Carbon monoxide (CO) is a toxic gas that is formed from the incomplete combustion of carbon-containing substances. Common sources are engines, non-electric space heaters, furnaces, and smoke from fires. Poisoning can also occur from inhalation or absorption of methylene chloride (dichloromethane) through the skin. This chemical which is commonly used in solvents for paint removing and is metabolized to CO in the liver. CO binds with a higher affinity than oxygen for the hemoglobin molecule in red blood cells forming carboxyhemoglobin (COHb), which impairs oxygen offloading resulting in hypoxia in the body’s tissues especially the brain and heart. CO also binds to the cytochrome proteins in mitochondrial impairing cellular respiration. Another mechanism of toxicity is from the precipitation of inflammation in the brain resulting in lipid peroxidation and the delayed neurologic sequelae (DNS). 
Because CO is an odorless, tasteless, and non-irritating gas, in a poorly ventilated area, people exposed to CO can quickly succumb to its toxicity. The result is a loss of consciousness and eventually death. Poisoning can be intentional, unintentional, or for unknown reasons. It is always best to be suspicious of attempted suicide until a complete history and further patient observation determine the cause of the poisoning. Unintentional poisoning is seasonal and weather-related; for example, during winter cold spells and after natural disasters when electric power service is interrupted, and people are inclined to run generators or heaters in enclosed areas, poisonings are more common. Cases of poisoning that are more random can occur because of faulty stoves, charcoal grills, and improperly ventilated or defective furnaces or heaters used indoors. Cases can often be clustered when entire families or many workers come down with similar symptoms of toxicity.
CO poisoning is a major health problem with up to 40 000 emergency department visits per year and up to about 6000 deaths per year in the United States alone. Most of the deaths are declared intentional, but about 500 are unintentional. Most inadvertent cases are from smoke inhalation. Common causes include working in enclosed areas with combustion engines running, faulty propane or natural gas heaters, and house fires. Workers using paint stripper containing methylene chloride in poorly ventilated areas and with the lack of protective respirators or gloves are at risk of inadvertent poisoning because of the metabolism of the methylene chloride to CO in the liver. Children and elderly with comorbid health problems, especially cardiac, neurological, and pulmonary disease, are at increased risk of poisoning because of lesser physiologic reserve to tolerate the exposure.
CO is readily absorbed through the lungs and binds to the hemoglobin molecule in red blood cells with an affinity that is 240 times that of oxygen. The amount of carboxyhemoglobin formed is dependent on the amount of CO in the environment, duration of exposure and minute ventilation. The COHb impairs the offloading of the remaining oxygen that is bound to the hemoglobin molecule (usually when fully oxygenated the hemoglobin molecule can carry four oxygen molecules). The result of this impairment is decreased oxygen delivery to the tissues and relative hypoxia. Not only is the oxygen delivery impaired, but 10% to 15% of the CO binds to myoglobin, cytochromes, and NADPH reductase impairing oxidative phosphorylation in the mitochondria of the cells. This is especially toxic to the heart muscle and the brain. This results in the classic symptoms of a headache, dizziness, nausea, vomiting, flu-like symptoms, fatigue, shortness of breath, impaired judgment, and cognitive dysfunction. The patient may also develop chest pain, agitation, abdominal pain, drowsiness, loss of consciousness, visual changes, and seizures. The sequelae of CO poisoning can include depression, memory problems, ataxia, behavior problems, and impaired cognitive function. The cause of the delayed neurologic sequelae is believed to be linked to the lipid peroxidation by toxic oxygen species that are generated by xanthine oxidase in the brain. This enzyme is released by neutrophils that are adherent to the damaged endothelial cells of the blood vessels in the brain as a result of the CO toxicity. This creates a common pathway that is seen with ischemia-reperfusion injuries.
CO is absorbed rapidly through the lungs, and most of it ends up bound to the hemoglobin molecules in red blood cells, resulting in the leftward shifting of the oxygen dissociation curve of the remaining oxygen bound to the hemoglobin, impairing oxygen delivery to the tissues. About 15% of the CO ends up bound to myoglobin, cytochromes, and NADPH reductase impairing cellular respiration. CO is eliminated from the body through oxygenation and to a lesser degree by ventilation. If the patient is on room air the half-life of the CO is about 300 minutes, but if the patient is on a non-rebreather, with high flow oxygen, this is reduced to 90 minutes, and under hyperbaric oxygen conditions, 100% oxygen at three atmospheres absolute pressure, the half-life is about 30 minutes.
The presenting signs and symptoms can be vague and nonspecific; therefore, one needs to have a high index of suspicion to avoid misdiagnosis. The incidence of CO poisoning is higher during cold weather months when there is also a higher incidence of viral gastroenteritis infections, and many of the symptoms can overlap. Common symptoms of CO toxicity are malaise, nausea, vomiting, dizziness, and headache. More severe cases can result in impaired cognitive function, loss of consciousness, seizures, and coma. It is important to suspect a loss of consciousness if the patient is unable to give a history of recent events. The classic “cherry red” lips and skin appearance associated with CO poisoning is insensitive and often not present. Cardiac toxicity can present with chest pain, myocardial ischemia, ventricular arrhythmia, acute congestive heart failure, and often profound metabolic lactic acidosis. Anyone with smoke inhalation, unexplained loss of consciousness, or coma should have the diagnosis of CO poisoning entertained. If the patient is conscious, it is helpful to test cognitive function with a mini-mental status exam. Dysdiadochokinesis (difficulty with rapid alternating movements) is impaired with CO poisoning and is a quick means of assessing cerebellar function. It is reasonable to assume that if the patient has cerebellar dysfunction from CO poisoning, then the rest of the brain function will probably be impaired.
CO-oximetry utilizes multiple wavelengths of light to measure oxyhemoglobin and carboxyhemoglobin, and some meters can also measure methemoglobin. Care should be taken in the placement of the meter probe and to make sure extraneous light or fingernail polish is not present, which could cause an erroneous reading.  If the clinical index of suspicion is high or if the reading is high, then carboxyhemoglobin should be done on a blood sample as close to the suspect CO exposure time as possible. The interpretation of the SpCO levels should be done in the setting of the clinical presentation:
CO-oximetry should not be relied upon solely because it has not been fully tested in clinical trials. Laboratory evaluation of suspected CO poisoning should include direct blood COHb level, complete blood count (CBC), comprehensive metabolic panel, lactic acid and blood gas can be very helpful. The examiner should also consider other concomitant toxicity such as alcohol intoxication, cyanide poisoning, inhalation of smoke (especially the burning of plastics in a fire), and urine toxicology if suspected drug abuse. All unconscious patients should have blood sugar levels tested and be closely monitored. The patient should have an ECG, and cardiac enzymes checked looking for cardiac toxicity. A chest x-ray is helpful if looking for pulmonary edema and is especially important if the patient is suffering from smoke inhalation. If the case is from a suspected suicide attempt, it is good practice to do a salicylate and acetaminophen level. CT scan of the brain is not helpful unless trying to exclude other causes of a headache and neurological impairment, such as intracranial bleeding. There has been an association of CO poisoning with hemorrhage of the globus pallidus and to a less extent white matter, but this is a rare finding.
Placing the patient on high-flow oxygen via non-rebreather mask is paramount in the treatment of suspected or confirmed CO poisoning, and removal from the source of the CO. If the patient is comatose or unable to protect their airway because neurologic impairment is severe, then the patient should be intubated and placed on 100% oxygen. Close monitoring is mandatory for cardiac dysrhythmia, and the treatment of concomitant conditions is essential to maximizing good outcomes. For example, with smoke inhalation and suspected cyanide poisoning should be treated with hydroxocobalamin kit or an emergency cyanide kit (note that hydroxocobalamin can interfere with the measurement of the COHb, but one should err on the side of treating for CO poisoning knowing that the result may be inaccurate). It is critical to differentiate intentional from unintentional poisoning, but if the reason for poisoning is unknown, it is better to be suspicious of a potential suicide attempt. If the patient is conscious, an attempt at assessing suicidal ideation should be done. This will help with patient management after the acute toxicity has been addressed. The decision to treat the patient with hyperbaric oxygen therapy (HBO2) can be controversial, but most experts agree that if there is loss of consciousness at the scene or in hospital, new neurologic deficits or mental status changes, end-organ ischemia (ECG changes, pH less than 7.1), or if the patient is pregnant (especially if COHb more than 20%), then treatment is indicated. 
Don't treat based only on how high the COHb is because levels do not always correlate with clinical symptoms, but if the COHb is greater than 25%, this is considered a severe poisoning, and most experts agree that treatment is justified. It is reasonable to err on the side of treating with HBO2 because it is difficult to predict which patients will go on to develop delayed neuropsychiatric syndrome (DNS), which is a significant problem in 40% of severe cases. This syndrome can arise in usually about 20 days after poisoning (range three to 240 days) and manifests as cognitive deficits, personality changes, movement disorders, and focal neurologic deficits. The deficits usually resolve after one year but may become permanent. HBO2 helps at reducing the risk of DNS by helping with the ischemia-reperfusion injury in the central nervous system. The mechanism of action of HBO2 is not only to cause more rapid displacement of the CO from hemoglobin and generate normal oxyhemoglobin in the red blood cell, but the oxygen delivery to the tissue is also increased by the oxygen that is dissolved in the plasma. The other benefit of the HBO2 is to reduce the adherence of neutrophils on the damaged endothelium of the blood vessels in the brain, which reduces tissue edema and reduces the lipid peroxidation. The extra oxygen that is delivered to the tissue also helps to reduce the CO bound to the cytochrome proteins in the mitochondria of the cells and contributes to restoring oxidative phosphorylation, which neurons are especially dependent upon for normal function and survival from the ischemic insult. It is also controversial as to the number of HBO2 treatments and what is the best treatment profile. Most hyperbaric oxygen practitioners would treat by the Weaver protocol or modification of this protocol.
Whichever treatment profile (dive table) is used, the goal of the therapy is the same: provide 100% oxygen under two to three atmospheres absolute pressure (ATA) for 60 to 90 minutes to clear the carbon monoxide faster and to help with an ischemia-reperfusion injury. With more severe cases of CO poisoning, treat with two to three treatments on a twice a day basis to help resolve the reperfusion injury. In moderately severe cases with milder neurologic deficits and symptoms, then one treatment is sufficient. The timing of the HBO2 is important and derives the most benefit if done within the first 6 hours after exposure. Once the patient has been adequately medically treated, then the disposition will depend on the clinical circumstances. Most patients with unintentional poisoning can be sent home with instructions to return if neurological defects develop. Have the patient follow-up with their primary care physician to do formal neurologic testing if deficits persist or recur. Patients with suicide attempt need a referral for formal psychiatric care. Most of the time, this done as inpatient care under involuntary retention.
Any patient suspected of intentional poisoning should be referred to a psychiatrist.
There is a lack of double-blind, placebo-controlled, randomized trials examining the treatment of CO poisoning, and the best treatment recommendations are controversial. What is problematic with most of the studies is that the patient's baseline neurologic function is not known and assumed to be normal. Some studies have shown worse outcomes with the HBO2 treated group than with the normobaric oxygen group (or sham treated group), but there is a potential bias in the studies because usually more severely poisoned patients tend to get treatment with HBO2. There is also insufficient neuropsychiatric testing, and long-term follow-up of the patients to differentiate long-term outcome. More studies are needed to determine the best practice guidelines for CO poisoning.
It is better to err on the side of treating with HBO2 because the benefits of treating outweigh the risks. The only major contraindication for HBO2 is a patient with untreated pneumothorax. The benefit of the HBO2 is faster restoring of normal oxygenation of hemoglobin and improved oxygen delivery, and reducing reperfusion injury to ischemic tissue. Also, there is a benefit to the reduction and prevention of delayed neuropsychiatric syndromes which are difficult to predict based on the poisoning. There is also the special case of treating pregnant patients in which the fetal hemoglobin has a greater affinity for CO than hemoglobin in the mother's red blood cells. The fetus acts as a sink for the CO, so pregnant patients may have fewer symptoms and lower COHb when compared to patients with similar exposures. Because of the increased toxicity of CO for the fetus and difficulty with evaluating, I recommend treating the pregnant patient with HBO2.
CO poisoning is not rare at all. This colorless and odorless often presents with a headache, confusion, coma and even death. Even though hyperbaric oxygen can lower the morbidity of the toxicity, a significant number of survivors end up with residual neurological and psychiatric deficits. In the majority of cases, CO poisoning can be prevented. Besides physicians, the role of the pharmacist and nurse is important. The nurse is usually the last professional to see the patient prior to discharge and thus has a great opportunity to educate the family on use of home CO detectors. The pharmacist gets the opportunity to see consumers regularly and should recommend cessation of smoking and ensuring that the family has a CO alarm at home. Also, both these professionals should educate the family on use of proper heating appliances during winter and having the home checked regularly for safety. (Level V)
The prognosis for patients with CO poisoning depends on the severity of the poisoning and clinical status at presentation. Patients who arrive at the ER in a coma or cardiac arrest usually have a poor outcome. In addition, patients who have a demonstrated abnormality on the MRI also tend to have residual neuropsychiatric deficits. (Level V)
|||Kim HH,Choi S, Therapeutic Aspects of Carbon Monoxide in Cardiovascular Disease. International journal of molecular sciences. 2018 Aug 13 [PubMed PMID: 30104479]|
|||Lettow I,Hoffmann A,Burmeister HP,Toepper R, [Delayed neuropsychological sequelae after carbon monoxide poisoning]. Fortschritte der Neurologie-Psychiatrie. 2018 Jun [PubMed PMID: 29954016]|
|||Ng PC,Long B,Koyfman A, Author Correction: Clinical chameleons: an emergency medicine focused review of carbon monoxide poisoning. Internal and emergency medicine. 2018 Aug [PubMed PMID: 29569089]|
|||Ng PC,Long B,Koyfman A, Clinical chameleons: an emergency medicine focused review of carbon monoxide poisoning. Internal and emergency medicine. 2018 Mar [PubMed PMID: 29435715]|
|||Sethuraman KN,Douglas TM,Bostick BB,Comer AC,Myers B,Rosenthal RE, Clinical Characteristics of Pediatric Patients With Carbon Monoxide Poisoning. Pediatric emergency care. 2018 Jan 2 [PubMed PMID: 29298249]|
|||Hanley ME,Patel PH, Toxicity, Carbon Monoxide null. 2018 Jan [PubMed PMID: 28613491]|
|||Otterness K,Ahn C,Nusbaum J,Gupta N, Emergency department management of smoke inhalation injury in adults [digest]. Emergency medicine practice. 2018 Mar 1 [PubMed PMID: 29489307]|
|||Wolf SJ,Maloney GE,Shih RD,Shy BD,Brown MD, Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Acute Carbon Monoxide Poisoning. Annals of emergency medicine. 2017 Jan [PubMed PMID: 27993310]|
|||Drinhaus H,Nüsgen S,Hinkelbein J, [Guidelines desirable for treatment of carbon monoxide poisoning]. Der Anaesthesist. 2016 Apr [PubMed PMID: 26961767]|
|||Mutluoglu M,Metin S,Ibrahim Arziman,Uzun G,Yildiz S, The use of hyperbaric oxygen therapy for carbon monoxide poisoning in Europe. Undersea [PubMed PMID: 27000013]|
|||Mintegi S,Clerigue N,Tipo V,Ponticiello E,Lonati D,Burillo-Putze G,Delvau N,Anseeuw K, Pediatric cyanide poisoning by fire smoke inhalation: a European expert consensus. Toxicology Surveillance System of the Intoxications Working Group of the Spanish Society of Paediatric Emergencies. Pediatric emergency care. 2013 Nov [PubMed PMID: 24196100]|
|||Akcan Yildiz L,Gultekingil A,Kesici S,Bayrakci B,Teksam O, Predictors of Severe Clinical Course in Children With Carbon Monoxide Poisoning. Pediatric emergency care. 2018 Aug 13 [PubMed PMID: 30106865]|
|||Kuo SC,Hsu CK,Tsai CT,Chieh MJ, [Hyperbaric Oxygen Therapy and Acute Carbon Monoxide Poisoning]. Hu li za zhi The journal of nursing. 2018 Aug [PubMed PMID: 30066318]|
|||Lee J,Fan NC,Yao TC,Hsia SH,Lee EP,Huang JL,Wu HP, Clinical spectrum of acute poisoning in children admitted to the pediatric emergency department. Pediatrics and neonatology. 2018 Apr 19 [PubMed PMID: 29748113]|
|||Hampson NB, Cost effectiveness of residential carbon monoxide alarms. Undersea [PubMed PMID: 29116693]|
|||Rose JJ,Nouraie M,Gauthier MC,Pizon AF,Saul MI,Donahoe MP,Gladwin MT, Clinical Outcomes and Mortality Impact of Hyperbaric Oxygen Therapy in Patients With Carbon Monoxide Poisoning. Critical care medicine. 2018 Jul [PubMed PMID: 29629990]|