Corticosteroids represent the most intensively investigated pharmaceutical therapy in the post-SCI setting. The preponderance of preclinical studies have demonstrated improved functional outcomes with steroid use in animal models of traumatic SCI, with these results thought secondary to drug neuroprotective effects at the cellular level. Such protective actions have been shown to include reduction in tumor necrosis factor-a, reduced calcium influx, attenuation of lipid peroxidation, and improvement in spinal cord perfusion (9, 10, 11 and 12). Attempts at translating these results into the clinical setting lead to the first randomized trials of any therapeutic modality for traumatic SCI.

The National Acute Spinal Cord Injury Studies (NASCIS I, II, and III) remain the landmark clinical trials in the field of acute SCI. The initial study, the NASCIS I, which was completed in 1984, compared a 10-day regimen of high-dose methylprednisolone sodium succinate (MPSS) to low-dose MPSS and found no difference in neurologic outcome scores at 6 weeks or 6 months (13). Overall mortality and complication rates were the same between the groups, with the exception of wound infection rates, which were significantly higher among patients who received the high-dose regimen. Subsequent laboratory evidence suggested that steroid doses higher than those administered in NASCIS I would be required to appreciate a clinical effect (14).
In response to this, the NASCIS II was undertaken to compare higher-dose MPSS administration, commenced within 24 hours and continued for 24 hours, to naloxone and placebo (15). According to the authors, there was an a priori hypothesis to compare patients treated before and after 8 hours of injury. When all time points within the 24-hour window were considered, there was significant sensory-related, but no motor-related, neurologic benefit in the treatment group. However, a subanalysis of the group receiving treatment within 8 hours from injury revealed significantly greater motor recovery among MPSS-treated patients. This positive treatment-related effect within an 8-hour window has been confirmed by one Japanese study that restricted enrollment to those receiving treatment within 8 hours (16). This study however has been criticized for small population size and poor methodology, hindering our ability to draw conclusions based on its findings.

Finally, the NASCIS III trial compared a 24-hour to a 48-hour MPSS infusion as well as to a third treatment group that received tirilazad mesylate, a 21-aminosteroid believed to be an antioxidant without glucocorticoid effects (17,18). Although there was a trend for greater motor function improvement in the patients treated for 48 hours, no significant differences were found between groups. However, in an a priori-defined subcomparison, patients treated between 3 and 8 hours improved more in neurologic function if treated with the 48-hour MPSS regimen as compared to the 24-hour regimen. The 48-hour regimen represents the highest dose of MPSS prescribed for any clinical condition. This dose was associated with a significantly higher rate of pneumonia and severe sepsis but no difference in overall mortality in the 48-hour administration group, when compared to the 24-hour group at 1-year follow-up.

In summary, the routine use of corticosteroids in traumatic SCI remains controversial in light of several observations. The first is that in NASCIS II and NASCIS III, the only positive steroid treatment effects were seen in post hoc analyses and not in the primary analysis involving the wholly defined study population. Since these subanalyses were not powered to answer the primary research question, the results must be interpreted with caution (19). An additional point of concern is that with the extremely high doses of steroid administered, treatment is not without risk, especially when considering the vulnerable patient population, with many unable to tolerate infection or additional metabolic derangement. In all three NASCIS studies, the incidence of infectious complications was higher in the high-dose MPSS groups than in the placebo or other treatment groups. Given these issues, the Canadian Neurosurgical Society, the Canadian Spine Society, and the Canadian Association of Emergency Physicians have adopted the recommendation that a high-dose, 24-hour infusion of methylprednisolone, started within 8 hours after an acute closed SCI, is not a standard treatment nor a guideline for treatment but, rather, a treatment option, for which there is level II and III evidence (20).

Gangliosides are complex glycosphingolipids that contribute to the cellular structure of the central nervous system (CNS). A particular ganglioside molecule called monosialotetrahexosylganglioside or GM-1 demonstrated in cord-transected rats to induce regeneration and reestablishment of axonal continuity as well as improvement in functional outcome (21). On the heels of these promising preclinical data, two efficacy trials investigating the role of GM-1 in human patients with traumatic SCI were undertaken. The first, a small single-center placebo-controlled randomized controlled trial (RCT) of 37 patients, demonstrated that patients receiving GM-1 experienced significantly greater Frankel grade and ASIA motor score (AMS) improvements at 1-year follow-up (22). Additional analysis showed that the greatest gains in motor recovery observed in the treatment group occurred in the muscles of the lower extremities, postulated by the authors to be secondary to enhanced recovery occurring in the white matter tracts passing through the level of injury. In 2001, with an aim to validate the results of the preliminary trial, a second much larger placebo-controlled RCT, the Sygen Multicenter Acute Spinal Cord Injury Study, including 750 patients from 28 institutions, was published (23). Overall, the results of this trial were contradictory to those of the original, with authors failing to detect a significant difference between the groups in the proportion of patients who achieved at least a 2-grade improvement on the modified Benzel walking scale (primary outcome measure) at 26 weeks follow-up. Subanalyses revealed trends favoring GM-1 with respect to AMS, light touch scores, pinprick scores, and bowel/bladder function; however, none of these differences reached statistical significance. Critics of this study have argued that the mean time to Sygen administration post-SCI was excessive at 55 hours and that if the drug had been administered in a more timely fashion, perhaps a more robust treatment effect would have been documented.

Based on laboratory evidence demonstrating the neurotoxic effects of endogenous opioids released post-SCI, naloxone, an opiate receptor antagonist, has been studied in the context of two human trials (24,25). Flamm et al. (26) published a phase 1 trial evaluating a high- versus low-dose naloxone regimen. This initial small and nonrandomized exploratory analysis suggested greater potential for neurologic improvement in the group receiving the high-dose regimen and demonstrated the safety of this therapy in the SCI population. To formally assess this drug in an efficacy trial, naloxone was compared to MPSS and placebo in the NASCIS II study but failed to demonstrate a significant treatment effect relative to these therapies (15).

In addition to its well-defined hypophysiotropic role, thyroid-releasing hormone (TRH) has been found to disrupt secondary injury mechanisms post-SCI, including antagonizing the effects of endogenous opioids, peptidoleukotrienes, platelet-activating factor, and excitotoxins (27). The only human trial was a small placebo-controlled RCT, in which a total of 20 patients were randomly allocated to either TRH bolus followed by a 6-hour infusion or to
saline bolus and infusion (28). At 4 months follow-up, those who had incomplete injuries and received TRH had significantly higher motor, sensory, and Sunnybrook scores than those incomplete patients who received placebo. There was no treatment effect for patients with complete injuries. Obviously, these results are to be cautiously interpreted given the small patient numbers and the lack of a larger follow-up confirmatory study.

Another potent mediator of secondary injury post SCI is glutaminergic excitotoxicity wherein injured cells release an abundance of the excitatory neurotransmitter glutamate, which serves to exacerbate neural tissue damage. Gacyclidine is a noncompetitive N-methyl-D-aspartic acid receptor antagonist that has demonstrated neuroprotective properties and a reduced side effect profile, compared to other glutamine receptor blockers, in preclinical studies evaluating the drug in rat SCI models (29,30). A multicenter French trial randomized over 200 patients to receive one of three escalating doses of Gacyclidine or placebo (31). At 1-year follow-up, there was a nonsignificant trend for greater motor improvements among those with incomplete cervical injuries receiving high doses; however, there were no significant treatment effects found overall.

Nimodipine, an L-type calcium channel blocker, when administered in animal SCI has been shown to attenuate the secondary injury cascade by reducing intracellular calcium entry, preventing the activation of destructive enzymes and reducing glutaminergic excitotoxicity (32). One potential drawback of this therapy is concomitant hypotension, secondary to vascular smooth muscle relaxation and vasodilatation, which places the spinal cord at risk for ischemic injury. When adequate spinal cord perfusion is maintained, nimodipine administration has been associated with functional improvements in rats with compression induced SCI (33). An attempt at translating this therapy to humans was performed in a French study of 100 patients randomly allocated to treatment with nimodipine, MPSS, both, or placebo (34). Unfortunately, there was no evidence of neurologic benefit over placebo in any of the treatment groups. To our knowledge, there have been no further clinical studies evaluating nimodipine in this context.

Minocycline, a metalloproteinase inhibitor and a chemically modified version of the antibiotic drug tetracycline, has been shown to have several actions that lead to the reduction of secondary injury after SCI. Such actions include inhibiting caspase gene expression, reducing mitochondrial cytochrome c release, and blocking excitotoxic glutamate release (35,36). Initial animal studies demonstrated improved locomotor outcomes and histologic tissue preservation with minocycline administration (37, 38, 39 and 40). However, a subsequent rat study, replicating the experimental conditions of preliminary work, failed to validate the positive findings of the original studies (41). A phase 1/II trial at the University of Calgary, investigating the clinical safety and tolerability of minocycline, began recruiting patients in 2007 and remains open for enrollment.