Carbon Monoxide Poisoning

Author: Zach Radwine, MD (EM Resident Physician, University of Illinois College of Medicine – Peoria) // Editors: Jennifer Robertson, MD, Lewis Nelson, MD and Alex Koyfman, MD (@EMHighAK)

Featured on the LITFL Review #180 – Thanks to Joe-Anthony Rotella and the LITFL Review group for the shout out!


Carbon monoxide (CO) poisoning leads to an estimated 50,000 ED visits per year in the US.1 It is also the leading cause of death by poisoning.2,3,4,5 CO causes toxicity via binding with hemoglobin to form carboxyhemoglobin (COHb).  This reduces the oxygen carrying capacity of blood because hemoglobin binds CO with 210 times greater affinity than does O2.4,6 In addition CO produces toxicity by direct cellular damage by numerous mechanisms, including lipid peroxidation, binding to intracellular proteins, and apoptosis.4 The most feared complications of CO poisoning survivors are persistent and delayed neuropsychiatric sequelae (PNS and DNS). These can range from mild intellectual impairment to Parkinsonian syndrome to seizure disorders.6,7 DNS have been reported to occur from 3 days to up to several weeks post-exposure.6


The diagnosis of CO poisoning requires the following:4

  1. History of recent CO exposure
  2. Presence of symptoms consistent with CO poisoning
  3. Elevated COHb level

The single most common presenting complaint in patients with mild to moderate CO poisoning is headache,8 occurring in 84% of patients.7  However, symptoms and clinical findings generally are varied and vague. They may also mimic gastroenteritis or viral illness, thus making the diagnosis difficult.6,7,8 Because of the non-specific nature of symptoms, the clinician must maintain a high index of suspicion.4,7 CO poisoning should always be considered when multiple patients present to the ED from a single location with appropriate findings.8

Physical exam is most often non-revealing, but should focus on a detailed neurologic exam, cardiovascular exam, and an assessment for signs of trauma.6,8 A mental status exam may reveal cognitive changes.8 Cherry-red skin is a rare and often post-mortem finding.4,6,7

A blood COHb is very helpful for diagnosis. However, symptoms do not always correspond with the COHb level.1 Also, neither peak nor ED levels correlate well with severity of poisoning or illness8 or with outcome.9 Importantly, a patient may have toxicity even with a normal level depending on the timing of exposure, relative to sampling, degree of exposure, and any oxygen therapy prior to sampling.8

 Nevertheless, the following levels, do suggest possible CO exposure:8

  1. >3% in non-smokers
  2. >10% in cigarette smokers

Pulse CO-oximetry

Standard pulse oximeters cannot distinguish the wavelengths of O2 and CO-bound hemoglobin, but more novel devices can measure CO separately.6 Studies are conflicting on the reliability of pulse CO-oximetry. One study found a negative predictive value of 100%, identifying 17 patients with CO poisoning out of 1,578 patients.10 Using the ROC, the optimal cutoff was 6.6% for both smokers and non-smokers. The authors of this study advocate the use of pulse CO-oximetry to screen large numbers of patients. Despite this, most studies suggest that it is useful for more rapid diagnosis but must always be followed by a blood COHb for confirmation, as false negatives occur at an unacceptably high rate.11,12,13,14 One study noted that pulse CO-oximetry was only accurate if the SpO2 is >85%.15

Adjunctive Testing8

-ECG and cardiac monitoring due to potential risk of myocardial ischemia and dysrhythmia

-Cardiac biomarkers for ECG changes, symptoms of ischemia, history of CAD, or age >65

-Pregnancy test

-For intentional exposure: acetaminophen and salicylate levels

-Venous or arterial COHb (no relevant clinical difference between arterial and venous COHb levels, with 95% of samples falling in the range 2.4% to -2.1% of each other)6,16

– Arterial blood gas (ABG) should be done if needed for another reason, but not necessary just for CO toxicity

-Severe metabolic acidosis correlates with a short-term mortality rate in CO-poisoned patients. If the source was a fire, consider concomitant cyanide (CN) poisoning.

-Lactate levels do not correlate with severity of poisoning but may be useful in possible CN toxicity as well as in sick patients with very high CO levels


The goal is removal of the CO as soon as possible. The mean half-life of COHb is 320 min on room air, 80 min on 100% O2 at one atmosphere, and 23 min on 100% O2 at three atmospheres in a hyperbaric chamber.5 Treatment thus begins as soon as possible with high flow, 100% oxygen, either by mask or endotracheal tube.4  The O2 should be continued until the COHb is normal (<3%) and the patient’s signs and symptoms have resolved, usually for about 6 hours.6

The use of hyperbaric oxygen (HBO), defined by the Undersea and Hyperbaric Medicine Society as breathing 100% oxygen at 2-3 atmospheres,6 is controversial. The mortality of CO-poisoned patients presenting to a hospital is about 3%, and to date, no study has clearly shown a reduction in mortality with HBO. Therefore, the goal of HBO therapy is the prevention of long-term and permanent neurologic dysfunction.4 A 2011 Cochrane review looked at six randomized controlled trials (RCTs) with a total of 1,361 patients. Of the six trials, two found benefit of HBO for the reduction of the incidence of DNS at one month, while four others did not find this benefit. One trial actually found worse outcomes with two vs one HBO treatments, suggesting that in some circumstances HBO might even worsen neurologic sequelae. These authors concluded that there is insufficient evidence to support the use of HBO for treatment of patients with CO poisoning.  Although they acknowledge that many experts in the hyperbaric medicine community strongly advocate that it has been established as effective, the reviewers maintain that more placebo-controlled clinical trials are warranted. 9

A 2008 American College of Emergency Physicians (ACEP) clinical policy makes similar conclusions:17

  1. HBO is a therapeutic option for CO-poisoned patients; however, its use cannot be mandated.
  2. No clinical variables, including COHb levels, identify a subgroup of CO-poisoned patients for whom HBO is most likely to provide benefit or cause harm (controversial; syncope patients likely benefit).

Emerging Therapies

  • Treatment of carbon monoxide poisoning with hyperbaric oxygen and therapeutic hypothermia18

A case series presents four patients all found unconscious, in respiratory distress, acidotic, with elevated COHb 28-45%. All received HBO x3 (per Weaver9 protocol) and therapeutic hypothermia x24 hours. All had excellent neurologic recovery.

  • Neuroprotective effects of erythropoietin in patients with carbon monoxide poisoning19

This RCT n=103 looked at the efficacy of erythropoietin (EPO) for CO poisoning.  They found significant decreases in delayed neurologic sequelae, increases in full recovery rate, and superior neurologic outcomes in EPO group.

Special Situations

Pregnancy and children

Pregnant patients and children are particularly at risk for permanent sequelae of CO poisoning but adult treatment criteria is still applied to these populations. In pregnancy, fetal distress and fetal death are special concerns in CO poisoning. HBO has been administered to pregnant women, but there are no adequate prospective studies of efficacy.4 All clinical trials of HBO excluded pregnant women. Although fetal outcomes in CO-exposed women have been described in several case series and one structured review, no study has compared pregnancy outcomes in women of similar poisoning severity treated with different therapeutic options.9 Steady-state levels of COHb in the fetus are higher than in maternal blood. It takes longer to reach steady state, and fetal elimination of CO is also prolonged. For these reasons some experts advocate HBO for pregnant women with a COHb level >15% or signs of fetal distress.8 Multiple studies have shown that children of CO-poisoned pregnant women with normal mental status and no history of LOC have good outcomes in terms of normal delivery and future development of their children.8

Cardiac Arrest

A Class III retrospective case series reported 18 consecutive patients who presented to a single institution after resuscitation from CO-associated cardiac arrest. Despite prompt and aggressive treatment of all patients with HBO, none survived to hospital discharge. However, survival, albeit with devastating neurologic injury, has been reported in one survivor of CO-associated cardiac arrest treated with HBO.9


Consensus among leading sources is that patients with resolution of symptoms, return of COHb level to normal, and without LOC or evidence of end-organ damage may be safely discharged home.8

Pearls and Pitfalls

  • Symptoms are variable and physical exam and pulse-oximetry are unreliable. Maintain high level of suspicion with emphasis on historical factors.
  • Always consider CO poisoning when two or more patients present from a single location, especially with non-specific symptoms.
  • CO-oximetry may be used as an initial test, but should be followed by a blood COHb level.
  • A normal COHb level does not exclude CO poisoning.
  • Start the patient on O2 as soon as the diagnosis is suspected.
  • Consider concomitant CN poisoning in fire-exposed patients
  • CO poisoning may present as cardiac ischemia, particularly in patients already with underlying risk factors.
  • HBO is recommended by many experts but is not the standard of care. When in doubt, consult with your nearest hyperbaric medicine specialist, medical toxicologist, or poison center.

Sources/Further Reading

  1. Hampson NB, Weaver LK. Carbon monoxide poisoning: a new incidence of an old disease. Undersea Hyperb Med 2007;34:269-276.
  2. (accessed 2/22/2015)
  3. (accessed 2/22/2015)
  4. Hampson NB, et al. Practice recommendations in the diagnosis, management, and prevention of carbon monoxide poisoning. Am J Resp Crit Care Med 2012;186(11):1095-1101.
  5. Varon J, Marik P, Fromm R, Gueler A. Carbon monoxide poisoning: a review for clinicians. J of Emerg Med 1999;17(1):87-93.
  6. Huzar T, George T, Cross J. Carbon monoxide and cyanide toxicity: etiology, pathophysiology and treatment in inhalation injury. Expert Rev Respir Med 2013;7(2):159-170.
  7. Prockop L, Chichkova R. Carbon monoxide: an updated review. J Neurological Sciences 2007;262:122-130.
  8. Nikkanen H, Skolnik A. Diagnosis and management of carbon monoxide poisoning in the emergency department. Emerg Med Practice 2011;13(2):1-14.
  9. Buckley NA, Juurlick DN, Isbister G, Bennett MH, Lavonas EJ. Hyperbaric oxygen for carbon monoxide poisoning. Cochrane Database Syst Rev [serial on the Internet]. 2011;4: CD002041. Available from:
  10. Roth D, et al. Accuracy of noninvasive multiwave pulse oximetry compared with carboxyhemoglobin from blood gas analysis in unselected emergency department patients. Ann Emerg Med 2011;58(1):74-79.
  11. Hampson N. Noninvasive pulse CO-oximetry expedites evaluation and management of patients with carbon monoxide poisoning. Am J of Emerg Med 2012;30:2021-24.
  12. Weaver LK, Churchill SK, Deru K, Cooney D. False positive rate of carbon monoxide saturation by pulse oximetry of emergency department patients. Respir Care 2013;58(2):232-240.
  13. Sebbane M, et al. Emergency department management of suspected carbon monoxide poisoning: role of pulse CO-oximetry. Respir Care 2013;58(10):1614-20.
  14. Piatkowski A, Ulrich D, Grieb G, Pallua N. A new tool for the early diagnosis of carbon monoxide intoxication. Inhal Toxicol 2009;21:1144–1147
  15. Feiner J, et al. Accuracy of carboxyhemoglobin detection by pulse co-oximetry during hypoxemia. Anesthesia Analgesia 2013;117(4):847-858.
  16. Touger M, Gallagher EJ, Tyrell J. Relationship between venous and arterial carboxyhemoglobin levels in patients with suspected carbon monoxide poisoning. Ann Emerg Med 1995;33:105-109.
  17. Wolf FJ, Lavonas EJ, Sloan EP, Jagoda AS. Clinical policy: critical issues in the management of adult patients presenting to the emergency department with acute carbon monoxide poisoning. Ann Emerg Med 2008;51:138–152
  18. Feldman J, Renda N, Markovitz G, Chin W, Sprau S. Treatment of carbon monoxide poisoning with hyperbaric oxygen. Undersea and Hyperbaric Med 2013;40(1):71-79
  19. Pang L, et al. Neuroprotective effects of erythropoietin in patients with carbon monoxide poisoning. J Biochem Molecular Toxicology 2013;27(5):266-271
  20. Poisoning/NonFireCarbonMonoxideDeathsAssociatedwiththeUseofConsumerProducts2011AnnualEstimatesSept2014.pdf (accessed 2/22/2015)
  21. Olson K. Carbon monoxide poisoning: mechanisms, presentation, and controversies in management. J of Emerg Med 1984;1:233-243.
  22. Hampson N, Dunn S. Symptoms of carbon monoxide poisoning do not correlate with the initial carboxyhemoglobin level. Undersea and Hyperbaric Medicine 2012;39(2):657-65.
  23. Touger M, Birnbaum A, Wang J, Chou K, Pearson D, Bijur P. Performance of the RAD-57 pulse CO-oximeter compared with standard laboratory carboxyhemoglobin measurement. Ann Emerg Med 2010;56:382–388

6 thoughts on “Carbon Monoxide Poisoning”

Leave a Reply

Your email address will not be published. Required fields are marked *