What makes UpToDate so powerful?

  • over 11000 topics
  • 22 specialties
  • 5,700 physician authors
  • evidence-based recommendations
See more sample topics
Find Patient Print
0 Find synonyms

Find synonyms Find exact match

Carbon monoxide poisoning
Official reprint from UpToDate®
www.uptodate.com ©2017 UpToDate, Inc. and/or its affiliates. All Rights Reserved.
The content on the UpToDate website is not intended nor recommended as a substitute for medical advice, diagnosis, or treatment. Always seek the advice of your own physician or other qualified health care professional regarding any medical questions or conditions. The use of this website is governed by the UpToDate Terms of Use ©2017 UpToDate, Inc.
Carbon monoxide poisoning
View in Chinese
All topics are updated as new evidence becomes available and our peer review process is complete.
Literature review current through: Sep 2017. | This topic last updated: Feb 23, 2017.

INTRODUCTION — Carbon monoxide (CO) is an odorless, tasteless, colorless, nonirritating gas formed by hydrocarbon combustion. The atmospheric concentration of CO is generally below 0.001 percent, but it may be higher in urban areas or enclosed environments. CO binds to hemoglobin with much greater affinity than oxygen, forming carboxyhemoglobin (COHb) and resulting in impaired oxygen transport and utilization. CO can also precipitate an inflammatory cascade that results in CNS lipid peroxidation and delayed neurologic sequelae.

CO poisoning will be reviewed here. A summary table to facilitate emergent management is provided (table 1). Related topics including smoke inhalation and hyperbaric oxygen therapy are presented separately. (See "Hyperbaric oxygen therapy" and "Inhalation injury from heat, smoke, or chemical irritants".)

EPIDEMIOLOGY — Fire-related smoke inhalation is responsible for most cases of carbon monoxide (CO) poisoning. Non-fire related CO poisoning is responsible for up to 50,000 emergency department (ED) visits and 1200 deaths per year, making it one of the leading causes of poisoning death in the United States [1-5]. Inadvertent, non-fire related CO poisoning likely causes around 400 deaths annually, while the number of intentional CO poisonings resulting in death is twice as high [1,4,6,7]. The case-fatality rate for non-fire CO poisoning ranges widely, but analysis of aggregated national data from the United States support an overall mortality of 1 to 3 percent [3,4,8,9]. The mortality rate is higher for intentional poisoning than for inadvertent exposure [4].

Unlike intentional poisoning, unintended poisoning demonstrates both seasonal and regional variation, and it is most common during the winter months in cold climates [7]. Morbidity, which is primarily related to late neurocognitive impairment, persists beyond initial stabilization in up to 40 percent of victims [3,8].

Potential sources of CO, other than fires, include poorly functioning heating systems, improperly vented fuel-burning devices (eg, kerosene heaters, charcoal grills, camping stoves [10], gasoline-powered electrical generators [11,12]), and motor vehicles operating in poorly ventilated areas (eg, ice rinks, warehouses, parking garages). CO poisonings following open air exposure to motorboat exhaust have also been reported [13]. In addition, underground electrical cable fires produce large amounts of CO, which can seep into adjacent buildings and homes [14]. An increase in carbon monoxide exposures has been reported to occur in the immediate aftermath of hurricanes [11,15,16].

Methylene chloride (dichloromethane) is an industrial solvent and a component of paint remover. Inhaled or ingested methylene chloride is metabolized to CO by the liver, causing CO toxicity in the absence of ambient CO [1,17]. The United States Occupational Safety and Health Administration lowered the workplace exposure limit for methylene chloride from 500 to 25 parts per million (ppm) based on concerns over the chronic effects of carboxyhemoglobinemia [18]. (See "Overview of occupational and environmental health".)

PATHOPHYSIOLOGY — Carbon monoxide (CO) diffuses rapidly across the pulmonary capillary membrane and binds to the iron moiety of heme (and other porphyrins) with approximately 240 times the affinity of oxygen. The degree of carboxyhemoglobinemia (COHb) is a function of the relative amounts of CO and oxygen in the environment, duration of exposure, and minute ventilation. (See "Oxygen delivery and consumption".)

Nonsmokers may have up to 3 percent carboxyhemoglobin at baseline; smokers may have levels of 10 to 15 percent [1]. Severe chronic obstructive pulmonary disease can cause a modest but significant elevation in carboxyhemoglobin levels, even among patients without exposure to tobacco smoke. The mechanism and clinical significance of this finding is unclear [19].

Once CO binds to the heme moiety of hemoglobin, an allosteric change occurs that greatly diminishes the ability of the other three oxygen binding sites to off-load oxygen to peripheral tissues (figure 1). This results in a deformation and leftward shift of the oxyhemoglobin dissociation curve, and compounds the impairment in tissue oxygen delivery (figure 2).

CO also interferes with peripheral oxygen utilization. Approximately 10 to 15 percent of CO is extravascular and bound to molecules such as myoglobin, cytochromes, and NADPH reductase, resulting in impairment of oxidative phosphorylation at the mitochondrial level (figure 3) [8,20]. The half-life of CO bound to these molecules is longer than that of COHb. The importance of these non-hemoglobin-mediated effects has been best documented in the heart, where mitochondrial dysfunction due to CO can produce myocardial stunning despite adequate oxygen delivery [21].

CO also interferes with peripheral oxygen utilization by inactivating cytochrome oxidase in a manner similar to, but clinically less important than, cyanide. CO and cyanide poisoning can occur simultaneously in patients following smoke inhalation, and their combined effects on oxygen transport and utilization appear to be synergistic [22,23]. (See "Inhalation injury from heat, smoke, or chemical irritants".)

The effects of CO on oxygen delivery and utilization, however, cannot account for the delayed neurologic sequelae (DNS) that may occur after CO poisoning. The mechanism of DNS is incompletely understood, but it probably involves lipid peroxidation by toxic oxygen species generated by xanthine oxidase. Xanthine oxidase is produced in situ from xanthine dehydrogenase via enzymes released by white blood cells that adhere to damaged endothelial cells [24-28]. During recovery from CO exposure, events analogous to ischemia-reperfusion injury and exposure to hyperoxia may exacerbate the initial oxidative damage [2,29].

KINETICS — Carbon monoxide (CO) is rapidly absorbed across the pulmonary endothelium. Elimination is dependent upon the degree of oxygenation and, to a lesser extent, minute ventilation. The half-life of CO while a patient is breathing room air is approximately 250 to 320 minutes, while breathing high-flow oxygen via a nonrebreathing face mask is about 90 minutes and with 100 percent hyperbaric oxygen is approximately 30 minutes.


Symptoms and signs — The clinical findings of carbon monoxide (CO) poisoning are highly variable and largely nonspecific [30,31] (table 1). Moderately or mildly CO-intoxicated patients often present with constitutional symptoms, including headache (the most common presenting symptom), malaise, nausea, and dizziness, and may be misdiagnosed with acute viral syndromes [29]. In addition to current symptoms, the clinician should specifically inquire (of the patient and/or witnesses) about loss of consciousness.

In the absence of concurrent trauma or burns, physical findings in CO poisoning are usually confined to alterations in mental status, so a careful neurologic examination is crucial. Patients may manifest symptoms ranging from mild confusion to coma. Specific cognitive testing, such as the Carbon Monoxide Neuropsychological Screening Battery, is usually not used in the acute setting and, in any event, is not universally endorsed due to its inability to discriminate between the effects of CO and those of other intoxicants [1,27]. Although some textbooks describe a "cherry red" appearance of the lips and skin as indicative of CO poisoning, this is an insensitive sign [30].

Severe CO toxicity can produce neurologic symptoms such as seizures, syncope, or coma, and also cardiovascular and metabolic manifestations such as myocardial ischemia, ventricular arrhythmias, pulmonary edema, and profound lactic acidosis.

Myocardial injury — Acute myocardial injury is common among CO-poisoned patients and is associated with increased long-term mortality. A retrospective study of 230 patients with moderate or severe CO poisoning referred to a specialized center found evidence of myocardial ischemia (characteristic electrocardiographic changes or elevated serum cardiac biomarkers) in one-third of all cases [32].

Long-term follow up (median 7.6 years) of this cohort noted a mortality rate of 24 percent among patients who sustained acute myocardial injury [33]. Mortality among patients with myocardial injury was more than twice that of poisoned patients without evidence of such injury, and was estimated to be triple that expected for a comparable unpoisoned cohort.

This study population was young (mean age 47 years) and had a low incidence of established cardiac disease or cardiac risk factors, other than smoking. It is probable that the incidence of myocardial ischemia among CO-poisoned patients with established cardiac disease is even higher.

Delayed neuropsychiatric syndrome — In up to 40 percent of patients with significant CO exposure, a syndrome of delayed neurologic sequelae (DNS) can arise 3 to 240 days after apparent recovery [34-37]. Characterized by variable degrees of cognitive deficits, personality changes, movement disorders, and focal neurologic deficits, DNS generally occur within 20 days of CO poisoning, and deficits may persist for a year or longer.

The development of DNS correlates poorly with COHb levels, although the majority of cases are associated with loss of consciousness during acute intoxication [1,8,38]. The incidence and severity of DNS have become increasingly important clinical end points in studies of treatment for CO poisoning. (See 'Hyperbaric oxygen' below and 'Hyperbaric oxygen and delayed neuropsychiatric syndrome (DNS)' below.)

DIAGNOSIS — Acute carbon monoxide (CO) poisoning is usually suspected on the basis of a suggestive history, while the diagnosis of chronic CO intoxication is notoriously difficult [1,31] (table 1). Standard pulse oximetry (SpO2) CANNOT screen for CO exposure, as it does not differentiate carboxyhemoglobin from oxyhemoglobin (figure 4) [39,40]. Eight-wavelength pulse oximeters capable of measuring carboxyhemoglobin and methemoglobin are being developed, but need further study and should not be used for diagnosis [41-43]. (See "Pulse oximetry in adults", section on 'Carboxyhemoglobin'.)

The diagnosis of CO poisoning is based upon a compatible history and physical examination in conjunction with an elevated carboxyhemoglobin level measured by cooximetry of an arterial blood gas sample. In hemodynamically stable patients, venous samples are accurate and commonly used [44,45]. Nonsmokers may have up to 3 percent carboxyhemoglobin at baseline; smokers may have levels of 10 to 15 percent. Levels above these respective values are consistent with CO poisoning.

Carboxyhemoglobin levels correlate imprecisely with the degree of poisoning and are not predictive of delayed neurologic sequelae (DNS). Venous samples may be used to determine the carboxyhemoglobin level [44,45], but they are less accurate in quantifying the associated acidosis. Venous samples are useful for screening large numbers of potential CO victims in a disaster situation and for monitoring changes in an individual’s carboxyhemoglobin level over time during treatment.

Case reports and animal and laboratory studies suggest that treatment with hydroxocobalamin (as might be performed for presumed cyanide exposure from smoke inhalation) can interfere with the measurement of carboxyhemoglobin leading to inaccurate results [46-50]. In patients with possible cyanide exposure (eg, brought to the hospital from a fire), clinicians assuming care should ask whether hydroxocobalamin was given. However, regardless of such treatment, clinicians should make a presumptive diagnosis of carbon monoxide poisoning on the basis of the history and clinical findings, and should err on the side of treatment if there is any doubt. If available, blood samples obtained before treatment with hydroxocobalamin may provide more accurate measurements.

Noninvasive pulse cooximeters capable of photo spectroscopic measurements of carboxyhemoglobin are in development [51]. However, preliminary observational studies question their accuracy [52,53]. Until well-performed trials demonstrate that these devices provide consistently accurate measurements, we cannot recommend their routine clinical use.

Blood PO2 measurements tend to be normal because PO2 reflects O2 dissolved in blood, and this process is not affected by CO. In contrast, hemoglobin-bound O2 (which normally comprises 98 percent of arterial O2 content) is profoundly reduced in the presence of COHb. (See "Oxygen delivery and consumption".)

Once the diagnosis of CO intoxication is confirmed, we recommend obtaining an electrocardiogram (ECG); cardiac biomarker evaluation is warranted in patients with ECG evidence of ischemia or a history of cardiac disease [32].

Computed tomography (CT) of the head is usually helpful only to rule out other causes of neurologic decompensation. Hemorrhagic infarction of the globus pallidus and, less frequently, the deep white matter have been reported following acute intoxication, but they are rare [54]. Imaging studies, including CT, magnetic resonance imaging (MRI), and positron-emission tomography (PET), suggest that abnormalities in the globus pallidus and deep white matter may be noted in the setting of DNS [35,55-58].


Initial treatment and disposition — Carbon monoxide (CO) is removed almost exclusively via the pulmonary circulation through competitive binding of hemoglobin by oxygen. The half-life of carboxyhemoglobin (COHb) in a patient breathing room air is approximately 250 to 320 minutes; this decreases to 90 minutes with high-flow oxygen provided via a nonrebreathing mask. Thus, the most important interventions in the management of a CO-poisoned patient are prompt removal from the source of CO and institution of high-flow oxygen by face mask. A summary table to facilitate emergent management is provided (table 1).

Comatose patients, or those with severely impaired mental status, should be intubated without delay and mechanically ventilated using 100 percent oxygen. For patients suffering from CO poisoning after smoke inhalation, it is important to consider concomitant cyanide toxicity, which can further impair tissue oxygen utilization and exacerbate cellular hypoxia [20,22]. (See "Cyanide poisoning" and "Inhalation injury from heat, smoke, or chemical irritants".)

Many patients can be managed in the emergency department, as most symptoms resolve with high-flow oxygen. Patients whose symptoms do not resolve, who demonstrate electrocardiogram (ECG) or laboratory evidence of severe poisoning, or who have other medical or social cause for concern should be hospitalized. Psychiatric assessment and determination of suicidality are crucial following intentional CO poisoning. (See "Suicidal ideation and behavior in adults".)

Many acute care hospitals lack the ability to measure carboxyhemoglobin concentrations [59]. Patients suspected of CO poisoning who are treated in hospitals without on-site cooximetry should be treated with 100 percent oxygen via nonrebreathing face mask. Clinicians at such hospitals should strongly consider the immediate transfer of patients at high risk for adverse outcomes (specifically those with acidemia, ischemic ECG changes, persistent chest pain, or altered mental status) to facilities capable of providing hyperbaric oxygen therapy.

Source identification is critical in cases of unintentional poisoning, in order to limit the risk to others. Local fire departments or other emergency management personnel can assist with an assessment of CO level in the suspected environment and removal of victims [1].

Hyperbaric oxygen — Hyperbaric oxygen therapy (HBO) involves exposing patients to 100 percent oxygen under supra-atmospheric conditions. This results in a decrease in the half-life of carboxyhemoglobin (COHb), from approximately 90 minutes on 100 percent normobaric oxygen to approximately 30 minutes during HBO. The amount of oxygen dissolved in the blood also rises from approximately 0.3 to 6.0 mL per dL, which substantially increases the delivery of non-hemoglobin-bound oxygen to the tissues. (See "Hyperbaric oxygen therapy".)

Despite the uncertainty in identifying patients who will benefit from HBO treatment, a broad set of recommendations has been established for therapy of CO-intoxicated patients (algorithm 1). In addition to a COHb level above 25 percent (see below), criteria for treatment with HBO include: evidence of ongoing end-organ ischemia (eg, profound metabolic acidosis (pH <7.1), myocardial ischemia), loss of consciousness, or in pregnant women a COHb >20 percent or evidence of fetal distress [1,2,31,60]. (See 'HBO during pregnancy' below.)

The COHb level at which HBO should be performed, independent of clinical status, is controversial. Many medical toxicologists routinely recommend HBO when the carboxyhemoglobin level is greater than 25 percent, whereas some societies use 40 percent as the appropriate threshold. There is no clear basis in the medical literature for choosing one level over the other. We routinely advocate HBO in patients with a COHb level greater than 25 percent, but recognize other clinicians may disagree based on their interpretation of existing studies.

The potential benefit of HBO is greater the more rapidly treatment is provided. Ideally, HBO should be initiated within six hours. Benefit for patients treated more than 12 hours after their CO exposure is unproven. Despite some positive studies, the use of HBO in mild to moderate CO poisoning is not routine, and we do not recommend HBO for patients other than those in the high-risk groups listed above. All patients selected to receive HBO should have at least one treatment at 2.5 to 3.0 atm as soon as possible to reverse the acute effects of CO intoxication, with possible additional therapy directed toward limitation or prevention of delayed neuropsychiatric syndrome [1,22,32,61].

The other indications, complications, and technique of HBO are discussed separately. (See "Hyperbaric oxygen therapy".)

Approximately 1500 patients with CO poisoning are treated with HBO in the United States every year [37]. A major impediment to the wider application of HBO in the management of CO poisoning is the limited availability of hyperbaric chambers. In the United States, approximately 250 hyperbaric facilities offer either single occupant ("monoplace") or multiple occupant ("multiplace") chambers. Information regarding the location of hyperbaric facilities can be accessed through the Undersea and Hyperbaric Medical Society website (www.uhms.org) or via the Divers Alert Network Emergency Hotline (1-919-684-8111) [1].

Hyperbaric oxygen and delayed neuropsychiatric syndrome (DNS) — Hyperbaric oxygen therapy (HBO) may be beneficial in preventing the late neurocognitive deficits associated with severe CO intoxication. If HBO is used, the available literature suggests that benefit is greatest if treatment begins as early as possible, ideally within six hours [62,63]. The indications for HBO are discussed above. (See 'Hyperbaric oxygen' above.)

The quality and results of clinical trials designed to assess the efficacy of HBO in reducing the severity of DNS have varied widely [3,34,43,64-70]. Of several such studies, the two most important, double-blinded trials that included all patients regardless of poisoning severity came to contradictory conclusions [64,68].

The first was a single-center, controlled trial that randomly assigned 152 patients within 24 hours of presentation to hyperbaric or normobaric oxygen therapy [64]. Treatment was administered during three sessions in a hyperbaric chamber, effectively blinding the therapy a patient was receiving. Six weeks after presentation, cognitive sequelae were more common in the group treated with normobaric oxygen (46 versus 25 percent). This advantage of hyperbaric therapy in terms of neurologic performance was maintained at six months and one year following initial presentation.

Discordant findings were noted in a randomized trial of 191 patients referred to a tertiary center with CO poisoning during a two-year period, which failed to document benefit for patients who received HBO [66]. Rather, delayed neurologic sequelae and poor performance on neuropsychiatric tests after one month were significantly more common among HBO-treated patients.

A subsequent dual-armed randomized controlled trial assessed the effect of HBO among patients with transient loss of consciousness or coma at the time of presentation following domestic CO poisoning. Among the patients with transient loss of consciousness (but without coma at presentation), the use of a single session of HBO showed no benefit in the rate of complete neurocognitive recovery at one month compared with 100 percent oxygen therapy (58 versus 61 percent; odds ratio (OR) 0.90, 95% CI 0.47-1.71). For patients presenting with coma, those undergoing two sessions of HBO experienced a significantly lower rate of recovery at one month compared with those treated with a single session of HBO (47 versus 68 percent; OR 0.42, 95% CI 0.23-0.79) [69]. This trial was stopped early based on an interim analysis suggesting harm associated with HBO among patients presenting with coma. Limitations of this trial include uncertainty surrounding the presence or absence of coma at the time of presentation (which was based upon bystander observation), the relatively large number of patients lost to follow up, and the use of 2 rather than 3 atmospheres of pressure for HBO therapy.

A systematic review of the HBO literature noted methodological flaws in all of the studies described above [70]. Limitations of the positive trial included a disparity in the severity of toxicity between patients randomized to normobaric oxygen therapy (NBO), (mean CO exposure of 22 hours), and patients treated with HBO, (mean CO exposure of 13 hours) [64]. Negative trials suffered from incomplete follow up [66,69], use of lower pressure (<2.5 atm) HBO [69], delays in treatment (mean time to HBO therapy over six hours) [66], and use of continuous NBO treatment for three days in patients not receiving HBO [66]. Such methodological problems make it difficult to draw conclusions about the ability of HBO to decrease DNS following CO poisoning.

HBO during pregnancy — The importance of pregnancy on the decision to initiate HBO is based on the greater affinity and longer half-life of CO bound to fetal hemoglobin, the inability to substantially increase placental perfusion, and the direct effects of hypoxemia and acidosis on the fetus. A prospective, multicenter study of fetal outcome following accidental CO poisoning found no physical or neurobehavioral deficits in 31 infants who were exposed to CO in utero when their mothers suffered mild to moderate CO poisoning [71]. Severe maternal poisoning, however, resulted in adverse outcomes in three of five patients treated with normobaric oxygen alone; HBO was used in two other cases, and those children did not demonstrate evidence of prenatal injury. Exposure to HBO does not seem to adversely affect the fetus, but the published experience is limited [72].

There is limited information on the characteristics of the fetal heart rate tracing of pregnant women with carbon monoxide poisoning in the third trimester. In the few cases with detailed reports, the initial tracing showed baseline fetal tachycardia of 160 to 190 beats per minute in three of four fetuses, and all four had minimal variability with no accelerations or decelerations [73,74]. After 60 to 90 minutes of maternal HBO therapy, all of the tracings became normal.

Isocapnic hyperpnea — The elimination half-life of COHb is, in part, a function of minute ventilation. Isocapnic hyperpnea is a technique by which an intubated patient is hyperventilated with a normobaric mixture of oxygen and a small amount of CO2, maintaining a PaCO2 of approximately 40 mmHg despite a sixfold increase in minute ventilation. Application of this technique in an animal model more than doubled the rate of CO elimination compared with conventional ventilation with 100 percent oxygen [75]. Furthermore, the technology required for isocapnic hyperpnea is much less expensive and cumbersome than HBO. Adding CO2 to the respiratory circuit MUST be accompanied by assisted hyperventilation or respiratory acidosis may occur, compounding the metabolic acidosis produced by carboxyhemoglobin.

A conceptually similar approach has been tested experimentally in nonintubated patients. One study randomly assigned seven patients with COHb levels of 10 to 12 percent to receive either 100 percent oxygen via face mask or FiO2 >95 percent with 4.5 to 4.8 percent CO2, in order to maintain normocapnia with a minute ventilation two to six times baseline [76]. The half-life of COHb was significantly reduced by the isocapnic hyperpnea approach (31 versus 78 minutes). To date, there are no clinical trials showing that isocapnic hyperpnea reduces the risk of delayed neurologic sequelae and larger trials are needed before this approach can be endorsed for all patients; however, the relative simplicity and low cost of this technique suggest that it might have a role in the early management of suspected CO poisoning.

PEDIATRIC CONSIDERATIONS — In young children, signs of carbon monoxide poisoning may be more subtle and nonspecific than those in adults. Infants and toddlers may present with complaints such as fussiness and feeding difficulty as the sole manifestation of carbon monoxide poisoning [77]. Because of their higher oxygen utilization and higher minute ventilation, young children may develop signs and symptoms of carbon monoxide poisoning before older children and adults who experience the same exposure (eg, family members living in a house with a faulty furnace).

In contrast, older children have symptoms similar to adults as they are able to verbalize feelings of headache and nausea. The incidence of delayed neuropsychological sequelae in the pediatric population falls between 3 and 17 percent (lower than that reported in adults) [78-80].

Patient age does not alter the management of carbon monoxide poisoning. Although theoretical reasons for both more or less aggressive treatment in young children exist, there are no human studies upon which to base a recommendation and treatment centers generally do not alter their approach to therapy based upon age.

When hyperbaric oxygen therapy (HBO) is administered to young children, specific concerns must be addressed:

Myringotomy should be performed in children with active otitis media who are younger than five years or who are unable to equalize their middle ear pressure [81].

It may be helpful to allow a family member to accompany a frightened child into the chamber.

When HBO is administered to an infant, care must be taken to keep the infant warm: Children in this age group easily become hypothermic.

Congenital abnormalities must also be taken in account:

A chest radiograph should be obtained to detect congenital anomalies such as lobar emphysema which could lead to a pneumothorax.

Patients with unpalliated ductal dependent cardiac lesions should undergo HBO with caution as oxygen may precipitate duct closure. In most cases, such patients are poor candidates for HBO. Consultation with a pediatric cardiologist is prudent.

PROGNOSIS — The prognosis of patients who sustain hypoxic-ischemic brain injury from any cause, including severe CO poisoning, is reviewed separately. (See "Hypoxic-ischemic brain injury: Evaluation and prognosis", section on 'Prognosis based on clinical findings'.)

PREVENTION — Home carbon monoxide (CO) monitors equipped with alarms are relatively inexpensive, widely available, and potentially life-saving. The United States Consumer Product Safety Commission (CPSC) recommends that every home have a CO monitor equipped with an alarm; further information is available online www.cpsc.gov or via their hotline (1-800-638-2772). Given the evidence that CO is capable of diffusing rapidly through standard wallboard and floorboard materials, we recommend that every home (even those without an obvious source of CO) be equipped with a CO monitor [82].

ADDITIONAL RESOURCES — Regional poison control centers in the United States are available at all times for consultation on patients who are critically ill, require admission, or have clinical pictures that are unclear (1-800-222-1222). In addition, some hospitals have clinical and/or medical toxicologists available for bedside consultation and/or inpatient care. Whenever available, these are invaluable resources to help in the diagnosis and management of ingestions or overdoses. The World Health Organization provides a listing of international poison centers at its website: www.who.int/gho/phe/chemical_safety/poisons_centres/en/index.html

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)

Basics topic (see "Patient education: Carbon monoxide (CO) poisoning (The Basics)")

SUMMARY AND RECOMMENDATIONS — Carbon monoxide (CO) poisoning is common, potentially fatal, and probably underdiagnosed because of its nonspecific clinical presentation. CO has profound effects on oxygen transport and, to a lesser degree, peripheral oxygen utilization. In addition to the extended summary below, a brief table to facilitate emergent management is provided (table 1).

CO poisoning is most common during winter in cold climates, but it may occur in all seasons and environments. Smoke inhalation is responsible for most unintentional cases. Other potential sources of CO include poorly functioning heating systems, improperly vented fuel-burning devices (eg, kerosene heaters, charcoal grills, camping stoves, gasoline-powered electrical generators), and motor vehicles operating in poorly ventilated areas. (See 'Epidemiology' above.)

CO diffuses rapidly across the pulmonary capillary membrane and binds to the iron moiety of heme with approximately 240 times the affinity of oxygen. The degree of carboxyhemoglobinemia (COHb) is a function of the relative amounts of CO and oxygen in the environment, duration of exposure, and minute ventilation. (See 'Pathophysiology' above.)

The clinical findings of CO poisoning are highly variable and largely nonspecific. Mild to moderate CO-intoxicated patients often present with constitutional symptoms, including headache (most common), malaise, nausea, and dizziness, and may be misdiagnosed with acute viral syndromes. In the absence of concurrent trauma or burns, physical findings in CO poisoning are usually confined to alterations in mental status, ranging from mild confusion to seizures and coma. A careful neurologic examination is crucial. (See 'Clinical presentation' above.)

Cardiac ischemia can occur. Once the diagnosis of CO intoxication is confirmed, we recommend obtaining an electrocardiogram (ECG); cardiac biomarker evaluation is warranted in patients with ECG evidence of ischemia, symptoms suggestive of ischemia, age greater than 65 years, or a history of cardiac disease or cardiac risk factors. (See 'Clinical presentation' above.)

The diagnosis of CO poisoning is based upon a compatible history and physical examination in conjunction with an elevated carboxyhemoglobin level. Diagnosis requires quantification by cooximetry of a blood gas sample; standard pulse oximetry (SpO2) is unable to distinguish between oxyhemoglobin and COHb. Blood PO2 measurements tend to be normal because PO2 reflects O2 dissolved in blood, and this process is not affected by CO. (See 'Diagnosis' above.)

Continual assessment and standard interventions to secure the airway, breathing, and circulation of the patient poisoned with CO are of paramount importance. We recommend that all comatose patients and those with severely impaired mental status be intubated without delay and mechanically ventilated (Grade 1B). (See 'Management' above.)

The most important interventions in the management of a CO-poisoned patient are prompt removal from the source of CO and institution of 100 percent oxygen by nonrebreathing face mask or endotracheal tube. We recommend initial treatment with 100 percent normobaric oxygen for all suspected victims of CO poisoning, regardless of pulse oximetry or arterial PO2 (Grade 1B). (See 'Management' above.)

In patients who have severe intoxication, there is controversy surrounding the ability of hyperbaric oxygen therapy (HBO) to decrease the incidence and severity of delayed neurocognitive deficits following CO poisoning. Expert guidelines regarding the judicious use of hyperbaric oxygen have been published (algorithm 1). We suggest treatment with HBO in the following circumstances (Grade 2B):

CO level >25 percent

CO level >20 percent in pregnant patient

Loss of consciousness

Severe metabolic acidosis (pH <7.1)

Evidence of end-organ ischemia (eg, ECG changes, chest pain, altered mental status) (see 'Management' above)

Many patients with mild symptoms from an unintentional poisoning can be managed in an emergency department and safely discharged. Patients whose symptoms do not resolve, who demonstrate ECG or laboratory evidence of severe poisoning, or who have other medical or social cause for concern should be hospitalized. Psychiatric evaluation must be obtained for all patients with self-inflicted CO poisoning. (See 'Management' above.)

Determination of the mechanism of exposure is critical in cases of unintentional poisoning in order to limit the risk to others. Local fire departments can assist with an assessment of CO level and removal of victims in the suspected environment. (See 'Management' above.)

Use of UpToDate is subject to the  Subscription and License Agreement.


  1. Ernst A, Zibrak JD. Carbon monoxide poisoning. N Engl J Med 1998; 339:1603.
  2. Weaver LK. Carbon monoxide poisoning. Crit Care Clin 1999; 15:297.
  3. Tibbles PM, Perrotta PL. Treatment of carbon monoxide poisoning: a critical review of human outcome studies comparing normobaric oxygen with hyperbaric oxygen. Ann Emerg Med 1994; 24:269.
  4. Hampson NB. U.S. Mortality Due to Carbon Monoxide Poisoning, 1999-2014. Accidental and Intentional Deaths. Ann Am Thorac Soc 2016; 13:1768.
  5. Rose JJ, Wang L, Xu Q, et al. Carbon Monoxide Poisoning: Pathogenesis, Management, and Future Directions of Therapy. Am J Respir Crit Care Med 2017; 195:596.
  6. www.cdc.gov/nceh/airpollution/carbonmonoxide/cofaq.htm (Accessed on August 09, 2005).
  7. Centers for Disease Control and Prevention (CDC). Unintentional non-fire-related carbon monoxide exposures--United States, 2001-2003. MMWR Morb Mortal Wkly Rep 2005; 54:36.
  8. Hardy KR, Thom SR. Pathophysiology and treatment of carbon monoxide poisoning. J Toxicol Clin Toxicol 1994; 32:613.
  9. Hampson NB, Hauff NM. Risk factors for short-term mortality from carbon monoxide poisoning treated with hyperbaric oxygen. Crit Care Med 2008; 36:2523.
  10. Thomassen Ø, Brattebø G, Rostrup M. Carbon monoxide poisoning while using a small cooking stove in a tent. Am J Emerg Med 2004; 22:204.
  11. Centers for Disease Control and Prevention (CDC). Carbon monoxide poisoning from hurricane-associated use of portable generators--Florida, 2004. MMWR Morb Mortal Wkly Rep 2005; 54:697.
  12. Hampson NB, Dunn SL. Carbon Monoxide Poisoning from Portable Electrical Generators. J Emerg Med 2015; 49:125.
  13. Centers for Disease Control and Prevention (CDC). Carbon monoxide poisonings resulting from open air exposures to operating motorboats--Lake Havasu City, Arizona, 2003. MMWR Morb Mortal Wkly Rep 2004; 53:314.
  14. Centers for Disease Control and Prevention (CDC). Carbon monoxide releases and poisonings attributed to underground utility cable fires--New York, January 2000-June 2004. MMWR Morb Mortal Wkly Rep 2004; 53:920.
  15. Centers for Disease Control and Prevention (CDC). Monitoring poison control center data to detect health hazards during hurricane season--Florida, 2003-2005. MMWR Morb Mortal Wkly Rep 2006; 55:426.
  16. Cukor J, Restuccia M. Carbon monoxide poisoning during natural disasters: the Hurricane Rita experience. J Emerg Med 2007; 33:261.
  17. Chang YL, Yang CC, Deng JF, et al. Diverse manifestations of oral methylene chloride poisoning: report of 6 cases. J Toxicol Clin Toxicol 1999; 37:497.
  18. Amsel J, Soden KJ, Sielken RL Jr, Valdez-Flora C. Observed versus predicted carboxyhemoglobin levels in cellulose triacetate workers exposed to methylene chloride. Am J Ind Med 2001; 40:180.
  19. Yasuda H, Yamaya M, Nakayama K, et al. Increased arterial carboxyhemoglobin concentrations in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2005; 171:1246.
  20. Vogel SN, Sultan TR, Ten Eyck RP. Cyanide poisoning. Clin Toxicol 1981; 18:367.
  21. Tritapepe L, Macchiarelli G, Rocco M, et al. Functional and ultrastructural evidence of myocardial stunning after acute carbon monoxide poisoning. Crit Care Med 1998; 26:797.
  22. Ellenhorn's Medical Toxicology: Diagnosis and Treatment of Human Poisoning, 2nd ed, Ellenhorn MJ, Schonwald S, Ordog G, Wasserberger J (Eds), Williams & Wilkins, Baltimore 1997.
  23. Norris JC, Moore SJ, Hume AS. Synergistic lethality induced by the combination of carbon monoxide and cyanide. Toxicology 1986; 40:121.
  24. Thom SR, Bhopale VM, Fisher D, et al. Delayed neuropathology after carbon monoxide poisoning is immune-mediated. Proc Natl Acad Sci U S A 2004; 101:13660.
  25. Goldbaum LR, Ramirez RG, Absalon KB. What is the mechanism of carbon monoxide toxicity? Aviat Space Environ Med 1975; 46:1289.
  26. Zhang J, Piantadosi CA. Mitochondrial oxidative stress after carbon monoxide hypoxia in the rat brain. J Clin Invest 1992; 90:1193.
  27. Thom SR. Carbon monoxide-mediated brain lipid peroxidation in the rat. J Appl Physiol (1985) 1990; 68:997.
  28. Thom SR, Xu YA, Ischiropoulos H. Vascular endothelial cells generate peroxynitrite in response to carbon monoxide exposure. Chem Res Toxicol 1997; 10:1023.
  29. Tomaszewski C. Carbon monoxide poisoning. Early awareness and intervention can save lives. Postgrad Med 1999; 105:39.
  30. Harper A, Croft-Baker J. Carbon monoxide poisoning: undetected by both patients and their doctors. Age Ageing 2004; 33:105.
  31. Kao LW, Nañagas KA. Carbon monoxide poisoning. Emerg Med Clin North Am 2004; 22:985.
  32. Satran D, Henry CR, Adkinson C, et al. Cardiovascular manifestations of moderate to severe carbon monoxide poisoning. J Am Coll Cardiol 2005; 45:1513.
  33. Henry CR, Satran D, Lindgren B, et al. Myocardial injury and long-term mortality following moderate to severe carbon monoxide poisoning. JAMA 2006; 295:398.
  34. 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; 25:474.
  35. Choi IS. Delayed neurologic sequelae in carbon monoxide intoxication. Arch Neurol 1983; 40:433.
  36. Kwon OY, Chung SP, Ha YR, et al. Delayed postanoxic encephalopathy after carbon monoxide poisoning. Emerg Med J 2004; 21:250.
  37. Hampson NB, Little CE. Hyperbaric treatment of patients with carbon monoxide poisoning in the United States. Undersea Hyperb Med 2005; 32:21.
  38. Seger D, Welch L. Carbon monoxide controversies: neuropsychologic testing, mechanism of toxicity, and hyperbaric oxygen. Ann Emerg Med 1994; 24:242.
  39. Bozeman WP, Myers RA, Barish RA. Confirmation of the pulse oximetry gap in carbon monoxide poisoning. Ann Emerg Med 1997; 30:608.
  40. Tremper KK, Barker SJ. Pulse oximetry. Anesthesiology 1989; 70:98.
  41. Barker SJ, Curry J, Redford D, Morgan S. Measurement of carboxyhemoglobin and methemoglobin by pulse oximetry: a human volunteer study. Anesthesiology 2006; 105:892.
  42. Roth D, Herkner H, Schreiber W, et al. Accuracy of noninvasive multiwave pulse oximetry compared with carboxyhemoglobin from blood gas analysis in unselected emergency department patients. Ann Emerg Med 2011; 58:74.
  43. American College of Emergency Physicians Clinical Policies Subcommittee (Writing Committee) on Carbon Monoxide Poisoning:, Wolf SJ, Maloney GE, et al. Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Acute Carbon Monoxide Poisoning. Ann Emerg Med 2017; 69:98.
  44. Touger M, Gallagher EJ, Tyrell J. Relationship between venous and arterial carboxyhemoglobin levels in patients with suspected carbon monoxide poisoning. Ann Emerg Med 1995; 25:481.
  45. Lopez DM, Weingarten-Arams JS, Singer LP, Conway EE Jr. Relationship between arterial, mixed venous, and internal jugular carboxyhemoglobin concentrations at low, medium, and high concentrations in a piglet model of carbon monoxide toxicity. Crit Care Med 2000; 28:1998.
  46. Livshits Z, Lugassy DM, Shawn LK, Hoffman RS. Falsely low carboxyhemoglobin level after hydroxocobalamin therapy. N Engl J Med 2012; 367:1270.
  47. Lee J, Mukai D, Kreuter K, et al. Potential interference by hydroxocobalamin on cooximetry hemoglobin measurements during cyanide and smoke inhalation treatments. Ann Emerg Med 2007; 49:802.
  48. Borron SW, Uhl W, Nolting A, Hostalek U. Interference of the cyanide antidote hydroxocobalamin with carboxyhemoglobin measurements should not limit clinical use in suspected or confirmed cyanide poisoning. Ann Emerg Med 2007; 50:624.
  49. Baud F. Clarifications regarding interference of hydroxocobalamin with carboxyhemoglobin measurements in victims of smoke inhalation. Ann Emerg Med 2007; 50:625.
  50. Pace R, Bon Homme M, Hoffman RS, Lugassy D. Effects of hydroxocobalamin on carboxyhemoglobin measured under physiologic and pathologic conditions. Clin Toxicol (Phila) 2014; 52:647.
  51. Plante T, Harris D, Savitt J, et al. Carboxyhemoglobin monitored by bedside continuous CO-oximetry. J Trauma 2007; 63:1187.
  52. O'Malley GF. Non-invasive carbon monoxide measurement is not accurate. Ann Emerg Med 2006; 48:477.
  53. Touger M, Birnbaum A, Wang J, et al. Performance of the RAD-57 pulse CO-oximeter compared with standard laboratory carboxyhemoglobin measurement. Ann Emerg Med 2010; 56:382.
  54. Finelli PF, DiMario FJ Jr. Hemorrhagic infarction in white matter following acute carbon monoxide poisoning. Neurology 2004; 63:1102.
  55. Zagami AS, Lethlean AK, Mellick R. Delayed neurological deterioration following carbon monoxide poisoning: MRI findings. J Neurol 1993; 240:113.
  56. Teksam M, Casey SO, Michel E, et al. Diffusion-weighted MR imaging findings in carbon monoxide poisoning. Neuroradiology 2002; 44:109.
  57. Kim JH, Chang KH, Song IC, et al. Delayed encephalopathy of acute carbon monoxide intoxication: diffusivity of cerebral white matter lesions. AJNR Am J Neuroradiol 2003; 24:1592.
  58. Chu K, Jung KH, Kim HJ, et al. Diffusion-weighted MRI and 99mTc-HMPAO SPECT in delayed relapsing type of carbon monoxide poisoning: evidence of delayed cytotoxic edema. Eur Neurol 2004; 51:98.
  59. Hampson NB, Scott KL, Zmaeff JL. Carboxyhemoglobin measurement by hospitals: implications for the diagnosis of carbon monoxide poisoning. J Emerg Med 2006; 31:13.
  60. Hampson NB, Dunford RG, Kramer CC, Norkool DM. Selection criteria utilized for hyperbaric oxygen treatment of carbon monoxide poisoning. J Emerg Med 1995; 13:227.
  61. Leach RM, Rees PJ, Wilmshurst P. Hyperbaric oxygen therapy. BMJ 1998; 317:1140.
  62. Ziser A, Shupak A, Halpern P, et al. Delayed hyperbaric oxygen treatment for acute carbon monoxide poisoning. Br Med J (Clin Res Ed) 1984; 289:960.
  63. Goulon M, Barois, Rapin M, et al. Carbon monoxide poisoning and acute anoxia due to breathing coal gas and hydrocarbons. J Hyperbar MEd 1986; 1:23. http://archive.rubicon-foundation.org/xmlui/handle/123456789/4302 (Accessed on October 06, 2011).
  64. Weaver LK, Hopkins RO, Chan KJ, et al. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med 2002; 347:1057.
  65. Ducassé JL, Celsis P, Marc-Vergnes JP. Non-comatose patients with acute carbon monoxide poisoning: hyperbaric or normobaric oxygenation? Undersea Hyperb Med 1995; 22:9.
  66. 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; 170:203.
  67. Raphael JC, Elkharrat D, Jars-Guincestre MC, et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989; 2:414.
  68. Juurlink DN, Buckley NA, Stanbrook MB, et al. Hyperbaric oxygen for carbon monoxide poisoning. Cochrane Database Syst Rev 2005; :CD002041.
  69. Annane D, Chadda K, Gajdos P, et al. Hyperbaric oxygen therapy for acute domestic carbon monoxide poisoning: two randomized controlled trials. Intensive Care Med 2011; 37:486.
  70. Buckley NA, Juurlink DN, Isbister G, et al. Hyperbaric oxygen for carbon monoxide poisoning. Cochrane Database Syst Rev 2011; :CD002041.
  71. Koren G, Sharav T, Pastuszak A, et al. A multicenter, prospective study of fetal outcome following accidental carbon monoxide poisoning in pregnancy. Reprod Toxicol 1991; 5:397.
  72. Elkharrat D, Raphael JC, Korach JM, et al. Acute carbon monoxide intoxication and hyperbaric oxygen in pregnancy. Intensive Care Med 1991; 17:289.
  73. Van Hoesen KB, Camporesi EM, Moon RE, et al. Should hyperbaric oxygen be used to treat the pregnant patient for acute carbon monoxide poisoning? A case report and literature review. JAMA 1989; 261:1039.
  74. Towers CV, Corcoran VA. Influence of carbon monoxide poisoning on the fetal heart monitor tracing: a report of 3 cases. J Reprod Med 2009; 54:184.
  75. Fisher JA, Rucker J, Sommer LZ, et al. Isocapnic hyperpnea accelerates carbon monoxide elimination. Am J Respir Crit Care Med 1999; 159:1289.
  76. Takeuchi A, Vesely A, Rucker J, et al. A simple "new" method to accelerate clearance of carbon monoxide. Am J Respir Crit Care Med 2000; 161:1816.
  77. Rudge FW. Carbon monoxide poisoning in infants: treatment with hyperbaric oxygen. South Med J 1993; 86:334.
  78. Meert KL, Heidemann SM, Sarnaik AP. Outcome of children with carbon monoxide poisoning treated with normobaric oxygen. J Trauma 1998; 44:149.
  79. Kim JK, Coe CJ. Clinical study on carbon monoxide intoxication in children. Yonsei Med J 1987; 28:266.
  80. Cho CH, Chiu NC, Ho CS, Peng CC. Carbon monoxide poisoning in children. Pediatr Neonatol 2008; 49:121.
  81. Waisman D, Shupak A, Weisz G, Melamed Y. Hyperbaric oxygen therapy in the pediatric patient: the experience of the Israel Naval Medical Institute. Pediatrics 1998; 102:E53.
  82. Hampson NB, Courtney TG, Holm JR. Diffusion of carbon monoxide through gypsum wallboard. JAMA 2013; 310:745.
Topic 322 Version 24.0

All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.