AANS Neurosurgeon | Volume 29, Number 1, 2020


The Evidence is Against the Use of Steroids for Acute Spinal Cord Injury Treatment: Part 2

Print Friendly, PDF & Email

Spinal cord injuries are all too common, even with countless efforts toward prevention. These patients suffer disability for the rest of their lives with huge costs to them, their family and society. For many reasons outlined in Part 1 of this article, glucocorticoids held great promise in the prevention of secondary injury. But, did it deliver?

Evidence for MPSS Use in SCI Lacking

Unfortunately, inherent flaws in animal experimental design limit direct translation to human SCI studies. The first animal experimentation in the field of pharmacologic intervention for spinal cord injury yielded fervor for therapies to be introduced to human subjects. Researchers sought to reproduce mechanical or ischemic insults in animal models via:

  • Weight drop
  • Focal or circumferential extradural balloon compression
  • Clip pressure
  • Photochemical trauma
  • Thermal injury
  • Distraction forces
  • Piston trauma

Therefore, animal injury models often omit the force of persistent compression that most humans suffer from anterior or circumferential vectors following spinal fractures. Similarly, the effects of repetitive mechanical trauma to the spinal cord from an unstable fracture or neurogenic shock are unaccounted for these animal models. There are also differences in drug metabolism among humans and animals. Metabolic differences compound with the timeliness of interventions performed in a controlled laboratory scenario in contrast with the variability inherent in clinical medicine. Altogether, the multiple limitations of animal studies are the foundation of why the results in human studies of SCI are not directly proportional with respect to expected outcomes.

National Acute Spinal Injury Studies


The National Acute Spinal Cord Injury Study (NASCIS) has become a three part series supporting the use of MPSS for acute spinal cord injury at face value. In NASCIS-I it was thought that not offering MPSS to patients with spinal cord injury was unethical and two arms of differing dosages were developed. Results published at one year of follow up failed to show a difference in the neurologic recovery of motor function, pinprick or touch. It was thought at the time that the dose administered was not sufficient to achieve therapeutic levels. 


The study group sought to correct the limitations of NASCIS-I with NASCIS-II. A placebo arm and a second treatment arm with naloxone were used to compare with MPSS results. Arbitrarily in this study, though bilateral neurological assessment was undertaken, only right sided data was used in analysis. In looking at the initial primary outcome analysis, neurologic outcome measures were not different between any of the treatment groups. However, an arbitrary eight-hour window from injury to administration of drug was chosen for post hoc analysis and it was in this subgroup of 12 patients only that an improvement in variable sensory or partial motor deficits was seen. Still when benefit was seen with respect to improvement in neurologic exam scores, the improvement did not correlate with improved functional status. Moreover, the rate of complications rose as the dose of MPSS escalated to show a:

  • Three fold rise in pulmonary embolus;
  • Two fold rise in wound infections; and
  • 5 fold increase in gastrointestinal hemorrhage.


Finally, NASCIS-III sought to evaluate the effect of a 24- versus 48-hour MPSS administration protocol with a comparison to another antioxidant tirilazad. A functional independence measure score was used to help translate neurologic improvement to interpretable functional improvement. In the initial data, at six weeks and six months, those receiving therapy 3-8 hours were reported to demonstrate improved neurologic function on the 48-hour protocol relative to the 24-hour protocol, though at one year this effect was lost. This study also reported two fold higher incidence of severe pneumonia, four fold higher severe sepsis and wound infection rate up trend on the longer steroid administration protocol.

NASCIS Critiqued

In the wake of publication, many criticisms of the data and the recommendation emerged. In 2000, John Hurlbert, MD, PhD, posited several criteria against which the studies in support of MPSS should be judged:

  • Evidence is obtained from a randomized double blind trial.
  • The study design and execution is optimal.
  • Data obtained is compelling with appropriate statistical analysis.
  • Changes yielded are meaningful to the patient.
  • Results are reproducible.

Though the NASCISs mostly meet the first two criteria, they fail in data analysis and the results yielded are not meaningful to the patients and have yet to be consistently reproduced.

Additional authors have used the Bracken protocol as a foundation for studies of acute SCI internationally. None have been able to reproduce the results seen to date.5 This may in fact be due to flawed statistical analysis in NASCIS-II and NASCIS-III. Multiple t-tests were used with analysis of variance and covariance, which predicates an assumption that the data set follows a Gaussian distribution. There were 66 subgroup comparisons in NASCIS-II and more than 100 in NASCIS-III introducing high likelihood of type I error, where non-parametric tests could have yielded cleaner results to interpret.

When reviewing the body of evidence-based medicine amassed between 2002 and 2012, the author group sponsored by the AANS/CNS Section on Spine and Peripheral Nerves deemed:

“Administration of methylprednisolone for the treatment of acute SCI is not recommended.”

This represented a reversal of the previous opinion in 2002. To date the FDA has not endorsed MPSS for this indication.

Still, there are physicians who advocate for the use of steroids to manage acute spinal cord injuries – some for fear of litigation and others because they truly believe in the benefit over risk of use. It is incumbent upon those physicians to explain what makes trauma to the spinal cord unique, since steroids have failed to show benefit for all other traumatic situations studied, including traumatic brain injury and traumatic optic neuropathy to date.


View All

1. Amar, AP and M Levy. “Pathogenesis and Pharmacological Strategies for Mitigating Secondary Damage in Acute Spinal Cord Injury.” Neurosurg. May 1999; 44 (5): 1027-1039.

2. Bracken, MB. “Steroids for acute spinal cord injury.” Coch Database Sys Rev 2012; Issue 1, Art. No.: CD001046.

3. Fehlings, MG. “Editorial: Recommendations Regarding the Use of Methylprednisolone in Acute Spinal Cord Injury: Making Sense Out of the Controversy.” Spine Dec 2001; 26 (24S): S56-S57.

4. Walters, BC, M Hadley, J Hurlbert, et al. “Guidelines for the Management of Acute Cervical Spine and Spinal Cord Injuries: 2013 Update.” Clin Neurosurg 2013; 60: 82-91.

5. Hurlbert, RJ. “The Role of Steroids in Acute Spinal Cord Injury: An Evidence-Based Analysis.” Spine Dec 2001; 26(24S): S39-S46.

6. Karsy, M and G Hawryluk. “Pharmacologic Management of Acute Spinal Cord Injury.” Neurosurg Clin N Am 2017; 28: 49-62.

7. Bracken MB, MJ Shephard, KG Hellenbrand, et al. “Methylprednisolone and neurological function 1 year after spinal cord injury. Results of the National Acute Spinal Cord Injury Study.” J Neurosurg 1985; 63: 704-713.

8. Bracken MB, MJ Shephard, WF Collins, et al. “A Randomized, Controlled Trial of Methylprednisolone or Naloxone in the Treatment of Acute Spinal-Cord Injury. Results of the National Acute Spinal Cord Injury Study.” NEJM 1990; 322 (20): 1405-1411.

9. Bracken MB, MJ Shephard, TR Holford, et al. “Administration of Methylprednisolone for 24 or 48 Hours or Tirilazad Mesylate for 48 Hours in the Treatment of Acute Spinal Cord Injury. Results of the National Acute Spinal Cord Injury Study Randomized Controlled Trial.” JAMA May 1997; 277 (20): 1597-1604.

10. Hurlbert, RJ, MN Hadley, B Aarabi, et al. “Pharmacological Therapy for Acute Spinal Cord Injury.” Neurosurg 2013; 72: 93-105.

11. Dumont, R, V Subodh, DO Okonkwo, et al. “Acute Spinal Cord Injury Part II: Contemporary Pharmacotherapy.” Clin Neuropharm 2001; 24 (5): 265-279.

12. S Miller. “Methylprednisolone in Acute Spinal Cord Injury: A Tarnished Standard.” J Neurosurg Anesthesiol 2008; 20 (2): 140-142.

13. Kwon, BK, W Tetzlaff, JN Grauer, et al. “Pathophysiology and pharmacologic treatment of acute spinal cord injury.” Spine J 2004: 451-464.


No upcoming events

Leave a Reply

Be the first to reply using the above form.