Secondary Logo

Journal Logo

The Role of Methylprednisolone in Acute Spinal Cord Injury

Methylprednisolone and Acute Spinal Cord Injury

An Update of the Randomized Evidence

Bracken, Michael B. PhD

Author Information
  • Free


It has been more than 10 years since publication of NASCIS II, 14,15 which led to the widespread use of high-dose methylprednisolone sodium succinate (MPSS) for the early treatment of acute spinal cord injury (SCI), and about 25 years since the founding of NASCIS, which was established to conduct and encourage trials of therapies for managing acute SCI. This article reviews the current state of evidence concerning the efficacy and safety of high-dose MPSS. Randomized trial evidence is well recognized to offer the most unbiased and highest quality evidence for therapeutic interventions and is the focus of the present update. The most recent systematic review of pharmacologic interventions for acute SCI, published in the Cochrane Library, 7 reviews this material in detail and is updated here.


The literature search for the current review was conducted in January 2001 using (all years) Medline, CINAHL, Best Evidence, and Current Contents databases. The MeSH headings methylprednisolone and acute spinal cord injury were searched with no other restrictions. The Cochrane Controlled Trials Register and the systematic review concerning pharmacologic interventions for acute spinal cord injury in the Cochrane Library 7 was also searched for trials. The Cochrane Library is a collection of peer-reviewed, evidence-based systematic reviews in medicine and surgery. The primary evidence for this review is based on human randomized, or quasi-randomized, controlled clinical trials. Trials were selected for inclusion if they investigated MPSS in any dose or regimen, and the comparison was placebo or nonuse of MPSS.

The quality of the trials was assessed by whether MPSS was administered in a truly random fashion against placebo or no therapy, whether the therapy was administered blind, whether the assessment of neurologic or functional recovery and sequelae was made blind to therapy, and whether a substantial number of patients (>90%) were followed up (Table 1). These criteria are the primary criteria promulgated by the Cochrane Collaboration. 20 Additional data to that published have been obtained from Petitjean et al 58 and Otani et al 56 for the Cochrane review 7 and are used in this report. Where necessary, standard deviations have been imputed by the method of Follmann et al. 29

Table 1
Table 1:
Assessment of Methodologic Quality of the Five MPSS Trials

A number of reports have been excluded from the review because the therapy was not administered randomly and comparisons are based on historical controls, contemporaneous but nonrandomized controls, or no controls. 31,34–36,48 An article by Pointillart et al 61 is excluded because it is a duplicate publication to Petitjean et al 58 with simply a translation of French to English, a change in first authorship, changing numbers to percent in one table, and without reference to the original article. An article by Sun et al 69 that compares high-dose MPSS and/or surgical decompression is excluded because the original is in Chinese, has not been translated, and its method of assignment to therapy is unclear. Other excluded reports are of penetrating SCI, typically from gunshot wounds. 45,50,62 This type of injury was excluded from the NASCIS and other trials, and no evidence from randomized controlled trials exists as to efficacy or safety.


Evidence for Efficacy

The evidence for the efficacy of MPSS comes from five randomized controlled trials conducted over a period of 20 years. The first trial, NASCIS I, 10,16 was based on a large body of animal literature suggesting the plausibility of MPSS as a therapy for the early management of acute SCI. 40 NASCIS I appears to have been the first randomized trial of any therapeutic modality for acute SCI and was a double-blind multicenter trial using a “double dummy” technique to mask the active and placebo arms of the trial. Patients needed to be randomized within 48 hours of injury and were administered either 100 or 1000 mg MPSS as a loading dose and 25 or 250 mg every 6 hours thereafter for 10 days. Neurologic examinations were conducted 6 weeks, 6 months, and 1 year after injury using a detailed neurologic examination that is almost exactly like the more recently adopted American Spinal Injury Association/International Medical Society of Paraplegia 2 criteria. NASCIS I was reported as a negative trial at all outcomes, and when inquiries were made about patients treated early, 19 an analysis of those treated within 6 hours showed no evidence of benefit, although very few patients were treated so early. 11 After subsequent interest focused on an 8-hour window after publication of NASCIS II, reanalysis of 112 NASCIS I patients treated within 8 hours showed some suggestion of benefit from moderate over low-dose MPSS at 6 months (P = 0.06) and 1 year (P = 0.20) (Figure 1). 7

Figure 1
Figure 1:
Summaries of five trials of the effect of high-dose methylprednisolone, administered within 8 hours of injury, on motor function. The height of each square is inversely proportional to the length of the confidence interval. Heterogeneity among trials is evaluated by the χ2 test. Statistical significance of the effect estimate is measured by a z test.

NASCIS II was the second multicenter randomized double-blind trial of acute SCI. 14,15 Treatment was either a 30 mg/kg bolus followed by 5.4 mg/kg/hr infusion of MPSS for 24 hours or placebo (a third treatment, naloxone, is not considered further here). Neurologic examinations followed those for NASCIS I. In NASCIS II an a priori hypothesis was proposed to study the effect of therapy commenced early or late after injury, and an 8-hour window was selected because it was closest to the median time of 8.5 hours, it was the only time window analyzed, and no post hoc alternative dichotomies were examined.

A Japanese randomized trial administered exactly the same MPSS regimen as NASCIS II to patients but the control group was absence of therapy rather than placebo, and it cannot be assumed that the trial was blind. There was differential loss to follow-up, and this seemed to occur more frequently in severely injured controls. Neurologic outcomes were assessed at 6 months using NASCIS II criteria. 56 A small French single center trial randomly administered MPSS according to NASCIS II criteria, except that the bolus dose was administered over 1 hour rather than 15 minutes and compared patients using American Spinal Injury Association/International Medical Society of Paraplegia at 1 year with those receiving no therapy rather than placebo. These interventions also cannot be assumed to be blind. 58 Both the Japanese and French trials used an 8-hour therapeutic window for eligibility to the trial.

The combined results of the three trials of high-dose MPSS versus placebo or no use are shown in Figure 1. Using standard statistical methods developed by the Cochrane collaboration, 20 the typical estimate of the size of effect is an improvement of 4.1 points on the motor score (P = 0.02) in patients treated with MPSS. An absence of significant heterogeneity in the trial estimates (P = 0.37) provides the rationale for a meta-analysis and the Cochrane weights (calculated automatically by the statistical program, shown by the box size and largely a function of trial sample size) show a preponderance of the weight of evidence to be from the NASCIS II and Japanese trials. It should be noted that this motor improvement is for only one side of the body, the 95% interval ranges from 0.6 to 7.6, and neurologically incomplete patients show greater improvement than completes (in NASCIS II motor improvement scores of 7.3 and 3.6, respectively).

A sensitivity analysis of the three trials indicates that if the Otani et al 56 trial were deleted (on grounds of unbalanced follow-up), the typical estimate of motor improvement is 4.2 (P = 0.07); if the Petitjean et al 58 trial is deleted (because the bolus dose was administered over a long period of time), the typical motor improvement is 4.7 (P = 0.01); and if both trials are deleted (because they are not placebo controlled), the effect from the NASCIS II trial is a motor improvement of 5.2 (P = 0.03).

Further analysis of NASCIS II suggested that the greatest degree of motor function recovery owed to MPSS within 8 hours of injury occurred in long spinal tracts as recorded below the level of lesion, although significant segmental recovery was also found at the level of the lesion. 12 The same analysis indicated that delayed initiation of administration of MPSS after 8 hours was associated with decreased neurologic recovery, suggesting that endogenous repair mechanisms may be hindered by MPSS without benefit of early administration. A possible mechanism for this delayed effect is the role of MPSS in depressing production of neurotrophins that are important for central nervous system (CNS) development and are normally enhanced after traumatic CNS injury. 43 Long tract recovery also accounted for the majority of total recovery in NASCIS III, for example, 85.6% and 89.9% of total motor function recovery at 6 months in the 24- and 48-hour MPSS-treated groups, respectively. 17

The fifth trial of MPSS (NASCIS III) examined the potential benefit of extending the NASCIS II regimen of 24-hour therapy to 48 hours using essentially identical protocols for drug administration and neurologic assessment as NASCIS II. 17,18 In all patients at the 1-year follow-up, a further 2.4 (95% confidence interval [CI] 1–7, 6.5, P = 0.30) statistically nonsignificant improvement in motor point scores was observed (Figure 1). Most benefit from the 48-hour regimen occurred in patients whose initial administration of MPSS was delayed from 3 to 8 hours after injury with motor improvement scores of 13.7 and 19.0 (P = 0.05) for 24- and 48-hour MPSS, respectively. The 3-hour dichotomy was based on the median time to treatment, was the only dichotomy examined, and was planned as an “early versus late” analysis in the study proposal.

FIM scores were only assessed directly in NASCIS III. At 1 year the total FIM in the 48-versus 24-hour MPSS-treated patients was improved by 2.6 points (P = 0.29) in the intent-to-treat analysis and 3.5 points in the compliers analysis (which excluded 18 patients with incomplete drug administration, P = 0.16) with self-care (P = 0.15 and P = 0.08, respectively) and sphincter control (P = 0.20 and P = 0.12, respectively) appearing to show the greatest effects of MPSS. Greater gains were seen in patients whose treatment could not be started until after 3 hours of injury, but as for the preceding data, none of these comparisons reached nominal levels of statistical significance. 18

To estimate the degree of functional recovery owed to MPSS that would be predicted from the improved motor scores observed in NASCIS II, the NASCIS II and III data have recently been analyzed to document the association between motor neurologic scores and the Functional Improvement Measure (FIM). 13 Based on the overall mean 5.1 motor improvement score owed to MPSS in NASCIS II, 18.6% of patients would improve 6+ FIM points and 9.0% 9+ FIM points. For complete patients the mean motor improvement of 3.6 predicted 63.9% would improve 3+ FIM points and 12.1% 6+ FIM points to a maximum of 8 points. Among incomplete patients the 7.3 mean improvement in motor function predicted 27.4% gaining 6+ FIM points and 21.0% gaining 9+ points to a maximum of 15 points. An improvement of 9+ FIM points is about a ≥10% improvement on the modified FIM scores used in this analysis (which omitted measures for comprehension and expression). These are clinically important degrees of recovery in the FIM score that are predicted by the motor function recovery scores in NASCIS II.

Related Randomized Trials

A Swedish trial of 40 men and women with whiplash injury Grade 2 and 3 by Quebec criteria and enrolled within 8 hours of injury, randomized patients to the NASCIS II dose of MPSS or placebo. Patients administered MPSS had fewer disabling symptoms (P = 0.047), fewer sick days (P = 0.01), and a healthier sick-leave profile (P = 0.003) 6 months after injury. 60 A second trial was of 32 patients undergoing lumbar discectomy for radicular symptoms and radiographically confirmed herniated nucleus pulposus. Twelve patients received 160 mg intramuscular Depo-medrol (methylprednisolone acetate) and 250 mg MPSS at the start of the procedure and a macerated fat graft soaked in 80 mg Depo-Medrol placed over the affected nerve root after discectomy. Another 10 patients received bupivacaine and 10 patients received neither bupivacaine nor corticoids. For patients given corticoids their hospital stay was significantly shorter than patients not so treated (1.4 vs. 4.0 days, P = 0.0004). 37 It is likely that both of these trials are primarily assessing nerve root recovery rather than long tract SCI.

Evidence for Safety

Wound infection among MPSS-treated patients was significantly increased in the NASCIS I trials, although in NASCIS II, in which MPSS was administered at a much higher dose but for 24 hours rather than 10 days, the absolute rates of infection were lower than in NASCIS I but still increased over placebo. Mortality from all causes at 6 months from the high-dose MPSS trials (Otani et al 56 and NASCIS II) was lower in the MPSS-treated patients compared with placebo or nothing (3.3%vs. 6.2%, relative risk [RR] = 0.54; 95% CI 0.24, 1.25). One-year mortality in the 48-versus 24-hour MPSS-treated patients showed essentially no difference between the two groups (6.0%vs. 5.4% respectively, RR = 1.11; 95% CI 0.46 2.66). 7

Results from the trials are in stark contrast to concerns raised in the recent literature about the safety of MPSS therapy. These commentaries are notable for being based on speculation rather than data, 46,55 being hypothesized from pharmacokinetics properties of MPSS, 63,65 being based on animal studies, 67 or using data but from uncontrolled or historically controlled small series of patients. 30,31,34,35

A recent systematic review of almost 2500 patients in 51 trials of the use of high-dose MPSS versus placebo or nothing by Sauerland et al 64 provides further reassurance of safety. High-dose MPSS was defined as any intravenous dose exceeding 15 mk/kg or 1 g MPSS given as a single or repeated dose within a maximum of 3 days and discontinued afterwards. The trials include trauma and elective and spine surgery considered by the authors to be of comparable severity and risk. No evidence was found for any increased risk of gastrointestinal bleeding (risk difference [RD] = 0.3%, P = 0.4), wound complication (RD = 1%, P = 0.2), pulmonary complication (for which MPSS was significantly protective RD = −3.5%, P = 0.003) or death (also moderately protective, RD = −0.9%, P = 0.10). No evidence of harm was found when spine surgery alone was considered, and citing specifically the acute SCI reports of Galandiuk et al 31 and Gerndt et al, 36 Sauerland et al noted that “. . .some nonrandomized studies have described serious complications after glucocorticoid administration, such as pneumonia. However, these findings can mainly be explained by the selection of more severely ill patients into an MPSS treatment regimen.”64 In another study long-term follow-up of avascular necrosis after high-dose MPSS, diagnosed by MRI of femoral and humeral heads assessed blind to steroid therapy, failed to find any increased risk. 73

There has been one report of two cases of steroid psychosis that were considered sufficiently severe to place the patient at risk of further serious injury. 70 The same authors suggest a “conservative” estimate of a 5.7% incidence for this condition. However, psychosis was not reported in the review of trials by Sauerland et al, 64 in which 981 patients received high-dose MPSS or in the 495 patients receiving high-dose MPSS in NASCIS II and III. A proposed incidence rate of 5–6% would predict 74–88 cases of psychosis, whereas none was reported. The randomized trial data suggest that if steroid psychosis is associated with high-dose MPSS, it is at a rate no higher than 1 per 1300 treated patients (adjusted because NASCIS II is counted in both series).

Analysis of NASCIS II found no evidence of compromised liver function as evaluated by serum glutamic-oxaloacetic transaminase, serum glutamic-pyruvic transaminase, alkaline phosphatase, and total bilirubin when measured 24 hours and 3 and 10 days after the end of drug infusion. 66 Even when controlling for drug protocol and severity of injury, variation in enzyme levels appeared to result from the SCI, not MPSS.

Plausibility of Methylprednisolone Therapy

Early work demonstrated that MPSS could deactivate free-radical oxidation products that progressively demyelinate neurons after SCI 25 and is supported in numerous studies, reviewed most recently by Hall. 41 These observations formed the theoretical rationale for the NASCIS II and III trials.

Since the publication of the first NASCIS trials, numerous additional animal studies have been reported, many looking at new biologic mechanisms. These studies contribute to the plausibility of MPSS improving recovery after acute SCI and to an understanding of new possible mechanisms for its efficacy. More recent animal experiments with MPSS have tended to either examine the effects of administering MPSS as a “cocktail” with another compound or to examine the effect of MPSS in other injury or disease conditions such as traumatic brain injury, 75 optic nerve damage, 5 and stroke. 23 These other indications for MPSS are not considered further in this review.

The effect of administering MPSS in conjunction with other compounds for acute SCI has included investigation of cyclosporin A, 26 riluzole, 53 the melanocortin melanotropin alphaMSH, 49 vitamin E, 24 basic fibroblast growth factor, 3 monosialic ganglioside GM1, 22 the TRH analogue YM-1473, 6 and thyrotropin-releasing hormone TRH. 1 These experiments have been inconclusive, showing either greater improvement in recovery from the combined therapy, equivalent improvement to mono-MPSS therapy, or a worsening of recovery. The experiments are remarkable for having no replication and none have gone on to human trials.

In addition to the widely studied therapeutic effects of MPSS on blood flow, inflammation, and edema, and the role of MPSS in alleviating lipid peroxidation after CNS injury, 40,41 other possible mechanisms observed for MPSS have included the following: a reduction in the “dieback” of vestibulospinal fibers and an increase in their transient regenerative sprouting 57; suppression of postinjury inflammation by inhibiting TNF-α NF-kB binding activity 74; inhibition of the proteinase calpain to prevent myelin destruction 4; failure to inhibit eicosanoids but inhibition of pial hyperfusion 42; and reduction of the release of excitatory amino acids. 51 One study of the role of MPSS in a rat model of nonmissile penetrating injury suggested a possible role for MPSS in prophylaxis for planned or incidental surgical trauma. 54 This concept has not been carried forward into randomized controlled trials.


The randomized trials of MPSS in the treatment of acute SCI provide evidence for a significant improvement in motor function recovery after treatment with the high-dose regimen within 8 hours of injury. This improvement is modest but does appear to have the potential to result in important clinical recovery in some patients. Even small changes in motor recovery, typically assessed in the MPSS trials on one side of the body, have the potential to be amplified into meaningful improvements in quality of life. 52,71 For example, an ability to move fingers may mean the difference between being employed or not. Sensory function improvement has been difficult to document, but in all trials and comparisons those patients administered MPSS recovered more sensory function than those not treated. 7

It is well understood that results of clinical trials do not predictably translate into experience recorded in normal clinical practice. 39 The enthusiasm and attention to a detailed clinical protocol, with attendant close monitoring of adherence to protocol and surveillance of protocol violations, which accompanies the modern randomized trial, are often not carried over to the vicissitudes of routine clinical practice. This is particularly likely to be the case when the protocol for therapy administration is not straightforward. There are only ad hoc reports of compliance to the NASCIS II protocol in routine practice, but they include reports to this author of therapy commencing >8 hours after injury, therapy being discontinued before 24 hours, the full 24-hour dose of MPSS being administered in the first hour, the maintenance therapy being administered at rates faster or slower than the recommended rate of 5.4 mg/kg/hr, and poor estimation of patient weight so that the administered dose is incorrect. Whether or not, and by how much, these protocol variations influence therapeutic efficacy and safety is, of course, unknown, but they may account for experiences different from those reported in the randomized trials.

A further difficulty in translating the results of trials investigating early pharmacologic therapy, which necessitate an emergency room baseline neurologic examination, is imprecision in such examinations for diagnosing complete neurologic loss because spinal shock may confound the diagnosis. If patients are incorrectly classified as having “complete” injuries in the emergency room, compared with an examination at 48 hours, for example, then neurologic recovery in this group of patients will be less than is found in patients classified more accurately as being complete in clinical practice. Because this error occurs in both MPSS-treated and nontreated patients, the validity of the overall trial results is not affected, only the result in patients categorized as being complete or incomplete. What is frequently not recognized is that neurologic recovery in incomplete patients will also be less than observed in clinical practice if more severe patients classified as complete in the trial really belong in the incomplete category. Again, because the overall result of a trial derives from summed weighted averages of the subgroups, the total trial effect is not influenced. This is an example of the Will Roger’s phenomenon, more properly called stage migration, 28 as exemplified by Will Rogers who is reported to have said that when the Okies moved from Oklahoma to California the average IQ of both States increased!

There is strong evidence for the safety of high-dose MPSS. The NASCIS II trial results provide evidence that complications are low. The number of patients who need to be treated to result in harm to one patient is 29 for wound infection, 67 for gastrointestinal bleeding, and 31 for pulmonary sequelae. 9 The review of 51 trials produced numbers needed to harm for wound infection of 100, 333 for gastrointestinal bleeding, and 29 for pulmonary complications. 64 There was some evidence in NASCIS III of a rise in mortality due to pneumonia, respiratory distress syndrome, or respiratory failure (grouped together) in the 48-hour methylprednisolone group. 8 This result was based on six deaths in the 48-hour group and one in the 24-hour group (RR = 6.0, 95% CI 0.73, 49.3), but in the Sauerland et al 64 analysis of 51 trials pulmonary complications were significantly reduced by 3.5% with a high degree of precision (95% CI −1.0, −6.1). 64

Animal studies offer evidence that the intervention might plausibly be expected to improve recovery and animal research forms some of the evidence base supporting trials. The NASCIS trials were all preceded by a substantial body of animal research that provided a rationale for therapeutic action. 40 After publication of the NASCIS trials we have seen new mechanisms of action postulated and considerable interest in trying to examine drug cocktails, which have typically added an experimental therapy to MPSS. This approach has been widely promoted, although the logic for it is weak.

First, although documenting proof of the principle that secondary CNS injury can be ameliorated by drug therapy, MPSS in the NASCIS trials has modest clinical effects and the development of newer single therapies that outperform MPSS is a more reasonable immediate strategy than searching for “cocktails.” New trials need to have MPSS therapy as one arm of investigation. If MPSS is barely effective, it will be readily challenged by a more efficacious drug. If MPSS provides some therapeutic benefit, then MPSS must be the standard of care to beat. This does not imply that patients in the experimental arms of new trials should be first treated with a full course of MPSS. In NASCIS III patients in the tirilazad mesylate arm of the trial were only given a bolus of MPSS and other trial groups may decide it is ethical to omit MPSS in the experimental arms of a new trial.

Second, sequencing therapies to more appropriately act on the biologic changes in the postinjury cascade that follow acute spinal cord trauma is a more rational approach than simply adding one drug to another. One drug may prevent the early onset of free radical damage, another may stabilize myelin and neuron architecture, whereas a third may express genes necessary for new neuron growth and regeneration, each at particular times during recovery. The GM1 trial represents this type of approach with GM1 therapy, hypothesized to have properties that encouraged neuron growth, after administration of MPSS to reduce early neuron damage. 33 The limited success of this particular GM1 treatment regimen 32 should not be a reason to abandon the principle it espoused.

It is difficult to know how widely MPSS is used in spinal cord patients because no objective current reports of use are available. MPSS is approved for use and routinely used in many countries and is an accepted standard of care as documented in numerous clinical textbooks and articles. A group of surgeons in Alberta, Canada, is reported to have recently recommended discontinuation of routine methylprednisolone administration. 47 Regrettably, the absence of a peer-reviewed published guideline makes it impossible to judge how systematically the evidence was reviewed, who reviewed it, what criteria were used for accepting studies under review, whether the literature studied was up to date, and whether cost–benefit analyses or a cost–utility analysis accounting for patient values was done, all of which are requisites for modern practice guideline statements. 21,38,44,72 Nor do we know if an economic analysis was conducted. 59 Even a superficial review of the cost of MPSS would indicate that it adds a trivial amount to total hospitalization costs for acute SCI and even minor clinical improvement because of its use would justify routine administration. It is unknown whether Alberta patient advocate groups were part of a consultative process or how they would react to patients being denied a therapy that they would most likely receive elsewhere.

Curiously, even though the rationale for use of MPSS is based on randomized trials, indeed, more so than any other form of management for acute SCI, it has also been the subject of unusually critical commentary. Silverman warned against the dangers of negativism: “In medicine most particularly, we are aware of the need to maintain a delicate balance between demolition and construction.”68

Perhaps expectations for a therapy for acute SCI have simply been too high. Again, Silverman offered wise counsel: “What needs to be overcome is a naïve ‘all-and-none concept’: the unrealistic expectation that an effective treatment will cure virtually all patients with a specified illness; and, conversely, practically none of the patients with this illness will improve without specific treatment.”68

No new data have emerged in the last decade to alter a recommendation reached after publication of NASCIS II: “After 20 years of clinical research and studies in animals, we now have evidence that a medication can reduce some of the damage that results from acute spinal injury. . .. Clearly the new treatment is not a cure, but corticosteroids should be used as soon as possible after spinal injury in the very high doses studied in the clinical trial.”27


1. Akdemir H, Pasaoglu H, Arman F, et al. Effects of TRH and high-dose corticosteroid therapy on evoked potentials, and tissue Na+, K+ and water content in experimental spinal injury. Res Exp Med 1993; 193: 297–304.
2. American Spinal Injury Association/International Medical Society of Paraplegia. Standards for neurological and functional classification of spinal cord injury, rev ed. Chicago, IL: American Spinal Injury Association, 1992.
3. Baffour R, Achanta K, Kaufman J, et al. Synergistic effect of fibroblast growth factor and methylprednisolone on neurological function after experimental spinal cord injury. J Neurosurg 1995; 83: 105–10.
4. Banik NL, Matzelle D, Terry E, et al. A new mechanism of methylprednisolone and other corticoids action demonstrated in vitro: inhibition of a proteinase (calpain prevents myelin and cytoskeletal protein degradation). Brain Res 1997; 748: 205–10.
5. Beck RW. The optic neuritis study group. Corticosteroid treatment for optic neuritis: a need to change treatment practices. Neurology 1992; 42: 1133–5.
6. Behrmann DL, Bresnahan JC, Beattie MS. Modeling of acute spinal cord injury in the rat: neuroprotection and enhanced recovery with methylprednisolone, U-74006F and YM-14673. Exp Neurol 1994; 126: 61–75.
7. Bracken MB. Pharmacological interventions after acute spinal cord injury (Cochrane Review). In: The Cochrane Library, Oxford: Update Software, 2000:Issue 3.
8. Bracken MB. Methylprednisolone and spinal cord injury. J Neurosurg Spine 2000; 93: 175–8.
9. Bracken MB, Aldrich EF, Herr DL, et al. Clinical measurement, statistical analysis, and risk benefit: controversies from trials of spinal injury. J Trauma Injury Infection Crit Care 2000; 48: 558–61.
10. Bracken MB, Collins WF, Freeman DF, et al. Efficacy of methylprednisolone in acute spinal cord injury. JAMA 1984; 251: 45–52.
11. Bracken MB, Collins WF, Eisenberg HM, et al. Author’s response. J Neurosurg 1986; 64: 986.
12. Bracken MB, Holford TR. Effect of timing of methylprednisolone or naloxone administration on recovery of segmental and long-tract neurological function in NASCIS 2. J Neurosurg 1993; 79: 500–7.
13. Bracken MB, Holford TR. Neurological and functional status one year after acute spinal cord injury: estimates of functional recovery in NASCIS 2 from results modeled in NASCIS 3. Submitted
14. Bracken MB, Shepard MJ, Collins WF, et al. A randomized controlled trial of methylprednisolone or naloxone in the treatment of acute spinal cord injury: result of the Second National Acute Spinal Cord Injury Study. N Engl J Med 1990; 322: 1405–11.
15. Bracken MB, Shepard MJ, Collins WF, et al. Methylprednisolone or naloxone treatment after acute spinal cord injury: 1-year follow-up data. Results of the Second National Acute Spinal Cord Injury Study. J Neurosurg 1992; 63: 704–13.
16. Bracken MB, Shepard MJ, Hellenbrand KG, 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: 704–13.
17. Bracken MB, Shepard MJ, Holford TR, 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 third National Acute Spinal Cord Injury randomized controlled trial. JAMA 1997; 277: 1597–604.
18. Bracken MB, Shepard MJ, Holford TR, et al. Methylprednisolone or tirilazad mesylate administration after acute spinal cord injury: 1-year follow-up. Results of the third National Acute Spinal Cord Injury randomized controlled trial. J Neurosurg 1998; 8: 699–706.
19. Braughler JM, Hall ED. “High-dose” methylprednisolone and CNS injury. J Neurosurg 1986; 64: 985–6.
20. Clarke M, Oxman AD, eds. Cochrane Reviewers’ Handbook 4.0 [updated July 1999]. In: Cochrane Library [database on CDROM]: The Cochrane Collaboration. Oxford: Update Software, 2000: Issue 1.
21. Cluzeau FA, Littlejohs P. Appraising clinical practice guidelines in England and Wales: the development of a methodologic framework and its application to policy. Jt Comm J Qual Improve 1999; 25: 514–21.
22. Constantini S, Young W. The effects of methylprednisolone and the ganglioside GM1 on acute spinal cord injury in rats. J Neurosurg 1994; 80: 97–111.
23. Courten-Myers GM, Kleinholz M, Wagner KR, et al. Efficacious experimental stroke treatment with high-dose methylprednisolone. Stroke 1994; 25: 487–93.
24. Daneyemez M. Silicone rubber microangiography of injured acute spinal cord after treatment with methylprednisolone and vitamin E in rats. Spine 1999; 24: 2201–5.
25. Demopoulos HB, Flamm ES, Seligman ML, et al. Further studies on free-radical pathology in the major central nervous system disorders: effect of very high doses of methylprednisolone on the functional outcome, morphology, and chemistry of experimental spinal cord impact injury. Can J Physiol Pharmacol 1982; 60: 1415–24.
26. Diaz-Ruiz A, Rios C, Duarte I, et al. Lipid peroxidation inhibition in spinal cord injury: cyclosporin-A vs methylprednisolone. Neuroreport 2000; 11: 1765–7.
27. Ducker TB. Treatment of spinal-cord injury [Editorial]. N Engl J Med 1990; 322: 1459–61.
28. Feinstein AR, Sosin DM, Wells CK. Stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. N Engl J Med 1985; 312: 1604–8.
29. Follmann D, Elliott P, Cutler J. Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol 1992; 45: 769–73.
30. Gabler C, Maier R. Clinical experiences and results of high-dosage methylprednisolone therapy in spinal cord trauma 1991 to 1993 [German]. Unfallchirurgie 1995; 21: 20–9.
31. Galandiuk S, Raque G, Appel S, et al. The two edged sword of large-dose steroids for spinal cord trauma. Ann Surg 1993; 218: 419–25.
32. Geisler FH. Sygen (GM-1) Acute SCI Study. 1999.
33. Geisler FH, Dorsey FC, Coleman WP. Past and current clinical studies with GM-1 ganglioside in acute spinal cord injury. Ann Emerg Med 1993; 22: 1041–7.
34. George ER, Scholten DJ, Buechler CM, et al. Failure of methylprednisolone to improve the outcome of spinal cord injuries. Am Surg 1995; 61: 659–63.
35. Gerhart KA, Johnson RL, Menconi J, et al. Utilization and effectiveness of methylprednisolone in a population-based sample of spinal cord injured persons. Paraplegia 1995; 33: 316–21.
36. Gerndt SJ, Rodriguez JL, Pawlik JW, et al. Consequences of high-dose steroid therapy for acute spinal cord injury. J Trauma Injury Infection Crit Care 1997; 42: 279–84.
37. Glasser RS, Knego RS, Delashaw JB, et al. The perioperative use of corticosteroids and bupivacaine in the management of lumbar disc disease. J Neurosurg 1993; 78: 383–7.
38. Grol R, Grimshaw J. Evidence-based implementation of evidence-based medicine. Jt Comm J Qual Improve 1999; 25: 503–13.
39. Guyatt GH, Sackett DL, Cook DJ (Evidence-Based Medicine Working Group). Users’ guides to the medical literature: II. How to use an article about therapy or prevention: B. What were the results and will they help me in caring for my patients? JAMA 1994; 271: 59–63.
40. Hall ED. The neuroprotective pharmacology of methylprednisolone. J Neurosurg 1992; 68: 462–5.
41. Hall ED. Neuroprotective actions of glucocorticoid and nonglucocorticoid steroids in acute neuronal injury. Cell Mol Neurobiol 1993; 13: 415–32.
42. Hall ED, Yonkers PA, Taylor BM, et al. Lack of effect of postinjury treatment with methylprednisolone or tirilazad mesylate on the increase in eicosanoid levels in the acutely injured cat spinal cord. J Neurotrauma 1995; 12: 245–56.
43. Hayashi M, Ueyama T, Senba E. Expression of neurotrophin and IL-1 beta mRNAs following spinal cord injury and the effects of methylprednisolone [in Japanese]. Kaibogaku Zasshi 1997; 72: 209–13.
44. Haywood RSA, Wilson MC, Tunis SR, et al. Users’ guide to the medical literature: VIII. How to use clinical practice guidelines: A. Are the recommendations valid? The Evidence-Based Medicine Working Group. JAMA 1995; 274: 570–4.
45. Heary RF, Vaccaro AR, Mesa JJ, et al. Steroids and gunshot wounds to the spine. Neurosurgery 1997; 41: 576–83.
46. Hurlbert RJ. Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. J Neurosurg (Spine 1) 2000; 93: 1–7.
47. Hurlbert RJ. Methylprednisolone and spinal cord injury. J Neurosurg (Spine 1) 2000; 93: 178–9.
48. Kiwerski J. Zastoswanie dexametazonu w leczeniu uszkodzen rdzenia kregowego we wczesnym okresie pourazowym. Neur Neurochir Pol 1992; 26: 518–27.
49. Lankhorst AJ, ter Laak MP, Hamers FP, et al. Combined treatment with alphaMSH and methylprednisolone fails to improve functional recovery after spinal injury in the rat. Brain Res 2000; 859: 334–40.
50. Levy ML, Gans W, Wijesinghe HS, et al. Use of methylprednisolone as an adjunct in the management of patients with penetrating spinal cord injury: outcome analysis. Neurosurgery 1996; 39: 1141–8.
51. Liu, McAdoo DJ. Methylprednisolone reduces excitatory amino acid release following experimental spinal cord injury. Brain Res 1993; 609: 293–7.
52. Marino RJ, Huang M, Knight P, et al. Assessing selfcare status in quadriplegia: comparison of the quadriplegia index of function (QIF) and the functional independence measure (FIM). Paraplegia 1993; 31: 225–33.
53. Mu X, Azbill RD, Springer JE. Riluzole and methylprednisolone combined treatment improves functional recovery in traumatic spinal cord injury. J Neurotrauma 2000; 17: 773–80.
54. Naso WB, Perot PL, Cox RD. The neuroprotective effect of high dose methylprednisolone in rat spinal cord hemisection. Neurosci Lett 1995; 189: 176–8.
55. Nesathurai S. Steroids and spinal cord injury: revisiting the NASCIS 2 and 3 trials. J Trauma Injury Infection Crit Care 1998; 45: 1088–93.
56. Otani K, Abe H, Kadoya S, et al. Beneficial effect of methylprednisolone sodium succinate in the treatment of acute spinal cord injury [in Japanese]. Sekitsui Sekizui 1994; 7: 633–47.
57. Oudega M, Vargas CG, Weber AB, et al. Long-term effects of methylprednisolone following transection of adult rat spinal cord. Eur J Neurosci 1999; 11: 2453–64.
58. Petitjean ME, Pointillart V, Dixmerias F, et al. Medical treatment of spinal cord injury in the acute stage [in French]. Ann Fr Anesth Reanim 1998; 17: 114–22.
59. Petitti DB. Meta-analysis, decision-analysis and cost-effectiveness analysis. In: Methods for Quantitative Synthesis in Medicine, 2nd ed. New York: Oxford University Press, 1994.
60. Pettersson K, Toolanen G. High dose methylprednisolone prevents extensive sick leave after whiplash injury: a prospective double blind randomized study. Spine 1998; 23: 984–9.
61. Pointillart V, Petitjean ME, Wiart L, et al. Pharmacological therapy of spinal cord injury during the acute phase. Spinal Cord 2000; 38: 71–6.
62. Prendergast MR, Saxe JM, Ledgerwood AM, et al. Massive steroids do not reduce the zone of injury after penetrating spinal cord injury. J Trauma Injury Infection Crit Care 1994; 37: 576–80.
63. Qian T, Campagnolo D, Kirshblum S. High-dose methylprednisolone may do more harm for spinal cord injury. Med Hypotheses 2000; 55: 452–3.
64. Sauerland S, Nagelschmidt M, Mallman P, et al. Risks and benefits of preoperative high dose methyprednisolone in surgical patients: a systematic review. Drug Saf 2000; 23: 449–61.
65. Segal JL, Maltby BF, Landorf MI, et al. Methylprednisolone disposition kinetics in patients with acute spinal cord injury. Pharmacotherapy 1998; 18: 16–22.
66. Shepard MJ, Bracken MB. The effect of methylprednisolone, naloxone, and spinal cord trauma on four liver enzymes: observations from NASCIS 2. National Acute Spinal Cord Injury Study. Paraplegia 1994; 32: 2236–45.
67. Short DJ, El Masry WS, Jones PW. High dose methylprednisolone in the management of acute spinal cord injury: a systematic review from a clinical perspective. Spinal Cord 2000; 38: 273–86.
68. Silverman WA. Where’s the Evidence: Debates in Modern Medicine. Oxford: Oxford University Press. 1998
69. Sun T, Xu S, Huang H. High methylprednisolone treatment in acute cervical spinal cord injury without fracture and dislocation [in Chinese]. Chung Hua Wai Ko Tsa Chih 1997; 35: 735–7.
70. Travlos A, Hirsch G. Steroid psychosis: a cause of confusion on the acute spinal cord injury unit. Arch Phys Med Rehabil 1993; 74: 312–5.
71. Welch RD, Lobley SJ, O’Sullivan SB, et al. Functional independence in quadriplegia: critical levels. Arch Phys Med Rehabil 1986; 67: 235–40.
72. Wilson MC, Hayward RSA, Tunis SR, et al. Users guide to the medical literature: VIII. How to use clinical practice guidelines: B. What are the recommendations and will they help you in caring for your patients? The Evidence-Based Medicine Working Group. JAMA 1995; 274: 1630–2.
73. Wing PC, Nance P, Connell DG, et al. Risk of avascular necrosis following short term megadose methylprednisolone treatment. Spinal Cord 1998; 36: 633–6.
74. Xu J, Fan G, Chen S, et al. Methylprednisolone inhibition of TNF-alpha expression and NF-kB activation after spinal cord injury in rats. Brain Res 1998; 59: 135–42.
75. Yates D, Farrell B, Teasdale G, et al. Corticosteroids in head injury: the CRASH trial. Accid Emerg Med 1999; 16: 83–4.

acute spinal cord injury; evidence-based medicine; methylprednisolone; randomized controlled trial; steroids; systematic review]Spine 2001;26:S47–S54

© 2001 Lippincott Williams & Wilkins, Inc.