Background
Traumatic brain injury is a leading cause of premature death and disability. Road crashes account for the majority of fatal head injuries (Jennett 1996). Although road death rates are falling in most industrialised countries, in the rapidly motorising Asian countries they are rising, and will almost certainly continue to do so. Road death rates per head in China are already similar to those in the United States, in spite of the fact that there are only five vehicles per 1,000 population in China, compared with 770 vehicles per 1,000 population in the US (Roberts 1995). Overall, about 75% of the estimated 850,000 road crash deaths each year occur in the developing world (Murray 1994).
In the US, the incidence of brain injury related disability is estimated to be 33 new cases/100,000 people per year (Kraus 1993). Since this often occurs in young people and is long term, traumatic brain injury related disability is a major cause of ill health worldwide.
In 1961 Galicich and French reported rapid and significant improvement in response to corticosteroids in 28 of 34 people with cerebral oedema either due to brain tumours, or post-operative (Galicich 1961). This led to their use in other intracranial problems characterized by raised intracranial pressure, including their use in severe head injury (Pickard 1993). Eighty percent of patients with fatal head injuries show evidence of increased intracranial pressure at necropsy (Miller 1992).
For a problem as common as brain injury, even a moderate reduction in mortality or disability from an intervention as widely practicable as corticosteroids would be important. There have been a number of randomised controlled trials of corticosteroids in head injury with apparently conflicting findings. Continuing uncertainty about the effects of corticosteroids for this indication is reflected in substantial variation in their use. A recent UK study found that corticosteroids were used in just under half of the intensive care units surveyed (Jeevaratnam 1996).
Objectives
- To quantify the effectiveness of corticosteroids in reducing mortality and morbidity in people with acute traumatic brain injury.
- To quantify the incidence of side effects of the use of corticosteroids.
- To quantify the economic effects of corticosteroid use in this situation.
Methods
Criteria for considering studies for this review
Types of studies
We sought to identify all randomised controlled trials of a corticosteroid drug versus any control in the treatment of acute traumatic brain injury. Studies using a quasi random form of allocation were excluded from the review.
Types of participants
People of all ages with clinically diagnosed acute traumatic brain injury secondary to head injury who were treated with steroids or control within seven days of the injury. All severities of head injury were included.
Types of interventions
The experimental intervention was corticosteroids (those steroids with predominantly glucocorticoid effects, namely prednisolone, betamethasone, cortisone, dexamethasone, hydrocortisone, methylprednisolone, prednisone and triamcinolone) administered in any dose by any route for any duration started within seven days of the injury. Trials with these interventions were included irrespective of other treatments used.
Types of outcome measures
All causes of case fatality, any valid and reliable measure of neurological functioning, any other valid and reliable quality of life measures and economic outcomes were considered relevant if available. We sought numbers of infections (however defined) and significant gastrointestinal bleeds (however defined).
Search methods for identification of studies
The searches were not restricted by date, language or publication status.
Electronic searches
We searched the following databases:
- CENTRAL (The Cochrane Library 2007, Issue 4);
- MEDLINE (Ovid SP) 1950 to Nov (week 2) 2007;
- PubMed [www.ncbi.nlm.nih.gov/sites/entrez/] (searched 7 Jan 2008: added to PubMed in the last 60 days);
- EMBASE (Ovid SP) 1980 to (week 1) Jan 2008;
- PsycINFO (Ovid SP) 1806 to April 2007.
The search strategies used for previous versions of this review can be found in Appendix 1. The strategies used for this update can be found in Appendix 2.
Searching other resources
We also searched specialised databases, handsearched journals and contacted trialists.
Data collection and analysis
We each extracted the following information independently from each trial: strategy for allocation concealment, number of randomised patients, duration of follow up and number lost to follow up. The major outcome data sought were numbers of deaths and numbers of people disabled at the end of the study period, using the Glasgow Outcome Scale (Jennett 1975) to assess the neurological outcome; the categories for persistent vegetative state and moderate disability were combined into 'disability' for this review. This enabled inclusion of the one trial which did not use the Glasgow Outcome Scale but a similar ordinal categorisation of function. We also extracted data on side effects or complications where these were reported, using the authors' definitions of these complications.
Since there is evidence that the quality of allocation concealment particularly affects the results of studies (Higgins 2008), each of us scored this quality on the scale used by Higgins (Higgins 2008) as shown below, assigning 'No' to poorest quality and 'Yes' to best quality:
- No = trials in which concealment was inadequate (such as alternation or reference to case record numbers or to dates of birth);
- Unclear = trials in which the authors either did not report an allocation concealment approach at all or reported an approach that did not fall into one of the other categories;
- Yes = trials deemed to have taken adequate measures to conceal allocation (i.e. central randomizations; numbered or coded bottles or containers; drugs prepared by the pharmacy; serially numbered, opaque, sealed envelopes; or other description that contained elements convincing of concealment).
If the method used to conceal allocation was not clearly reported, we contacted the author whenever possible, for clarification. We then compared the scores allocated and resolved differences by discussion.
We calculated relative risks and 95% confidence intervals for mortality for each trial on an intention to treat basis. Heterogeneity between trials was tested using a chi-squared test, where P less than or equal to 0.05 was taken to indicate significant heterogeneity. As long as statistical heterogeneity did not exist, for dichotomous data, we calculated summary relative risks and 95% confidence intervals using a fixed-effect model.
2004 update
For the October 2004 update of the review, the Cochrane Injuries Group staff searched the Group's register for more trials but found none. In October 2004, the initial results of the CRASH trial, previously listed as an ongoing study, were published. Since Ian Roberts was a principal investigator on this trial, Phil Alderson extracted data and updated the review, with Ian Roberts checking for correctness.
The inclusion of the CRASH trial introduced significant heterogeneity, and the original review's methods section did not clearly specify how heterogeneity would be investigated. It was therefore decided that the only investigation of heterogeneity would be to undertake a sensitivity analysis, removing those trials with less than adequate allocation concealment. If that failed to remove heterogeneity, the trials would not be pooled in the review, and reasons for heterogeneity suggested but not examined formally.
2006 update
Searches were repeated by the Cochrane Injuries Group in November 2005. The final results of CRASH are now available and have been included. No other new data were identified by the searches. Phil Alderson added the new data and updated the text of the review. As the results of CRASH dominate the other trial results for death or severe disability, the results were not pooled.
2009 update
The search was updated by the Cochrane Injuries Group in January 2008. No new trials were found; the results and conclusions remain the same.
Results
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies.
The combined search strategies identified 19 trials which satisfied the inclusion criteria. The earliest was from 1972 and the most recent from 1995. There were two reports of the same trial (see Fanconi 1988). Two were previously unpublished studies for which outcome data were obtained (Hernesniemi 1979; Pitts 1980). For one other unpublished study (Tahara 1972) the authors were unable to provide outcome data, and in one other we have been unable to trace the author (Hoyt 1972). The 2004 update added one more trial (CRASH 2005), which is the largest of all the trials in the review.
A total of 11,792 participants are included in outcome of death.
Risk of bias in included studies
Methodological quality was variable − see table 'Characteristics of included studies' for details.
Effects of interventions
The effect of corticosteroids on the risk of death was reported in 17 included trials. There was significant heterogeneity for this outcome when using a fixed-effect risk ratio model (Chi
The CRASH trial reported a relative risk of 1.05 (95%CI 0.99 to 1.10) for death or severe disability. There was no significant heterogeneity between the results from the 10 trials reporting this outcome, but it was decided not to pool the results as the data from CRASH dominate the other results.
Five trials reported infections, and the pooled relative risk was 1.03 (95% CI 0.99 to 1.07) making a decrease in infectious complications unlikely with steroids. For the ten trials reporting gastrointestinal bleeding the pooled relative risk was 1.23 (95% CI 0.91 to 1.67) which neither confirms nor excludes an important increase or decrease in this complication. For both these outcomes, data on about 2% of the CRASH trial participants were missing, and these patients have been excluded from the analysis. Sensitivity analysis was not undertaken due to the low proportion and because it had not been prespecified.
No study included economic data.
Discussion
This systematic review summarises the evidence from randomised controlled trials of corticosteroids in acute traumatic brain injury.
Methodological issues
The inclusion of an EMBASE search identified one study not found on MEDLINE (Tahara 1972). Contact with trialists enabled us to include data from two large unpublished studies (Hernesniemi 1979; Pitts 1980), but not from others (Hoyt 1972; Tahara 1972).
The addition of the CRASH 2005 trial's early results introduced significant heterogeneity. A sensitivity analysis based on the quality of allocation concealment failed to remove this. The most obvious reason for the presence of heterogeneity is the contrasting results of the Faupel 1976 and CRASH 2005 trials. It is difficult to come up with a convincing reason post hoc for excluding either of these trials from a meta-analysis. We noted in a previous version of this review that in the Faupel trial "...the outcome was assessed at discharge", yet overall 19% of the participants were classified as "unconscious stabilized". The apparently short follow up period may account for the incongruous result. Other sources of variation between trials may include severity and pathology of the head injury, variations in corticosteroid regimens (e.g. drug, dose, route) and temporal trends in the use of other interventions. The CRASH trial was stopped early because of the apparent harmful effect of steroids, and stopping trials because of extreme results may select extreme results by chance. Trials of the use of corticosteroids in spinal cord injury suggest that the timing of administration is important (Bracken 1990), but the CRASH trial followed the protocol from spinal cord trials closely, so this is unlikely to explain the excess mortality in the CRASH trial. However, as none of these hypotheses was listed in the original review methods, we think it is safer not to pool the trial results.
There is no clear evidence of a difference in the occurrence of infectious complications and the risk of gastrointestinal bleeding with steroids, making these an unlikely explanation of the increased mortality in the CRASH trial.
In the absence of a meta-analysis, interpretation of the whole body of evidence has to be qualitative. The high methodological quality and large size of the CRASH trial suggest that its result should be the main basis of a summary. This is reinforced by its finding of a significant increase in mortality, which should not be ignored despite the more optimistic results in some other trials.
Authors' conclusions
Implications for practice The results of the large CRASH trial suggest that steroids should not be used in head injury, as they appear to increase mortality. Despite the heterogeneity, we feel that the results of this trial are the most relevant to current practice and should be the basis for clinical decisions, rather than any of the earlier trials or their meta-analysis. |
Implications for research The mechanism of harm is unclear and there will need to be a reassessment of the understanding of the pathophysiology of traumatic brain injury. There seems no reason to start other trials. |
Acknowledgements
We wish to acknowledge help and advice from Dr Iain Chalmers in preparing this review, Julia Langham, Leah Lepage and Yoichi Nagayama in identifying trials, Dr Anthony Rodgers and Dr Colin Baigent for comments on a draft, Liz Norman and Jini Hetherington for proof reading and those trialists who we were able to contact for their cooperation.
Data and analyses
- Top of page
- Background
- Objectives
- Methods
- Results
- Discussion
- Authors' conclusions
- Acknowledgements
- Data and analyses
- Appendices
- What's new
- History
- Contributions of authors
- Declarations of interest
- Sources of support
- Index terms
| |||||||||||||||||||||||||||||||||||||||||||||
Appendices
Appendix 1. Previous search strategies
CENTRAL SEARCH STRATEGY (The Cochrane Library 2005, Issue 4)
#1 (head or crani* or capitis or brain* or forebrain* or skull* or hemisphere* or intracran* or orbit*) in ab or ti
#2 (injur* or trauma* or lesion* or damag* or wound* or destruction* or oedema* or edema* or fracture* or contusion* or commotion* or pressur*) in ab or ti
#3 (#1 and #2)
#4 BRAIN INJURIES
#5 DIFFUSE AXONAL INJURY
#6 CRANIOCEREBRAL TRAUMA
#7 #3 or #4 or #5 or #6
#8 (steroid* or glucocorticoid* or prednisolone* or betamethasone* or cortisone* or dexamethasone* or hydrocortisone* or methylprednisolone* or prednisone* or triamcinolone* or corticosteroid*) in ab or ti
#9 GLUCOCORTICOIDS
#10 ADRENAL CORTEX HORMONES
#11 (#8 or #9 or #10)
#12 (#7 and #11)
MEDLINE SEARCH STRATEGY (1966 to November 2005)
1. explode "Craniocerebral-Trauma" / all SUBHEADINGS in MIME,MJME
2. (head or crani* or capitis or brain* or forebrain* or skull* or hemisphere* or intracran* or orbit*) near (injur* or trauma* or lesion* or damag* or wound* or destruction* or oedema* or edema* or fracture* or contusion* or commotion* or pressur*)
3. 1 or 2
4. explode "Adrenal-Cortex-Hormones" / all SUBHEADINGS in MIME,MJME
5. explode "Glucocorticoids-" / all SUBHEADINGS in MIME,MJME
6. steroid* or glucocorticoid* or prednisolone* or betamethasone* or cortisone* or dexamethasone* or hydrocortisone* or methylprednisolone* or prednisone* or triamcinolone* or corticosteroid*
7. 4 or 5 or 6
8. 3 and 7
9. 8 and (Cochrane highly sensitive RCT strategy)
EMBASE SEARCH STRATEGY (1980 to November 2005, week 46)
1. exp Head Injury/
2. ((head or crani$ or capitis or brain$ or forebrain$ or skull$ or hemisphere$ or intracran$ or orbit$) adj3 (injur$ or trauma$ or lesion$ or damag$ or wound$ or destruction$ or oedema$ or edema$ or fracture$ or contusion$ or commotion$ or pressur$)).mp.[title,abstract]
3. 1 or 2
4. exp CORTICOSTEROID THERAPY/
5. (steroid$ or glucocorticoid$ or prednisolone$ or betamethasone$ or cortisone$ or dexamethasone$ or hydrocortisone$ or methylprednisolone$ or prednisone$ or triamcinolone$ or corticosteroid$).mp. [title, abstract]
6. Glucocorticoid/dt [Drug Therapy]
7. 4 or 5 or 6
8. 3 and 7
Appendix 2. Search strategy: 2008 update
CENTRAL (The Cochrane Library 2007, Issue 4)
#1MeSH descriptor Craniocerebral Trauma explode all trees
#2MeSH descriptor Cerebrovascular Trauma explode all trees
#3MeSH descriptor Brain Edema explode all trees
#4(brain or cerebral or intracranial) near3 (oedema or edema or swell*)
#5MeSH descriptor Glasgow Coma Scale explode all trees
#6MeSH descriptor Glasgow Outcome Scale explode all trees
#7MeSH descriptor Unconsciousness explode all trees
#8glasgow near3 (coma or outcome) near3 (score or scale)
#9(Unconscious* or coma* or concuss* or 'persistent vegetative state') near 3 (injur* or trauma* or damag* or wound* or fracture*)
#10"Rancho Los Amigos Scale"
#11(head or crani* or cerebr* or capitis or brain* or forebrain* or skull* or hemispher* or intra-cran* or inter-cran*) near3 (injur* or trauma* or damag* or wound* or fracture* or contusion*)
#12Diffuse near3 axonal near3 injur*
#13(head or crani* or cerebr* or brain* or intra-cran* or inter-cran*) near3 (haematoma* or hematoma* or haemorrhag* or hemorrhag* or bleed* or pressure)
#14(#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13)
PubMed [www.ncbi.nlm.nih.gov/sites/entrez/] (searched 7 Jan 2008 (added to PubMed in the last 60 days)
#1Craniocerebral Trauma [mh] OR Brain Edema [mh] OR Glasgow Coma Scale [mh] OR Glasgow Outcome Scale [mh] OR Unconsciousness [mh] OR Cerebrovascular Trauma [mh] OR ((head OR cranial OR cerebral OR brain* OR intra-cranial OR inter-cranial) AND (haematoma* OR hematoma* OR haemorrhag* OR hemorrhage* OR bleed* OR pressure)) OR (Glasgow AND scale) OR ("diffuse axonal injury" OR "diffuse axonal injuries") OR ("persistent vegetative state") OR ((unconscious* OR coma* OR concuss*) AND (injury* OR injuries OR trauma OR damage OR damaged OR wound* OR fracture* OR contusion* OR haematoma* OR hematoma* OR haemorrhag* OR hemorrhag* OR bleed* OR pressure))
#2(randomised OR randomized OR randomly OR random order OR random sequence OR random allocation OR randomly allocated OR at random OR randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh]) NOT ((models, animal[mh] OR Animals[mh] OR Animal Experimentation[mh] OR Disease Models, Animal[mh] OR Animals, Laboratory[mh]) NOT (Humans[mh]))
#3Search #1 AND #2 Limits: published in the last 60 days
MEDLINE (Ovid SP) 1950 to Nov (week 2) 2007
1.exp Craniocerebral Trauma/
2.exp Brain Edema/
3.exp Glasgow Coma Scale/
4.exp Glasgow Outcome Scale/
5.exp Unconsciousness/
6.exp Cerebrovascular Trauma/
7.((head or crani$ or cerebr$ or capitis or brain$ or forebrain$ or skull$ or hemispher$ or intra-cran$ or inter-cran$) adj3 (injur$ or trauma$ or damag$ or wound$ or fracture$ or contusion$)).ab,ti.
8.((head or crani$ or cerebr$ or brain$ or intra-cran$ or inter-cran$) adj3 (haematoma$ or hematoma$ or haemorrhag$ or hemorrhag$ or bleed$ or pressure)).ti,ab.
9.(Glasgow adj3 (coma or outcome) adj3 (scale$ or score$)).ab,ti.
10."rancho los amigos scale".ti,ab.
11.("diffuse axonal injury" or "diffuse axonal injuries").ti,ab.
12.((brain or cerebral or intracranial) adj3 (oedema or edema or swell$)).ab,ti.
13.((unconscious$ or coma$ or concuss$ or 'persistent vegetative state') adj3 (injur$ or trauma$ or damag$ or wound$ or fracture$)).ti,ab.
14.or/1-13
15. (randomised or randomized or randomly or random order or random sequence or random allocation or randomly allocated or at random or controlled clinical trial$).tw,hw.
16.clinical trial.pt.
17.randomized controlled trial.pt.
18.17 or 18 or 19
19.exp models, animal/
20.exp Animals/
21.exp Animal Experimentation/
22.exp Disease Models, Animal/
23.exp Animals, Laboratory/
24.or/21-25
25.Humans/
26.20 not 25
27.18 not 26
28.14 and 27
29.2007$.ed.
30.28 and 29
EMBASE (Ovid SP) 1980 to (week 1) Jan 2008
1.exp Brain Injury/
2.exp Brain Edema/
3.exp Glasgow Coma Scale/
4.exp Glasgow Outcome Scale/
5.exp Rancho Los Amigos Scale/
6.exp Unconsciousness/
7.((brain or cerebral or intracranial) adj3 (oedema or edema or swell$)).ab,ti.
8.((head or crani$ or cerebr$ or capitis or brain$ or forebrain$ or skull$ or hemispher$ or intra-cran$ or inter-cran$) adj3 (injur$ or trauma$ or damag$ or wound$ or fracture$ or contusion$)).ab,ti.
9.(Glasgow adj3 (coma or outcome) adj3 (scale$ or score$)).ab,ti.
10.Rancho Los Amigos Scale.ab,ti.
11.((unconscious$ or coma$ or concuss$ or 'persistent vegetative state') adj3 (injur$ or trauma$ or damag$ or wound$ or fracture$)).ti,ab.
12.Diffuse axonal injur$.ab,ti.
13.((head or crani$ or cerebr$ or brain$ or intra-cran$ or inter-cran$) adj3 (haematoma$ or hematoma$ or haemorrhag$ or hemorrhag$ or bleed$ or pressure)).ab,ti.
14.or/1-13
15.exp animal model/
16.Animal Experiment/
17.exp ANIMAL/
18.exp Experimental Animal/
19.1 or 2 or 3 or 4
20.Human/
21.5 not 6
22.(randomised or randomized or randomly or random order or random sequence or random allocation or randomly allocated or at random or controlled clinical trial$).tw,hw.
23.exp clinical trial/
24.8 or 9
25.10 not 7
26.14 and 25
27.2007$.em.
28.26 and 27
PsycINFO (Ovid SP) 1806 to April 2007
1.explode "Head-Injuries" in MJ,MN
2.explode "Brain-Damage" in MJ,MN
3.explode "Traumatic-Brain-Injury" in MJ,MN
4.explode "Brain-Concussion" in MJ,MN
5.explode "Coma-" in MJ,MN
6.Unconscious* or coma* or concuss* or "persistent vegetative state"
7.(head or crani* or cerebr* or capitis or brain* or forebrain* or skull* or hemispher* or intra-cran* or inter-cran*) near (injur* or trauma* or damag* or wound* or fracture* or contusion*)
8.(head or crani* or cerebr* or brain* or intra-cran* or inter-cran*) near (haematoma* or hematoma* or haemorrhag* or hemorrhag* or bleed* or pressure)
9.#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8
What's new
Last assessed as up-to-date: 6 January 2008.
| |||||||||
History
Protocol first published: Issue 3, 1997
Review first published: Issue 3, 1997
| |||||||||||||||||||||
Contributions of authors
Both authors selected studies for inclusion, extracted data and wrote the text. Both authors contacted trialists for further information.
Declarations of interest
Ian Roberts collaborated on the design and conduct of the CRASH trial of corticosteroids in head injury, funded by the UK Medical Research Council.
Philip Alderson has no known conflict of interest.
Sources of support
Internal sources
- Institute of Child Health, University of London, UK.
- London School of Hygiene and Tropical Medicine, UK.
- NHS R&D Programme, UK.
External sources
- NHS R&D Programme, Mother and Child Health, UK.
Index terms
Medical Subject Headings (MeSH)
Brain Injuries [*drug therapy]; Glucocorticoids [*therapeutic use]; Neuroprotective Agents [*therapeutic use]; Randomized Controlled Trials as Topic
MeSH check words
Humans
* Indicates the major publication for the study