Web site designed and maintained by Chris Cairns  © SICS EBM Group 2004                                  

Steroids in Head Injury

Citation/s:
CRASH group. Effect of intravenous corticosteroids on death within 14 days in 10,008 adults with clinically significant head injury (MRC CRASH Trial): randomised, placebo-controlled trial. Lancet 2004:364;1321-1328

Lead author's name and fax: CRASH group: crash@lshtm.ac.uk

Three-part Clinical Question: Does the use of steroids in adult patients with head injury improve mortality at 14 days.

Search Terms: Steroids, head injury, therapy, outcome.

The Study: Double-blinded concealed randomised controlled trial with intention-to-treat.

The Study Patients:  Adults (≥16years); within 8 hours of head injury; GCS ≤14/15; no clear indication or contraindication to steroids. Allocation of patients to study groups was balanced for sex, age, time since injury, GCS, pupil reactiveness and country. Patients were enrolled from April 1999 to May 2004, 239 hospitals in 49 countries were involved.

 
Control group (N = 5001; 4979 analysed): Identically looking vials and solutions prepared in the same way and delivered at the same rate as the experimental group.

Experimental group (N = 5007; 4985 analysed): Prepared packs containing methylprednisolone (and solutions); loading dose 2g in 100ml 0.9% NaCl followed by 0.4g methylprednisolone per hour infusion rate 20ml/hr for 48hours.

The Evidence:

Comments:

1. Do the methods allow accurate testing of the hypothesis? Yes. There were some limitations of the study: no data was collected to establish the main cause of death (as this is often difficult to establish when multiple factors related to trauma are present); The cause of the increased death in the corticosteroid group remains unclear; There was no statistically significant increase for a large rise in either infectious complications or gastrointestinal bleeding in the corticosteroid group.

2.  Do the statistical tests correctly test the results to allow differentiation of statistically significant results? Yes

3. Are conclusions valid in light of the results? Yes

4. Did results get omitted, and why? Powered to show a 2% survival difference at more than 90% power to achieve a p‹0.01; intended to collect 20 000 patients. Patient data collection stopped when the data was revealed by an independent committee to the steering committee after 10008 patients as results revealed “proof beyond reasonable doubt” of difference in outcome between the two treatment groups; which may alter the choice of treatment. 22 patients from both the methylprednisolone and placebo groups were lost to follow up (99.6% data complete at 2 weeks for each group). 62 patients were found to be younger than 16 years; 21 were enrolled ›8 hours after head injury and the trial infusion was stopped at the request of a relative in 3 patients. Intention to treat analysis performed.   

5.  Did they suggest areas of further research?

(a) The effect of corticosteroid treatment on disability 6 months after head injury will be reported soon from the same study.

(b) They suggest further research into corticosteroids after spinal cord injury: NASCIS-2⁽¹⁾ demonstrated some evidence of neurological benefit seen in a subgroup of patients treated within 8 hours of spinal cord injury. As numbers in trials were small (total ≈ 500 patients)  and subgroup effects have been emphasised this area should remain open for debate.

(c) This trial has demonstrated it is possible to enrol many trauma patients into clinical trials in the emergency setting. However consent procedures do affect the number recruited and time to randomisation after head injury⁽²⁾.  Other treatments of uncertain effectiveness in head injury could be studied.

(d) Hypotension after head injury is one the strongest predictors of poor outcome after head injury that is amenable to therapeutic modification. Blood loss from trauma accounts for nearly half of in-hospital trauma deaths. A large placebo-controlled trial of the effects of an antifibrinolytic drug on death and transfusion requirements in patients with clinically significant haemorrhage after trauma is in progress⁽³⁾.

6. Did they make any recommendations based on the results and were they appropriate? Corticosteroids should not be used routinely to treat head injury, whatever the severity. Appropriate from results.

7. Is the study relevant to my clinical practice? Yes.

8. What level of evidence does this study represent? 1⁺⁺

9. What grade of recommendation can I make on this result alone? A

10.  What grade of recommendation can I make when this study is considered along with other available evidence? A

11. Should I change my practice because of these results? Should not use steroids if currently doing so for treatment of head injury (Most units stopped in the 1990’s).

12. Should I audit my current practice because of these results? Yes; if any doubt that steroids are still being used to treat head injuries in your unit.

References:

1. Bracken MB et al. A randomised controlled trial of methylprednisolone or naloxone in the treatment of acute spinal cord injury. N Engl J Med 1990; 322:1405-11.

2. The CRASH Trial Management Group. Research in emergency situations: with or without relatives consent. Emerg Med J 2004; 21:703.

3. http://www.crash2.lshtm.ac.uk

Appraised by: Dr Cathie Wallace. Spr, Western Infirmary, Glasgow, 22 October 2004.

Email: cathiewallace@doctors.org.uk

Citation: EBM Critical Appraisals. Scottish Intensive Care Society EBM Group. Wallace C.2004. CRASH group. Effect of intravenous corticosteroids on death within 14 days in 10,008 adults with clinically significant head injury (MRC CRASH Trial): randomised, placebo-controlled trial. Lancet 2004:364;1321-1328 .

Kill or Update By: October 2009

Reviewed & Edited by CC & VL

Printer Friendly Version

©SICS EBMG 2004