Effectiveness of methylprednisolone in the acute phase of spinal cord injuries: a systematic review of randomized controlled trials

Abstracts

Steroid therapy has been tested as a protector in spinal cord injury. Multicenter studies evaluating the methylprednisolone (MP) in post traumatic neurological recovery have shown promising results according to NASCIS. A large number of critical studies related to the NASCIS results have been published. OBJECTIVE: To review literature related to use of methylprednisolone compared with placebo. METHODS: This analysis added the average improvement achieved in groups of patients who used MP and placebo (PL) to the average scores of groups at baseline, before treatment, resulting in the final neurological outcome for both groups. RESULTS: The motor score of the MP group was only 2.5 points higher than the PL in a one year follow-up. In neurologically intact patients, the total score is 70 points. Improvement in sensitive scores was also discrete (1.1 and 1.7 points for the pinprick and light touch respectively). A high rate of complications was observed in a group of patients about 60 years old who used MP. CONCLUSION: Differences in the clinical magnitude of benefit obtained (not confirmed by other studies) with the use of MP or PL are not significant, in comparison with the potential for complications when using methylprednisolone.

Methylprednisolone Hemisuccinate; Spinal cord injuries

A corticoterapia tem sido testada como protetor neuronal no trauma. Os estudos multicêntricos avaliando a metilprednisolona (MP) na recuperação neurológica após trauma revelaram-se promissores (NASCIS). Porém, várias, críticas às suas metodologias foram publicadas. OBJETIVO: Revisar a literatura relacionada ao uso da metilprednisolona comparada com placebo. MÉTODOS: A análise feita somou a média da melhora obtida nos grupos de pacientes que usaram MP e placebo (PL) à média dos escores dos grupos na linha de base, antes de tratamento, obtendo-se o resultado neurológico final de ambos os grupos. RESULTADOS: O escore motor do grupo MP foi apenas 2,5 pontos maior que o PL em um ano de seguimento. O escore motor em indivíduos intactos é de 70 pontos. A melhora nos escores sensitivos foi igualmente discreta (1,1 e 1,7 pontos para sensibilidade a picadas e tato respectivamente). A taxa de complicações observada em grupo de pacientes em torno de 60 anos foi alta com o uso de MP. CONCLUSÃO: As diferenças na magnitude do benefício clínico obtido (não confirmado por outros estudos) com o uso da MP e com o PL não são significativas em confronto com o potencial de complicações do uso da medicação.

Traumatismos da medula espinhal; Metilprednisolona; Hemissuccinato de metilprednisolona

ORIGINAL ARTICLE

Effectiveness of methylprednisolone in acute spinal cord injury - A systematic review of randomized controlled trials

Ricardo Vieira BotelhoI, ^*; Jefferson Walter Daniel^II; José Luis Romeo Boulosa^III; Benedicto Oscar Colli^IV; Ronald de Lucena Farias^V; Osmar José Santos Moraes^VI; Wilson Eloy Pimenta Jr.^VII; Carlos Henrique Ribeiro^VIII; Francisco Ricardo Borges Ribeiro^IX; Mario Augusto Taricco^X; Marcio Vinhal de Carvalho^XI; Wanderley Marques Bernardo^XII

^IMédico neurocirurgião; Orientador do programa de pós-graduação em Ciências da Saúde-IAMSPE; Doutor em Ciências pela Universidade Federal de São Paulo-Escola Paulista de Medicina - UNIFESP-EPM- Departamento de Psicobiologia; Coordenador do grupo de Cirurgia da Coluna Vertebral dos Serviços de Neurocirurgia do Hospital do Servidor Público Estadual Francisco Morato de Oliveira - HSPE-FMO e do Conjunto Hospitalar do Mandaqui, São Paulo, SP

^IIMédico - Assistente de Neurocirurgia e professor Instrutor da Faculdade de Ciências Médicas da Santa Casa de São Paulo, São Paulo, SP

^IIIDoutor em Neurocirurgia pela Faculdade de Medicina de Ribeirão Preto - USP e médico assistente do Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto - USP, Ribeirão Preto, SP

^IVProfessor Titular do Departamento de Cirurgia e Anatomia e Chefe da Disciplina de Neurocirurgia da Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo.

^VMédico neurocirurgião e professor de Neuroanatomia da Universidade Federal da Paraíba, João Pessoa, PB

^VIMédico - Chefe de serviço de neurocirurgia do Hospital Santa Marcelina, São Paulo, SP

^VIIMédico - Assistente do serviço de Cirurgia da Coluna Vertebral do departamento de Ortopedia e Traumatologia da Faculdade de Medicina da Universidade Federal de Goiás - DOT-FM/UFG, Goiânia, GO

^VIIIMestre em neurocirurgia pela Universidade Federal de São Paulo - Escola Paulista de Medicina - Unifesp-EPM; Chefe de serviço de Neurocirurgia do Hospital Municipal Salgado Filho, Rio de Janeiro-RJ

^IXMédico - Coordenador de departamento de coluna da Sociedade Brasileira de Neurocirurgia -SBN e membro da Coordenação da educação Neuro. Fundação AOSPINE, São Paulo, SP

^XProfessor livre-docente em Neurocirurgia e professor associado de disciplina de Neurocirurgia pela Faculdade de Medicina da Universidade de São Paulo - FMUSP e coordenador do Grupo de Coluna da Neurocirurgia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, SP

^XIMédico - Membro titular Sociedade Brasileira Neurocirurgia e membro do grupo coluna do Hospital de Base do Distrito Federal; Coordenador departamento Coluna da SBN e membro titular da Sociedade Brasileira de Coluna.

^XIIDoutor em Medicina - Coordenador técnico do Projeto Diretrizes da Associação Médica Brasileira e do Conselho Federal de Medicina; Professor de graduação e pós-graduação da Faculdade de Medicina da Universidade de São Paulo - FMUSP e Centro Universitário Lusíada - UNILUS, São Paulo, SP

ABSTRACT

Steroid therapy has been tested as a neuronal protector in spinal cord injury. Multicenter studies evaluating the efficacy of methylprednisolone (MP) in posttraumatic neurological recovery have shown promising results (NASCIS). However, several critical studies related to NASCIS results have been published.

OBJECTIVE: To review the literature concerning the use of methylprednisolone compared with placebo.

METHODS: This analysis added the mean improvement obtained by groups of patients using MP and placebo (PL) to their mean scores at baseline, before treatment, resulting in the final neurological outcome for both groups.

RESULTS: The motor score in the MP group was only 2.5 points higher than in the PL group after a one-year follow-up. In neurologically intact individuals, the motor score is 70 points. Improvement in sensory scores was also discrete (1.1 and 1.7 points for pinprick and light touch sensation, respectively). A high rate of complications was observed in a group of patients aged about 60 years who used MP.

CONCLUSION: Differences in the clinical magnitude of the benefits obtained (not confirmed by other studies) with the use of MP and PL are not significant against the potential complications when using this medication.

Key words: Spinal cord injuries. Methylprednisolone. Methylprednisolone hemisuccinate.

INTRODUCTION

Spinal trauma (ST) is the major cause of spinal cord injury damage in young adults. The resulting neurological damage is caused by primary mechanical lesion and, secondarily, by a series of subsequent cell-cell and biochemical interactions that perpetuate and amplify the initial lesion. Pharmacological treatment usually aims at reducing secondary injuries, being considered as a neuronal protector in trauma.

Steroid therapy has been tested as a neuronal protector in trauma. In 1984, a study conducted in seven U.S. states was published comparing two distinct methylprednisolone (MP) regimens in the acute phase of trauma: 100-mg bolus followed by daily 100 mg for 10 days compared with 1,000-mg bolus followed by daily 1,000 mg for 10 days.¹ The results after 12 months of study were published in 1985.² None of the therapy regimens showed differences in neurological protection (NASCIS I). New experimental studies suggested that the dosage in neuroprotective steroid therapy should be higher than that used until that moment.² Therefore, studies using high-dose MP were then designed.^3,4

Preliminary results described high-dose MP as efficient in the treatment of trauma in the United States; however, the publishing strategy for MP studies was viewed with some skepticism by part of the scientific community. The results were largely spread even before their initial publication, associated with advertising MP use in several trauma centers in the United States.⁵

Several critical studies regarding MP use have then been published by various Specialty Societies from different countries.^{6,7,8,9,10,11}

OBJECTIVES

The objective of this study is to review randomized controlled trials comparing the effectiveness of methylprednisolone with placebo in the acute phase of spinal cord injury. The authors aimed to answer the following clinical question: does the use of methylprednisolone bring benefits and/or risks to patients with spinal cord injury after spinal fracture?

METHODS

In order to guide our search for evidence within scientific information databases, the clinical question was structured according to the acronym PICO (P- Patients; I- Intervention; C- Comparison; O- Outcome). In addition to organizing the literature search, these components defined the inclusion criteria related to populations, interventions and outcomes to be considered.

Types of participants: Patients that suffered spinal trauma with spinal cord injury. Studies examining spinal trauma without spinal cord injury were excluded, as well as those analyzing victims of whiplash injury without neurological damage.

Types of intervention: Comparison between the use of methylprednisolone (regardless of dose) and placebo in the acute phase of ST with spinal cord injury.

Types of outcomes studied: a) Neurological status (motor and sensory functions) after spinal cord injury; b) Complications of MP therapy.

Conflict of interest: None declared.

Study designs included: In an attempt to retrieve studies with greater strength to answer the clinical question, only randomized controlled trials were included in the literature search.

Databases searched: MEDLINE, LILACS and EMBASE.

Reference review: All articles were reviewed by two independent reviewers.

Search strategy

A) MEDLINE database: A systematic review was conducted using PubMed web-based search engine (www.pubmed.com), with the question structured by "PICO" according to the descriptors below, applying the sensitive and specificity filters of PubMed Clinical Queries tool and MeSH Browser. Two distinct searches were performed for the effect of MP on SCI-associated neurological damage and complications of MP use (therapy) compared to placebo. The following descriptors were used:

1) Effect of MP on SCI-associated neurological damage

Search tool: Clinical Queries: ("Methylprednisolone")AND("Spinal Cord Injuries")AND(Randomized controlled trial [Publication Type]OR(randomized [Title/Abstract]ANDcontrolled [Title/Abstract]ANDtrial[Title/Abstract])) - 38.

Search tool: MeSH Browser: "Methylprednisolone" [Mesh]AND"Spinal Cord Injuries" [Mesh]AND("Randomized controlled trial" [Publication Type]OR"Randomized controlled trialas Topic[Mesh]) - 43.

2) Complications of MP therapy

Search tools: Clinical Queries and MeSH Browser: "Methylprednisolone"[Mesh] AND "complications"[Subheading] OR "adverse effects"[Subheading] AND "Randomized Controlled Trial[ptyp] AND "Spinal Cord Injuries"[Mesh] AND "Randomized Controlled Trial[ptyp] - 11.

B) LILACS database:

Search: "methylprednisolone" or "methylprednisolone hemisuccinate" or "methylprednisolone succinate" [Subject descriptor] and "spinal cord injuries" or "spinal cord trauma" [Subject descriptor] - 9.

C) EMBASE database:

Search: 1) "methylprednisolone" and"Spinal cord injuries" and"controlled clinical trial" (limits: human/lim and 1974-2008) - 19.2) "methylprednisolone" and "Spinal cord injuries" and randomized controlled trial/lim AND [humans]/lim AND [embase]/lim AND [1974-2008] - 4.

Among all authors, 11 were neurosurgeons. We consulted with the authors on whether any other scientific works, not uncovered by our web-based search, should be incorporated into the results of the present review, but no additional studies were suggested.

Critical appraisal and selection of studies

The studies were classified according to the Jadad score, from one to five. The Jadad score¹² assigns one point to each positive answer if the article describes the method of randomization and blinding to treatment and evaluates and describes withdrawals and dropouts in the study sample. Furthermore, an additional point is assigned if the method of randomization is appropriate, the method of blinding is appropriate, and withdrawals and dropouts are less than or equal to 20% of the initial sample. The total score may range from 0 to 5. Scores greater than or equal to three reveal a study with adequate methodological quality.

Studies scoring Jadad 3 or greater were included for critical appraisal according the checklist below.

The checklist analyzed methodological quality and internal validity performed by the authors considering: the study design, whose objective is a randomized controlled trial; the level of evidence for inclusion was established as 1b according to the Oxford table (http://www.projetodiretrizes.org.br/projeto_diretrizes/texto_introdutorio.pdf); inclusion criteria for selection (patients with spinal cord injury that were randomized within 12 hours of injury); exclusion criteria were nerve root or cauda equina injury, gunshot wound, pregnancy, life-threatening morbidity, drug-addicted persons, individuals using corticoids for any reason, aged less than 13 years, using 100 mg of methylprednisolone or 1 mg of naloxone, and those cases in which follow-up was considered difficult; description of the method of randomization and presence of blinded allocation; blinding of raters; therapy protocol consisting of an intervention group using methylprednisolone at any dose and/or administration regimen and a control group using placebo; participants were blinded to which drug was being administered; the outcomes considered included neurological function (as the main outcome), assessed between six weeks and one year after injury; neurological function was assessed concerning components of motor function, pain appreciation (pinprick), and superficial sensation (light touch); instruments to measure the outcome of interest (final standardized motor score and the sensory score related to pinprick-evoked pain and light touch); period of patient follow-up and description of losses; presence or not of migration between groups; sample size calculation and estimated differences between groups; significance level was set at p<0.05; presence of intention-to-treat analysis and a summary of the benefits and risks.

Statistical analysis: The difference between motor and sensory scores obtained from the sample at admission (baseline) and at the end of the follow-up (posttreatment) was assessed, the mean score being expressed as the proportion of the approximate normal score for whole numbers, using the chi-square test.

RESULTS

The three search strategies uncovered 72 scientific works. These studies were evaluated according to PICO inclusion and exclusion criteria and to study design.

1) Effect of MP on SCI-associated neurological damage

Evidence retrieved from PubMed - 1st quality analysis of published articles - Selection by study design

The literature review via PubMed (PICO) returned 41 articles: 10 review studies,^{5,6,8,10,13,15,16,17,18,20} 13 commentaries and discussion,^{7,9,11,19,21,23,24,25,26,27,28,29,30} two experimental animal studies,^29,30 three studies on the use of GM-1 ganglioside,^33,34,35 one article analyzing the role of surgery in neurological recovery,¹⁷ one article performing a parallel evaluation of the effect of MP on liver enzymes,³⁸ and one secondary analysis.³⁹ All of these studies were excluded. Ten articles were described as randomized clinical trials and were selected for analysis.^{1,2,3,4,40,41,42,43,44, 45}

These 10 randomized clinical trials were set apart for reanalysis:

Evidence retrieved from PubMed - 2nd quality analysis of articles - Critical appraisal of selected evidence

Articles excluded by PICO:

The articles by Bracken et al. published in 1984¹ and in 1985² were described as the National Acute Spinal Cord Injury Studies (NASCIS I) at their first publication and after a one-year follow-up. These two articles did not compare MP with placebo.

The articles by Petitjean et al.⁴³ and Pointillart et al.⁴¹ can be considered superposable, since they describe the same topics in English and French. These studies did not include a placebo control group, comparing the MP group only to a group without treatment. Similarly, the studies known as NASCIS III⁴² did not use placebo as an intervention control group and, therefore, were not analyzed.

The article by Epstein et al.⁴⁵ compared the effect of methylprednisolone, heparin and cimetidine on gastrointestinal bleeding in patients with spinal cord injuries. This study did no perform an appropriate randomization of subgroups and, therefore, was excluded (Jadad score <3).

Studies included with a Jadad score greater than three:

Only two articles compared methylprednisolone with placebo in the outcome neurological improvement six weeks and six months after injury (NASCIS II³) and one year after injury (NASCIS II⁴). NASCIS II³ was described by means of two publications: one article describing the results after six weeks and six months and another article describing the results after one year.⁴ Of the initial sample of 487 patients, 97% of the participants were assessed one year after injury, and 87% of the sample received the predetermined pharmacological therapy.

NASCIS II generated randomization lists by computer and described the method of randomization, blinded allocation to treatment, analysis, and withdrawals and dropouts. There were 20% of losses, that is, the withdrawal/dropout threshold for final analysis. The study was classified as Jadad score 5, being then analyzed. Sample size was not calculated prior to the study. There was no migration between treatment groups. There was no temporary evaluation of possible unacceptable injuries in either treatment group. Intention-to-treat was not analyzed, although the injured-patient group as a whole was compared to the placebo group (with no differences between treatments) and a subgroup analysis of patients treated within eight hours of injury was performed.

2) Studies on complications of MP therapy

Evidence retrieved - 1st quality analysis of articles - Selection according to type of publication

The articles by Bracken 1984,¹ 1991,³⁰ and 1993²⁶ did not use a placebo control group¹ or were classified as a discussion; these articles were, therefore, excluded. The articles by Petitjean et al.,⁴³ Bracken 1997,⁴⁴ Pointillart et al.,⁴⁰ and Bracken 1998⁴² were not controlled against placebo, being, therefore, excluded. The study by Epstein et al.⁴⁵ was not actually randomized, being, therefore, excluded.

Evidence retrieved - 2nd quality analysis of published articles

Of the 11 trials initially selected, two were randomized clinical trials and were set apart for reanalysis. Only the study by Matsumoto et al.⁴⁰ had not been analyzed in the selection of works related to MP effect on bone marrow damage. Both NASCIS II^3,4 and the study by Matsumoto et al.⁴⁰ were analyzed for complications. The quality analysis of both NASCIS II was described above.

The study by Matsumoto et al.⁴⁰ was randomized, did not calculate sample size and did not show an intention-to-treat analysis. There was no migration between treatment groups and no withdrawals. The article was classified as Jadad score 5.

Evidence retrieved from LILACS - 1st quality analysis of published articles - Selection by study design

LILACS database search returned nine articles: six review studies,^{46,47,48,49,50,51} one experimental animal study,⁵⁵ one case series analysis,⁵³ and one case report.⁵⁴ LILACS database search did not uncover any randomized study.

Evidence retrieved from EMBASE - 1st quality analysis of published articles - Selection by study design

EMBASE database search returned 16 articles, in addition to those obtained from MEDLINE database: the study by Cengiz⁵⁵ analyzed the effect of timing of surgery on trauma, 13 articles were review studies,^{56,57,58,61,60,61,62,63,64,65,66,67,68} the study by Pettersson⁶⁹ analyzed whiplash injury, and the article by Trembly⁷⁰ was classified as a conference paper.

RESULTS

1) Effect of MP on SCI-associated neurological damage

A) Results for motor function

Motor function was assessed according to an expanded score, ranging from 0 to 70 points (ranging between no muscle contraction and all normal responses, tested in 14 muscle groups). The results for motor function were described as the magnitude of change in neurological status after injury due to treatment, instead of being described as the patient's pre and posttreatment absolute scores. A cutoff point for clinical benefit was not defined. In this study, the mean motor score for the whole study sample at admission (baseline) was compared to the final score of patients treated with methylprednisolone and placebo. The results were initially described for the whole sample and, subsequently, for the subgroup treated within eight hours of injury.

Taking into consideration the whole sample of patients using methylprednisolone, regardless of the duration of MP administration, there was no difference in motor recovery between the MP group and the placebo group. In addition, sensory changes did not reach the significance level set for this study. Similarly, patients treated after eight hours of injury did not show differences in neurological outcomes after treatment.

Analysis of motor improvement after six weeks:

Table 1 summarizes the improvement in motor function in NASCIS II. The subgroup treated within eight hours of injury accounted for 37.9% of the total study sample (185/487).

Patients receiving methylprednisolone (MP) had an initial motor score of 23.7 points, whereas those receiving placebo (PL) had a score of 24 points. After six weeks, the MP group improved 10.6 and the PL group 7.2 points. When final motor scores were compared (initial score plus improvement), MP score corresponded to 34.3 and PL score to 31.2 points of the total 70 points, with a difference of 3.1 points (4.4% of the total 70 points for patients without neurological deficit).

Analysis of motor improvement after six months:

Patients receiving methylprednisolone (MP) had an initial motor score of 23.7 points, whereas those receiving placebo (PL) had a score of 24 points. After six months, the MP group improved 16 and the PL group 11.2 points. When final motor scores were compared (initial score plus improvement), MP score corresponded to 39.7 and PL score to 35.2 points of the total 70 points, with a difference of 4.5 points.

Analysis of motor improvement after one year:

The evaluation of the whole sample of randomized patients, regardless of the start date of treatment, did not show any differences between MP and placebo after one year in either of the three neurological functions (motor function, pinprick response, and touch sensation).

Of the total patients in the initial group receiving MP and PL, 37.5% were analyzed in the subgroup that received medication within eight hours of injury (183/487). Motor function improved 17 points in the MP group and 12 points in the PL group (difference of 5 points in improvement). The difference in the final neurological status of the MP group in comparison with the PL group was of 2.5 points (38.3-35.5), 3.57% of the total 70 points.

B) Results for sensory function

In the six-week analysis, the mean score for pinprick response at baseline was 53 points (60.9% of the total points) in the MP group and 54.4 points (62.5%) in the PL group, with a difference of 1.4 points (1.6%). The MP group improved 8.9 points (group treated within 8 hours of injury) and the PL group improved 4 points, with a difference of 5.9 points. The final score was 61.9 points in the MP group and 58.4 points in the PL group, with a final difference of 3.5 points (4% of the total points) (chi-square test; p=0.80). Regarding touch sensation after six weeks, the MP group scored 54.3 points (62.4%) and the PL group scored 55.7 points (64%) at baseline, with a difference of 1.4 points. The MP group improved 7.1 points and the PL group improved 4 points, with a difference of 3.1 points. The final score was 61.4 points (70%) in the MP group and 59.7 points (68%) in the PL group, with a difference of 1.7 points (1.9%).

Tables 2 and 3 describe the improvement in sensory scores after six months and one year, respectively, and follow the same design of the above-mentioned table for motor improvement, showing initial (baseline) scores and improvement at the end of each study period. The difference in final sensory scores for pain appreciation (pinprick) between groups was 3.5, 5.6 and 1.1 points after six weeks, six months and one year, respectively. Regarding superficial sensation (light touch), the difference for the same study periods was 1.7, 3.8 and 1.7 points, respectively.

2) Results- Studies on complications of MP therapy:

NASCIS II: NASCIS II reported no statistically significant differences between complications in the MP and PL groups, but data were not available for reanalysis.

Study by Matsumoto et al.: MP and PL groups were composed of 23 patients each. The percentage of complications between the MP and PL groups was 56.5 vs. 34.8 (95%CI 0.498-0.064), with no significant difference. When complications were analyzed according to type, the occurrence of pulmonary and gastrointestinal complications was higher in the MP group than in the PL group. There was a trend toward bronchopneumonia (BCP) in the group of patients older than 60 years. There were eight BCP cases in the MP group and one case in the PL group. Probability of BCP in the MP group was 52.2 vs. 4.3% in the PL group. Absolute BCP risk in the MP group was 34.7 vs. 4.3% in the PL group. Increased absolute BCP risk due to use of MP was 47.9% (0.25-0.69), and the necessary number for the occurrence of BCP was two patients. It is worth mentioning that, in that study, the patients' mean age was about 60 years.

Regarding gastrointestinal bleeding (GIB), there were four cases in the MP group and none in the PL group (odds of 17.4 and 0%, respectively). Increased absolute GIB risk due to use of MP was 17.4% (95%CI 0.01-0.32), and the necessary number for the occurrence of damage was six patients.

DISCUSSION

Critical appraisal

Studies using a placebo control group:

Although there is substantial literature on MP use, the number of randomized studies comparing MP with placebo in neurological protection is restricted to two studies (NASCIS II³ and Matsumoto et al.⁴⁰). NASCIS II was described in two publications six months and one year after injury.^3,4.

The studies by Pointillart et al.⁴¹ and Petitjean et al.⁴³ were written in English and French, respectively, and have the same results, describing the same study in two different languages and being, therefore, considered as one single study. The study by Pointillart et al.⁴¹ did not use a placebo control group and was, therefore, excluded.

Report of results in NASCIS II:

NASCIS II³ described the results as the amount of change from baseline neurological examination, instead of measuring neurological status at baseline and at the final assessment. In baseline (pretreatment) groups, there were less neurologically intact patients in the MP group than in the PL group (5 vs. 8.8%). This(ese) study(ies) did not show differences between MP and PL groups. A neurological function improvement was observed only when the subgroup of patients treated within eight hours of injury was analyzed, which accounts only for 38% of the total sample. This effect represented a difference of 3.1 points in neurological improvement between MP and PL groups concerning motor function after six weeks, 4.5 points after six months, and 2.5 points after one year. In neurologically intact individuals, the motor score is 70 points.

No improvement was observed in patients treated with MP after eight hours of injury and in patients with complete motor injury (plegic patients) with sensory preservation (Frankell B) at any moment. The difference revealed to be too small to express a real clinically significant benefit. We attempted to describe the odds ratio for improvement in functional classification, but the original data were not available and the 95% confidence interval (CI) was too wide (-2.9 to 25.5). Studies with wide 95%CI have been classified as level of evidence II.

Subgroup analysis:

Findings based on subgroup analyses should not describe results, but rather formulate hypotheses. NASCIS did not provide the results as absolute data, hindering recalculation of risk increase or decrease and of the necessary number of treatment patients to show benefits or risks.

Definition of the amount of clinically significant improvement:

There is no such a definition in NASCIS II of a cutoff point, which prevented us from defining whether there was presence or absence of clinical improvement. The analysis of results from subgroups without previous sample size calculation yields the possibility of producing results caused by random effects.

The results were obtained from a sample less than 40% of the original study sample.

Both NASCIS II were not associated with a functional independence scale that could reveal an improvement in the patient's functional status. This limitation was corrected in NASCIS III. The functional independence measure (FIM) was used; however, a placebo control group was not used. The real benefit of the increase of 2.5 points in the final motor score one year after injury seems to be clinically insignificant.

Displaying the results as changes from baseline neurological status, before performing an analysis of treatment-induced neurological status in heterogeneous samples, may have contributed to the results.

Analysis model used in the present review:

The method chosen to quantify a possible difference between groups included adding the patient's improvement score to the mean score for groups at baseline, resulting in the final neurological outcome for both groups. Similar to motor scores, improvement in sensory scores was discrete.

Real clinical benefits of MP therapy:

The main discussion about these results refers to the real clinical significance of the difference in the improvement obtained with the use of MP. The results from NASCIS II were not reproduced, and the original data have never been made available for reanalysis.

The study by Matsumoto et al.⁴⁰ aimed to investigate complications resulting from MP therapy by means of a randomized, double-blind study comparing complications of MP use with placebo, in patients treated within 8 hours of injury. Mean age was high (60.6 years [18-84]). Inclusion criteria were the same as those used in NASCIS II. The study included only patients treated without surgery for SCI in the cervical spine, with predominantly centro-medullary neurological damage. The use of medication increased the risk of pulmonary and gastrointestinal complications in that study sample.

CONCLUSION

Due to the modest differences found in the treatments, the results do not suggest clinical benefits.

The use of MP is associated with an increased risk of pulmonary complications and gastrointestinal bleeding in patients aged about 60 years.

SUMMARY OF EVIDENCE

The differences between motor and sensory final scores of patients in the MP and PL groups were not significant and were minimal in relation to the maximum normal score possible, thus failing to suggest clinical benefits. In addition, the use of MP may be associated with an increased risk of pulmonary complications and gastrointestinal bleeding.

FINANCIAL SUPPORT:

This study received partial financial support from the Brazilian Society of Neurosurgery and Congresso de Cirurgia Espinhal.

REFERENCES

1. Bracken MB, Collins WF, Freeman DF, Shepard MJ, Wagner FW, Silten RM, et al. Efficacy of methylprednisolone in acute spinal cord injury. JAMA. 1984;251(1):45-52.

2. Bracken MB, Shepard MJ, Hellenbrand KG, Collins WF, Leo LS, Freeman DF, et al. Methylprednisolone and neurological function 1 year after spinal cord injury. Results of the National Acute Spinal Cord Injury Study. J Neurosurg. 3. Bracken MB, Shepard MJ, Collins WF, Holford TR, Young W, Baskin DS, et al. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med. 1990;322(20):1405-11.

4. Bracken MB. Pharmacological treatment of acute spinal cord injury: current status and future prospects. Paraplegia. 1992;30(2):102-7.

5. Hurlbert RJ. Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. J Neurosurg. 2000;93(1 Suppl):1-7.

6. Kronvall E, Sayer FT, Nilsson OG. Methylprednisolone in the treatment of acute spinal cord injury has become more and more questioned. Lakartidningen. 2005;102(1):24-5.

7. Hugenholtz H. Methylprednisolone for acute spinal cord injury: not a standard of care. CMAJ. 2003;168(9):1145-6.

8. Citerio G, Cormio M, Sganzerla EP. Steroids in acute spinal cord injury. An unproven standard of care. Minerva Anestesiol. 2002;68(5):315-20.

9. Short D, El Masry WS, Jones PW. High dose methyprednisolone in the management of acute spinal cord injury: a systematic review from a clinical perspective. Spinal Cord. 2000;38(5):273-86.

10. Short D. Is the role of steroids in acute spinal cord injury now resolved? Curr Opin Neurol. 2001;14(6):759-63.

11. Nesathurai S. Steroids and spinal cord injury: revisiting the NASCIS II and NASCIS 3 trials. J Trauma. 1998;45(6):1088-93.

12. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17(1):1-12.

13. Nesathurai S. The role of methylprednisolone in acute spinal cord injuries. J Trauma. 2001;51(2):421-3.

14. Coleman WP, Benzel D, Cahill DW, Ducker T, Geisler F, Green B, et al. A critical appraisal of the reporting of the National Acute Spinal Cord Injury Studies (II and III) of methylprednisolone in acute spinal cord injury. J Spinal Disord. 2000;13(3):185-99.

15. Miller SM. Methylprednisolone in acute spinal cord injury: a tarnished standard. J Neurosurg Anesthesiol. 2008;20(2):140-2.

16. Rozet I. Methylprednisolone in acute spinal cord injury: is there any other ethical choice? J Neurosurg Anesthesiol. 2008;20(2):137-9.

17. Zeidman SM, Ling GS, Ducker TB, Ellenbogen RG. Clinical applications of pharmacologic therapies for spinal cord injury. J Spinal Disord. 1996;9(5):367-80.

18. Ducker TB. Treatment of spinal-cord injury.N Engl J Med. 1990;322(20):1459-61.

19. Ducker TB, Zeidman SM. Spinal cord injury. Role of steroid therapy. Spine. 1994;19(20):2281-7.

20. Young W, Bracken MB. The Second National Acute Spinal Cord Injury Study. J Neurotrauma. 1992;9(Suppl 1):S397-405.

21. Young W. Secondary injury mechanisms in acute spinal cord injury. J Emerg Med. 1993;11(Suppl 1):13-22.

22. Bracken MB, Aldrich EF, Herr DL, Hitchon PW, Holford TR, Marshall LF, et al. Clinical measurement, statistical analysis, and risk-benefit: controversies from trials of spinal injury. J Trauma. 2000;48(3):558-61.

23. Bracken MB. Methylprednisolone and spinal cord injury. J Neurosurg. 2000;93(1 Suppl):175-9.

24. Savitsky E. Role of glucocorticosteroids in treatment of acute spinal cord injury. West J Med. 1996;164(1):69-70.

25. Rosner MJ. Treatment of spinal cord injury.J Neurosurg. 1994;80(5):954-5.

26. Bracken MB. Pharmacological treatment of acute spinal cord injury: current status and future projects. J Emerg Med. 1993;11(Suppl 1):43-8.

27. Hanigan WC, Anderson RJ. Commentary on NASCIS-2. J Spinal Disord. 1992;5(1):125-31.

28. Rosner MJ. National acute spinal cord injury study of methylprednisolone or naloxone. Neurosurgery. 1991;28(4):628-9.

29. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. N Engl J Med. 1990;323(17):1207-9.

30. Bracken MB. Treatment of acute spinal cord injury with methylprednisolone: results of a multicenter, randomized clinical trial. J Neurotrauma. 1991;8(Suppl 1):S47-50.

31. Borgens RB, Toombs JP, Breur G, Widmer WR, Waters D, Harbath AM, et al. An imposed oscillating electrical field improves the recovery of function in neurologically complete paraplegic dogs. J Neurotrauma. 1999;16(7):639-57.

32. Perez-Espejo MA, Haghighi SS, Adelstein EH, Madsen R. The effects of taxol, methylprednisolone, and 4-aminopyridine in compressive spinal cord injury: a qualitative experimental study. Surg Neurol. 1996;46(4):350-7.

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(6):1041-7.

34. Geisler FH. GM-1 ganglioside and motor recovery following human spinal cord injury. J Emerg Med. 1993;11(Suppl 1):49-55.

35. Geisler FH. Clinical trials of pharmacotherapy for spinal cord injury. Ann N Y Acad Sci. 1998;19(845):374-81.

36. Bracken MB, Holford TR. Effects of timing of methylprednisolone or naloxone administration on recovery of segmental and long-tract neurological function in NASCIS II. J Neurosurg. 1993;79(4):500-7.

37. Duh MS, Shepard MJ, Wilberger JE, Bracken MB. The effectiveness of surgery on the treatment of acute spinal cord injury and its relation to pharmacological treatment. Neurosurgery. 1994;35(2):240-8;

38. Shepard MJ, Bracken MB. The effect of methylprednisolone, naloxone, and spinal cord trauma on four liver enzymes: observations from NASCIS II. National Acute Spinal Cord Injury Study. Paraplegia. 1994;32(4):236-45.

39. Duh M-S, Shepard MJ, Wilberger JE, Bracken MB, Tator CH, Marshall LF.The effectiveness of surgery on the treatment of acute spinal cord injury and its relation to pharmacological treatment. Neurosurgery. 1994;35(2):240-9.

40. Matsumoto T, Tamaki T, Kawakami M, Yoshida M, Ando M, Yamada H. Early complications of high-dose methylprednisolone sodium succinate treatment in the follow-up of acute cervical spinal cord injury. Spine. 2001;26(4):426-30.

41. Pointillart V, Petitjean ME, Wiart L, Vital JM, Lassié P, Thicoipé M, et al. Pharmacological therapy of spinal cord injury during the acute phase. Spinal Cord. 2000;38(2):71-6.

42. Bracken MB, Shepard MJ, Holford TR, Leo-Summers L, Aldrich EF, Fazl M, 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;89(5):699-706.

43. Petitjean ME, Pointillart V, Dixmerias F, Wiart L, Sztark F, Lassié P, et al. Medical treatment of spinal cord injury in the acute stage. Ann Fr Anesth Reanim. 1998;17(2):114-22.

44. 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. National Acute Spinal Cord Injury Study. JAMA. 1997;277(20):1597-604.

45. Bracken MB. Pharmacological treatment of acute spinal cord injury: current status and future prospects. Paraplegia. 1992;30(2):102-7.

46. Bracken MB, Shepard MJ, Hellenbrand KG, Collins WF, Leo LS, Freeman DF, 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(5):704-13.

47. Bracken MB, Collins WF, Freeman DF, Shepard MJ, Wagner FW, Silten RM, et al. Efficacy of methylprednisolone in acute spinal cord injury. JAMA. 1984;251(1):45-52.

48. Epstein N, Hood DC, Ransohoff J. Gastrointestinal bleeding in patients with spinal cord trauma. Effects ofsteroids, cimetidine, and mini-dose heparin. J Neurosurg. 1981;54(1):16-20.

49. Postigo TR. Metilprednisolona en el tratamiento del trauma2006 raquimedular: analisis de la evidencia. Methylprednisolone in the treatment of the spinal cord injury: analysis of the evidence. Rev Med Clin Condes. 2006;17(1):12-9.

50. Barros Filho TEP. Tratamento medicamentoso no traumatismo raquimedular/Pharmacological management of spinal cord injury. Rev Bras Ortop. 2000;35(5):143-6.

51. Gianchino LM. Consideraciones anestesicas en el paciente con traumatismo raquimedular. Rev Argent Anestesiol.1998;56 (4): 274-80.

52. Gerbrin AS, Cristante AF, Marcon RM, Da-Silva CF. Intervenções farmacológicas no trauma raquimedular: uma nova visão terapêutica. Pharmacological interventions in spinal cord trauma: a therapeutic approach. Acta Ortop Bras.1997;5(3):123-36, .

53. San Roman E, Peralta H, Gruenberg M. Trauma de raquis/Spine trauma. Med Intens. 1994;11(2):23-6.

54. Zambrano D. Recientes adelantos en la evalualuacion y tratamiento del traumatismo craneoencefalico-espinal severo. Prensa Med Argent. 1982;69(14):575-81.

55. Tebet MA, Barros Filho, TEP, Machado IR, Carvalho, MOP, Hanania FR, Daci K. Efeito da metilprednisolona na lesão medular em ratos: analise functional e histological. Effect of methylprednisolone in medullar injury in rats: a functional and histological analysis. Acta Ortop Bras. 2003;11(2):80-7.

56. Pérez Berrios J, Jiménez Garcia R. Analises de 76 casos de trauma espinal agudo en el Hospital San Juan de Dios.1995-1997. Acta Méd Costarica. 1999;41(2):32-8.

57. Evora PRB, Alegre EMP, Freitas JNA, Nakao RMS, Conceição MC. Edema pulmonary unilateral após trauma medular com comprometimento contralateral do sistema nervosa sintomático J Pneumol.1985;11(4):206-10.

58. Cengiz SL, Kalkan E, Bayir A, Ilik K, Basefer A. Timing of thoracolomber spine stabilization in trauma patients; impact on neurological outcome and clinical course. A real prospective (rct) randomized controlled study. Arch Orthop Trauma Surg. 2008;128(9):959-66.

59. Koyanagi I, Murakami T, Houkin K. Management of acute cervical spine and spinal cord injury using clinical guidelines. Jap J Neurosurg. 2007;16(1):18-25.

60. Martiñon S, Ibarra A. Pharmacological neuroprotective therapy for acute spinal cord injury:Stage of the art. Mini-reviews. Med Chem. 2008;8(3):222-30.

61. Adibhatla RM, Hatcher JF. Role of lipids in brain injury and diseases. Future Lidology. 2007;2(4):403-22.

62. Ewart MC, Marshalll RW. Perioperaative management of the unstable cervical spine. Care Critically Ill. 2006;22(4):82-90.

63. Klussmann S, Martin-Villalba A. Molecular targets in spina cord injury. J Mol Med. 2005;83(9):657-71.

64. Bossche LV, Vanderstraten G. Heterotopic ossification: a review. J Rehabil Med. 2005;37(3):129-36.

65. Bernhard M, Gries A, Kremer P, Martin-Villalba A, Bottiger BW. Prehospital management of spinal cord injuries. Anaesthesist. 2005;54(4):357-76.

66. Eck JC, Hodges SD, Humphreys SC. Whiplash: a review of a commonly misunderstood injury. Am J Med. 2001;110(8):651-6.

67. Gruen JP, Weiss M. Management of complicated neurologic injuries. Surg Clin North Am. 1996;76(4):905-22.

68. Faden AI. Pharmacological treatment of central nervous system trauma. Pharmacol Toxicol. 1996;78(1):12-7.

69. Cronstein BN. Clinical use of methylprednisolone sodum succinate: a review. Curr Ther Res Clin Exp. 1995;56(1):1-15.

70. Cortez R, Levi AD. Acute spinal cord injury. Curr Treat Options Neurol. 2007;9(2):115-12.

71. Ramer LM, Ramer MS, Steeves JD. Setting the stage for functional repair of spinal cord injuries: A cast of thousands. Spinal Cord. 2005;43(3):134-61.

72. Pettersson K, Toolanen G. High-dose methylprednisolone prevents extensive sick leave after whiplash injury. Spine. 1998;23(9):984-9.

73. Trembly B. Clinical potential for the use of neuroprotective agents. A brief overview. Ann Academy Sci. 1995;765(1):1-20.

Artigo recebido: 28/02/09

Aceito para publicação:11/08/09

Conflict of interest: No conflicts of interest declared concerning the publication of this article.

This study received research grant support from the Spinal Department of the Brazilian Society of Neurosurgery and was supported by the Brazilian Medical Association (Guidelines Project) and Congresso de Cirurgia Espinhal.

*

Correspondence: Rua Haberbeck Brandão, nº 68 - apt 92 - Indianópolis. Tel.: [phone]-Fax: [phone]. São Paulo, SP, Brazil. CEP: 4027040

Publication Dates

• Publication in this collection
  25 Feb 2010
• Date of issue
  2009

History

• Accepted
  11 Aug 2009
• Received
  28 Feb 2009