Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Hyperbaric (HBO) or normobaric oxygen (NBO) in the treatment of carbon monoxide poisoning

Three Part Question

In [adult patients with carbon monoxide poisoning] does [hyperbaric oxygen when compared to normobaric oxygen] reduce [mortality and the incidence of persistent (PNS) or delayed neuropsychological sequelae (DNS)]?

Clinical Scenario

A patient is brought in after attempting suicide by inhaling exhaust fumes from their car. He/she has a GCS of 3 and has required intubation but is otherwise stable. He/she has a carboxyhaemoglobin level of 40%. They have been receiving high flow oxygen, but you wonder if transfer for hyperbaric oxygen therapy is indicated.

Search Strategy

Medline 1966-08/01 using the OVID interface.
({exp carbon monoxide poisoning OR carbon monoxide} AND {exp hyperbaric oxygenation OR hyperbaric oxygen$.mp}) LIMIT to human AND english.

Search Outcome

Altogether 299 papers were found of which 294 were irrelevant or of insufficient quality for inclusion. The remaining 5 papers are shown in the table.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Mathieu D et al,
575 Non-comatose adults with CO poisoning Interim analysis <12 hours HBO 90 min 2.5 ATA v NBO (100%) 12 hrs prospective randomized controlled trialNeurological signs and symptomsSignificantly fewer persistent neurological manifestations in HBO group at 3 months, no difference at 6-12 monthsNot blinded, no sham treatments Completed study not available
Raphael JC et al,
629 adults with CO poisoning 1. Patients with no LOC-6 hrs NBO (100%) V 2 hrs HBO 2. Patients with initial or persisting LOC-HBO x2 hrs v NBO (100%) x 4 hoursprospective randomized controlled trialself assessment questionnaire or postal/telephone questionnaireNo difference in % recovery between 2 groups at 1 month 54% v 52% (p=0.75). 2 deaths in each groupNot blinded, no sham treatments Outcome measures not standardised
% recovery at 1 month - Symptoms, Physical signs No difference in % recovery between 2 groups at 1 month 66% v 68% (p=0.75).
Ducasse JL et al,
26 non-comatose adults with CO poisoning HBO 2 hours (+NBO 100% 4 hrs, 50% 6 hrs) versus NBO (100% 6 hrs, 50% 6 hrs) Prospective randomised controlled trialSymptoms and physical signs at 2 hours+12 hoursSignificantly slower recovery in NBO group. All asymptomatic at dischargeSmall numbers Not blinded, no sham treatments
EEGEEG- More abnormalities at 3 weeks in NBO group
Thom SR et al,
63 adults with CO poisoning NBO (100%) v HBO 2.8 ATA 30 min + 2 ATA 90 min Prospective randomised controlled trial% with delayed neurological sequelae (signs and symptoms and psychometric evaluation) post treatment at 1 week and 1 monthDNS seen in 8 (23%) NBO group. None in HBO group. All fully recovered 41 days.Not blinded, no sham treatments Small numbers Psychometric evaluation compared with only 8 healthy volunteers
Scheinkestel CD et al,
191 adult patients with CO poisoning (comatose and non comatose) NBO (100%) 3 days + sham treatments v HBO 1 hour for 3 days 2.8 ATA Prospective randomised controlled trial. Double-blinded outcome assessment.Neuropsychological performance on treatment completion + 1 monthPNS 62% at follow up - no difference between 2 groups. DNS (5) only seen in HBO group (p=0.03). Neuropsychological testing - significantly lower number of abnormal tests in NBO group at completion of treatmentCluster randomisation Only 46% attended 1 month follow up HBO treatment delay (mean 7.1 hrs)
Mortality ratesMortality- no significant difference between 2 groups


Studies vary in their inclusion criteria (presence of symptoms, COHb level, all emergency referrals) and in their treatment regimes and time to treatment. Whilst there are studies (non-randomised and randomised) suggesting benefit of HBO over NBO therapy in carbon monoxide poisoning, only one study (Scheinkestel et al) was double-blinded, utilitzing sham therapies. This study showed no benefit of HBO therapy over NBO treatment. The conclusions of this BET are similar to that of a Cochrane Review (6).

Editor Comment

This BET is currently in the process of an update 13/06/05 KW

Clinical Bottom Line

There is no evidence of an advantage in terms of mortality and neurological sequelae of hyperbaric oxygen therapy over 3 days of high flow normorbaric oxygen for patients with all grades of CO poisoning.


  1. Mathieu D, Nolf M, Durocher A, et al. Acute carbon monoxide poisoning. Risk of late sequelae and treatment by hyperbaric oxygen. J Toxicol Clin Toxicol 1985;23(4-6):315-24.
  2. Raphael JC, Elkharrat D, Jars-Guincestre MC, et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989;2(8660):414-9.
  3. Ducasse JL, Celsis P, Marc-Vergnes JP. Non-comatose patients with acute carbon monoxide posioning: hyperbaric or normobaric oxygenation? Undersea Hyperb Med 1995;22(1):9-15.
  4. Thom SR, Taber RL, Mendiguren II, et al. Delayed neuropsychologic sequelae after carbon monoxide poisoning: prevention by treatment with hyperbaric oxygen. Ann Emerg Med 1995;25(4):474-80.
  5. Scheinkestel CD, Bailey M, Myles PS, et al. Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomised controlled clinical trial. Med J Aust 1999;170(5):203-10.
  6. Juurlink DN, Stanbrook MB, McGuigan MA. Hyperbaric oxygen for carbon monoxide poisoning (Cochrane Review). The Cochrane Library Issue 3, 2001.