical validation of novel therapies for acute spinal cord injury (SCI). The scientific development of novel therapies would be facilitated by a better understanding of the acute pathophysiology of human SCI. Clinical validation of such therapies would be facilitated by the availability of biomarkers with which to stratify injury severity and predict neurological recovery. Cerebrospinal fluid (CSF) samples were obtained over a period of 72 h in 27 patients with complete SCI (ASIA A) or incomplete SCI (ASIA B or C). Cytokines were measured in CSF and serum samples using a multiplex cytokine array system and standard enzyme-linked immunosorbent assay (ELISA) techniques. Neurological recovery was monitored, and patient-reported neuropathic pain was documented. IL-6, IL-8, MCP-1, tau, S100beta, and glial fibrillary acidic protein (GFAP) were elevated in a severity-dependent fashion. A biochemical model was established using S100beta, GFAP, and IL-8 to predict injury severity (ASIA A, B, or C). Using these protein concentrations at 24-h post injury, the model accurately predicted the observed ASIA grade in 89% of patients. Furthermore, segmental motor recovery at 6 months post injury was better predicted by these CSF proteins than with the patients' baseline ASIA grade. The pattern of expression over the first 3 to 4 days post injury of a number of inflammatory cytokines such as IL-6, IL-8, and MCP-1 provides invaluable information about the pathophysiology of human SCI. A prediction model that could use such biological data to stratify injury severity and predict neurological outcome may be extremely useful for facilitating the clinical validation of novel treatments in acute human SCI.