Traumatic brain injury: no clear benefit from hyperventilation, mannitol, steroids, barbiturates or CSF drainage

Clinical bottom line (level 1a-)

  1. Patients with severe head injury who receive hyperventilation, mannitol, or barbiturates are not clearly less likely to die than those who do not.
  2. Patients who receive steroids are not clearly less likely to die or suffer disability than those who do not.
Roberts et al: Journal of Neurology, Neurosurgery and Psychiatry 1998; 65: 729-733
Expires November 2003

The study

Systematic review of all unconfounded randomised trials of
  • Patients: acute traumatic brain injury
  • Intervention: hyperventilation, mannitol, CSF drainage, barbiturates or corticosteroids compared with normoventilation, no mannitol, no CSF drainage, no barbiturates and no corticosteroids respectively
  • Outcome: death, disability


    • Articles found in all languages using Cochrane Controlled Trials Register, Medline, Embase, (search terms: not given ) and by hand-searching 43 international journals, conference abstracts and the proceedings of 16 neurosurgical meetings. Reference lists of trials and review articles were reviewed and authors of trials were contacted to identify additional studies.

    Selection criteria: see above
    Appraisal criteria: detailed in text: extracted by 2 independent reviewers
    Articles excluded if:
    • quasi-randomisation
    • randomisation failed to preclude knowledge of the next treatment


    One study on hyperventilation involving 77 patients; 1 study on mannitol involving 41 patients; 2 studies on barbiturate therapy involving 126 patients; 13 studies on corticosteroids involving 2073 patients. No studies on CSF drainage were found.
    The studies were not found to be significantly heterogeneous.

    The evidence

    Outcome Time to outcome CER OR
    (95% CI)    		 NNT
    (95% CI)
    hyperventilation: death unknown 14/41
    (34.1%)    		 0.73
    (0.36 to 1.49)     		15
    (NNT = 5 to infinity;
    NNH = 11 to infinity)
    mannitol: death unknown 3/21
    (14.3%)    		 1.75
    (0.48 to 6.38)     		-12
    (NNT = 15 to infinity;
    NNH = 3 to infinity)
    barbiturates: death unknown 32/62
    (51.6%)    		 1.12
    (0.81 to 1.54)     		-35
    (NNT = 19 to infinity;
    NNH = 9 to infinity)
    steroids: death unknown 296/836
    (35.4%)    		 0.95
    (0.84 to 1.07)     		86
    (NNT = 26 to infinity;
    NNH = 64 to infinity)
    steroids: death or disability unknown -/-
    (-%)    		 1.01
    (0.91 to 1.11)

    Comments

    1. This study is based on Cochrane systematic reviews of hyperventilation, mannitol, barbiturates and steroids.
    2. Currently, short-term hyperventilation before CT scans define any injury is practiced. There is no clear evidence to support or refute this practice - this study supports a lack of benefit from the long term effects of hyperventilation.

    Citation

    1. Roberts I, Schierhout G, Alderson P: Absence of evidence for the effectiveness of five interventions routinely used in the intensive care management of severe head injury: a systematic review. Journal of Neurology, Neurosurgery and Psychiatry 1998; 65: 729-733
    Search Terms: found in 'Other articles noted' in ACP Journal Club
    Contributor: Chris Ball and Clare Wotton, November 1999
    Reviewer: Malcolm Daniel

    Clinical Question.
    Patient head injury, acute traumatic brain injury
    Intervention or Exposure mannitol, steroids, hyperventilation, CSF drainage, barbiturates
    Outcome death, disability