CRASH: Corticosteroids in Significant Head Injury (2004)

“Our results show that, in patients with head injury, corticosteroids should not be used routinely. The results of this trial provide no evidence that corticosteroid treatment is beneficial, and indeed suggest that it could be harmful.”

  • The CRASH Trial Collaborators

1. Publication Details

  • Trial Title: 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
  • Citation: CRASH trial collaborators. 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(9442):1321-1328. DOI: 10.1016/S0140-6736(04)17188-2
  • Published: October 9, 2004, in The Lancet
  • Author: The CRASH trial collaborators
  • Funding: UK Medical Research Council

2. Keywords

  • Traumatic Brain Injury (TBI), Head Injury, Corticosteroids, Methylprednisolone, Randomized Controlled Trial

3. The Clinical Question

  • In adult patients with clinically significant head injury (Population), does a 48-hour infusion of high-dose corticosteroids (Intervention) compared to placebo (Comparison) reduce 14-day all-cause mortality (Outcome)?

4. Background and Rationale

  • Existing Knowledge: Corticosteroids were widely used in the management of severe head injury based on the physiological rationale that they could reduce cerebral edema and secondary brain injury.
  • Knowledge Gap: Despite their common use, the evidence for corticosteroids was based on numerous small and inconclusive trials. A large, definitive, multicenter randomized controlled trial was urgently needed to determine if this common practice was truly beneficial or potentially harmful.
  • Proposed Hypothesis: The authors hypothesized that a 48-hour infusion of corticosteroids would reduce all-cause mortality in patients with significant head injury.

5. Study Design and Methods

  • Design: A very large, international, multicenter, prospective, randomized, double-blind, placebo-controlled trial (used to test the effectiveness of interventions).
  • Setting: 239 hospitals in 49 countries.
  • Trial Period: Enrollment ran from April 1999 to May 2004.
  • Population:
    • Inclusion Criteria: Adult patients (≥16 years) with a clinically significant head injury (Glasgow Coma Scale score ≤ 14) who were within 8 hours of the injury.
    • Exclusion Criteria: Included patients with clear contraindications to corticosteroids.
  • Intervention: Patients received a 48-hour intravenous infusion of methylprednisolone (2g bolus over 1 hour, followed by 400mg/hr for 47 hours).
  • Control: Patients received a matching intravenous placebo (0.9% saline).
  • Management Common to Both Groups: All other aspects of care for traumatic brain injury were at the discretion of the treating clinicians according to local guidelines.
  • Power and Sample Size: The trial was originally planned to enroll 20,000 patients, which would provide 90% power to detect a 2% absolute risk reduction in mortality. (Power is a study’s ability to find a real difference between treatments if one truly exists; 90% power means the study had a 90% chance of detecting the specified effect, which is considered very high).
  • Outcomes:
    • Primary Outcome: All-cause mortality within 14 days of randomization.
    • Secondary Outcomes: Included death or severe disability at 6 months.

6. Key Results

  • Enrollment and Baseline: 10,008 patients were randomized (5007 to corticosteroids and 5001 to placebo). The groups were well-matched at baseline.
  • Trial Status: The trial was stopped early by the data monitoring committee after enrolling 10,008 patients because of a clear finding of harm in the corticosteroid group.
  • Primary Outcome: 14-day mortality was significantly higher in the corticosteroid group: 1052 of 5007 patients (21.1%) died, compared with 893 of 5001 patients (17.9%) in the placebo group (p=0.0001).
  • Secondary Outcomes: The risk of death or severe disability at 6 months was also significantly higher in the corticosteroid group.
  • Adverse Events: The primary adverse event was death, which was significantly more common in the corticosteroid group.

7. Medical Statistics

  • Analysis Principle: The trial was analyzed using an intention-to-treat principle.
  • Statistical Tests Used: The primary outcome was analyzed using a chi-square test.
  • Primary Outcome Analysis: The primary outcome was a comparison of the proportions of death between the two groups.
  • Key Statistic(s) Reported: Relative Risk (RR) for death at 14 days: 1.18 (95% CI, 1.09 to 1.27; P-value: 0.0001).
  • Interpretation of Key Statistic(s):
    • Relative Risk (RR):
      • Formula: Conceptually, RR = (Risk in Intervention Group) / (Risk in Control Group).
      • Calculation: The paper reports the result as 1.18.
      • Clinical Meaning: The RR of 1.18 means that patients in the corticosteroid group had an 18% higher relative risk of dying within 14 days compared to the placebo group.
    • Confidence Interval (CI):
      • Formula: Conceptually, CI = (Point Estimate) ± (Margin of Error).
      • Calculation: The 95% CI was 1.09 to 1.27.
      • Clinical Meaning: Since this entire range is above the line of no effect (1.0), it confirms that the result is highly statistically significant and demonstrates a clear signal of harm.
    • P-value: The p-value of 0.0001 is extremely low, indicating the result is highly statistically significant (a result is conventionally considered statistically significant if the p-value is less than 0.05).
  • Clinical Impact Measures:
    • Absolute Risk Increase (ARI):
      • Formula: ARI = (Risk in Intervention Group) – (Risk in Control Group)
      • Calculation: ARI = 21.1% – 17.9% = 3.2%.
      • Clinical Meaning: For every 100 patients with head injury treated with corticosteroids, about 3 additional deaths occurred.
    • Number Needed to Harm (NNH):
      • Formula: NNH = 1 / ARI
      • Calculation: NNH = 1 / 0.032 = 31.25, which is rounded down to 31.
      • Clinical Meaning: You would only need to treat 31 patients with this steroid regimen to cause one additional death.
  • Subgroup Analyses: The harm was consistent across all pre-specified subgroups.

8. Strengths of the Study

  • Study Design and Conduct: The very large, multicenter, randomized, double-blind, placebo-controlled design is the gold standard and provided a massive amount of high-quality data.
  • Generalizability: The pragmatic design with broad inclusion criteria and enrollment across 49 countries makes the findings highly generalizable to a wide range of patients with head injury.
  • Statistical Power: The enormous sample size provided definitive power to detect even a small but clinically important difference, ultimately revealing a clear signal of harm.
  • Patient-Centered Outcomes: The primary outcome of all-cause mortality is a robust, objective, and highly relevant endpoint.

9. Limitations and Weaknesses

  • Internal Validity (Bias): The study was well-conducted with a very low risk of bias.
  • External Validity (Generalizability): The findings are highly generalizable to the broad population of patients with clinically significant traumatic brain injury.
  • Other: The trial was stopped early for harm, which was ethically necessary.

10. Conclusion of the Authors

  • The authors concluded that corticosteroids should not be used in the treatment of head injury, as they are associated with an increase in mortality.

11. To Summarize

  • Impact on Current Practice: This was a profoundly practice-changing trial. It provided definitive evidence that a common and long-standing practice was not just ineffective, but actively harmful. It led to the immediate and universal de-adoption of corticosteroids for head injury.
  • Specific Recommendations:
    • Patient Selection: For adult patients with clinically significant head injury.
    • Actionable Intervention: Do not administer high-dose corticosteroids.
    • Expected Benefit: Avoiding this therapy prevents one additional death for every 31 patients treated.
  • What This Trial Does NOT Mean: This trial does NOT mean that corticosteroids are harmful in all neurological conditions. Its findings are specific to traumatic brain injury.
  • Implementation Caveats: The key takeaway is to “first, do no harm.” This trial is a powerful example of why even physiologically plausible therapies must be tested in rigorous trials.

12. Context and Related Studies

  • Building on Previous Evidence: The CRASH trial (2004) was designed to definitively resolve decades of uncertainty from numerous smaller, conflicting, and underpowered studies on the use of steroids in head injury.
  • Influence on Subsequent Research: The definitive finding of harm in this trial effectively ended this line of research for traumatic brain injury. It also set the stage for the subsequent successful CRASH trials (CRASH-2 and CRASH-3) which investigated tranexamic acid.

13. Unresolved Questions & Future Directions

  • Unresolved Questions: This trial definitively answered its primary question with a clear finding of harm.
  • Future Directions: The results of this trial shifted the focus of neurotrauma research away from corticosteroids and towards other potential neuroprotective strategies.

14. External Links

15. Framework for Critical Appraisal

  • Clinical Question: The research question was highly relevant, testing a very common and accepted clinical practice for a deadly condition.
  • Methods: The very large, multicenter, double-blind RCT design was of the highest quality and was essential for uncovering this unexpected harm. Without a placebo group, the excess deaths would have been attributed to the patients’ underlying disease.
  • Results: The study reported a statistically significant and clinically important increase in harm (NNH of 31). The finding was so clear and robust that the trial was stopped early.
  • Conclusions and Applicability: The authors’ conclusion is strongly supported by the data. The trial is a landmark in evidence-based medicine, serving as a powerful example of how a logical, physiologically-based therapy can be harmful, and how only a very large, rigorous, placebo-controlled trial can reveal such a truth.

16. Disclaimer and Contact

  • This summary is provided by the Academic Committee of ESBICM (ACE) to facilitate the understanding of this study; readers are advised to refer to the original trial document for a deeper understanding. If you find any information incorrect, or missing, or it needs an update or have a request for a specific critical care trial summary, kindly write to us at academics[at]esbicm.org.
Scroll to Top