CARBON MONOXIDE

Classification | Detailed evidence-based information

Substances Included Synonyms

Therapeutic Toxic Class

Specific Substances

1) Carbonic oxide
2) Carbon monoxide
3) Carbon oxide
4) Molecular Formula: CO
5) Exhaust gas
6) Flue gas
7) CAS 630-08-0
8) CARBON MONOXIDE, CRYOGENIC LIQUID

1.2.1) MOLECULAR FORMULA
1) CO

Available Forms Sources

1) Carbon monoxide is sold for commercial and industrial use in grades of purity ranging from 98 to 99.9 percent (CGA, 1999).

1) SUMMARY

a) Natural sources of carbon monoxide in the environment include oxidation of atmospheric methane, release from the ocean (originating from deep-sea organisms), volcanoes, grass and forest fires, marsh gases, and electrical storms (Harbison, 1998). These processes account for approximately 90 percent of atmospheric CO (HSDB , 2001).
b) Incompletely combusted organic fuels will form carbon monoxide (Baselt, 2000). Automobile exhaust accounts for over half of manmade carbon monoxide. Industrial processes, space and water heaters, furnaces, and solid waste disposal procedures are also contributing sources (Harbison, 1998).

2) SOURCES OF EXPOSURE

a) SUMMARY

1) In one year, out of 90 cases of carbon monoxide exposure reported to one poison center, the sources of exposure were (Spiller, 1987):

Home	53.4%
School bus	29.4%
Automobile	13.3%
Work	7.8%
Other	1.1%

2) According to the National Electronic Injury Surveillance System All Injury Program (NEISS-AIP) from 2001 to 2003, 64.3% of CO exposures were reported to occur in the home and 21.4% occurred in public facilities and areas (Anon, 2005).

b) ANESTHESIA

1) Three hospitals reported elevated blood COHb levels, detected by co-oximetry, in 26 patients undergoing surgery. Risk factors determined during a case-control analysis included operations performed on Monday or Tuesday, operations in a room routinely inactive during weekends, and use of anesthesia equipment that had not been used for 24 hours or more.

a) Exposure was believed to be secondary to an interaction between the anesthetic gas and the CO2 absorbent canister. Subsequently canisters were flushed with high flow oxygen for 60 seconds before each procedure and replaced if not used within 24 hours; no further exposures were documented (CDC, 1991).

2) In an animal study, inhalation of desflurane through a breathing circuit with dried CO2 adsorbents produced carbon monoxide poisoning (Woehlck & Dunning, 1998).
3) An adult developed CO poisoning after being ventilated with desflurane (Berry et al, 1999). Poisoning was considered to be secondary to CO being produced when desiccated carbon dioxide absorbent reacted with the anesthetic agent.

c) AUTOMOBILES

1) CATALYTIC CONVERTERS: The use of catalytic converters in automobiles has lessened the likelihood of death resulting from a suicide attempt via inhalation of exhaust fumes (Vossberg & Skolnick, 1999). Severity of intoxication may be less related to carboxyhemoglobin than to the duration of CO exposure and to the other constituents present in exhaust fumes.
2) PICK-UP TRUCKS: Children riding in the backs of pickup trucks may become exposed to CO fumes from the trucks' exhaust. Hampson & Norkool (1992) reported 20 cases. Fifteen became unconscious, one died, and one had permanent neurologic damage (Hampson & Norkool, 1992).
3) CLOGGED EXHAUST SYSTEMS: Carbon monoxide poisoning in humans has been associated with sitting in cars with snow-obstructed vehicle exhaust systems while the engine was running (CDC, 1996; Rao et al, 1997).
4) INADEQUATE VENTILATION: Unintentional deaths have been reported from motor vehicle exhaust exposure which resulted from the following: inadequately ventilated passenger compartment, operation of a motor vehicle in an enclosed space (e.g. garage, tunnel or parking garage) with poor ventilation, use of an auxiliary fuel-burning heater inside a passenger compartment or camper, and operation of a motor vehicle with a damaged or malfunctioning exhaust system (Anon, 1996; Omaye, 2002; Handa & Tai, 2005).
5) CAR WASHES: Four deaths due to CO poisoning have been associated with car washes in which the bay doors were closed by the patron during winter months (Carson & Stephens, 1999). In all cases, the individuals had taken mind-altering substances (e.g. alcohol, THC, amphetamines). One death was an apparent suicide in which the method appeared to be a copy of previously reported deaths in the local area.
6) According to the National Electronic Injury Surveillance System All Injury Program (NEISS-AIP) from 2001 to 2003, involving 778 cases of CO exposures, 9% of cases were associated with motor vehicles (Anon, 2005).
7) A retrospective review, from the Virginia Mason Center for Hyperbaric Medicine in Seattle Washington, of 250 pediatric patients with CO poisoning over a 26-year period revealed that the most common source of carbon monoxide exposure was motor vehicles (34%) (Mendoza & Hampson, 2006).

d) CHARCOAL-BURNING BARBECUE

1) Food preparation using a charcoal-burning barbecue indoors may lead to carbon monoxide poisoning (Gasman et al, 1990; Sternbach & Varon, 1991). Clinical presentation and history may suggest a presumptive diagnosis of food poisoning.
2) Dozens of individuals developed symptoms of CO poisoning after visiting a Chinese restaurant where charcoal-pans were used to prepare fondue in an area with limited ventilation (Gauzere et al, 1999). Approximately 60 minutes after exposure the average COHb level ranged from 12% to 2% or less.
3) INTENTIONAL EXPOSURE: Several case reports of children intentionally exposed to carbon monoxide via burning charcoal (in a closed room or dwelling) have occurred in Hong Kong and Taiwan, in which parents sought to harm the child(ren) and themselves (Cho et al, 2008; Lee et al, 2002).
4) According to a retrospective review of pediatric carbon monoxide exposure, conducted by the Virginia Mason Center for Hyperbaric Medicine in Seattle, Washington, involving 250 pediatric patients over a 26-year-period, burning of charcoal briquettes was the second-most common source of CO exposure (24%), behind motor vehicles, however it was the most common source of CO exposure among 0 to 2-year-olds (40%; n=42) (Mendoza & Hampson, 2006).
5) Intentional exposure to carbon monoxide via charcoal burning is one of the most common methods of suicide in Asian countries, accounting for approximately 20% to 30% of all suicides in Hong Kong and Taiwan (Chen et al, 2015; Lee et al, 2014).

e) BOATING

1) OPEN AIR EXPOSURES/CASE SERIES: Individuals (both employees and boaters) were exposed to carbon monoxide during close proximity to operating motorboats located in a crowded channel. During the study, employees had CO exposures equal to or exceeding the NIOSH recommended exposure limit and 42 employees had post-shift %COHb levels that exceeded the ACGIH biologic exposure index (guidelines established to assist in the control of health hazards) of 3.5%. Cases of carbon monoxide poisoning, some fatal, had been reported in vacationers wading or boating in this channel (MMWR, 2003).
2) A 4-year-old girl developed carbon monoxide poisoning (venous COHb level of 22.2 percent on admission) after swimming behind a house boat anchored with a factory installed, gasoline fueled, electrical generator that vented 1 to 1.5 inches above the water surface at the stern of the boat (Easley, 2000).

f) COMBUSTION

1) Internal combustion engine exhaust fumes, malfunctioning home heating systems, gas hot water heaters, gas clothes dryers, charcoal and poorly vented wood/coal stoves, space heaters, gas and kerosene lanterns, and fires in buildings are common sources of carbon monoxide poisoning (ACGIH, 1991; Cook et al, 1995; Girman et al, 1998; Hathaway et al, 1991; Hawkes et al, 1998; Jones, 1997; Jumbelic, 1998; Lawrence L, Seger D & Bonfiglio F et al, 1995; 221.; Silvers & Hampson, 1995; Handa & Tai, 2005; Cho et al, 2008).
2) Winter storms which cause prolonged loss of heat and electricity can result in winter epidemics of CO poisoning from use of combustible materials indoors (particularly gasoline-powered generators or charcoal grills) (Lawrence L, Seger D & Bonfiglio F et al, 1995; 221.).
3) Common occupational exposures to CO occur in workers who operate internal combustion engines in areas lacking proper ventilation and in those who operate blast furnaces in steel manufacturing (ACGIH, 1991). Other such exposures occur in workers in mines, petroleum refineries, pulp mills, and boiler rooms (Bingham et al, 2001).
4) According to the National Electronic Injury Surveillance System All Injury Program (NEISS-AIP) from 2001 to 2003, involving 778 cases of CO exposure, 18.5% of cases were associated with faulty furnaces, including oil, gas, and unspecified furnaces, 2.8% were associated with generators, and 1.9% were associated with space heaters(Anon, 2005).
5) According to a retrospective study, conducted at Tan Tock Seng Hospital in Singapore and involving 12 patients with a diagnosis of carbon monoxide poisoning from 1999 to 2003, the most common source of intentional poisoning was the gas stove (n=5) (Handa & Tai, 2005).
6) GAS GEYSER: A 21-year-old woman was found dead of carbon monoxide poisoning in a bathroom where a liquid petroleum gas fitted water heater, known as a gas geyser, had been installed. The bathroom was poorly ventilated without proper exhausts available for the release of combustion gases (Mohankumar et al, 2012).
7) UNDERGROUND UTILITY CABLE FIRES

a) Unintentional burning of underground utility cables (usually occurs when the rubber coating around the cable fails from normal wear, weather changes or deicing agents {eg, salt} used on roads) has resulted in a small number of CO poisonings. Due to the generally oxygen-poor environment found underground, incomplete combustion occurs which results in development of CO. Several case reports of underground utility cable fires in New York have resulted in CO seeping into surrounding buildings, which forced the evacuation of homes, public facilities, hospitals and a nursing center. Of note, because of the noxious smell given off by the burning rubber, the number of exposures is likely decreased as compared to more traditional CO-induced combustion exposures (MMWR, 2004).

g) FORMIC ACID COMBINED WITH SULFURIC ACID

1) CASE REPORT: A 26-year-old man was found dead in his car in a parking lot with the windows closed, the doors locked, and the car idling with occlusion of the gas exhaust. A 5-gallon bucket containing an unidentified clear boiling liquid was on the front passenger side floor, with a subsequent pH test of the liquid indicating a pH of 0. An empty container of a sulfuric acid-based drain opener and an empty container suspected to contain formic acid were also found in the car. Toxicologic analysis of the patient's heart blood indicated 85% carboxyhemoglobin. It is believed that the chemical reaction of formic acid and sulfuric acid in addition to the running engine and obstruction of the gas exhaust may have contributed to carbon monoxide poisoning of this patient (Lin & Dunn, 2014).

h) METABOLISM

1) Metabolism itself produces small quantities of carbon monoxide in the body, usually resulting in carboxyhemoglobin levels less than 1 percent of total hemoglobin (Bingham et al, 2001).

i) METHYLENE CHLORIDE

1) Inhalation of methylene chloride (dichloromethane) can produce carbon monoxide poisoning because methylene chloride is metabolized to CO in the liver. Carbon monoxide toxicity can occur some time after exposure from the slow release of methylene chloride from adipose tissue. Paint strippers are common methylene chloride-containing products (Bingham et al, 2001; Omaye, 2002).
2) PERSONAL DEFENSE SPRAYS: In some European countries, including Spain, a personal defense spray containing 49% methylene chloride as a solvent and 0.8% o-chlorobenzylidene malononitrile has been manufactured (Duenas et al, 2000). Four individuals (1 adult, 3 children) developed CO poisoning following accidental discharge of the units and all required 100% normobaric oxygen. No permanent sequelae was reported.

j) OCCUPATIONAL

1) FIRE FIGHTERS: A high risk of CO poisoning exists for fire fighters who often enter enclosed spaces in structural fires. Use of respiratory protective gear can prevent lethal CO exposure, but are not routinely used in all phases of fire fighting (Bingham et al, 2001).
2) ACETYLENE GAS WELDING

a) A 23-year-old welder died of carbon monoxide poisoning following inhalation of carbon monoxide that had been trapped and then released from a previously acetylene gas welded pipe that he was venting after it had been installed as part of a heating system in a residential dwelling. The patient's carboxyhemoglobin concentration was 46% (Antonsson et al, 2013).

3) FORKLIFTS

a) WAREHOUSE WORKERS HEADACHE: Thirty workers in a warehouse in which a propane-fueled forklift was operating developed headache, difficulty concentrating, and confusion associated with increased levels of expired CO (Ely et al, 1994).
b) Liquefied petroleum gas-powered forklifts used in light industry were responsible for causing CO poisoning in workers (Anon, 1999). The forklifts were found to emit high CO concentrations while the warehouse was inadequately ventilated. Due to the vague symptoms, many workers did not seek appropriate treatment.

4) VEHICULAR: Virtually any work on or near an internal-combustion engine can involve carbon monoxide exposure. Some occupations included here are taxi, ambulance, bus and truck drivers, mechanics, toll booth operators, garage attendants and police officers (Bingham et al, 2001).

k) RESIDENTIAL DWELLINGS

1) CO poisoning occurred in two adults after carbon monoxide was found trapped in a pocket under the foundation of their home which was presumed to be due to the use of explosives at a nearby rain sewer construction site (Auger et al, 1999).
2) A fatal case of CO poisoning was reported in an adult due to a faulty electrical storage heater. Examination revealed that the cast-iron core or "blocks" and insulating material (ceramic, not shown to contain significant levels of carbon) showed evidence of heat damage with significant areas devoid of carbon. The oxygen source was the air passing through the heater unit, thus favoring the production of carbon monoxide within the heater unit. The heater itself (i.e., the "blocks", rather than fuel) was considered to be the source of carbon which produced the carbon monoxide within the dwelling (Morgan et al, 2001).

l) SEASONALITY

1) Poisonings are often seasonal with a higher incidence in the fall and winter. In one Michigan study, 87 percent of accidental deaths occurred in November to March.
2) According to the National Electronic Injury Surveillance System All Injury Program (NEISS-AIP) from 2001 to 2003, both fatal and non-fatal CO exposures occurred more often during fall and winter with the greatest occurrence during December (56 fatal and 2,157 non-fatal exposures) and January (69 fatal and 2,511 non-fatal exposures) (Anon, 2005).

m) SMOKING

1) Mainstream cigarette smoke, that which is inhaled into the smoker's lungs, can contain as much as 5% carbon monoxide by volume. Sidestream smoke, the source of environmental exposures, contains between 70 and 90% of the total CO per cigarette (Bingham et al, 2001).
2) In indoor areas where smoking is permitted, carbon monoxide levels can exceed 11 ppm; this compares to less than 2 ppm in most nonsmoking areas.
3) Narghile (also known as hooka, shisha, goza, hubble bubble, argeela, waterpipe), is a traditional method, particularly in the Eastern Mediterranean region, for smoking tobacco, and has been associated with acute CO poisoning (Wang et al, 2015; La Fauci et al, 2012; Uyanik et al, 2011; Lim et al, 2009). In one case, a 25-year-old man experienced 3 syncopal episodes after smoking narghile. Arterial blood gas analysis revealed a carboxyhemoglobin level of 31.1% (Cavus et al, 2010). Higher concentrations of carbon monoxide are absorbed because narghile smokers inhale more deeply due to the less irritating nature of the narghile smoke, as compared to cigarette smoke, and the duration of each smoking session tends to be longer. Also, because of the charcoal that is used to burn the narghile tobacco, CO concentrations in the inhaled vapors are higher as well (La Fauci et al, 2012).

a) CASE REPORT: A 24-year-old man developed carbon monoxide poisoning after lighting approximately 30 to 40 hookahs filled with flavored tobacco while working at a hookah bar. After placing a butane lighter next to coals, he inhaled through the hookah hose in order to draw the flame to light the coals. After lighting the coals, the patient initially lost consciousness for approximately 10 minutes, then subsequently developed nausea, headache, generalized weakness, substernal chest pressure, and bilateral paresthesias of his shoulders. An ECG revealed incomplete right bundle branch block, left anterior fascicular block, and nonspecific ST-T segment changes. Arterial blood gas analysis revealed a carboxyhemoglobin level of 33.8%. The patient recovered following hyperbaric oxygen therapy (Misek & Patte, 2014).

n) SPRAY PAINT

1) CO poisoning has been reported in one adult with a history of cigarette smoking using a spray paint outdoors which contained methylene chloride (a halogenated hydrocarbon) (Nager & O'Connor, 1998).

1) Carbon monoxide is used in chemical manufacturing to produce methanol and phosgene and in plastic synthesis. It also plays a role in organic synthesis, especially in the oxo reaction and the Fischer-Tropsch processes for petroleum-type products (Budavari, 2000; CGA, 1999; HSDB , 2001; OHM/TADS , 2001).
2) Carbon monoxide is used in metallurgy in the Mond process to recover high-purity nickel from crude ore and for reducing oxides and special steels. It is used to obtain highly pure powdered metals and to form particular metal catalysts used to synthesize hydrocarbons or organic oxygenating compounds. It is also used in the manufacture of white zinc pigments (Budavari, 2000; CGA, 1999).
3) It is used in the manufacturing process for acids, esters, and hydroxy acids (acetic and propionic acids and their methyl esters), and glycolic acid (CGA, 1999).

Overview

Life Support

Clinical Effects

0.2.1)

A) SOURCES: Odorless, colorless gas produced by the incomplete combustion of any carbon containing substance. Common sources include household fires, home furnaces, stoves and water heaters, and vehicle exhaust. Another potential source is methylene chloride (often used as a paint stripper or degreaser) that is absorbed through inhalation, ingestion, or dermal contact and is subsequently metabolized by the liver to carbon monoxide.
B) PHARMACOLOGY: Binds to hemoglobin with an affinity approximately 250 times greater than that of oxygen.
C) TOXICOLOGY: Impairs oxygen delivery, producing cellular hypoxia and ischemia.
D) EPIDEMIOLOGY: Common poisoning; one of the leading toxicologic causes of death.
E) WITH POISONING/EXPOSURE

1) MILD TO MODERATE TOXICITY: Headache, nausea, dizziness, vomiting, weakness, and confusion are often reported, with headache being the most common.
2) SEVERE TOXICITY: Coma, syncope, seizures, cardiac dysrhythmias, myocardial ischemia, and death result from more severe poisonings and reflect damage to the organ systems (brain and heart) with the highest oxygen demand. Delayed neurocognitive effects, which include dementia, amnestic syndromes, psychosis, parkinsonism, chorea, apraxia, neuropathies, difficulty concentrating, and personality changes, can occur from 2 to 40 days following the initial exposure. Unfortunately, there are no good predictive markers for who will develop neurocognitive sequelae, including the initial carboxyhemoglobin level or the severity of the initial poisoning.

0.2.3)

A) WITH POISONING/EXPOSURE

1) Alterations in blood pressure may occur following exposure.

0.2.20)

A) CO exposure during pregnancy is teratogenic, depending upon the stage of pregnancy. The fetus is more vulnerable to CO poisoning than the mother.

Laboratory Monitoring

Treatment Overview

0.4.3)

A) MANAGEMENT OF MILD TO MODERATE TOXICITY

1) The primary treatments for carbon monoxide (CO) poisoning are oxygen supplementation and supportive care. An initial carboxyhemoglobin level should be obtained as soon as possible. Treat with 100% oxygen until asymptomatic and carboxyhemoglobin levels are below 5%. Perform a Mini-Mental State Exam and careful neurologic exam. An abnormal Mini-Mental State Exam or ataxia or other neurological abnormalities may be a subtle sign of more severe poisoning. Signs or symptoms of metabolic acidosis would indicate a more severe poisoning. Identifying the source of the exposure is critical to prevent recurrent poisoning.

B) MANAGEMENT OF SEVERE TOXICITY

1) In addition to oxygen supplementation, patients who are comatose or patients with smoke inhalation injuries may need mechanical respiratory support. Standard treatment protocols for hypotension and cardiac dysrhythmias should be followed. The role of hyperbaric oxygen therapy is controversial. While data are conflicting regarding improved outcome with hyperbaric oxygen therapy, patients with severe poisoning should generally receive hyperbaric oxygen. Some of the suggested indications include syncope, altered mental status or neurologic deficits, evidence of cardiac injury, and persistent or severe metabolic acidosis. Pregnancy, especially with fetal distress, is generally considered an indication for hyperbaric therapy because fetal carboxyhemoglobin concentration is generally higher, and its elimination slower, than corresponding maternal carboxyhemoglobin. Other experts have suggested patients with milder poisoning may also benefit.

C) DECONTAMINATION

1) PREHOSPITAL: Remove from the source and administer high-flow oxygen.

D) AIRWAY MANAGEMENT

1) Patients who are comatose or who have lung injury from smoke inhalation injuries may need mechanical respiratory support and orotracheal intubation.

E) ANTIDOTE

1) 100% oxygen should be administered to the patient via nonrebreather and continued until the patient is asymptomatic and carboxyhemoglobin levels are below 5%. Pregnant women need to be treated for a longer time since CO elimination may be slower in the fetus. Some evidence supports the use of hyperbaric oxygen to prevent delayed cognitive and neurologic sequelae; however, there is controversy around the use of hyperbaric therapy. In general, the risks of hyperbaric oxygen treatment (which include seizures and barotrauma) are low. In complex cases, a poison center or hyperbaric center should be contacted.

F) ENHANCED ELIMINATION

1) Oxygen increases the elimination of CO.

G) PATIENT DISPOSITION

1) HOME CRITERIA: Asymptomatic, nonpregnant patients can be observed at home, provided it is certain that the source of exposure has been eliminated and the patient has been moved to a CO-free environment. Most area fire departments and utility companies can perform ambient CO monitoring to determine the source of exposure.
2) OBSERVATION CRITERIA: Mildly to moderately symptomatic patients should be sent to a healthcare facility for evaluation and treated until symptoms resolve.
3) ADMISSION CRITERIA: Patients who remain symptomatic despite adequate treatment with oxygen, patients with abnormal neurological exams or myocardial injury, and pregnant women with fetal distress should be admitted.
4) CONSULT CRITERIA: Consult a poison center or medical toxicologist for assistance in managing severe poisonings and for recommendations on determining the need for hyperbaric oxygen treatment. Contact your nearest hyperbaric chamber for recommendations and arrangements for hyperbaric oxygen therapy.

H) PITFALLS

1) Pulse oximetry will measure normal oxygen saturations even in the presence of significant carboxyhemoglobin levels. Carboxyhemoglobin levels and initial severity of the patient's symptoms are not good predictors for delayed neurocognitive sequelae. The clinical manifestations of mild toxicity are nonspecific and often mistaken for viral illnesses. A high index of suspicion is necessary; the diagnosis of CO poisoning should be considered in patients presenting with nausea, vomiting, or headache, especially if these occur in family clusters or at the beginning of cold weather. Failure to determine the source of exposure may result in recurrent poisoning.

I) TOXICOKINETICS

1) The half-life of carboxyhemoglobin in a patient breathing ambient air is approximately 4 to 6 hours. In a patient breathing 100% oxygen at sea level, the half-life is approximately 60 to 75 minutes, and during hyperbaric oxygen treatment, it is approximately 20 minutes.

J) DIFFERENTIAL DIAGNOSIS

1) The differential diagnosis for the nonspecific symptoms of CO poisoning is extremely broad and includes influenza, viral syndrome, headache (eg, migraine or tension), and food poisoning. Cyanide gas poisoning should also be considered when a patient is comatose due to a house fire.

Range Of Toxicity

Clinical Effects

Summary Of Exposure

Vital Signs

3.3.1)

A) WITH POISONING/EXPOSURE

1) Alterations in blood pressure may occur following exposure.

3.3.4)

A) WITH POISONING/EXPOSURE

1) Hypotension may occur secondary to vasodilatation or myocardial depression (Lowe-Ponsford & Henry, 1989).
2) Hypertension may occur transiently (Whorton, 1976).

Heent

3.4.3)

A) WITH POISONING/EXPOSURE

1) RETINAL FINDINGS: Papilledema, optic atrophy, retinopathy, retinal hemorrhage, and engorgement of the retinal veins may occur (Ferguson et al, 1985; Kelley & Sophocleus, 1978; Dempsey et al, 1976; Garland & Pearce, 1967).

a) It has been suggested that a thorough examination of the eye (ie, electrodiagnostic tests) may reveal retinal hemorrhages, which can occur superficially or deeper in the nerve fiber layer (flame hemorrhage) and are often peripapillary. The venous changes that develop include engorgement and tortuosity, while edema of the optic disc may be observed. All of these changes reflect the hypoxic injury to the retina due to CO poisoning (Denniston, 2001).

2) VISUAL FIELD DEFICITS: Paracentral scotomata, homonymous hemianopia, tunnel vision, temporary blindness, and permanent blindness are known sequelae (Dempsey et al, 1976) Uraneck et al, 1993).

a) Late-onset vision loss was reported in 2 patients with CO poisoning. Vision improved significantly in both patients over several months (Ersanli et al, 2004).
b) Late-onset visual disturbances (visual acuity and field deficits) occurred in 3 adolescent girls (ages 11 to 14) approximately 5 to 6 days following CO exposure. All 3 patients had been given oxygen via a nasal cannula (4 L/minute) for 2 days immediately following exposure. Following development of visual disturbances, MRI and T2- and diffusion-weighted imaging revealed cortical ribbon lesions in the temporal and parietal lobes, suggestive of cortical (laminar) infarctions. Resolution of visual deficits occurred within 3 weeks following exposure, and there appeared to be no neurologic sequelae at 2 months postexposure (Hon et al, 2006).

3) RETROBULBAR NEURITIS: In a case report, retrobulbar neurits was reported as a late effect (Reynolds & Shapiro, 1979).
4) CORTICAL BLINDNESS: In a series of 2360 patients with CO poisoning, 2 developed cortical blindness (Choi, 1983).
5) VISUAL DISTURBANCES: In a cohort of 38 patients with CO poisoning, the presence or absence of visual disturbances was an independent predictor for at least 1 late symptom, other than headache, occurring 1 year postexposure (Henz & Maeder, 2005).

3.4.4)

A) WITH POISONING/EXPOSURE

1) ACUTE TOXICITY

a) DEAFNESS: Transient sensorineural hearing loss has been reported following acute poisoning (Baker & Lilly, 1977; Garland & Pearce, 1967).
b) BAER ABNORMALITIES: Abnormal brainstem auditory evoked response (BAER) potentials have been described in patients with carbon monoxide (CO) poisoning (Choi, 1985).

1) BAERs were studied in 32 patients with acute CO poisoning. Peripheral abnormalities were found in 6 cases and central abnormalities in 2 . The incidence of BAER abnormalities was increased in patients who were unconscious for more than 24 hours (Rybak, 1992).

2) CHRONIC TOXICITY

a) Although sensorineural hearing loss is associated with acute CO poisoning, chronic low-dose exposure to CO may result in similar toxicity. In a case report, a 61-year-old woman presented with a 2-week history of bilateral hearing loss along with dizziness and lethargy. A gas appliance in the home was found to emit large levels of CO, and the patient had elevated carboxyhemoglobin levels. Following high-flow oxygen therapy, the patient's CNS symptoms rapidly resolved, and her hearing gradually improved (Hassan et al, 2003).

Cardiovascular

3.5.2)

A) CARDIOVASCULAR FINDING

1) WITH POISONING/EXPOSURE

a) The heart is sensitive to the hypoxia caused by carbon monoxide (CO). Increased coronary blood flow occurs as the carboxyhemoglobin level in the blood rises (Adams et al, 1973; Ayres et al, 1973).
b) Clinical effects of CO poisoning include tachycardia, hypotension, peripheral vasodilation, cyanosis, shock, and cardiac arrest (Vacchiano & Torino, 2001; Thompson & Henry, 1983). Early cardiac arrhythmias are thought to be rare (Lasater, 1986). Tachycardia is a compensatory mechanism for cellular hypoxia (Ernst & Zibrak, 1998).
c) COHORT STUDY: A prospective cohort study, involving 230 patients with moderate to severe CO poisoning, assessed the relationship of myocardial injury with long-term mortality in these patients. Eighty-five of the 230 patients (37%) developed myocardial injury (identified as a cardiac troponin I level of at least 0.7 ng/mL or a CK-MB level of at least 5 ng/mL and/or diagnostic ECG changes) as a result of CO poisoning. The patients' clinical courses were then followed for a median of 7.6 years (range: in hospital only to 11.8 years postdischarge). Overall, 32 of the 85 patients (38%) with myocardial injury died as compared with 22 of 145 patients (15%) with CO poisoning who did not develop myocardial injury (adjusted hazard ratio, 2.1; 95% confidence interval, 1.2 to 3.7; p=0.009), indicating that myocardial injury in patients with moderate to severe CO poisoning may be a significant indicator of mortality (Henry et al, 2006)

1) Predictors of myocardial injury within this same study included the male gender, a Glasgow Coma Scale score of 14 or less, and the presence of hypertension (Satran et al, 2005).

d) A prospective study was conducted involving 20 patients with CO poisoning. On admission to the emergency department, ECG was performed, and troponin I, B-type natriuretic peptide (BNP), creatine kinase, CK-MB, and carboxyhemoglobin levels were obtained. All measurements were repeated at 6, 12, 24, and 48 hours after admission. The results showed that cardiac markers were elevated in 6 patients (CK-MB 94 +/- 60 units/L, troponin I 2.7 +/- 2.3 mcg/mL) and normal in 14 patients. The patients with elevated cardiac markers also had higher carboxyhemoglobin levels (37% +/-7% vs 25% +/-8%; p=0.01) and were exposed to CO longer (8.2 +/- 1.6 hours vs 4.2 +/- 1.3 hours; p less than 0.01) than the patients with normal cardiac markers. The left ventricular ejection fraction (LVEF) values were less than 45% in 8 patients (mean LVEF 38% +/- 5%) and greater than 45% in 12 patients. The lower LVEF values also appeared to directly correlate with higher carboxyhemoglobin levels (36% +/- 6% vs 24% +/- 9%; p=0.003) and BNP levels (77 +/- 50 ng/mL vs 16 +/- 6 ng/mL; p less than 0.01). Of the patients with decreased LVEF, 2 patients had global wall motion impairment and 5 had regional wall motion impairment in the left ventricle (Kalay et al, 2007).

B) MYOCARDIAL ISCHEMIA

1) WITH POISONING/EXPOSURE

a) Myocardial ischemia may be precipitated or aggravated by CO exposure and has been reported even with low CO levels in patients with preexisting coronary artery disease.
b) Chest pain, dyspnea, diaphoresis, syncope, or palpitations may occur (Allred et al, 1989; Kleinman et al, 1989; Atkins & Baker, 1985; Aronow & Isbell, 1973).
c) CASE REPORT: Reversible cardiac failure with myocardial ischemia was reported in a 15-year-old who had a blood level of carboxyhemoglobin of only 10% (Diltoer et al, 1995).
d) INCIDENCE: In a retrospective review of 18 patients with burns and carboxyhemoglobin levels of 10% or higher, 5 (28%) had ECG evidence of myocardial ischemia; 3 also had elevated CK-MB enzymes. Only 3 of 105 controls (burn patients without increased carboxyhemoglobin) had ECG abnormalities (Williams et al, 1992).
e) Eighty-three healthy men and women exposed to CO from heating stoves or gas-operated hot water heaters were studied. CO exposure ranged between 20 minutes and 480 minutes (average 52.5 minutes). Carboxyhemoglobin levels ranged from 9.4% to 66% (average 34.4% +/- 15.9%). CK results were elevated on admission for 6 patients and at 6 hours post arrival for 15 patients; CK-MB results were high for 18 patients on arrival and again 6 hours post arrival (Aslan et al, 2006).

1) Sinus tachycardia was observed in 26.5% of patients, and ischemic changes were found in 14.4% of patients. Myocardial perfusion single photon emission computed tomography (SPECT) showed ischemia in 9 patients (10.8%), which was confirmed by ECHO in 6 of these patients. The relationship between ischemia with gender, smoking status, level of confusion, level of consciousness, presence of palpitations, and carboxyhemoglobin level showed significant relationships at varying degrees (Aslan et al, 2006).

f) Myocardial infarction in the absence of fixed coronary artery obstruction has been reported (Marius-Nunez, 1990).
g) CASE REPORT: A 36-year-old man presented to the hospital following deliberate inhalation of gas heater exhaust in a bathroom and sustained a brief loss of consciousness (less than a minute). Approximately 1 hour post-exposure, the patient complained of chest pain. An initial ECG showed slight ST-T changes, suggesting myocardial ischemia. Laboratory data revealed a carboxyhemoglobin concentration of 22% and an elevated troponin I concentration (2.19 mcg/L; cutoff value 0.01). With high-flow oxygen therapy (hyperbaric oxygen therapy was not available), the patient's chest pain improved, with slight improvement in ST-T changes; however, cardiac markers showed an increase in troponin I (5.94 mcg/L), creatine kinase (CK) of 1544 International Units/L, and a CK-MB of 160 International Units/L. An echocardiogram demonstrated apex hypokinesia with an ejection fraction of 50%. A coronary angiography was performed indicating acute occlusion of the left anterior descending coronary artery, and a primary angioplasty was subsequently performed. Following angioplasty, the patient's symptoms resolved and cardiac markers decreased. An echocardiogram performed before discharge showed a left ventricular ejection fraction of 65% and mild apex hypokinesia; however, the ST-T changes persisted, suggesting myocardial injury (Dziewierz et al, 2013).

C) ELECTROCARDIOGRAM ABNORMAL

1) WITH POISONING/EXPOSURE

a) CASE REPORT: An increase in QT interval (max QTc 542 ms lead V5; min QTc 433 ms lead V2; QTdc 109 ms) was reported in a 59-year-old man following an attempted suicide by CO poisoning (initial carboxyhemoglobin level 34%). After several hours of mechanical ventilation with 100% oxygen, carboxyhemoglobin levels decreased to 8.5%, and a repeat ECG showed a QTdc of 55 ms (max QTc 524 ms, min QTc 469 ms); no other cardiac abnormalities were reported. Following successful extubation at 24 hours, the patient refused further medical care and was lost to follow-up (MacMillan et al, 2001).
b) Electrocardiographic changes include ST-segment depression or elevation, T-wave abnormalities, atrial fibrillation, and intraventricular conduction block (Satran et al, 2005; Hayes & Hall, 1964).
c) QT DISPERSION: A prospective study was conducted, involving patients with carbon monoxide poisoning, to determine the relationship between QT dispersion, defined as the difference between the maximum QT interval (QTmax) and the minimum QT interval (QTmin), and carbon monoxide intoxication. A total of 127 patients were included in the study (79 patients with CO intoxication and 48 controls). An ECG at admission demonstrated a significantly increased QT dispersion (QTd) in the study group as compared to the control group (39 +/- 10.8 ms vs 24.4 +/- 6.2 ms; p<0.001). One week later, a repeat ECG showed a significant decrease in the QTd of the study group (23.6 +/- 7 ms; p<0.001). Similarly, at admission, the corrected QT dispersion (cQTd) was also significantly increased in the study group as compared to the control group (46.2 +/- 14.7 ms vs 25.3 +/- 6.2 ms; p<0.001). One week later, a repeat ECG showed a significant decrease in the cQTd of the study group (27.1 +/- 8.7 ms; p<0.001), all of which indicates a direct correlation between CO intoxication and increased QT dispersion (Atescelik et al, 2012).
d) CASE REPORT: A 34-year-old woman, exposed to carbon monoxide after starting her vehicle in a closed garage, developed difficulty in breathing, nausea and vomiting, abdominal pain, and bloody diarrhea. Her initial carboxyhemoglobin level was 23%. Treatment included hyperbaric oxygen therapy. Following therapy, her carboxyhemoglobin level decreased to 2.6%, however, her troponin I level was 1.49 ng/L. An ECG revealed sinus tachycardia with T-wave inversions, and an echocardiogram, obtained 7 hours after CO exposure, demonstrated a reduced ejection fraction of 30% to 35% with akinesis. Over the next 24 hours, her ejection fraction increased to 42% and her troponin level peaked at 3.7 ng/mL. An adenosine cardiac stress MRI, performed 2 days later, revealed anterior and anteroseptal hypokinesis, with basal hypercontractility, resulting in normalization of her ejection fraction (60%). With continued observation, the patient recovered and was discharged to an inpatient psychiatric unit on day 4 (Weaver & Deru, 2016).

D) CONDUCTION DISORDER OF THE HEART

1) WITH POISONING/EXPOSURE

a) Dysrhythmias and angina may be precipitated or aggravated by CO exposure via increased cardiac output caused by cellular hypoxia (Ernst & Zibrak, 1998).
b) Low-level CO exposure (100 parts per million (ppm) and 200 ppm) was not demonstrated to be arrhythmogenic in patients with coronary artery disease and no ventricular ectopy at baseline, in a randomized, double-blind, crossover trial in 10 patients (Hinderliter et al, 1989).

1) All patients continued antianginal medications (beta blockers, long-acting nitrates, and calcium antagonists) during the study (Hinderliter et al, 1989).

c) Low levels of carboxyhemoglobin during graded exercise exacerbated myocardial ischemia in participants with coronary artery disease (Allred et al, 1991; Allred et al, 1989; Kleinman et al, 1989).
d) The number and complexity of ventricular dysrhythmias increased significantly during exercise after CO exposure, producing 6% carboxyhemoglobin compared with room air in 41 nonsmokers who had documented coronary artery disease, in a randomized, double-blind, crossover study (Sheps et al, 1990).
e) Patients with frequent ventricular ectopic activity who were exposed to low-level CO (producing 3% to 5% carboxyhemoglobin) did not have an increase in dysrhythmia frequency of single or multiple beats during rest or exercise (Dahms et al, 1993).
f) In a series of children with CO poisoning, tachycardia was common and bradycardia occurred less often (Meert et al, 1998).
g) SUPRAVENTRICULAR TACHYCARDIA: A 17-year-old girl presented to the emergency department with palpitations, headache, and nausea. Her O2 saturation via pulse oximeter was 97%. An ECG indicated supraventricular tachycardia with a heart rate of 170 bpm. Laboratory data revealed a carboxyhemoglobin level of 19%. With supportive treatment, the patient's symptoms and ECG abnormalities resolved, without recurrence at her 1-year follow-up (Cetin et al, 2011).
h) CASE REPORT: A 42-year-old woman developed syncope, nausea and vomiting, and tachycardia (120 bpm) following 3 hours of exposure to carbon monoxide leaking from a stove. An ECG demonstrated atrial fibrillation with a rapid ventricular heart beat of 130 bpm. Laboratory data revealed a carboxyhemoglobin concentration of 13.9% (reference range, 0.5% to 2.5%) and an oxygen saturation of 54.6%. Following oxygen therapy, the patient's carboxyhemoglobin concentration decreased to 4%, her oxygen saturation increased to 96%, and her cardiovascular status normalized (Akdemir et al, 2014).

E) HYPOTENSIVE EPISODE

1) WITH POISONING/EXPOSURE

a) Hypotension may develop with severe poisoning (Chamberland et al, 2004; Meert et al, 1998; Tritapepe et al, 1998).
b) CASE REPORT: A 25-year-old woman with severe CO poisoning developed reversible cardiac failure. On presentation, her blood pressure was 70/40 mmHg, ejection fraction was 35%, cardiac index was 1.8 L/min/m(2), and pulmonary artery occlusion pressure was 21 mmHg. She had clinical evidence of pulmonary edema. Her clinical condition and hemodynamics improved by day 5 (Tritapepe et al, 1998).

F) CARDIAC ARREST

1) WITH POISONING/EXPOSURE

a) CASE SERIES: In a retrospective review of patients who had been found in cardiac arrest following severe CO poisoning, all patients were resuscitated at the scene and then treated with hyperbaric oxygen. The patients ranged in age from 3 to 72 years, and 10 were female and 8 male. Carboxyhemoglobin levels averaged 31.7% (+/- 11%) and arterial pH averaged 7.14 (+/- 0.19) at the time of admission. Bradydysrhythmias were the underlying presenting rhythms in 10 of 18 patients. Hyperbaric oxygen therapy was initiated within 4.3 hours on average after poisoning (less than 3 hours in 10 patients, and less than 6 hours in 15 patients). All 18 patients died during their hospitalization (Hampson & Zmaeff, 2001).
b) Based on a review of the literature, a survival rate of 1.4% to 26% (average rate is 20%) has been reported in CO-exposed patients with an out-of-hospital cardiac arrest. Two factors associated with a poorer prognosis were noncardiac cause for arrest and asystole or bradydysrhythmia at the time of discovery (Hampson & Zmaeff, 2001).

G) MYOCARDITIS

1) WITH POISONING/EXPOSURE

a) WALL MOTION ABNORMALITIES: Abnormal left ventricular wall motion has been reported in 3 of 5 cases studied. Papillary muscle dysfunction can also be found (Corya et al, 1976).

H) ACUTE LEFT VENTRICULAR FAILURE

1) WITH POISONING/EXPOSURE

a) CASE REPORT: An 82-year-old woman developed recurrent episodes of acute left ventricular failure following CO exposure in her home. Initial symptoms include progressive shortness of breath and palpitations 6 days prior to admission. On admission, ECG showed atrial fibrillation (without ischemic changes) and chest radiograph showed pulmonary edema. Following supportive care, the patient improved and returned to her apartment. The symptoms recurred 2 more times until the source of the leak was found (Webber, 2003).
b) CASE REPORT: Sinus tachycardia (heart rate of 100 bpm), hypotension (systolic blood pressure of 80 mmHg), bradypnea (8 breaths/minute), and a Glasgow Coma Scale of 6 were reported in a 49-year-old woman who was found in an idling vehicle in a closed garage. Laboratory analysis revealed an initial carboxyhemoglobin level of 35%, a CK level of 3161 mg/dL, a CK-MB level of 30.7 mg/dL, and elevated troponin I levels (peaked at 23.4 ng/mL on hospital day 2). An ECG showed right bundle branch block, and an echocardiogram showed a left ventricular ejection fraction (LVEF) of 20% with global hypokinesis and reduced function of the right ventricle. With supportive care, the patient's cardiac dysfunction gradually improved; a myocardial perfusion scan on hospital day 8 showed no cardiac ischemia and a LVEF fraction of 70% (Chamberland et al, 2004).
c) Echocardiogram studies in a report showed a decrease in ejection fraction (EF) values (less than 46%) in 12 of 46 cases (26%) and a decrease in cardiac output in 30 cases (65.2%) of acute CO poisoning. Both parameters improved following hyperbaric oxygen therapy. Likewise, cardiac enzymes (in particular CPK and CK-MB) were increased following acute CO poisoning and were improved following hyperbaric oxygen (Yiqun et al, 2002).
d) COHORT STUDY: Echocardiograms showed left ventricular dysfunction in 30 of 53 patients (57%) with CO poisoning. Fourteen of the 30 patients had global LV dysfunction and 16 of the 30 patients had regional wall motion abnormality (Satran et al, 2005).
e) CASE REPORT: A 28-year-old man presented with altered mental status following intentional exposure to CO. Lab analysis showed creatine kinase (CK) of 412 units/L, a CK-MB of 6.9 ng/mL, a troponin I level of 0.96 ng/mL, and a carboxyhemoglobin level of 27%. A chest radiograph revealed pulmonary edema and mild cardiomegaly, an ECG indicated sinus tachycardia (120 bpm), and a transthoracic echocardiogram demonstrated severe left ventricular systolic dysfunction with global hypokinesia. Following supportive therapy with oxygen, diuretics, an ACE inhibitor, and urinary alkalinization, repeat cardiac biomarker measurements, obtained on hospital day 3, indicated elevated CK (5994 units/L peaking to 15,951 units/L due to rhabdomyolysis), CK-MB (38.6 ng/mL), and troponin I (11.7 ng/mL) levels. With continued supportive therapy, the patient recovered, with resolution of pulmonary edema and normalization of cardiac size and function (Jang & Park, 2010).

I) CARDIOGENIC SHOCK

1) WITH POISONING/EXPOSURE

a) Two patients (a 29-year-old woman and her 8-year-old daughter) had a prolonged exposure to sublethal levels of CO (longer than 24 hours; carboxyhemoglobin levels of 20.4% and 22.6%), and acute cardiac failure developed despite evidence of neurologic recovery in both patients. The authors suggested that massive binding of the toxin to myocardial myoglobin and mitochondrial cytochrome chain enzymes might explain their protracted cardiac failure. It was suggested that stunned myocytes contributed to ventricular insufficiency, but that metabolic viability remained, which may explain the normal ECG findings and cardiac enzyme levels (Yanir et al, 2002).

1) Following hyperbaric oxygen therapy, the 29-year-old woman had a low cardiac output (2.4 L/min) and high systemic vascular resistance, with a CK level of 2000 units/L and a CK-MB fraction of 3%. Transthoracic echocardiography revealed a LVEF of 15%, right ventricular dysfunction, and moderate mitral valve insufficiency. No evidence of ischemic disease was present, and the ECG showed sinus tachycardia only. Despite regaining full consciousness, the patient remained hemodynamically unstable and required inotropic support (dopamine, dobutamine, and nitroprusside) for approximately 5 days. An echocardiogram on day 10 revealed normal left and right ventricular function. The patient was discharged to home on day 11 (Yanir et al, 2002).
2) The 8-year-old patient developed a similar clinical course. Her initial ECG showed T-wave inversion at L1, AVL, and V6, and CK was 7000 units/L with a CK-MB fraction of 2%. Following hyperbaric oxygen therapy and intravenous mannitol, the child was successfully extubated on day 2 of admission. Proximal muscle weakness was reported. On day 3, a transthoracic echocardiography demonstrated a moderate reduction in global left ventricular function, with an ejection fraction of less than 30% and moderate mitral regurgitation. Captopril was given orally to reduce afterload. Repeat echocardiograms showed gradual improvement, and a normal study was obtained around day 24; the ECG was also normal. The child was discharged to home on day 24 (Yanir et al, 2002).

J) CARDIOMYOPATHY

1) WITH POISONING/EXPOSURE

a) CASE REPORT: A 21-year-old woman presented with weakness, dyspnea, and nausea and vomiting after CO exposure (companion in her house died with a tissue carboxyhemoglobin level of 64%). Laboratory tests revealed a CK measurement of 3087, CK-MB index of 0.6, troponin I level of 1.13 (normal less than 0.2), and carboxyhemoglobin level of 6. An ECG showed sinus tachycardia, QTc interval prolongation, left atrial abnormality, and diffuse ST-T wave changes. An echocardiography revealed reduced left ventricular function with an ejection fraction of 25% and global hypokinesis. With supportive therapy, including administration of oxygen, digoxin, enalapril, and carvedilol, the patient gradually recovered, with normalization of cardiac function 6 weeks postpresentation (Swank et al, 2004).
b) RETROSPECTIVE STUDY: The medical records of children (younger than 17 years of age) diagnosed with CO poisoning from 2004 to 2007 were reviewed in order to determine the presence of myocardial injury and its associated factors secondary to CO poisoning. Of 131 patients identified with CO poisoning, 107 had cardiac biomarkers drawn and 27 patients underwent an echocardiogram (ECHO). Sixteen of the 107 patients had elevated cardiac biomarkers indicative of myocardial injury. All of these patients had elevated troponin-t levels and 11 patients had elevated CK-MB levels. The ECHO demonstrated a low ejection fraction and abnormal left ventricular function in 9 of the 27 patients. Eight of the 9 patients had global left ventricular dysfunction and 1 patient had regional wall motion abnormality without any ECG abnormality. Eight of the 16 patients with elevated cardiac biomarkers had hypotension and a Glasgow Coma Score (GCS) of 14 or less, as compared with 1 patient (n=91) with normal cardiac enzyme levels (p<0.05), indicating that hypotension and a GCS of 14 or less were associated factors in children with myocardial injury secondary to CO poisoning (Teksam et al, 2010).

3.5.3)

A) ANIMAL STUDIES

1) MYOCARDITIS

a) Selective findings include degeneration of the myocardium and coronary artery atherosclerosis (Wright & Shephard, 1979).

Respiratory

3.6.2)

A) HYPERVENTILATION

1) WITH POISONING/EXPOSURE

a) Hyperventilation and dyspnea may occur. In a series of children with carbon monoxide (CO) poisoning (many from smoke inhalation), tachypnea was common (Meert et al, 1998). Tachypnea is a compensatory mechanism for cellular hypoxia (Ernst & Zibrak, 1998).

B) RESPIRATORY FAILURE

1) WITH POISONING/EXPOSURE

a) Mechanical ventilation was required in 22 of 286 subjects (7.7%) who had initial impairment of consciousness following acute CO exposure (Raphael et al, 1989).
b) In a series of children with CO poisoning (many from smoke inhalation), bradypnea and apnea were common (Meert et al, 1998).

C) ACUTE LUNG INJURY

1) WITH POISONING/EXPOSURE

a) Severe CO intoxication may be associated with an increase in alveolar-epithelial permeability and increased cardiac output with resultant pulmonary edema (Ernst & Zibrak, 1998) (Sawa, 1981) (Fein et al, 1980; Ogawa et al, 1974).
b) Acute respiratory distress syndrome (ARDS) was reported in 8.3% of patients (n=12) with CO poisoning. The ARDS in 1 of the patients, who exhibited bilateral pulmonary infiltrates and a PaO2/FiO2 of less than 200 mmHg, was thought to be due to either smoke inhalation or aspiration (Handa & Tai, 2005).

D) PULMONARY ASPIRATION

1) WITH POISONING/EXPOSURE

a) Aspiration pneumonia and respiratory alkalosis (Hart et al, 1988; Skorodin et al, 1986) have been reported.

E) RESPIRATORY ALKALOSIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Respiratory alkalosis was reported in an elderly patient taking diuretics and a beta-blocker for hypertension (Skorodin et al, 1986).
b) CASE REPORT: Respiratory alkalosis (pH 7.441, pCO2 37.3 mmHg, PO2 501 mmHg) was reported in a 19-year-old man who developed carbon monoxide poisoning after smoking shisha, an Eastern Mediterranean traditional method of smoking tobacco (Lim et al, 2009).

F) PARALYSIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Transient unilateral diaphragmatic paralysis was associated with acute CO poisoning (carboxyhemoglobin 10.2 g% 2 hours postexposure) in a 56-year-old man (Joiner et al, 1990).

G) PULMONARY EMBOLISM

1) WITH POISONING/EXPOSURE

a) A retrospective study was conducted to evaluate the risk of deep vein thrombosis (DVT) and pulmonary embolism (PE) in patients with carbon monoxide poisoning. Review of cases from the Taiwan National Health Insurance Research Database from 2000 to 2011 identified 8316 patients newly diagnosed with carbon monoxide poisoning, and 33,264 patients without carbon monoxide poisoning as the comparison cohort. After adjusting for age, gender, and comorbidities of hypertension, diabetes, cerebral vascular disease, heart failure, cancer, pregnancy, and a lower leg fracture or surgery, the incidence of DVT was significantly higher in the patients with carbon monoxide poisoning compared to the control group (HR 3.85 [95% CI 2.17 to 6.83]; p < 0.001). The incidence of PE was also higher in the patients with carbon monoxide poisoning compared to the control group, but was not significant (HR 1.66 [95% CI 0.70 to 3.92]) (Chung et al, 2015).

Neurologic

3.7.2)

A) CENTRAL NERVOUS SYSTEM FINDING

1) WITH POISONING/EXPOSURE

a) Symptoms and signs of acute poisoning may include headache, dizziness, disorientation, abnormal reflexes, difficult concentration, memory loss, fainting, cerebral edema, coma, seizures, and death (Uyanik et al, 2011; Anon, 2005; Henz & Maeder, 2005) (Turner & Clark, 1999) (Coric et al, 1998; Meert et al, 1998; CDC, 1995; Raphael et al, 1989; Durnin, 1987; Thompson & Henry, 1983; Burney et al, 1982; Larcan & Lambert, 1981).
b) In a prospective study of 38 Swiss soldiers with CO poisoning and 46 unexposed controls within the same military unit, dizziness was the most common symptom within the first 2 weeks following exposure, occurring in 92% of CO poisoning patients and 13% of controls (p less than 0.0001). Other neurologic symptoms that were commonly reported among the CO poisoning patients included headache, confusion, weakness, and clumsiness. Dizziness, weakness, and clumsiness generally lasted 1 day; however, headaches and confusion appeared to last up to 8 days (Henz & Maeder, 2005).
c) Cellular hypoxia and cerebral vasodilatation due to CO toxicity may result in presyncope, syncope, and seizures. Cerebral edema may further result (Ernst & Zibrak, 1998).
d) Children may be more susceptible than adults to the neurological effects of CO (Crocker & Walker, 1985), but no statistical comparisons exist to support this claim. In a study of 140 children with CO poisoning (age 15 years or less) that compared clinical symptoms and outcome to adults with CO poisoning, lethargy and flaccidity were more frequent findings, especially in the youngest children (Mathieu & Mathieu-Nolf, 2001).
e) A woman moved into an apartment that had a natural gas space heater and within 3 weeks developed daily headaches, fatigue, difficulty getting out of bed, lightheadedness, nausea, anorexia, diarrhea, severe forgetfulness, mental confusion, and difficulty concentrating due to chronic CO poisoning. Although the patient was removed from the apartment, approximately 2 months after the exposure she had persistent mild bilateral cerebral dysfunction (attention deficit, inability to concentrate, poor problem solving, and difficulty with visual perception and visual motor integration). The patient gradually improved but had some difficulty with concentration and memory (Meggs, 2001).
f) According to a retrospective review of 30 pediatric cases of CO exposure which occurred between 1996 and 2007, the most common neurologic presenting signs and symptoms were disturbance of consciousness (86.7%), seizures (23.3%), headaches (10%), and dizziness (10%) (Cho et al, 2008).
g) CASE REPORT: A 25-year-old man experienced 3 syncopal episodes after smoking narghile, a traditional method, particularly in the Eastern Mediterranean region, of smoking tobacco. Arterial blood gas analysis revealed a carboxyhemoglobin level of 31%, indicating acute CO poisoning. The patient recovered following treatment with high-flow oxygen, and was discharged without sequelae (Cavus et al, 2010).
h) CASE REPORT: A 19-year-old man presented to the emergency department after experiencing dizziness and subsequently hitting his head after falling, resulting in retrograde amnesia. Approximately 4 hours prior to presentation, the man had been smoking shisha, an Eastern Mediterranean traditional method of smoking tobacco. In light of the shisha smoking, a carboxyhemoglobin level was obtained, revealing a level of 27.8%. After receiving 100% oxygen for 4 to 6 hours, the patient's dizziness resolved and a repeat carboxyhemoglobin level, obtained 8 hours later, had decreased to 1.1%. A CT scan of the brain showed no evidence of intracranial hemorrhages or skull fractures, and the patient was discharged 2 days after admission (Lim et al, 2009).
i) CASE REPORT: A 20-year-old woman presented with nausea, mild headache, and severe light-headedness approximately 1 hour after using a waterpipe. The patient frequently used a waterpipe several days per week, with each session lasting 45 to 60 minutes, and had experienced similar symptoms in the past. At presentation, the patient was lethargic, but the rest of her physical exam was normal. Venous blood gas levels, obtained at presentation, and arterial blood gas levels, obtained 20 minutes after presentation, revealed carboxyhemoglobin concentrations of 25.4% and 23.9%, respectively, leading to a diagnosis of carbon monoxide poisoning secondary to waterpipe use. Following 24 hours of observation and supportive care, the patient recovered and was discharged (Wang et al, 2015).

B) TOXIC ENCEPHALOPATHY

1) WITH POISONING/EXPOSURE

a) POST-INTERVAL SYNDROME (SEVERE DELAYED NEUROTOXICITY): Severe residual or delayed neurologic effects ("interval" form of CO poisoning) may occur after acute CO poisoning. Demyelination in the CNS and other effects may occur 48 to 72 hours after exposure. The patient should be observed carefully for CNS and other hypoxic effects. The most commonly involved regions of the brain are the globus pallidus and the deep white matter.

1) SIGNS/SYMPTOMS: Delayed effects may include mental deterioration, irritability, aggressive behavior, apathy, disorientation, hypokinesia, akinetic mutism, distractibility, confusion, severe memory loss, loss of consciousness, coma, gait disturbances, fecal and urinary incontinence, speech disturbance, tremor, mania, bizarre behavior, visual hallucinations, visual loss, movement disorders, paresis, chorea, peripheral neuropathy, Tourette syndrome, and parkinsonian syndrome (Mizuno et al, 2014; Park & Kim, 2014; Kondo et al, 2007; Aslan et al, 2004 ; Choi, 2002; Gillespie et al, 1999; Kawanami et al, 1998; Ernst & Zibrak, 1998; Sakamoto et al, 1998; Inagaki et al, 1997; Lugaresi et al, 1990; Gordon & Mercandetti, 1989; Hart et al, 1988; Krantz et al, 1988; Davous et al, 1986; Min, 1986; Schwartz et al, 1985; Choi, 1983; Pulst et al, 1983; Lacey, 1981; Sawada et al, 1980; Smith & Brandon, 1973; Grinker, 1926).
2) PHYSICAL FINDINGS: Physical findings may include masked face, glabella sign, grasp reflex, increased muscle tone, short-stepped gait, retropulsion, intention tremor, hyperreflexia, clonus, flaccid paresis, Babinski sign, ataxia, and choreoathetosis movements (Mizuno et al, 2014; Hart et al, 1988; Davous et al, 1986; Min, 1986; Choi, 1983; Lacey, 1981).
3) PATHOPHYSIOLOGY: Although the exact cause is unknown, several theories have been developed. CO exposure may result in brain lipid peroxidation and leukocyte-mediated changes in the brain. The latency between anoxic insult and onset of neurological sequelae may be related to the time required for remyelination, oxidation reactions, central synaptic reorganization, transsynaptic neuronal degeneration, and diaschisis mediated by collateral sprouting, and denervation supersensitivity that follows. At present, no clear marker has been found that can predict the prognosis of sequelae (Choi, 2002).
4) INCIDENCE: In a large study of 2360 patients who had CO poisoning, 65 (2.75%) developed severe delayed neurologic effects (Choi, 1983), although older studies report a somewhat lower incidence (Shillito et al, 1936).

a) In another series of 79 patients with more severe acute toxicity (cardiopulmonary insufficiency or altered mental status after 1 hour of normobaric 100% oxygen), 11 (14%) patients developed significant brain damage (Krantz et al, 1988).
b) Incidence is probably related to initial level of consciousness (Smith & Brandon, 1973) and coma duration. Incidences of 1.3% in patients without coma and 30% in patients with coma lasting 6 days or more have been reported (Choi, 1983).
c) Incidence may be lower in younger patients (Shillito et al, 1936), but occurrence has been reported in all age groups (Lacey, 1981).

5) ONSET: Patients may be asymptomatic for 2 to 240 days (mean 22) after exposure before presentation with severe delayed neurologic sequelae (Mizuno et al, 2014; Aslan et al, 2004 ; Gallerani et al, 2000; Min, 1986; Choi, 1983).
6) CT AND MRI FINDINGS: Bilateral basal ganglia lesions, particularly of the globus pallidus, are common and may be demonstrated on CT or MRI results (Kondo et al, 2007; Gordon & Mercandetti, 1989; Davous et al, 1986; Min, 1986; Handa & Tai, 2005; Schwartz et al, 1985; Choi, 1983; Jellinger, 1968).

a) Diffuse white matter changes, including demyelination, may also occur (Lugaresi et al, 1990; Hart et al, 1988; Min, 1986; Choi, 1983; Lacey, 1981). Ischemia in the posterior temporoparietal area has also been reported in an adult exposed to CO (Aslan et al, 2004 ).
b) Bilateral cerebellar swelling causing acute hydrocephalus has been observed on CT scan (Prabhu et al, 1993). Interval development of hydrocephalus with dilatation of lateral and third ventricles and a normal appearing aqueduct and fourth ventricle occurred in a 2.5-year-old child on the third hospital day following CO exposure in a fire (So et al, 1997).

7) EEG FINDINGS: Widespread slow-wave activity, especially in the bilateral frontotemporal region, has been reported in a young adult with parkinsonian symptoms starting 20 days after acute exposure (Gallerani et al, 2000).
8) In a retrospective review of 89 patients hospitalized with acute CO intoxication, 12 patients were identified with delayed encephalopathy. Delayed encephalopathy occurred 14 to 45 days following the acute phase of CO poisoning and consisted of cognitive impairment, akinetic mutism, sphincter incontinence, gait ataxia, and extrapyramidal signs and symptoms (ie, chorea, dystonia, and parkinsonism). Initial brain MRI studies showed multiple hyperintensity lesions, primarily in the basal ganglia (n=12) but also in the subcortical white matter (n=9) and in the midbrain (n=1). In the basal ganglia, the globus pallidus was primarily involved (n=10). Serial brain MRI studies were conducted in 6 of the 12 patients, and in all 6 patients there was prominent improvement in the subcortical white matter and globus pallidus lesions, indicating that the delayed encephalopathy may be a reversible disorder (Hsiao et al, 2004).
9) CASE REPORT/CHILD: A 5-year-old child developed delayed neuropsychiatric effects, including lethargy, visual and gait disturbances, and hemiparesis, approximately 2 days after recovery following initial CO exposure. Initial MRI and fluid-attenuated inversion recovery (FLAIR) imaging indicated high signal intensity in the hippocampus and occipital and frontal cortex. Following 10 days of hyperbaric oxygen therapy , the patient gradually began talking and walking normally; however, left-hand paresis and left hemianopsia continued. A follow-up MRI demonstrated high signal intensity of the globus pallidus, although the hippocampal and cortical lesions were no longer detected. Single-photon emission computed tomography (SPECT) revealed patchy hypoperfusion in the cerebral cortex, which persisted 4 months later (Kondo et al, 2007).
10) In 7 patients suffering from CO poisoning and 7 suffering from organophosphate (OP) poisoning, SPECT images showed hypoperfused brain lesions in the frontal, temporal, and parietal lobes of all CO poisoning patients and 4 OP poisoning patients. Neuropsychological sequelae, including disorientation problems and confusion, were reported in 6 CO patients and 5 OP patients, including 3 who had normal SPECT scans.(Ozyurt et al, 2008).
11) DIFFUSION TENSOR IMAGING FINDINGS: A 45-year-old woman presented to the hospital comatose (Glasgow coma scale (GCS) of 8) with a diagnosis of carbon monoxide poisoning (carboxyhemoglobin concentration 23.5%). Following several sessions of HBO therapy, the patient became alert (GCS 15) on hospital day 4; however, an initial MRI of the brain revealed bilateral globus pallidus necrotic lesions. Diffusion tensor imaging (DTI), performed on day 23, showed a decline of fractional anisotropy (FA) and apparent diffusion coefficient (ADC), although a repeat MRI showed no apparent change from the initial MRI. On hospital day 24, the patient's mental status rapidly deteriorated, and a third MRI detected white matter abnormalities, resulting in a diagnosis of delayed encephalopathy (DE). Despite supportive therapy and treatment with HBO, the patient continued to decline neurologically. On hospital day 35, she was at a GCS of 11, combined with parkinsonism. Her ADC and FA levels were at their lowest values on day 38 and day 54, respectively. With continued supportive therapy, the patient's mental status improved and, following 3 months of rehabilitation, she was discharged without neurologic sequelae (Kuroda et al, 2012).

a) The decline of FA values that occurred prior to the patient's clinical symptoms of DE, appeared to directly correspond to the patient's neurologic deficits, and the decrease and subsequent increase of ADC values, directly corresponding to the patient's decrease in level of consciousness with subsequent consciousness recovery, suggests that DTI may be useful in early detection of DE following carbon monoxide poisoning (Kuroda et al, 2012).

12) PROGNOSIS

a) In a large study, 60% of patients with severe delayed neurologic effects recovered fully; another 15% had minor memory disturbances (Choi, 1983).
b) Comatose patients with abnormal results on head CT scans appear to have a worse prognosis than those with normal scans (Sawada et al, 1980).
c) Hyperbaric oxygen therapy is thought to improve the prognosis of patients by attenuating or eliminating delayed neurologic sequelae (Thom et al, 1995). However, normal neuropsychologic and functional outcomes are possible after severe CO poisoning, even without the use of hyperbaric oxygen therapy (Weaver et al, 1996).
d) PROGNOSTIC INDICATOR: A prospective cohort study, involving 5 patients with CO poisoning, was conducted to determine the value of measuring myelin basic protein (MBP) in cerebrospinal fluid (CSF) for use as a predictive marker of delayed encephalopathy in patients with CO poisoning. Of the 5 patients, 3 patients developed delayed encephalopathy (DE) and were classified as group DE, and the other 2 patients did not develop delayed encephalopathy and were classified as group non-DE. Within group DE, the MBP concentration was elevated during the acute stage in patient 1, then decreased below the detection limit before becoming elevated prior to the clinical manifestations of DE. In patients 2 and 3, the MBP was below the limit of detection in the acute stage but markedly increased prior to development of DE. Within group non-DE, the MBP concentrations in both patients remained below the limit of detection. Although the sample size of this study was small, based on the results, it is suggested that measurement of MBP in the CSF may be a useful predictive marker for the development of CO-associated DE; further studies are warranted (Ide & Kamijo, 2008).
e) A prospective cohort study, involving 21 patients with CO poisoning and 31 healthy controls, was conducted to determine an association between changes in dopamine transporter (DAT) availability and cognitive function in patients with CO poisoning. At baseline, DAT availability in the CO-poisoned patients was significantly lower in the left and right striatum (2.30 +/-0.71 and 2.29+/- 0.64, respectively), as compared to the healthy controls (3.15 +/- 0.87 and 3.08 +/-0.85, respectively). In addition, cognitive function, based on performance of several psychological tests administered that measured attention, verbal memory, visual memory, and executive functions, was also significantly worse in the CO-poisoned patients as compared to the healthy controls. At a six-month follow-up examination, there was no significant change in DAT availability in the CO-poisoned patients; however, cognitive function showed significant improvement in the majority of psychological tests, with the exception of the Wisconsin Card Sorting Test (WCST), measuring executive function (Yang et al, 2015).
f) CLINICAL GRADING/OUTCOME: A prospective observational study was conducted in a single tertiary medical center in Japan from 2006 to 2014 to determine the factors predicting the outcome of patients with delayed neurologic sequelae (DNS) following carbon monoxide (CO) poisoning. Of the 102 patients with CO poisoning who were enrolled in the study, 100 patients completed the study. Of the 100 patients, 20 patients developed DNS. The patients with DNS were then graded according to their clinical course: Grade 1 (n=6) was consistent independence with development of mild neurologic deficits that resolved within a few months; Grade 2 (n=10) was transient dependence with moderate cognitive and motor deficits with gradual recovery and independence within 1 year; Grade 3 (n=4) was persistent dependence with severe cognitive and motor deficits and the patients remained dependent despite any neurologic improvements. Analysis of the patient demographics revealed that Grade 3 patients were significantly older than Grade 1 patients (median age 77 years vs 41 years; p < 0.01). Clinically, the onset of DNS was significantly later in Grade 1 patients than in Grade 3 patients or Grade 2 patients (median interval after acute CO poisoning: 35 days vs 10 days [p<0.001] and 25 days [p < 0.05], respectively). Analysis of the cerebrospinal fluid (CSF) in the DNS patients demonstrated that, in the acute phase (defined as within 7 days after CO exposure), there were no significant differences between the 3 grades; however, in the recovery phase (defined as 1 month after CO exposure), the CSF-MBP levels of the Grade 3 patients were significantly higher than in the Grade 1 patients (median 474 pg/mL vs 86 pg/mL; p < 0.05) and the myelin basic protein (MBP) index in the Grade 3 patients was also significantly higher than in the Grade 1 patients (median: 37.2 vs 3.2; p < 0.01). Based on the results of this study, poor DNS outcomes (Grade 3) are associated with advanced age, earlier onset of DNS, a higher 1-month CSF-MBP, and a higher MBP index (Kuroda et al, 2015).
g) PROGNOSTIC INDICATOR/30-DAY NEUROLOGIC SEQUELAE: A retrospective analysis was conducted of patients with carbon monoxide poisoning, who reported to one of 4 hospitals in the Shandong Province, China from 1990 to 2011, to determine predictors of 30-day neurological sequelae in these patients. The study consisted of 258 patients, with ages ranging from 6 to 97 years. Of these patients, 139 patients (53.9%) had 30-day neurological sequelae. Univariate analysis demonstrated a higher risk (p<0.1) for 30-day neurological sequelae in patients with the following variables: age, respiratory rate, occupational exposure, altered mental status, lack of pupil reflex, loss of consciousness, headache, dizziness, positive Babinski's reflex, and urinary incontinence. Following a multivariate logistic regression analysis of these variables, it was determined that the lack of pupil reflex and loss of consciousness were the only independent predictors of 30-day neurological sequelae in patients with carbon monoxide poisoning. With either variable present, the sensitivity was 77%, the specificity was 47.1%, and the positive predictive (PPV) and negative predictive values (NPV) were 62.9% and 63.6%, respectively. With both variables present, the sensitivity, specificity, PPV, and NPV were 11.5%, 99.2%, 94.1%, and 49%, respectively (Zou et al, 2015).
h) LONG-TERM OUTCOME

1) Although the data appear to be inconclusive and inconsistent, it has been estimated that approximately 30% of patients with severe poisoning have a fatal outcome. In a study, 11% of patients developed long-term neuropsychiatric deficits, of which 3% were delayed. Ongoing morbidity in the form of lasting subtle memory deficits or personality changes may occur in up to 33% of CO poisonings (Varon et al, 1999).
2) PERMANENT DISABILITY/AMNESIA: Anterograde and retrograde amnesia were reported in a 22-year-old man following CO poisoning. The patient had an intact memory up until the age of 18. MRI showed bilateral hippocampal infarcts, while the remaining neurological exams were within normal limits. The patient was transferred to a rehabilitation facility following medical discharge (Bourgeois, 2000). Similar symptoms of amnesia were reported in another young adult following acute exposure (Gallerani et al, 2000).
3) TIC DISORDER: An adult female developed a permanent tic disorder several weeks after CO poisoning in which she was comatose for 3 days. The patient had continual symptoms over the course of 11 years. Neurologic exam was normal except for repetitive stereotypic movements. Brain MRI showed bilateral symmetric high signal intensities on T2-weighted images and low signal intensities on T1-weighted images in the globus pallidus, and this was thought to be cavitary necrosis secondary to CO poisoning. Clonazepam was beneficial in relieving most symptoms (Ko et al, 2004).
4) INCREASED RISK OF PARKINSON DISEASE: A retrospective analysis of patients with carbon monoxide poisoning was conducted to determine an association between carbon monoxide intoxication and the development of Parkinson disease. The study included 9012 patients hospitalized with carbon monoxide poisoning between 2000 and 2011. For each person with carbon monoxide poisoning, 4 comparison persons without carbon monoxide poisoning were randomly identified and matched by age, sex, and index date of hospitalization (n=36,048). The majority of patients were younger than 34 years (42.9% in both groups). Comorbidities were significantly greater at baseline in the carbon monoxide group (p<0.001) and included diabetes, hypertension, hyperlipidemia, head injury, depression, stroke, dementia, and chronic kidney disease. After 12 years of follow-up, the cumulative incidence of Parkinson disease was 1.47% greater in the carbon monoxide cohort than in the non-carbon monoxide cohort and, after adjusting for age, sex, and comorbidities, there was a 9.08-fold higher risk for the patients with carbon monoxide poisoning to develop Parkinson disease than in the non-carbon monoxide group (Lai et al, 2015).
5) CASE REPORT: A 32-year-old woman, who developed pallidal lesions following CO poisoning, experienced diurnal bruxism, depression, and psychic akinesia, with a marked decrease in spontaneous activities and speech, 2 years later. Although her Mini-Mental Status Examination score was 30, she exhibited a decrease in her thinking, verbal fluency, and intellectual processes. A cerebral perfusion single photon emission computed tomograph (SPECT) indicated decreased perfusion in the left frontal lobe. Despite various treatments with dopaminergic agents, anticholinergics, antidepressants, antipsychotics, and psychotherapy, her symptoms persisted (Liang et al, 2011).
6) CASE REPORT: A 42-year-old man presented as comatose (Glasgow coma scale of 3) after being found in his room that was filled with burning charcoal smoke. Laboratory data revealed a carboxyhemoglobin concentration of 30.9%. Following intubation and mechanical ventilation with 100% oxygen, the patient's carboxyhemoglobin concentration decreased with an improvement in consciousness. After the patient was extubated, he was noted to have mild right-sided weakness. An MRI of the brain indicated a left subcortical ischemic infarction. With supportive care, he recovered approximately 2 weeks after admission; however, 6 weeks post-hospitalization, delayed neurologic sequelae were noted, including memory loss, slowness in walking and talking, spasticity in his extremities, personality changes, and dysphagia. A repeat brain MRI revealed symmetrical white matter hyperintensities. The patient's condition remained unchanged 12 months later (Ahmad & Sharma, 2012).
7) CASE REPORT/GRINKER MYELINOPATHY: A 41-year-old woman became comatose after intentional inhalation of fumes from a charcoal barbecue. Her Glasgow Coma Scale (GCS) score was 5 and her blood carboxyhemoglobin concentration was 18.7%. Following intubation and transfer to the hospital, hyperbaric oxygen therapy was initiated, however she remained comatose (GCS 6). An MRI revealed bilateral hemorrhage in her globi pallidi. Three weeks post-admission, she was weaned off the ventilator; however, despite early and supportive care, she remained in a vegetative state (GCS 7) with decorticated limb rigidity. A repeat MRI revealed diffuse demyelination lesions in the white matter. The patient's clinical condition in combination with her MRI findings were consistent with Grinker myelinopathy, also known as delayed posthypoxic leukoencephalopathy (Hoffmann et al, 2016).

b) DELAYED SUBTLE NEUROPSYCHOLOGIC SEQUELAE: Another syndrome of delayed subtle neuropsychologic effects has been described.

1) SYMPTOMS: Effects include headache, anorexia, nausea, apathy, lethargy, forgetfulness, subtle personality changes and memory problems, irritability, and dizziness (Siddiqi et al, 1997; Myers & Britten, 1989) (Myers et al, 1985) (Smith & Brandon, 1973).
2) PHYSICAL FINDINGS: Generally, no gross abnormalities are found on physical or neurologic exam. Neuropsychometric testing is usually required to identify abnormalities (Myers et al, 1995)(Thom et al, 1995).
3) INCIDENCE: With more severe acute toxicity or higher carboxyhemoglobin levels (greater than 20%), even with minimal symptoms, the incidence is estimated to be 8% to 12% in patients not treated with hyperbaric oxygen (Myers et al, 1985) (Myers, 1984).

a) CASE STUDY: In a prospective, randomized, nonblinded study of patients with moderate CO poisoning presenting within 6 hours of exposure, 7 of 30 (23%) patients treated with ambient oxygen developed delayed neurologic sequelae, compared with 0 of 30 patients treated with hyperbaric oxygen (Thom et al, 1995). The development of delayed neurologic sequelae was determined by neuropsychologic testing and questionnaire.

1) This study has been criticized because the lack of double blinding permits investigator bias in assessing the results of neuropsychologic tests and placebo effect in the subjects (Weaver et al, 1995).

b) CASE STUDY: In another prospective, controlled, nonblinded study, 41 adults were evaluated after suicide attempt by CO poisoning, along with 20 matched controls. Cognitive function was tested initially and at a 2-month follow-up with psychological tests and other instruments to assess orientation, attention, concentration, speed of information processing, verbal memory, premorbid intellect, executive function, and mood disorder. (Hay et al, 2002).

1) The findings suggested that a minority of people with high levels of carboxyhemoglobin soon after suicide attempt may suffer severe memory and executive function deficits. However, in most cases, levels of cognitive impairment and deficits were moderate to small and were not greater than the control group of depressed suicide subjects (Hay et al, 2002).
2) The statistical power of this study was limited by the fact that the authors were unable to recruit a large number of control subjects, and they were less likely to return for follow-up. The control group was also found to have higher levels of depression than the CO-exposed group (Hay et al, 2002).

4) ONSET: Onset of subtle delayed neuropsychological impairment was 1 to 21 days (mean 5.7) after exposure (Myers et al, 1985).
5) PEDIATRICS: During a 2-year period, children (n=140 less than 15 years) admitted with CO poisoning were enrolled into a study to determine the possible differences in clinical symptoms and outcome as compared to adults with CO poisoning. Almost all children were free of symptoms 1 year later (1 died of a cardiac arrest and 1 had secondary neurological manifestations), and late neurological findings were less frequent than in adults (Mathieu & Mathieu-Nolf, 2001).

c) VARIATIONS AMONG INDIVIDUALS

1) CASE REPORT: A faulty water heater exhaust system resulted in 2 pairs of adult siblings being exposed to CO that resulted in similar levels of CO poisoning (17% to 29%) . All patients experienced a decrease in cognitive function and ongoing emotional-behavioral problems. The authors noted that significant variability in the extent of cognitive, vocational, and affective results occurred in each patient despite similar levels of CO exposure, age, and preexposure cognitive function (Dunham & Johnstone, 1999).

C) LEUKOENCEPHALOPATHY

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Hemorrhagic leukoencephalopathy was reported in a 20-year-old woman following CO exposure. Following presentation to the emergency department, her carboxyhemoglobin level was 35% and she was subsequently treated with hyperbaric oxygen therapy. A CT scan, performed the next day, revealed symmetrical hypodensities within the globus pallidus and hyperdensities at the anterior and posterior vascular perfusion boundaries; an EEG demonstrated generalized slowing consistent with diffuse encephalopathy. An MRI, obtained 7 days later, demonstrated T1-hyperintensities; fluid-attenuated inversion recovery (FLAIR) imaging showed extensive multifocal hyperintensities within the corpus callosum, subcortical white matter, and deep gray matter; and diffusion-weighted imaging showed restricted diffusion in those same areas, all of which are indicative of hemorrhagic leukoencephalopathy (Seet et al, 2008).

D) LOSS OF CONSCIOUSNESS

1) WITH POISONING/EXPOSURE

a) Eighty-three healthy men and women exposed to CO from heating stoves or gas-operated hot water heaters were studied. CO exposure ranged between 20 minutes and 480 minutes (average 52.5 minutes). The most common complaint was loss of consciousness (n=52; 62.7%), followed by palpitations (n=33; 39.8%), headache (n=27; 32.5%), confusion (n=20; 24.1%), dizziness (n=18; 21.7%), dyspnea (n=12; 14.5%), and weakness and fatigue (n=11; 13.3%) (Aslan et al, 2006).

E) DEPRESSIVE DISORDER

1) WITH POISONING/EXPOSURE

a) A prospective study was conducted to determine if severity of CO poisoning was directly proportional to cognitive and affective outcomes. The study involved 55 patients with less severe CO poisoning, as defined by no loss of consciousness and a carboxyhemoglobin level of 15% or less, and 201 patients with more severe CO poisoning, as defined by carboxyhemoglobin levels greater than 15% or loss of consciousness. Cognitive sequelae, depression, and anxiety outcomes were measured at 6 weeks, 6 months, and 12 months postexposure. There appeared to be no difference in either cohort with regards to the prevalence of cognitive sequelae. At 6 months, there was a significantly greater prevalence of depression in the less severe group as compared with the more severe group (19% (n=47) vs 11% (n=132; p=0.04). There was a significantly greater prevalence of anxiety in the less severe group at 6 weeks than in the more severe group (30% (n=44) vs 11% (n=114); p=0.008); however, anxiety appeared to decrease over time in the less severe group. Cognitive sequelae, depression, and anxiety appear to be independent of CO poisoning severity (Chambers et al, 2008).

F) CEREBRAL INFARCTION

1) WITH POISONING/EXPOSURE

a) CASE REPORT: A 42-year-old man presented as comatose (Glasgow coma scale of 3) after being found in his room that was filled with burning charcoal smoke. Laboratory data revealed a carboxyhemoglobin concentration of 30.9%. Following intubation and mechanical ventilation with 100% oxygen, the patient's carboxyhemoglobin concentration decreased with an improvement in consciousness. After the patient was extubated, he was noted to have mild right-sided weakness. An MRI of the brain indicated a left subcortical ischemic infarction. With supportive care, he recovered approximately 2 weeks post-admission; however, 6 weeks post-hospitalization, delayed neurologic sequelae were noted, including memory loss, slowness in walking and talking, spasticity in his extremities, personality changes, and dysphagia. A repeat brain MRI revealed symmetrical white matter hyperintensities. The patient's condition remained unchanged 12 months later (Ahmad & Sharma, 2012).

G) LESION OF BRAIN

1) WITH POISONING/EXPOSURE

a) CEREBELLAR LESION/CASE REPORT: An 11-year-old boy presented to the emergency department comatose (Glasgow Coma Scale score of 7) due to suspected carbon monoxide poisoning. A cranial MRI demonstrated diffusion restriction of right-dominant cerebellar cortical and vermis, and hyperintense bilateral supratentorial white matter, globus pallidus, and periventricular deep white matter. The patient's carboxyhemoglobin concentration was 33.7%. He regained consciousness after the first session of hyperbaric oxygen therapy; he received daily therapy for a total of 5 days (Zhu et al, 2014).

H) INJURY OF BRACHIAL PLEXUS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: A 42-year-old man developed brachial weakness associated with facial and upper extremity edema after being unconscious for approximately 3 hours following unintentional carbon monoxide poisoning. Three days later, a neurological examination revealed brachial diplegia, hypoesthesia, and deep tendon areflexia of his upper extremities. Laboratory data revealed an elevated creatine kinase concentration of 693 units/L. Electroneuromyography indicated a pattern consistent with a diagnosis of bilateral brachial axonal plexus injury primarily on the left side. Following 10 sessions of hyperbaric oxygen therapy, the patient improved with remission of sensory and motor deficits as well as resolution of edema. At follow-up 4 months later, the patient had completely recovered without sequelae (Rahmani et al, 2013).

3.7.3)

A) ANIMAL STUDIES

1) BRAIN STEM DISORDER

a) Selective findings include degeneration of the cerebral cortex, limbic system, and brainstem (Wright & Shephard, 1979).

Gastrointestinal

3.8.2)

A) NAUSEA AND VOMITING

1) WITH POISONING/EXPOSURE

a) Nausea and vomiting are common (Uyanik et al, 2011; Henz & Maeder, 2005; Swank et al, 2004; CDC, 1995). Symptoms may mimic acute gastroenteritis (Sternbach & Varon, 1991; Gasman et al, 1990; Gemelli & Cattani, 1985).
b) INCIDENCE: According to the National Electronic Injury Surveillance System All Injury Program (NEISS-AIP), in 405 cases of carbon monoxide (CO) exposure which transpired from 2001 to 2003, nausea occurred in 17.3% of cases and vomiting, in 7.7% (Anon, 2005).
c) CASE REPORT: A 20-year-old woman presented with nausea, mild headache, and severe light-headedness approximately 1 hour after using a waterpipe. The patient frequently used a waterpipe several days per week, with each session lasting 45 to 60 minutes, and had experienced similar symptoms in the past. At presentation, the patient was lethargic, but the rest of her physical exam was normal. Venous blood gas levels, obtained at presentation, and arterial blood gas levels, obtained 20 minutes post-presentation, revealed carboxyhemoglobin concentrations of 25.4% and 23.9%, respectively, leading to a diagnosis of carbon monoxide poisoning secondary to waterpipe use. Following 24 hours of observation and supportive care, the patient recovered and was discharged (Wang et al, 2015).

B) VASCULAR INSUFFICIENCY OF INTESTINE

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Bowel necrosis has been reported, manifested by abdominal pain and loose bloody stools 12 hours after exposure (Watson & Williams, 1984).

C) SERUM AMYLASE RAISED

1) WITH POISONING/EXPOSURE

a) Elevations in amylase (salivary-type) levels in the blood were reported in 40% of patients in a study (Tahahashi et al, 1982).

D) ISCHEMIC COLITIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: An 87-year-old woman experienced diffuse abdominal pain with blood in the stools approximately 30 minutes after presenting to the emergency department with CO poisoning. Initially her hemoglobin and carboxyhemoglobin levels were 11.7 g/dL and 19.7%, respectively. A colonoscopy revealed mucosa with edema and erythema, submucous hemorrhages, and ulcerations within the sigmoid colon, indicative of ischemic colitis. With supportive therapy, the patient recovered without sequelae (Duenas-Laita et al, 2008).
b) CASE REPORT: A 34-year-old woman, exposed to carbon monoxide after starting her vehicle in a closed garage, developed difficulty in breathing, nausea and vomiting, abdominal pain, and bloody diarrhea. Her initial carboxyhemoglobin level was 23%. Treatment included hyperbaric oxygen therapy, resulting in a decrease in her carboxyhemoglobin level to 2.6%; however, following therapy, her abdominal pain became worse and she continued to have severe nausea and vomiting refractory to antiemetic agents. A CT scan of her abdomen and pelvis, performed 6 hours later, revealed mesenteric ischemia manifested by diffuse colonic thickening, a sigmoidoscopy demonstrated edematous, friable pale mucosa extending from her rectum to her distal sigmoid colon, and a colon biopsy revealed acute inflammatory cells, fibrin present in the lamina propria, and mucosal erosions, all of which was consistent with the diagnosis of ischemic colitis. With supportive care, the patient gradually recovered and was discharged to an inpatient psychiatric unit on day 4 (Weaver & Deru, 2016).

Hepatic

3.9.2)

A) HEPATIC NECROSIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Hepatic necrosis has been reported, with prolonged prothrombin time noted 12 hours postexposure and jaundice and elevated transaminases 48 hours after exposure (Watson & Williams, 1984).

Genitourinary

3.10.2)

A) RENAL FAILURE SYNDROME

1) WITH POISONING/EXPOSURE

a) Oliguric and nonoliguric acute renal failure have been reported (Bessoudo & Gray, 1978; Leavell et al, 1969; Jackson et al, 1959), sometimes associated with rhabdomyolysis (Kade et al, 2012; Vossberg & Skolnick, 1999; Florkowski et al, 1992).

B) MYOGLOBINURIA

1) WITH POISONING/EXPOSURE

a) Myoglobinuria may occur (Vossberg & Skolnick, 1999).

C) ALBUMINURIA

1) WITH POISONING/EXPOSURE

a) Albuminuria, hematuria, and acute tubular necrosis may develop (Leavell et al, 1969; Jackson et al, 1959).

Acid-Base

3.11.2)

A) METABOLIC ACIDOSIS

1) WITH POISONING/EXPOSURE

a) CASE SERIES: Metabolic acidosis (pH 7.28 to 7.34; standard bicarbonate 17 to 19 mmol/L; pCO2 35 to 38 mmHg) was reported in 5 of 12 patients with carbon monoxide (CO) poisoning (Handa & Tai, 2005).
b) According to a retrospective review of cases that occurred from 1996 to 2007, 5 pediatric patients out of 30 (16.7%) developed severe metabolic acidosis after CO exposure (Cho et al, 2008).
c) Metabolic acidosis (pH 7.25, pCO2 33 torr, pO2 110 torr, base excess -11.5 mmol/L) occurred in a 34-year-old woman who was exposed to carbon monoxide after starting her vehicle in a closed garage (Weaver & Deru, 2016).

Hematologic

3.13.2)

A) HYPOXEMIA

1) WITH POISONING/EXPOSURE

a) OXYGEN CARRYING CAPACITY: Carbon monoxide (CO) combines with hemoglobin to form carboxyhemoglobin. This results in a decreased oxygen carrying capacity and a shift to the left of the oxygen-dissociation curve (reducing the unloading of oxygen to the tissues).

B) THROMBOCYTOPENIC PURPURA

1) WITH POISONING/EXPOSURE

a) Thrombocytopenic purpura has been reported in association with CO exposure (Stonesifer et al, 1980).

C) HEMOLYTIC ANEMIA

1) WITH POISONING/EXPOSURE

a) Hemolytic anemia has been reported following acute exposure (Nagy et al, 1979; Leavell et al, 1969).

D) ERYTHROCYTOSIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Polycythemia has been reported in a 62-year-old woman with chronic CO exposure (DiMarco, 1988).

E) BLEEDING

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Acute postpartum hemorrhage with mental status alterations was reported in a 41-year-old woman 2 weeks after a normal vaginal delivery. Symptoms were associated with acute CO poisoning (ie, a faulty gas heater in the home). Mental status changes and bleeding resolved after she received 100% oxygen over a 4-hour period. Of note, other family members were also treated for high carboxyhemoglobin levels (Ramsey et al, 2001).

F) LEUKOCYTOSIS

1) WITH POISONING/EXPOSURE

a) Leukocytosis has been reported following acute exposure (Vossberg & Skolnick, 1999).

G) ATYPICAL LYMPHOCYTES

1) WITH POISONING/EXPOSURE

a) In a small convenience sample of patients who had CO poisoning, oxidative damage of circulating lymphocytes increased after poisoning. Alteration in lymphocytes may have been due in part to inhibition by mitochondrial cytochrome c oxidase (COX). The pathophysiological impact has yet to be determined (Miro et al, 1999).

H) VENOUS THROMBOSIS

1) WITH POISONING/EXPOSURE

3.13.3)

A) ANIMAL STUDIES

1) POLYCYTHEMIA

a) A compensatory increase in red blood cell mass has been a consistent finding (Stewart, 1975).

Dermatologic

3.14.2)

A) BULLOUS ERUPTION

1) WITH POISONING/EXPOSURE

a) Skin changes may include bullae, vesicles, areas of edema, and erythematous patches (Myers et al, 1985a). An association between skin changes and poisoning severity has been suggested.
b) CASE REPORT: Bullous lesions on the left heel with copious serous drainage (estimated 4 L) were reported in a patient following a failed suicide attempt with CO. Under the initial bullae, profound epidermal and dermal lysis occurred. Complete resolution occurred after 8 months of local wound care (Johnson et al, 1999).

B) DISORDER OF SKIN

1) WITH POISONING/EXPOSURE

a) Cherry red skin and mucous membranes are NOT commonly found (Norkool & Kirkpatrick, 1985; Burney et al, 1982; Larcan & Lambert, 1981).

C) ALOPECIA

1) WITH POISONING/EXPOSURE

a) Delayed localized alopecia may develop rarely (Nagy et al, 1979; Leavell et al, 1969; Long, 1968).

D) FROSTBITE

1) WITH POISONING/EXPOSURE

a) Rapid release of compressed gas may cause frostbite (NFPA, 1991).

Musculoskeletal

3.15.2)

A) NECROSIS

1) WITH POISONING/EXPOSURE

a) Muscle necrosis may occur even in the absence of local pressure necrosis. Elevated CPK and myoglobinuria are characteristic (Florkanski et al, 1992).
b) CASE REPORT: Despite therapy with hyperbaric oxygen, myonecrosis with a peak CPK of 65,988 international units developed in a patient who had been rendered unconscious by CO (Herman et al, 1988).
c) CASE SERIES: Myonecrosis was not seen in any of 81 patients with elevated CO levels; however, the report did not indicate the severity of poisoning (Shapiro et al, 1989).
d) Serum LDH and CPK levels may rise transiently following acute CO poisoning (Penney & Maziarka, 1976). Following a failed suicide attempt, CPK of 131,805 international units/L was reported, with no myoglobinuria (Vossberg & Skolnick, 1999).

B) RHABDOMYOLYSIS

1) WITH POISONING/EXPOSURE

a) CASE REPORT: Rhabdomyolysis was accompanied by absent or decreased glycolytic enzyme activities in 2 cases of severe CO poisoning, indicating a direct toxic effect on muscle (Florkowski et al, 1992).
b) Rhabdomyolysis was reported in a 40-year-old man following acute CO intoxication from a defective vehicle heating system. Myoglobinuria was present, with elevated CPK and LDH levels (12,900 units/L and 974 units/L, respectively) (Vacchiano & Torino, 2001).
c) Rhabdomyolysis (creatinine kinase level 15,951 units/L) was reported in a 28-year-old man following intentional exposure to CO (Jang & Park, 2010).
d) CASE REPORT: Acute renal failure with mild rhabdomyolysis (peak creatine phosphokinase level 2500 units/L) was reported in a 33-year-old man with carbon monoxide poisoning following exposure to a propane gas heater in a bathroom. With supportive treatment, including intermittent hemodialysis sessions, the patient completely recovered and was discharged approximately 3 weeks post-admission (Kade et al, 2012).

C) MOVEMENT DISORDER

1) WITH POISONING/EXPOSURE

a) CASE SERIES: Delayed movement disorders were diagnosed in 32 (13.2%) of 242 patients who had CO poisoning. Disorders reported included parkinsonism (n=23), dystonia (n=5), chorea (n=3), and myoclonus (n=1) (Choi & Cheon, 1999).

1) Onset of symptoms among the various disorders was between 4 and 51 weeks. No correlation was found between radiological findings and the occurrence of movement disorders. Despite the occurrence of delayed CO encephalopathy, all patients recovered. Those with abnormal dyskinesias recovered in 2 to 8 weeks and those with parkinsonism, within 6 months (Choi & Cheon, 1999).
2) INCIDENCE: The incidence of parkinsonism after CO poisoning in a study was 9.5% (Choi, 2002)
3) ONSET: The age of onset of parkinsonism following CO poisoning ranged from 16 to 69 (mean 45.8) years. However, the peak incidence occurred in patients between the ages of 50 and 60 (Choi, 2002).
4) LATENCY PERIOD: The appearance of symptoms of parkinsonism varied from 2 to 26 weeks, with most symptoms developing within 4 weeks following acute anoxia (Choi, 2002).
5) CHARACTERISTIC SIGNS/SYMPTOMS: The most frequent symptoms of parkinsonism with encephalopathy gait disturbances, impaired mental ability, and urinary incontinence (Choi, 2002).
6) CASE SERIES/CHOREA: Chorea is a rare finding following CO poisoning and occurs as a delayed effect. In a series of 6 patients with chorea following CO poisoning, the latency period was 10 to 30 days (mean 21.7 days). The duration of symptoms was 14 to 90 days (mean 39.8 days). Neuroleptic agents usually alleviate symptoms without relapse. Prognosis is usually considered good (Park & Choi, 2004).

D) COMPARTMENT SYNDROME

1) WITH POISONING/EXPOSURE

a) Compartment syndrome may also occur, but it appears to be uncommon (Herman et al, 1988; Slevin, 1979; Finley et al, 1977; Jackson et al, 1959) Shapiro, 1989).
b) Compartment syndrome that required an urgent fasciotomy has been reported. The patient was exposed for 8 to 10 hours to the exhaust of a garaged automobile with a catalytic converter (Vossberg & Skolnick, 1999).

Endocrine

3.16.2)

A) HYPERGLYCEMIA

1) WITH POISONING/EXPOSURE

a) Hyperglycemia may be associated with carboxyhemoglobin levels greater than 25% (Leiken et al, 1988).

3.16.3)

A) ANIMAL STUDIES

1) HYPOGLYCEMIA

a) INTRACELLULAR GLUCOSE: Acute carbon monoxide (CO) exposure in experimental animals resulted in a rapid decrease in intracellular CNS glucose levels in the presence of unchanged or increased blood glucose levels (MacMillan, 1977).

Reproductive

3.20.1)

A) CO exposure during pregnancy is teratogenic, depending upon the stage of pregnancy. The fetus is more vulnerable to CO poisoning than the mother.

3.20.2)

A) HUMANS

1) CONGENITAL ANOMALY

a) Carbon monoxide (CO) appears to have teratogenic and embryotoxic potential when exposures are sufficient to cause significant increase in maternal carboxyhemoglobin concentrations or moderate to severe maternal toxicity (Norman & Halton, 1990).
b) Exposure of the fetus can result in permanent brain damage, including mental retardation, limb malformation, hypotonia, areflexia, basal ganglia damage, neuronal loss in the cerebral cortex, microcephalus, low infant birth weight (Wouters et al, 1987; Alderman et al, 1987; Cramer, 1982), telencephalic dysgenesis, seizures, and stillbirth (Barlow & Sullivan, 1982; Caravati et al, 1987; Farrow et al, 1990; Woody & Brewster, 1990; Caravati et al, 1988; Brown et al, 1992). Fetal death due to nonlethal CO poisoning has been reported. Maternal carboxyhemoglobin level was 7% and fetal level was 61% (Farrow et al, 1990).
c) CASE REPORT: Telencephalic dysgenesis was chronologically associated with repeated maternal exposures to intoxicating concentrations of CO around 6 weeks of fetal life (Woody & Brewster, 1990).
d) CASE REPORT, MILD CHRONIC POISONING: Fetal exposure from the end of the second month of gestation through the seventh month has been associated with multiple fetal abnormalities consisting of cleft lip and palate, low-set ears, hypoplastic external genitalia, bilateral retinal colobomata, and multiple fatal cardiac lesions (Hennequin et al, 1993).
e) CASE REPORT: A 2.5-year-old child developed dystonic cerebral palsy following nonlethal maternal exposure to CO at 20 weeks' gestation. An MRI at 2 months of age revealed hypoxic-ischemic lesions in the globus pallidus, and continuous follow-up of the infant showed progressive neurologic deterioration with development of gross motor retardation, spasticity and severe dystonia of the upper extremities, and mild mental retardation (Alehan et al, 2007).
f) CASE REPORT: A 40-year-old woman at 22 weeks' gestation and her 9-year-old son were exposed to CO. Her son died, but she recovered following 2 days of supportive therapy. At the time, an ultrasound examination of her fetus was normal. Four weeks later, a repeat ultrasound examination revealed fetal intracranial cystic lesions. Because of a poor neurological prognosis, the pregnancy was terminated. Pathological examination of the fetal brain revealed porencephalic cysts (likely due to vascular occlusion) within the cerebral parenchyma and excessive ventricular dilatation (Gul et al, 2009).

B) ANIMAL STUDIES

1) CO was not teratogenic in mice or rabbits (Schwetz, 1979). CO was teratogenic in rabbits exposed to an airborne concentration of 90 parts per million (ppm) or 180 ppm (Astrup, 1972). Pregnant rats exposed by inhalation to airborne concentrations of 60 ppm, 125 ppm, 250 ppm, or 500 ppm delivered pups with enlarged hearts (Penney, 1980; Prigge & Hochrainer, 1977). Other studies in undefined species reported effects on cardiac development (Fechter, 1980) and on fetal development (Astrup, 1972a).

3.20.3)

A) HUMANS

1) PLACENTAL BARRIER

a) PLACENTAL TRANSFER: CO readily crosses the placenta. Fetal carboxyhemoglobin concentration is 10% to 15% greater than the corresponding maternal carboxyhemoglobin level at steady state. Elimination of CO from the fetal circulation occurs at a slower rate than from the maternal circulation. Fetal hemoglobin may retain up to 5 times longer than maternal hemoglobin, which has both toxicity and treatment implications (Longo & Hill, 1977; Margulies, 1986).

2) STILLBIRTH

a) Fetal death due to nonlethal CO poisoning has been reported. Maternal carboxyhemoglobin level was 7% and fetal concentration was 61% (Farrow et al, 1990).
b) CASE SERIES: In a series of 86 acute CO poisonings in pregnant women, 77 (89.5%) had a successful outcome. Fetal death occurred in 5 cases, a 4-fold increased relative risk compared with controls. All women were treated with hyperbaric oxygen (Mathieu et al, 1992).
c) CASE SERIES: In a report of 6 consecutive cases of CO poisoning during pregnancy, with carboxyhemoglobin levels of 2.8% to 39%, three resulted in fetal death. Two of these were stillborn within 36 hours of exposure. One fetus, exposed at 13 weeks' gestation, was delivered at 33 weeks with multiple anomalies and was nonviable (Caravati et al, 1988).

3) BIRTH WEIGHT SUBNORMAL

a) CASE-CONTROL STUDY: In a case-control study, no association was found between maternal exposure to CO (assessed by ambient monitoring in the neighborhood) and the risk of delivering a low-birth-weight infant, after adjusting for maternal race. CO exposure in this study ranged from less than 1 ppm to approximately 5 ppm (Alderman et al, 1987).

4) BIRTH PREMATURE

a) If the mother is rendered unconscious from CO poisoning, there is a high risk of fetal death or brain damage (Bankl & Jellinger, 1967). Maternal exposure to 100 ppm of CO for 4 hours has been associated with fetal neurological damage, lower birth weight, spontaneous abortions, stillbirths, and developmental problems (Waterman, 1984). Premature delivery has also been associated with CO poisoning (Aubard & Magne, 2000).

B) FETAL DISTRESS

1) CASE REPORT: A 31-year-old woman at 35 weeks' gestation was exposed to CO in her home from a faulty heating system. She presented with sudden onset of severe headache, dizziness, visual disturbance, nausea, vomiting, palpitations, and a general unwell feeling. Her carboxyhemoglobin level was 25.3%; fetal cardiotochograph (CTG) recording showed variable decelerations and marked bradycardia with slow recovery. She was treated with hyperbaric oxygen. After treatment, her symptoms resolved completely, and CTG recording showed a normal reactive trace. A healthy baby boy was delivered 6 weeks later (Mandal et al, 2001).
2) CASE REPORTS: Three pregnant women (gestational ages ranging from 32 to 37 weeks) with CO poisoning were placed on continuous fetal heart monitoring. Initial monitoring revealed baseline fetal heart rates ranging from the 140s to the 180s, with minimal variability and an absence of accelerations and decelerations. Initial carboxyhemoglobin levels of the 3 women ranged from 28% to 36%. After initiation of supportive therapy, including hyperbaric oxygen treatment, the fetal heart rates normalized, with moderate variability and heart rate accelerations. All 3 patients subsequently had uneventful deliveries. Follow-up 16 to 24 months after delivery showed normal infant development, without sequelae (Towers & Corcoran, 2009).
3) CASE REPORT: A 29-year-old woman, 34 weeks pregnant, presented with a decreased level of consciousness following exposure to CO. Her carboxyhemoglobin level was 25%. Following administration of oxygen, the patient was discharged 1 day later. Approximately 1 week post-presentation, the patient returned to the hospital because of decreased fetal movements. An emergent cesarean section was performed due to fetal distress. The apgar scores of the infant were 1 and 3 at 1 and 5 minutes, respectively, requiring CPR and subsequent mechanical ventilation. The infant was diagnosed with severe hypoxic ischemic encephalopathy believed to be secondary to maternal CO poisoning. Blood gas analysis of the infant revealed metabolic acidosis and a carboxyhemoglobin level of 5%, and an MRI demonstrated high signal intensity in T1-weighted images and low signal intensity in T2-weighted images at the thalamus bilaterally and at the lentiform nucleus, with encephalomalacic changes at the temporal and frontoparietal cortex and degeneration at the corpus callosum. Despite continued supportive therapy, the infant's condition continued to deteriorate, and he subsequently died approximately 56 days later (Yildiz et al, 2010).

C) ANIMAL STUDIES

1) Rats chronically exposed in utero to 150 parts per million (ppm) of CO demonstrated permanent memory deficits (Mactutus & Fechter, 1985).
2) Growth retardation was seen in fetal rats who had been exposed to 1100 ppm to 1200 ppm of CO for 2 hours daily throughout gestation; CO may be the main component of cigarette smoke that is responsible for low birth weight in infant humans (Leichter, 1993).
3) In rats, prenatal exposure to CO may induce reversible immunologic changes. Rats exposed prenatally to relatively low levels (150 ppm) of CO were immunosuppressed at the age of 15 to 21 days, but not at 60 days (Giustino et al, 1993).
4) Inhalation exposure to CO at an airborne concentration of 3400 ppm prolonged the estrus cycle and reduced fertility in rats (Barlow & Sullivan, 1982). It increased absorptions and stillborns in rats exposed to airborne concentrations of 600 or 1000 ppm (RTECS). Increased absorptions and lower newborn weights were seen in rats exposed to 500 ppm (Moon & Cha, 1976). CO at an airborne concentration of 750 ppm for 3 hours a day, given on days 7 to 9 of gestation, produced increased resorptions and stillbirths in rats (Choi & Oh, 1975). Exposure to an airborne concentration of 5000 ppm for 20 minutes produced fetal wastage in rats (Yun, 1980).
5) Chronic prenatal exposure produced permanent brain damage in rats; female offspring were more sensitive than males (pp 17-33; Swartzwedler, 1979)). Exposures to 150 ppm during pregnancy produced behavioral and learning deficiencies in the offspring of rats (Mactutus & Fechter, 1984; Fechter & Annau, 1980; Annau, 1977)). CO inhalation exposures at or above 125 ppm retarded fetal development in mice (Singh & Scott, 1984; Annau, 1977).
6) Rats exposed to low CO levels (75 ppm and 150 ppm) from days 0 to 20 of gestation had differences in inactivation kinetics of the sodium current in the peripheral nervous system, compared with unexposed animals, which were largely reversible by day 270 after birth (Carratu et al, 1993).

Carcinogenicity

3.21.1)

A) IARC Carcinogenicity Ratings for CAS630-08-0 (International Agency for Research on Cancer (IARC), 2016; International Agency for Research on Cancer, 2015; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2010; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2010a; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2008; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2007; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2006; IARC, 2004):

1) Not Listed

Laboratory Monitoring

Monitoring Parameters Levels

4.1.1)

A) Monitor vital signs and mental status and perform a Mini-Mental State Exam.
B) Carboxyhemoglobin level from either venous or arterial blood should be obtained as soon as possible after arrival. Nonsmokers generally have carboxyhemoglobin levels of less than 2%, while chronic smokers will routinely have levels of carboxyhemoglobin between 6% to 10%.
C) Arterial or venous blood gas may be needed in moderate to severe poisonings to assess the adequacy of oxygenation, ventilation, and the acid-base status.
D) Obtain lactate and serum electrolyte levels in patients with moderate to severe poisoning. Obtain a pregnancy test.
E) Obtain an ECG in adult patients, and measure cardiac enzymes in patients if cardiac injury is suspected.
F) CT of the brain is indicated if there are signs or symptoms of cerebral edema or coma or if there is a persistent abnormal neurologic exam.
G) Fetal monitoring should be performed in pregnant patients.

4.1.2)

A) ACID/BASE

1) Obtain blood carboxyhemoglobin levels in any symptomatic patient. Venous samples are adequate in most patients, as venous and arterial carboxyhemoglobin levels have been shown to correlate well at levels of less than 25% (Tougher et al, 1994).

a) This should be determined using blood obtained prior to the initiation of oxygen therapy, as long as this does not delay the administration of oxygen.
b) Carbon monoxide (CO) levels do not necessarily correlate with symptoms or degree of exposure and must not be taken as an absolute index of toxicity, particularly when the levels are low and the patient has significant signs or symptoms of exposure (Sanchez et al, 1988).

1) CASE SERIES: In a study of 43 patients who were mass-exposed to CO, carboxyhemoglobin levels were obtained 5 to 330 minutes after the end of the exposure. Regression analysis showed a relationship between carboxyhemoglobin levels and time of blood sampling. The authors concluded that carboxyhemoglobin levels are very dependent on the time of blood sampling and do NOT correlate with CO poisoning severity in the acute phase (Butera et al, 2002).

c) In a retrospective review of 1407 patients who received hyperbaric oxygen treatment for acute CO poisoning, carboxyhemoglobin concentrations were not useful in predicting complications or outcome of poisoning (Hampson & Hauff, 2008).
d) NORMAL CO levels (Wright & Shephard, 1979):

Category	Normal CO level
Endogenous production	0.4%-0.7%
Pregnant women	0.4%-2.6%
Normal infants	0.5%-4.7%
Average adult	1%-5%
In hemolytic anemia	up to 6%
Cigarette smoker (1 pack/day)	3%-7%

e) Carboxyhemoglobin levels are fairly stable in stored blood for 2 weeks and slowly decline thereafter (Diaz & Roberts, 1997).

2) ARTERIAL BLOOD GASES: Obtain measurement in patients with severe symptoms or carboxyhemoglobin level greater than 20%. ABG analyzers without CO-oximeters can only calculate (rather than directly measure) oxyhemoglobin, which can result in inaccurate readings (Bozeman, 1998; Wright, 1998). Similar to pulse oximetry, confounding results may occur by the presence of unexpected dyshemoglobinemias like carboxyhemoglobin when oxyhemoglobin saturation is calculated.

a) ABG and electrolyte levels may demonstrate acidosis or alkalosis, but they DO NOT correlate with carboxyhemoglobin levels or poisoning severity (Lebby, 1989) (Myers & Britten, 1989).

B) BLOOD/SERUM CHEMISTRY

1) Monitor electrolytes, BUN, creatinine, LDH, CPK, AST, and ALT in all severely symptomatic patients and in asymptomatic patients with a carboxyhemoglobin level greater than 20%.
2) Monitor serial cardiac biomarkers (troponin I, CK-MB levels) in all patients with CO poisoning severe enough to require hospitalization. In CO poisoning patients who had a low incidence of significant cardiac risk factors or known coronary artery disease, 81 of 183 patients had cardiac biomarkers indicative of myocardial injury (CK-MB of at least 5 ng/mL or a cardiac troponin I level of at least 0.7 ng/mL) (Satran et al, 2005).
3) PLASMA LACTATE: Patients with severe CO poisoning may have elevated lactate concentrations (Inoue et al, 2008).

a) A prospective observational study questioned the utility of lactate measurements as a biomarker in the setting of pure CO poisoning. The study, involving 146 patients with pure CO poisoning, showed that plasma lactate concentrations were only mildly elevated, even in the setting of severe CO poisonings (defined as development of transient loss of consciousness or coma) in comparison with asymptomatic patients (2.8 mmol/L (range 1.7 to 6.1 mmol/L) vs 2 mmol/L (range 1 to 2.8 mmol/L); p less than 0.001) (Benaissa et al, 2003). However, a retrospective cross-sectional study, involving 74 patients with CO poisoning, showed that initial lactate levels had a positive correlation with carboxyhemoglobin values and the need for hyperbaric oxygen (HBO) therapy. Twenty-four of the 74 patients with a mean carboxyhemoglobin level of 35.5% received HBO treatment and had a mean lactate level of 2.3 mmol/L compared to the 50 patients with mean carboxyhemoglobin and lactate levels of 14% and 1 mmol/L, respectively, receiving normobaric oxygen therapy (p less than 0.001). Based on a receiver operating characteristics (ROC) curve analysis, it was determined that a lactate level of 1.85 mmol/L had the sensitivity and specificity of 70.8% and 78%, respectively, to accurately predict the need for HBO treatment in patients with CO poisoning (Dogan et al, 2015).

4.1.3)

A) URINALYSIS

1) Urinalysis is useful in conjunction with BUN and creatinine to monitor renal function in patients with severe poisoning.

B) OTHER

1) MYOGLOBIN: If rhabdomyolysis is suspected, myoglobin measurement is useful (Florkowski et al, 1992).

C) CEREBROSPINAL FLUID

1) PROGNOSTIC INDICATOR: A prospective cohort study involving 5 patients with CO poisoning was conducted to determine the value of measuring myelin basic protein (MBP) in cerebrospinal fluid (CSF) for use as a predictive marker of delayed encephalopathy. Of the 5 patients, 3 developed delayed encephalopathy (DE) and were classified as group DE; the other 2 patients did not develop delayed encephalopathy and were classified as group non-DE. Within group DE, the MBP concentration was elevated during the acute stage in patient 1, then decreased below the detection limit before becoming elevated prior to the clinical manifestations of DE. In patients 2 and 3, the MBP was below the limit of detection in the acute stage, but the level was markedly increased prior to development of DE. Within group non-DE, the MBP concentrations in both patients remained below the limit of detection. Although the sample size of this study was small, based on the results, it is suggested that measurement of MBP in the CSF may be a useful predictive marker for the development of CO-associated DE; further studies are warranted (Ide & Kamijo, 2008).

4.1.4)

A) OTHER

1) NEUROPSYCHOMETRIC TESTING

a) Neuropsychometric testing is indicated following moderate to severe poisoning. Abnormal test results were associated with higher carboxyhemoglobin levels in one study (Myers & Britten, 1989).

1) Evaluated parameters included general orientation, digit span, trailmaking, digit symbols, aphasia screening, and block design. Equipment for doing this test included the WAIS set of nine blocks for block design testing (8991-135).
2) Useful tools for psychometric testing include the WAIS Spiral booklet for picture completion and block design (8991-192), both of which are available from The Psychological Corporation, 7560 Old Oak Boulevard, Cleveland, Ohio 44130, and the Aphasia Test booklet (102), which may be obtained from the Reitan Neuropsychology Laboratory, 1338 South Edison 57, Tucson, Arizona, 85719.

b) OPHTHALMOSCOPIC EXAM: Retinal hemorrhage is a common finding in CO poisoning. It has been suggested that careful eye exam in the emergency department may provide useful diagnostic information. Findings may include superficial or deep retinal hemorrhage, venous changes (ie, engorgement and tortuosity), and edema of the optic disc.

1) Although generally not available in an emergency setting, electrodiagnostic tests may be helpful in detecting subtle changes in visual function that can occur with CO poisoning (Denniston, 2001).

2) ECG

a) Adult patients who were found unconscious or who have high carboxyhemoglobin levels should be monitored to detect signs of myocardial damage (Hampson, 1990).

3) EEG

a) EEG readings have been found to be relatively well correlated with clinical signs/symptoms in the acute or intermittent period of CO poisonings, but they are inconsistent as a guide for long-term prognosis (Sakamoto et al, 1998).

4) OXYGEN SATURATION

a) Pulse oximetry is an inaccurate measurement of oxygen saturation in the presence of carboxyhemoglobin, since oxyhemoglobin and carboxyhemoglobin are absorbed at the same wavelength. A multiple-band CO-oximeter is preferred (Graybeal et al, 1991; Vegfors & Lennmarken, 1991) (Gonzally et al, 1990) (Hampson, 1998).
b) RETROSPECTIVE REVIEW: A study has been conducted to determine the degree to which pulse oximetry overestimates actual oxyhemoglobin (O2Hb) saturation in individuals with CO poisoning (Bozeman et al, 1997).

1) The pulse oximetry gap, defined as the difference between pulse oximetry saturation and measured oxyhemoglobin saturation (10.5% +/- 9.7%; range, 0% to 40.6%), increased linearly with carboxyhemoglobin level and approximated carboxyhemoglobin level. Pulse oximetry should be considered inaccurate in the setting of CO poisoning; direct ABG measurements are needed (Buckley et al, 1994).

c) PULSE CO-OXIMETRY

1) A prospective observational study was conducted to determine the role of utilizing pulse CO-oximetry for noninvasive identification of patients with unsuspected CO toxicity. Of 14,818 patients who presented to the emergency department over a period of 2 months, 10,856 patients (75.2%) had carboxyhemoglobin levels measured with a pulse CO-oximeter. The overall mean carboxyhemoglobin level, when measured with CO-oximetry, was 3.6% (range 0% to 33%). Of those cases, 28 cases of CO toxicity were identified, 11 of which were unexpected and only identified with the help of pulse CO-oximetry. Confirmation was conducted using venous measurement of carboxyhemoglobin levels , which demonstrated a good correlation between the results of measurements done by CO-oximeter and confirmatory venous measurements (r=0.72). There was also a good correlation between CO-oximeter measurements and venous measurements in all patients who had both tests performed, regardless of CO toxicity (n=64; r=0.72). Using venous carboxyhemoglobin as the gold standard, the sensitivity and specificity of CO-oximeter measurements were 94% and 54%, respectively, in patients who underwent both (Suner et al, 2008).
2) A prospective study was conducted, evaluating the utility of pulse co-oximetry as a reliable method for early diagnosis of CO poisoning. Twenty patients admitted for CO poisoning were included in this study. In each patient, carboxyhemoglobin (COHb) levels were measured via standard blood gas analysis, and SpCO (an estimate of the carboxyhemoglobin percentage level) was measured via pulse co-oximetry. Overall, pulse co-oximetry overestimated the carboxyhemoglobin levels , with relatively higher SpCO levels as compared to the COHb values. A mean error of approximately 3.15% occurred with pulse co-oximetry compared to the results from utilizing standard blood gas analysis. With COHb levels greater than 10%, he mean error didn't vary significantly (3.43%). Testing various blood gas analyzers using the same patient sample, produced a mean error of 2.4%, which was only 1% lower than the mean error from the pulse co-oximeter, indicating that utilizing pulse co-oximetry may be a valuable tool in early diagnosis of CO poisoning (Piatkowski et al, 2009).

Radiographic Studies

1) CT scan may be useful in assessing the degree and location of neurologic injury. Consider CT scan or MRI if neurologic abnormalities develop or persist; results may correlate with poisoning severity and outcome (Sawada et al, 1980; LaGasse et al, 1991).
2) CT scan and MRI have shown sequential changes in brain tissue after CO poisoning. Changes are often correlated with prognosis (Gotoh et al, 1993; Hayashi et al, 1993; Silverman et al, 1993).
3) In a series of 111 patients with CO poisoning, CT scan immediately after injury revealed cortical and/or subcortical atrophy in 78 (Kroch et al, 1995).
4) Lucencies of the basal ganglia, particularly the globus pallidus, are characteristic of severe CO poisoning (Gordon & Mercandetti, 1989).
5) Low-density lesions of subcortical white matter, representing demyelination or necrosis, may also develop (Hart et al, 1988; Lugaresi et al, 1990).

1) In a study of 16 patients, with CO poisoning, who had developed coma at the time of exposure, routine MRI was completed between 1 and 10 years (mean 3.4 years) following intoxication. Persistent neuropsychiatric findings were reported in 8 patients; 8 were asymptomatic. MRI findings in all patients displayed varying degrees of bilateral symmetric white matter hyperintensities that were significant in the centrum semiovale, with relative sparing of the temporal lobes and anterior parts of the frontal lobes, on T2-weighted and fluid-attenuated inversion recovery (FLAIR) images. Bilateral globus pallidus lesions were observed in 3 patients. In the chronic phase, persistent cerebral changes may occur in patients with severe CO poisoning, even in the absence of neuropsychiatric findings (Durak et al, 2005).
2) CASE SERIES: In a study of 62 patients, with CO poisoning, MRI scans performed 6 months after exposure detected a 15 mm(2) loss in the cross-sectional surface area of the corpus callosum, compared with MRI images obtained on the day of CO exposure. The effects appeared to be generalized atrophy, rather than subregion specific alterations. Neuropsychological tests done at baseline and 6 months did not correlate with the level of corpus callosum atrophy. The authors suggested that long-term brain effects of CO poisoning may be underestimated (Porter et al, 2003).
3) T2-weighted MRI may demonstrate abnormalities of the basal ganglia, particularly the globus pallidus (Stuppaeck et al, 1995) (Kawanami et al, 1998; Inagaki et al, 1997). Hyperdense areas appearing in the deep white matter on T2-weighted MRI persisted for lengthy periods after improvement of clinical symptoms and did not appear to be an accurate indicator of clinical prognosis (Sakamoto et al, 1998).

a) An adult female who worked as a cook was exposed to long-term chronic CO. Fifteen months after the exposure her initial MRI was considered to be normal, but subsequent blinded reviews of the same scan identified multiple bilateral lesions in the basal ganglia, which appeared consistent with chronic CO exposure. The authors suggested that MRI along with a complete neuropsychological exam can assist in detecting CNS dysfunction related to CO exposure (Devine et al, 2002). MRI can detect occult changes not visible by CT (Kanaya et al, 1992).

4) PROSPECTIVE STUDY: Basal ganglia volumes were measured via MRI during a prospective longitudinal study to determine their presence in the presence or absence of basal ganglia lesions and to determine their correlation with cognitive sequelae in patients with CO poisoning. Seventy-three patients with CO poisoning underwent MRI, measuring basal ganglia volumes, on day 1 of exposure, 2 weeks postexposure, and 6 months postexposure. Neuropsychological tests, assessing cognitive sequelae, were also administered to these patients during the same timeframes. Results showed that the basal ganglia volumes of the caudate, putamen, and globus pallidus decreased over time, from day 1 to 6 months postexposure. There did not appear to be any significant correlations between the basal ganglia volumes and the overall presence or absence of cognitive sequelae at 2 weeks or 6 months postexposure; however, there did seem to be a significant correlation between the smaller basal ganglia volumes and lower neuropsychological scores on 2 individual tests: the Digit Symbol test, assessing mental processing speed, and the Story Recall test, assessing verbal memory. Only 1 of the 73 patients actually had basal ganglia lesions at 2 weeks and 6 months postexposure, indicating that volume reduction in CO poisoning can occur in the absence of observable lesions (Pulsipher et al, 2006).

1) GENERAL: Diffusion MRI has been used as a more specific diagnostic aid following CO poisoning in some adults and children (Kawada et al, 2004; Finelli & DiMario, 2004; Sener, 2003).
2) CASE REPORT: A 9-year-old comatose girl with acute CO poisoning had diffusion MRI done approximately 12 hours after exposure. This showed widespread symmetrical hyperintense changes that involved mainly the subcortical white matter in both hemispheres (frontal and temporal regions). The basal ganglion, thalamus, periventricular white matter, and hippocampus were unaffected. She received therapy with 100% hyperbaric oxygen. MRI was done 16 days later and revealed extensive involvement of the basal ganglion structure; however, no remaining white matter lesions were seen (Sener, 2003).

a) The authors concluded that during the early stages of CO poisoning, white matter may be more sensitive than gray matter to ischemia, which can be reversible. Basal ganglion changes are not seen immediately following exposure, but occur at later time period (Sener, 2003).

3) CASE REPORT/ACUTE FINDING: In 3 patients acutely exposed to CO, each had an MRI following exposure (range, approximately 2 to 5 days) which showed either selective or prominent diffuse white matter abnormalities. Subcortical T2-weighted MRI showed abnormalities in all cases, with hemorrhagic infarction found in the white matter of 1 patient. The authors concluded that these early findings suggested that white matter may be more sensitive than previously thought to CO exposure (Finelli & DiMario, 2004).
4) CASE SERIES: Diffusion-weighted MRI (DWI) findings were consistent with clinical features of delayed cytotoxic edema in chronic relapsing types of CO poisoning in 5 patients. In addition, the DWI findings correlated with most areas of hypoperfusion in 99m Tc-HMPAO single-photon emission computed tomography (SPECT). The authors concluded that DWI was able to identify larger areas of cytotoxic edema, as compared to conventional MRI. It was also suggested that this type of imaging may be a better indicator of the extent of tissue damage, and it also may be beneficial in differentiating between other white matter diseases (Chu et al, 2004).

a) In another CO poisoning, an elderly female had a diffusion MRI approximately 50 hours after exposure, which showed the restricted lesion of the bilateral globus pallidus that is suggestive of early cytotoxic edema (Kawada et al, 2004).

1) Abnormal images on SPECT scans were reported in 70% of a series of 9 patients with severe CO poisoning (Bedry et al, 1994). It was found that N-isopropyl-p-[123I]iodoamphetamine ([123I]IMP) SPECT findings did not appear to be accurate indicators of clinical prognosis in CO poisonings (Sakamoto et al, 1998).
2) Proton magnetic resonance spectroscopy (1H-MRS) appears to be the best indicator in clinical evaluations of patients with interval forms of CO poisoning. Frontal hyperperfusion is seen on 1H-MRS studies of the interval form of CO poisonings. A significant correlation between a prolonged lactate peak on 1H-MRS and the development of sequelae and severity of CO poisoning was noted (Sakamoto et al, 1998).

1) POSITRON EMISSION TOMOGRAPHY/CHRONIC TOXICITY: In a study of 2 adults, 3 years after chronic CO poisoning, PET scan images were consistent with the residual neurological deficits observed in each patient. Findings indicated significant metabolic decreases in the orbitofrontal and dorsolateral prefrontal cortex as well as areas of the temporal lobe. The authors suggested that PET imaging may be helpful in detecting the neuropathologic sequelae associated with chronic, nonlethal CO poisoning (Pinkston et al, 2000).

1) MYOCARDIAL PERFUSION SCINTIGRAPHY

a) (99m)Tc-MIBI SPECT is a noninvasive diagnostic tool that may aid in the detection of myocardial ischemia related to CO poisoning. (99m)Tc-MIBI SPECT has provided information on location and extent of cardiac injury. This may be a useful diagnostic tool along with laboratory and ECG monitoring to diagnose often clinically unclear symptoms or lack of symptoms related to potential CO-related cardiac toxicity (Hubalewska-hola et al, 2003).

Methods

1) Laboratory determination of the carboxyhemoglobin level should be performed. This value is reported as the percent saturation of hemoglobin by CO (See RANGE OF TOXICITY). For example, 50% saturation (or a level of 0.5) means that a patient with a total of 16 g hemoglobin/100 mL blood has 8 g of carboxyhemoglobin, and 25% (or a level 0.25) saturation indicates that 4 g of carboxyhemoglobin are present.
2) SPOT TESTS: Four different spot tests for carboxyhemoglobin have been shown to be very inaccurate and unreliable (Otten et al, 1985).
3) BREATH ANALYSIS: A breath sampling method has been described using an MSA/CATALYST CO Air Monitor. Measurements by this technique showed a correlation coefficient of 0.98 compared with that obtained by a CO-oximeter. Alcohol can alter the CO measurement by this method (Kurt et al, 1990).
4) A similar method using a MiniCO CO Monitor had a correlation coefficient of 0.8 with the carboxyhemoglobin level, with the breath analyzer tending to read higher than the measured carboxyhemoglobin (Turnbull et al, 1988).
5) For end-tidal CO measurement, a semiportable CO analyzer with an electrochemical detector has a sensitivity of 0.1 +/- 0.01 V per 1 mcL/L CO in air; results in 108, 1-day-old infants compared favorably with those from more difficult gas chromatography analysis (Vreman et al, 1994).
6) RESIDENTIAL CO DETECTORS

a) Since CO is colorless, odorless, and tasteless, it is virtually undetectable until patients become symptomatic and have a high index of suspicion that CO is responsible, or unless the odor of other by-products of combustion alert individuals that CO may be present. However, devices exist which can detect CO in ambient air and warn of its presence.
b) Essentially, there are two types of detectors:

1) DETECTORS WITH AN AUDIBLE ALARM: While there is some controversy regarding the sensitivity threshold of the CO detectors with audible alarms, all UL (Underwriters Laboratories) certified detectors will alarm before an equivalent of 10% carboxyhemoglobin is attained on the Coburn curve. In one study CO detectors with audible alarms alerted most victims before they became symptomatic and reduced the number of patients treated in a healthcare facility. In the detector group the mean CO concentration was 18.6 parts per million (ppm) compared with 96.6 ppm in residences without detectors. In 13.3% of homes with detectors, people became symptomatic, compared with 63.4% of people in homes without detectors. Only 2 individuals were hospitalized in homes where detectors were utilized, compared with 24 people in homes without detectors (Krenzelok et al, 1996).
2) CARD-TYPE DETECTORS: A study which evaluated the visual detectors without an audible alarm revealed that they lacked sensitivity and were not dependable indicators of CO (Softley et al, 1996). The most significant limitation of these detectors is that they rely upon visual detection of a color change and do not have an audible alarm to alert or awaken potential victims.

7) SPECTROPHOTOMETRY

a) FALSE ELEVATION: Measurement of a carboxyhemoglobin level in an infant may be falsely elevated due to interference of fetal hemoglobin when using standard spectrophotometric methods of analysis.

1) In one case, a 3-month-old girl presented to the emergency department with smoke inhalation caused by a faulty furnace. An initial exam showed normal cardiorespiratory and neurological function. As a precaution, the patient was administered high-flow oxygen through an oxygen hood. An initial carboxyhemoglobin (COHb) level, obtained 1.5 hours after removal from the source and measured via a standard, automated spectrophotometer, was 2.2%. Six hours later, a repeat level was 11.2%. Administration of normobaric oxygen was continued and the patient remained asymptomatic. A third COHb level, drawn 12 hours after admission, was 5.5%. The patient was discharged approximately 14 hours after admission (Mehrotra et al, 2011).
2) From birth to approximately 6 months of age, fetal hemoglobin constitutes approximately 70% of total hemoglobin. At 6 months of age, fetal hemoglobin is approximately 1% to 2% of the total hemoglobin. It is believed that the elevated COHb level of 11.2% was due to interference of the patient's fetal hemoglobin when the COHb level was measured with spectrophotometry because of the slightly different spectral properties of fetal hemoglobin as compared to adult hemoglobin (Mehrotra et al, 2011).

8) OTHER

a) SOOT DEPOSITS: In a prospective study of residential fire victims, soot deposits were monitored and were not found to be predictive of CO poisoning. Although the absence of soot makes carboxyhemoglobinemia less likely, this study indicated that specificity was low in determining actual CO poisoning (Borron et al, 1999).

Treatment

Life Support

Patient Disposition

6.3.3)

6.3.3.1) ADMISSION CRITERIA/INHALATION

A) Patients who remain symptomatic despite adequate treatment with oxygen, patients with abnormal neurological exam results or myocardial injury, and pregnant women with fetal distress should be admitted.

6.3.3.2) HOME CRITERIA/INHALATION

A) Asymptomatic, nonpregnant patients can be observed at home, provided it is certain that the source of exposure has been eliminated and the patient has been moved to a CO-free environment. Most area fire departments and utility companies can perform ambient CO monitoring to determine the source of exposure.

6.3.3.3) CONSULT CRITERIA/INHALATION

A) Consult a poison center or medical toxicologist for assistance in managing severe poisonings and for recommendations on determining the need for hyperbaric oxygen treatment. Contact your nearest hyperbaric chamber for recommendations and arrangements for hyperbaric oxygen therapy.

6.3.3.5) OBSERVATION CRITERIA/INHALATION

A) Mildly to moderately symptomatic patients should be sent to a healthcare facility for evaluation and treated until symptoms resolve.

Monitoring

Inhalation Exposure

6.7.1)

A) DECONTAMINATION: Move patient from the toxic environment to fresh air. Monitor for respiratory distress. If cough or difficulty in breathing develops, evaluate for respiratory tract irritation, bronchitis, or pneumonitis.
B) OBSERVATION: Carefully observe patients with inhalation exposure for the development of any systemic signs or symptoms, and administer symptomatic treatment as necessary.
C) INITIAL TREATMENT: Administer 100% humidified supplemental oxygen with assisted ventilation as required.

6.7.2)

A) OXYGEN

1) ADMINISTER 100% OXYGEN by tight-fitting face mask to reduce the carbon monoxide (CO) biological half-life. This treatment should be started in any patient as soon as the diagnosis is suspected. Treatment with oxygen may not reverse all symptoms completely (Burney et al, 1982).

B) PATIENT CURRENTLY PREGNANT

1) OXYGEN: Administer 100% oxygen by tight-fitting face mask. Oxygen should be continued for 5 times as long as is necessary to reach acceptable maternal CO levels to assure elimination from fetal circulation. Administration of oxygen to the mother should not adversely affect the fetus, since fetal umbilical vein PO2 is never higher than the uterine vein PO2 (usually 30 to 40 Torr).
2) HYPERBARIC OXYGEN: The decision to use hyperbaric oxygen during pregnancy must be based on several factors: the maternal need for hyperbaric oxygen, the proven fetotoxicity of CO, the theoretical fetotoxicity of hyperbaric oxygen, and the absence of demonstrated efficacy of hyperbaric oxygen to prevent the fetotoxicity of CO. Consider consultation with a physician experienced in hyperbaric oxygen therapy and a physician experienced in fetotoxicity when evaluating a pregnant patient.
3) Hyperbaric oxygen therapy may decrease uterine blood flow (Assali et al, 1968) and is theoretically teratogenic.

a) CASE REPORTS

1) Precipitation of uterine activity was reported in a 38-week pregnant woman with CO poisoning after 46 minutes of hyperbaric oxygen at 3 atmospheres, resulting in delivery of a normal neonate (Hollander et al, 1987).
2) A good outcome was seen after the use of 2.4 atmospheres for 90 minutes in a 37-week pregnant woman. In this case, uterine activity precipitated by CO exposure was relieved after hyperbaric oxygen therapy (Van Hoesen et al, 1989).
3) Hyperbaric oxygen has been used without adverse effects in a few pregnant women after diving accidents (Anon, 1987).
4) The outcome of 44 pregnant women who sustained an acute CO poisoning at home has been reported. The women were treated within 1 to 12 hours (mean 5.3 +/- 3.7 hours) of intoxication with a combination of 2 hours of hyperbaric oxygen at a pressure of 2 atmospheres absolute and 4 hours of normobaric oxygen (Elkharrat et al, 1991).

OBSTETRIC FOLLOW-UP
Lost to follow-up	13.6%
Normal delivery of normal infants	72.7%
Spontaneous abortion	4.5%
Medical abortion (unrelated to CO exposure)	2.3%
Premature delivery of normal infant	2.3%
Labor induced (36 weeks' gestation)	2.3%
Baby with Down syndrome	2.3%

MATERNAL FOLLOW-UP
Lost to follow-up	20.5%
Recovered	56.8%
Moderate sequelae	22.7%
Severe sequelae	0

MATERNAL FOLLOW-UP
Lost to follow-up	20.5%
Recovered	56.8%
Moderate sequelae	22.7%
Severe sequelae	0

a) INCLUSION CRITERIA

1) End of exposure to treatment shorter than 24 hours
2) Carboxyhemoglobin concentration at admission greater than or equal to 10% in smokers and 5% in nonsmokers

b) EXCLUSION CRITERIA

1) Multiple intoxications (CO combined with another toxic gas or with drug poisoning)
2) Contraindication to hyperbaric oxygen in a monoplace unit (hemodynamic instability or pulmonary edema)
3) Patient's refusal of hyperbaric oxygen technically not feasible within 24 hours of intoxication

b) ANIMALS

1) In atmospheric pressures and durations exceeding those used in human CO poisonings, animal studies have shown teratogenic effects (Ferm, 1964; Fujikura, 1964; Telford et al, 1969; Miller et al, 1971).
2) Animal studies using the usually recommended procedures (2.4 to 3 atmospheres for 20 to 180 minutes) failed to show harmful effects (Ferm, 1964; Cho & Yun, 1982; Gilman et al, 1983).

C) SEIZURE

1) SUMMARY

a) Attempt initial control with a benzodiazepine (eg, diazepam, lorazepam). If seizures persist or recur, administer phenobarbital or propofol.
b) Monitor for respiratory depression, hypotension, and dysrhythmias. Endotracheal intubation should be performed in patients with persistent seizures.
c) Evaluate for hypoxia, electrolyte disturbances, and hypoglycemia (or, if immediate bedside glucose testing is not available, treat with intravenous dextrose).

2) DIAZEPAM

a) ADULT DOSE: Initially 5 to 10 mg IV, OR 0.15 mg/kg IV up to 10 mg per dose up to a rate of 5 mg/minute; may be repeated every 5 to 20 minutes as needed (Brophy et al, 2012; Prod Info diazepam IM, IV injection, 2008; Manno, 2003).
b) PEDIATRIC DOSE: 0.1 to 0.5 mg/kg IV over 2 to 5 minutes; up to a maximum of 10 mg/dose. May repeat dose every 5 to 10 minutes as needed (Loddenkemper & Goodkin, 2011; Hegenbarth & American Academy of Pediatrics Committee on Drugs, 2008).
c) Monitor for hypotension, respiratory depression, and the need for endotracheal intubation. Consider a second agent if seizures persist or recur after repeated doses of diazepam .

3) NO INTRAVENOUS ACCESS

a) DIAZEPAM may be given rectally or intramuscularly (Manno, 2003). RECTAL DOSE: CHILD: Greater than 12 years: 0.2 mg/kg; 6 to 11 years: 0.3 mg/kg; 2 to 5 years: 0.5 mg/kg (Brophy et al, 2012).
b) MIDAZOLAM has been used intramuscularly and intranasally, particularly in children when intravenous access has not been established. ADULT DOSE: 0.2 mg/kg IM, up to a maximum dose of 10 mg (Brophy et al, 2012). PEDIATRIC DOSE: INTRAMUSCULAR: 0.2 mg/kg IM, up to a maximum dose of 7 mg (Chamberlain et al, 1997) OR 10 mg IM (weight greater than 40 kg); 5 mg IM (weight 13 to 40 kg); INTRANASAL: 0.2 to 0.5 mg/kg up to a maximum of 10 mg/dose (Loddenkemper & Goodkin, 2011; Brophy et al, 2012). BUCCAL midazolam, 10 mg, has been used in adolescents and older children (5-years-old or more) to control seizures when intravenous access was not established (Scott et al, 1999).

4) LORAZEPAM

a) MAXIMUM RATE: The rate of intravenous administration of lorazepam should not exceed 2 mg/min (Brophy et al, 2012; Prod Info lorazepam IM, IV injection, 2008).
b) ADULT DOSE: 2 to 4 mg IV initially; repeat every 5 to 10 minutes as needed, if seizures persist (Manno, 2003; Brophy et al, 2012).
c) PEDIATRIC DOSE: 0.05 to 0.1 mg/kg IV over 2 to 5 minutes, up to a maximum of 4 mg/dose; may repeat in 5 to 15 minutes as needed, if seizures continue (Brophy et al, 2012; Loddenkemper & Goodkin, 2011; Hegenbarth & American Academy of Pediatrics Committee on Drugs, 2008; Sreenath et al, 2010; Chin et al, 2008).

5) PHENOBARBITAL

a) ADULT LOADING DOSE: 20 mg/kg IV at an infusion rate of 50 to 100 mg/minute IV. An additional 5 to 10 mg/kg dose may be given 10 minutes after loading infusion if seizures persist or recur (Brophy et al, 2012).
b) Patients receiving high doses will require endotracheal intubation and may require vasopressor support (Brophy et al, 2012).
c) PEDIATRIC LOADING DOSE: 20 mg/kg may be given as single or divided application (2 mg/kg/minute in children weighing less than 40 kg up to 100 mg/min in children weighing greater than 40 kg). A plasma concentration of about 20 mg/L will be achieved by this dose (Loddenkemper & Goodkin, 2011).
d) REPEAT PEDIATRIC DOSE: Repeat doses of 5 to 20 mg/kg may be given every 15 to 20 minutes if seizures persist, with cardiorespiratory monitoring (Loddenkemper & Goodkin, 2011).
e) MONITOR: For hypotension, respiratory depression, and the need for endotracheal intubation (Loddenkemper & Goodkin, 2011; Manno, 2003).
f) SERUM CONCENTRATION MONITORING: Monitor serum concentrations over the next 12 to 24 hours. Therapeutic serum concentrations of phenobarbital range from 10 to 40 mcg/mL, although the optimal plasma concentration for some individuals may vary outside this range (Hvidberg & Dam, 1976; Choonara & Rane, 1990; AMA Department of Drugs, 1992).

6) OTHER AGENTS

a) If seizures persist after phenobarbital, propofol or pentobarbital infusion, or neuromuscular paralysis with general anesthesia (isoflurane) and continuous EEG monitoring should be considered (Manno, 2003). Other anticonvulsants can be considered (eg, valproate sodium, levetiracetam, lacosamide, topiramate) if seizures persist or recur; however, there is very little data regarding their use in toxin induced seizures, controlled trials are not available to define the optimal dosage ranges for these agents in status epilepticus (Brophy et al, 2012):

1) VALPROATE SODIUM: ADULT DOSE: An initial dose of 20 to 40 mg/kg IV, at a rate of 3 to 6 mg/kg/minute; may give an additional dose of 20 mg/kg 10 minutes after loading infusion. PEDIATRIC DOSE: 1.5 to 3 mg/kg/minute (Brophy et al, 2012).
2) LEVETIRACETAM: ADULT DOSE: 1000 to 3000 mg IV, at a rate of 2 to 5 mg/kg/min IV. PEDIATRIC DOSE: 20 to 60 mg/kg IV (Brophy et al, 2012; Loddenkemper & Goodkin, 2011).
3) LACOSAMIDE: ADULT DOSE: 200 to 400 mg IV; 200 mg IV over 15 minutes (Brophy et al, 2012). PEDIATRIC DOSE: In one study, median starting doses of 1.3 mg/kg/day and maintenance doses of 4.7 mg/kg/day were used in children 8 years and older (Loddenkemper & Goodkin, 2011).
4) TOPIRAMATE: ADULT DOSE: 200 to 400 mg nasogastric/orally OR 300 to 1600 mg/day orally divided in 2 to 4 times daily (Brophy et al, 2012).

7) PHENYTOIN/FOSPHENYTOIN

a) Benzodiazepines and/or barbiturates are preferred to phenytoin or fosphenytoin in the treatment of drug or withdrawal induced seizures (Wallace, 2005).
b) PHENYTOIN

1) PHENYTOIN INTRAVENOUS PUSH VERSUS INTRAVENOUS INFUSION

a) Administer phenytoin undiluted, by very slow intravenous push or dilute 50 mg/mL solution in 50 to 100 mL of 0.9% saline.
b) ADULT DOSE: A loading dose of 20 mg/kg IV; may administer an additional 5 to 10 mg/kg dose 10 minutes after loading dose. Rate of administration should not exceed 50 mg/minute (Brophy et al, 2012).
c) PEDIATRIC DOSE: A loading dose of 20 mg/kg, at a rate not exceeding 1 to 3 mg/kg/min or 50 mg/min, whichever is slower (Loddenkemper & Goodkin, 2011; Prod Info Dilantin(R) intravenous injection, intramuscular injection, 2013).
d) CAUTIONS: Administer phenytoin while monitoring ECG. Stop or slow infusion if dysrhythmias or hypotension occur. Be careful not to extravasate. Follow each injection with injection of sterile saline through the same needle (Prod Info Dilantin(R) intravenous injection, intramuscular injection, 2013).
e) SERUM CONCENTRATION MONITORING: Monitor serum concentrations over next 12 to 24 hours for maintenance of therapeutic concentrations. Therapeutic concentrations of 10 to 20 mcg/mL have been reported (Prod Info Dilantin(R) intravenous injection, intramuscular injection, 2013).

c) FOSPHENYTOIN

1) ADULT DOSE: A loading dose of 20 mg phenytoin equivalent/kg IV, at a rate not exceeding 150 mg phenytoin equivalent/minute; may give additional dose of 5 mg/kg 10 minutes after the loading infusion (Brophy et al, 2012).
2) CHILD DOSE: 20 mg phenytoin equivalent/kg IV, at a rate of 3 mg phenytoin equivalent/kg/minute, up to a maximum of 150 mg phenytoin equivalent/minute (Loddenkemper & Goodkin, 2011).
3) CAUTIONS: Perform continuous monitoring of ECG, respiratory function, and blood pressure throughout the period where maximal serum phenytoin concentrations occur (about 10 to 20 minutes after the end of fosphenytoin infusion) (Prod Info CEREBYX(R) intravenous injection, 2014).
4) SERUM CONCENTRATION MONITORING: Monitor serum phenytoin concentrations over the next 12 to 24 hours; therapeutic levels 10 to 20 mcg/mL. Do not obtain serum phenytoin concentrations until at least 2 hours after infusion is complete to allow for conversion of fosphenytoin to phenytoin (Prod Info CEREBYX(R) intravenous injection, 2014).

D) HYPERBARIC OXYGEN THERAPY

1) INDICATIONS

a) The need for hyperbaric oxygen is more accurately predicted by neurologic, metabolic, and cardiovascular findings than by carboxyhemoglobin measurements (Myers, 1986; Coric et al, 1998).

1) The following criteria for hyperbaric oxygen therapy have been recommended: unconsciousness upon presentation, altered sensorium, ECG changes consistent with ischemia, or carboxyhemoglobin levels in excess of 25% (Myers, 1986).
2) Another group has recommended hyperbaric oxygen therapy if at least 1 of the following indications exist (Hamilton-Farrell & Nathanson, 1990).

1) The patient is or was at some time unconscious.
2) The patient has had neurological signs or symptoms other than a simple headache.
3) The patient has had a carboxyhemoglobin concentration greater than 20%.
4) The patient has had cardiac complications.
5) The patient is pregnant.

b) Severely symptomatic patients should be referred to a facility with a hyperbaric chamber.

1) Severe symptoms include coma (at any point during the exposure), syncope, seizures, focal neurologic deficits, severe metabolic acidosis (pH less than 7.25), pulmonary edema, or cardiovascular involvement (hypotension, shock, angina, ECG evidence of ischemia) (Myers et al, 1981; Myers, 1984; Kindwall, 1977; Myers et al, 1979; Goulon et al, 1969).

c) According to the Undersea and Hyperbaric Medical Society, in addition to severely symptomatic patients, hyperbaric oxygen therapy is recommended to reduce the risk of cognitive sequelae for patients who are 36 years of age or older, who were exposed to CO for at least 24 hours, or who have a carboxyhemoglobin level of 25% or more (Weaver, 2009).
d) Patients with signs of neurologic impairment (abnormal results of psychometric testing, mental disorientation, confusion, irritability, aggressive behavior) are generally recommended for treatment irrespective of carboxyhemoglobin level, due to the potential for delayed sequelae, estimated to be 8% to 10% in patients with levels in the range of 25% to 30% and minimal symptoms (Myers, 1984).

1) Delayed neurologic sequelae can occur after an asymptomatic interval (Werner et al, 1985).

e) Severely ill patients should NOT be transferred to a facility with a hyperbaric chamber until they have been stabilized: an airway should be secured, ventilation should be adequate, convulsions should be controlled, and blood pressure and perfusion should be acceptable (Haddad, 1986).
f) PREVENTION OF DELAYED OR PERMANENT NEUROLOGIC DAMAGE is the primary goal of hyperbaric oxygen therapy.
g) HIGH LEVELS/MINIMAL SYMPTOMS

1) The need for hyperbaric oxygen therapy in patients with high carboxyhemoglobin levels (greater than 30%) but presenting with minimal (ie, nausea, headache only) or no symptoms is controversial. Some patients will do well with 100% oxygen.
2) CONCLUSION: Patients with levels above 25% to 40% are recommended for referral irrespective of symptoms by many centers (Kindwall, 1977; Myers, 1986).

h) PREGNANCY: Hyperbaric oxygen therapy may cause vasoconstriction and is theoretically teratogenic.

1) The decision to use hyperbaric oxygen during pregnancy must be based on several factors: the maternal need for hyperbaric oxygen, the proven fetotoxicity of CO, the theoretical fetotoxicity of hyperbaric oxygen, and the absence of demonstrated efficacy of hyperbaric oxygen to prevent the fetotoxicity of CO.
2) Consider consultation with a physician experienced in hyperbaric oxygen therapy and a physician experienced in fetotoxicity when evaluating a pregnant patient.

2) ADMINISTRATION

a) INITIAL THERAPY

1) 100% oxygen at 3 atmospheres reduces the biologic half-life of CO to 23.5 minutes and may awaken the patient very soon after treatment is initiated or within 24 to 48 hours, if it is effective (Kindwall, 1977; Myers et al, 1979) Stewart, 1976).
2) Hyperbaric oxygen should be instituted with 30 minutes of 100% oxygen at 3 atmospheres absolute pressure, followed by 2 atmospheres absolute pressure for 60 minutes or until a carboxyhemoglobin level less than 10% is achieved.

b) REPEAT THERAPY

1) Normally a dramatic recovery of consciousness is seen during hyperbaric treatment.
2) Patients remaining unconscious may be given further hyperbaric oxygen treatments. Treatments may be continued daily until there is no further improvement in level of consciousness (Myers, 1986).
3) Repeated treatments of 2.4 atmospheres twice daily for 3 days has resulted in reversal of coma on the seventh day, with no residual neurologic effects (Dean et al, 1992).
4) Anecdotal evidence suggests that recurring symptoms may benefit from additional hyperbaric oxygen treatment (Meyers et al, 1985b; (Gibson et al, 1991; Prybys et al, 1995).

3) EFFICACY

a) At the time of this review, controlled studies in humans to evaluate the effects of hyperbaric oxygen therapy are being conducted. The following is a listing of research (various study designs) and case series/uncontrolled studies that detail the results of hyperbaric oxygen therapy:

1) RESEARCH STUDIES

a) An interim analyses of 2 randomized controlled trials indicated that hyperbaric oxygen therapy did not improve outcome compared with normobaric oxygen therapy. Both trials enrolled comatose patients. In addition, the benefit of using hyperbaric oxygen on long-term neurological outcome remains unknown (Weaver, 1999).
b) DOUBLE-BLIND, RANDOMIZED STUDY: A double-blind randomized trial was conducted to evaluate the effect of hyperbaric oxygen therapy on cognitive sequelae. A total of 152 (n=76 for each therapy) symptomatic acute CO poisoning patients were randomly assigned to treatment groups within 24 hours of exposure. Treatment consisted of either 3 hyperbaric oxygen treatments or 1 normobaric oxygen treatment plus 2 sessions of exposure to normobaric room air (Weaver et al, 2002).

1) METHOD: Patients in the hyperbaric oxygen group were exposed to 100% oxygen at 3 and then 2 atmospheres absolute (304 kilopascals (kPa) and 203 kPa, respectively) during the first chamber session and then to 100% oxygen at 2 atmospheres absolute for chamber sessions 2 and 3. Patients in the normobaric oxygen group were exposed to air at 1 atmosphere absolute for all 3 chamber sessions. The chamber was pressurized to sea level in order to maintain blinding of both the patients and investigators (Weaver et al, 2002).
2) RESULTS: Baseline characteristics were similar in both groups; however, cerebellar dysfunction was more frequent in the normobaric group (15%;4% in the hyperbaric group). The investigators stopped the trial at the third interim analysis, which included 150 patients, because hyperbaric oxygen was found to be more efficacious (p less than 0.01). Cognitive sequelae at 6 weeks occurred less frequently in the hyperbaric group (25%) than in the normobaric group (46.1%). Based on further adjustments by logistic regression for cerebellar dysfunction before treatment and for stratification variables, hyperbaric oxygen was considered the more effective therapy (odds ratio 0.45; 95% confidence interval, 0.22 to 0.92). Other outcomes reported in the hyperbaric group included fewer difficulties with memory and less reported depression (not statistically significant). Ability to complete activities of daily living was the same for both groups. Outcomes at 6 and 12 months showed that cognitive sequelae were less in the hyperbaric group, based on both the intention-to-treat analysis and the efficacy analysis (Weaver et al, 2002). A follow-up study identified age and interval of exposure to CO as significant risk factors for CO-related cognitive sequelae. Univariate analysis of 163 patients not treated with hyperbaric oxygen therapy showed that 55% of patients 36 years or older (n=74) developed cognitive sequelae as compared with 32% of patients less than 36 years old (n=89) (p=0.003). Also, 27 of 46 patients (59%) who were exposed to CO for at least 24 hours developed cognitive sequelae as compared with 42 of 117 patients (36%) who were exposed to CO less than 24 hours (p=0.01) (Weaver et al, 2007).
3) FUTURE STUDIES: It has been suggested that future research should focus on the optimal number of treatments and treatment pressures of hyperbaric oxygen. In addition, the maximum treatment delay (less than 6 hours still appears optimal) from CO poisoning for hyperbaric oxygen to be efficacious in patients who have well-defined risk factors for poor outcome needs to be defined. These risk factors include individuals identified with neurologic abnormalities and loss of consciousness following CO poisoning. General recommendations still support hyperbaric oxygen therapy for patients with a carboxyhemoglobin above 25%. Extremes of age (ie, infants, children, and the elderly) still warrant special consideration, and firm recommendations have yet to be determined (Hampson et al, 2001; Thom, 2002).

c) STUDY/LACK OF EFFECT: In a prospective, controlled, double-blind study of patients with all grades of CO poisoning, after randomization to either three 100-minute normobaric oxygen sessions or three 100-minute hyperbaric oxygen sessions, more patients in the hyperbaric oxygen group required additional treatments (28% vs 15%, p=0.01 for all patients; 35% vs 13%, p=0.001 for severely poisoned patients). Patients received sham or hyperbaric treatment on each of 3 consecutive days and were treated with continuous oxygen by nonocclusive face mask at 14 L/minute between treatments(Scheinkestel et al, 1999).

1) At the completion of the test, hyperbaric oxygen patients had a worse outcome in the learning test and a greater number of abnormal test results. Only treated patients developed delayed neurological sequelae (p=0.03). These authors found no benefit of hyperbaric oxygen therapy (Scheinkestel et al, 1999).
2) Other authors have pointed out flaws in the above study, including Mini-Mental State Scores of patients which suggested that a significant number may have had only mild CNS impairment at presentation. The above study did not take into account numbers of severely poisoned patients with long delays to treatment. Thus, a number of the patients were probably treated at a length of time following CO exposure when hyperbaric oxygen was likely to be ineffective. In addition, patients were "cluster" randomized (ie, patients that presented in a group were randomized together) to receive the same therapy, thus introducing the potential bias that patients with similar poisoning severity were assigned to the same treatment arm (Moon & DeLong, 1999).
3) Another author also analyzed the above study and suggested similar limitations including that 73% of the patients had severe poisoning, with 76% of the cases resulting from intentional exposure. The delay of up to 7 hours in most patients before receiving hyperbaric oxygen potentially reducing its effectiveness. The use of high concentrations of oxygen continuously in both groups for 3 days may have minimized the ability to compare the groups, and the study had limited follow-up. It was suggested that further multicenter, prospective, randomized trials are needed because information to date does NOT provide clear evidence that hyperbaric oxygen therapy provides substantial advantage when treating clinical poisoning (Weaver, 1999).

d) NONBLINDED STUDY: In a prospective, randomized, nonblinded study of patients with moderate CO poisoning presenting within 6 hours of exposure, 7 of 30 (23%) patients treated with ambient oxygen developed delayed neurologic sequelae, compared with 0 of 30 patients treated with hyperbaric oxygen (Thom et al, 1995). The development of delayed neurologic sequelae was determined by neuropsychologic testing and questionnaire. This study has been criticized because the lack of double-blinding permits investigator bias in assessing the results of neuropsychologic tests and placebo effect in the subjects (Weaver et al, 1995).
e) SYSTEMATIC REVIEW: Systematic reviews have been conducted, involving randomized controlled trials with or without blinding, in order to determine the effectiveness of hyperbaric oxygen therapy as compared to normobaric oxygen therapy for the prevention of neurologic symptoms in patients with acute CO poisoning. Review of several trials showed that severity of CO poisoning was inconsistent between trials and that different hyperbaric and normobaric oxygen regimens were employed. However, pooled analysis of the trials indicated that there was no significant difference in the frequency of neurologic symptoms in patients 1 month after therapy. It was suggested that further research may better define the role of hyperbaric oxygen therapy in the treatment of CO poisoning (Buckley et al, 2011; Buckley et al, 2005; Juurlink et al, 2004).
f) RECOMMENDATION: According to the clinical policy presented by the American College of Emergency Physicians (ACEP), hyperbaric oxygen therapy is considered a treatment option in patients with CO poisoning; however, due to inconclusive and conflicting evidence, its use cannot be mandated. Further research, involving large, multicenter, randomized controlled clinical trials with strict blinding of patients and evaluators, is warranted (Wolf et al, 2008).
g) CASE SERIES/UNCONTROLLED STUDIES

1) STUDY: In a trial comparing 148 subjects who received normobaric oxygen therapy for 6 hours and 159 subjects who received hyperbaric oxygen therapy for 2 hours at 2 atmosphere absolute plus normobaric oxygen therapy for 4 hours, a lower incidence of behavioral impairment (1.2% vs 6.7%) and visual disturbance (1.9% vs 6.7%) were reported in the hyperbaric-treated group (Raphael et al, 1989). All subjects in these groups had no initial impairment of consciousness. Other authors have suggested that severely intoxicated patients may not have received adequate therapy, and follow-up assessment was not adequate to detect subtle delayed effects. No statistics were reported comparing these findings (Brown & Piantadosi, 1989).
2) STUDY: In a trial comparing 145 subjects who received 1 treatment of hyperbaric oxygen and 141 subjects who received 2 treatments of hyperbaric oxygen, there was no statistical difference in outcome at 1 month follow-up. The authors did not compare 2 treatments of hyperbaric oxygen with more than 2 treatments. All subjects had initial impairment of consciousness (Raphael et al, 1989).
3) CASE SERIES: Among 115 patients with CO poisoning and treated with hyperbaric oxygen, only 2 of 104 survivors had neurologic sequelae (hemiplegia and decreased mental acuity in 1 patient, cortical blindness in another) (Norkool & Kirkpatrick, 1985).
4) CASE SERIES: Hyperbaric oxygen was NOT effective at preventing delayed symptoms in 3 children (chronic headache, seizures and memory difficulties, and poor school performance) (Crocker & Walker, 1985).
5) CASE SERIES: Of 203 patients with varying levels of consciousness who were treated for CO poisoning with hyperbaric oxygen, 9 (4.4%) developed persistent or delayed neurologic sequelae. Twenty-seven patients with symptoms of only headache and nausea who were not treated with hyperbaric oxygen did not develop delayed neurologic sequelae. The mean carboxyhemoglobin level in this group was 18.5% +/- 15.1%. Except for the group with only headache and vomiting, the authors concluded that neither the clinical status nor the carboxyhemoglobin level can predict the development of delayed changes if the patient presents with neurologic abnormalities (Mathieu et al, 1985).
6) STUDY: An uncontrolled study demonstrated that, among 131 patients receiving hyperbaric oxygen, there were no delayed sequelae; of 82 patients not receiving hyperbaric oxygen, 10 (12.1%) developed headaches, irritability, personality changes, confusion, and memory loss. These delayed symptoms were reversible with later administration of hyperbaric oxygen (Myers et al, 1985b).
7) CASE SERIES: Improvement in neurological status following hyperbaric oxygen therapy has been reported in 6 patients (5 who had negligible carboxyhemoglobin levels at the time of treatment) who had failed to respond to 100% oxygen therapy. However, there were no controls in this report (Myers et al, 1981a).
8) CASE SERIES: In a retrospective review of patients who had been found in cardiac arrest following severe CO poisoning, all patients were resuscitated at the scene and then treated with hyperbaric oxygen. The patients ranged in age from 3 to 72 years, and 10 were female and 8 male. Carboxyhemoglobin levels averaged 31.7% (+/- 11%) and arterial pH averaged 7.14 (+/- 0.19) at the time of admission. Bradydysrhythmias were the underlying presenting rhythms in 10 of 18 patients. Hyperbaric oxygen therapy was initiated within 4.3 hours on average after poisoning (less than 3 hours in 10 patients and less than 6 hours in 15 patients). All 18 patients died during their hospitalization (Hampson & Zmaeff, 2001).

a) Based on the literature, the authors concluded that factors associated with a poorer prognosis were noncardiac cause for arrest and asystole or bradydysrhythmia at the time of discovery (Hampson & Zmaeff, 2001).
b) ANIMAL STUDY: In a mouse study, hyperbaric oxygen therapy was not effective in the prevention of neurologic sequelae and there was no benefit of hyperbaric oxygen over normobaric therapy following severe CO neurotoxicity.

1) Histologic examination found that the poisoned mice receiving atmospheric air had significantly reduced viable cells and a larger percentage of pyknotic cells than the nonpoisoned group (p less than 0.05). The hyperbaric and normobaric oxygen groups had slightly more viable cells; if the total number of viable cells and total number of pyknotic cells are considered, no significant neuroprotection was seen histologically with hyperbaric or normobaric oxygen therapy compared with ambient air. The results also found no significant difference between hyperbaric and normobaric oxygen therapy (Gilmer et al, 2002).

b) DELAYED HBO2 THERAPY

1) CASE REPORT: Hyperbaric oxygen therapy was begun 1 month after exposure to CO in a patient who had been initially treated with normobaric oxygen therapy. The patient did not improve with normobaric oxygen, but showed significant symptom relief and improved neuropsychological testing after four 90-minute hyperbaric oxygen treatments (Coric et al, 1998).
2) CASE REPORT: A 65-year-old woman was given hyperbaric oxygen therapy (HBO2) greater than 50 days after onset of signs and symptoms indicating delayed leukoencephalopathy (abnormal behavior, gait disturbances, echolalia, chorea, and bilateral symmetric hyperintensities in the subcortical white matter observed on MRI), following chronic exposure to low levels of carbon monoxide. Prior to hyperbaric oxygen, the patient received IV corticosteroid pulse therapy (methylprednisolone 1000 mg daily) which resolved the echolalia and chorea. Following 19 HBO2 therapy sessions over a 1-month period and continued supportive care, including sedation with quetiapine for a subsequent 4 months, she became more independent with regards to her activities of daily living within 6 months, and almost complete normalization in cognitive function within 14 months (Mizuno et al, 2014).

4) MECHANISM OF ACTION

a) Hyperbaric oxygen increases the concentration of oxygen that is dissolved in the plasma and displaces CO from hemoglobin (Pace et al, 1950; Kindwall, 1977; Anderson, 1978).

5) COMPLICATIONS OF THERAPY

a) Major complications include barotrauma (eg, ear or sinus trauma, tympanic membrane rupture, pneumothorax, or air embolism), oxygen toxicity (eg, seizures, ARDS, or respiratory arrest), fire, reversible myopia, and claustrophobia (Fernau et al, 1992; Murphy et al, 1991). Other reported complications include constriction of the visual fields, sweating, palpitations, and syncope (Chen & Kuo, 1995).

E) CEREBRAL EDEMA

1) Patients with signs of increased intracranial pressure should be hyperventilated with 100% oxygen via an endotracheal tube to keep the arterial pCO2 level at 25 mmHg to 30 mmHg.

a) Parenteral fluids should be limited to 2/3 to 3/4 of normal maintenance.
b) Dexamethasone administration does NOT appear to be beneficial in preventing or reducing hypoxia-induced cerebral edema.

2) Refractory cerebral edema is due to cell death and although mannitol, urea, glycerol, or other methods to reduce life-threatening cerebral edema may be employed, they are unlikely to affect the outcome.

F) ACUTE LUNG INJURY

1) ONSET: Onset of acute lung injury after toxic exposure may be delayed up to 24 to 72 hours after exposure in some cases.
2) NON-PHARMACOLOGIC TREATMENT: The treatment of acute lung injury is primarily supportive (Cataletto, 2012). Maintain adequate ventilation and oxygenation with frequent monitoring of arterial blood gases and/or pulse oximetry. If a high FIO2 is required to maintain adequate oxygenation, mechanical ventilation and positive-end-expiratory pressure (PEEP) may be required; ventilation with small tidal volumes (6 mL/kg) is preferred if ARDS develops (Haas, 2011; Stolbach & Hoffman, 2011).

a) To minimize barotrauma and other complications, use the lowest amount of PEEP possible while maintaining adequate oxygenation. Use of smaller tidal volumes (6 mL/kg) and lower plateau pressures (30 cm water or less) has been associated with decreased mortality and more rapid weaning from mechanical ventilation in patients with ARDS (Brower et al, 2000). More treatment information may be obtained from ARDS Clinical Network website, NIH NHLBI ARDS Clinical Network Mechanical Ventilation Protocol Summary, http://www.ardsnet.org/node/77791 (NHLBI ARDS Network, 2008)

3) FLUIDS: Crystalloid solutions must be administered judiciously. Pulmonary artery monitoring may help. In general the pulmonary artery wedge pressure should be kept relatively low while still maintaining adequate cardiac output, blood pressure and urine output (Stolbach & Hoffman, 2011).
4) ANTIBIOTICS: Indicated only when there is evidence of infection (Artigas et al, 1998).
5) EXPERIMENTAL THERAPY: Partial liquid ventilation has shown promise in preliminary studies (Kollef & Schuster, 1995).
6) CALFACTANT: In a multicenter, randomized, blinded trial, endotracheal instillation of 2 doses of 80 mL/m(2) calfactant (35 mg/mL of phospholipid suspension in saline) in infants, children, and adolescents with acute lung injury resulted in acute improvement in oxygenation and lower mortality; however, no significant decrease in the course of respiratory failure measured by duration of ventilator therapy, intensive care unit, or hospital stay was noted. Adverse effects (transient hypoxia and hypotension) were more frequent in calfactant patients, but these effects were mild and did not require withdrawal from the study (Wilson et al, 2005).
7) However, in a multicenter, randomized, controlled, and masked trial, endotracheal instillation of up to 3 doses of calfactant (30 mg) in adults only with acute lung injury/ARDS due to direct lung injury was not associated with improved oxygenation and longer term benefits compared to the placebo group. It was also associated with significant increases in hypoxia and hypotension (Willson et al, 2015).

G) EXTRACORPOREAL MEMBRANE OXYGENATION

1) CASE REPORT: A 38-year-old woman presented with suspected carbon monoxide poisoning from a faulty exhaust system in her recreational vehicle. Initial laboratory data revealed severe anion gap metabolic acidosis and a carboxyhemoglobin level of 13.6% (obtained approximately 2 to 4 hours after receiving oxygen [FiO2 of 1]). Despite aggressive supportive therapy, including IV fluids, and volume control ventilation, she progressively worsened, developing refractory shock, and was too unstable to receive hyperbaric oxygen therapy. Refractory hypoxemia occurred with multiple-organ failure, including myocardial ischemia, rhabdomyolysis, cardiogenic pulmonary edema, coagulopathy, severe renal and hepatic failure, and severe hemodynamic instability with a blood pressure of 87/53 mmHg and a heart rate of 144 beats/min. In addition to treatment with vasopressors and inotropes, venoarterial extracorporeal membrane oxygenation (ECMO) was started. A chest X-ray revealed diffuse alveolar opacities, a chest CT demonstrated lung consolidation with interstitial edema, and pleural effusions, and an echocardiogram indicated myocardial dysfunction with an ejection fraction of less than 20%. On day 2 of ECMO, an intra-aortic balloon pump was used to support cardiac function, and dialysis was initiated. On day 4, a repeat echocardiogram revealed significant improvement in cardiac function with an increased left ventricular ejection fraction of greater than 55%, resulting in removal of the intra-aortic balloon pump and cessation of ECMO therapy. Following gradual improvement hemodynamically, with improved oxygenation and kidney function, all other supportive therapies were also discontinued, and the patient was eventually discharged to home 30 days later. A follow-up exam 6 weeks later indicated a small left-sided pleural effusion; however at 24 weeks post-discharge, the patient's lung function tests were normal (Teerapuncharoen et al, 2015).

H) HYPOTHERMIA

1) This treatment has not proven effective in preventing CO-induced cerebral edema poisoning.

I) DEXTROSE

1) Administer supplemental glucose to prevent hypoglycemia.

J) ACIDOSIS

1) Severe acidosis should be treated. However, a slight acidosis may be beneficial by shifting the oxygen-dissociation curve to the right, allowing more oxygen to be released to the tissues. Therefore, alkalemia should be avoided.

K) TOBACCO USE AND EXPOSURE - FINDING

1) It has been reported that symptoms, carboxyhemoglobin levels, and rate of elimination of CO did not differ in smokers and nonsmokers who experienced CO poisoning, but smokers felt "back to normal" sooner than nonsmokers (Burney et al, 1982).

L) EXPERIMENTAL THERAPY

1) ALLOPURINOL/N-ACETYLCYSTEINE

a) A 26-year-old was administered N-acetylcysteine (NAC) 150 mg/kg IV over 15 minutes, then 50 mg/kg over 4 hours, and then 100 mg/kg over 16 hours beginning on the fifth day of coma following CO poisoning (Howard et al, 1987).

1) His carboxyhemoglobin level was 25% over 40 hours after exposure.
2) Allopurinol 100 mg nasogastrically was administered daily for 14 days beginning on the fifth day of coma.
3) The patient opened his eyes and tracked passing visitors at 8 hours post NAC infusion.
4) He made steady recovery and at 6 weeks follow-up, neurological and mental examination were normal.

b) This therapy was used based on animal data that indicated that allopurinol combined with a sulfhydryl-based free-radical scavenger was protective for ischemic tissue resulting from oxidative damage during hypoxic/reperfusion injury.
c) This therapy should be considered investigational. More studies are needed to demonstrate the efficacy of allopurinol and N-acetylcysteine in CO poisoning.

2) ANTIOXIDANTS

a) In a review article, the author suggests that the use of antioxidant supplements may be beneficial because oxidative stress plays a role in the progression of CO-induced tissue damage. There have been no clinical studies evaluating the use of antioxidants in CO poisoning; further study is warranted. (Omaye, 2002).

3) CORTICOSTEROID PULSE THERAPY

a) CASE REPORT: A 65-year-old woman, who developed delayed leukoencephalopathy (abnormal behavior, gait disturbances, echolalia, chorea, and bilateral symmetric hyperintensities in the subcortical white matter observed on MRI) following chronic exposure to low levels of carbon monoxide, received IV corticosteroid pulse therapy with methylprednisolone 1,000 mg daily for 3 days approximately 51 days post-onset of symptoms. Following therapy, the patient's echolalia and chorea resolved; however hyperbaric oxygen therapy (19 sessions over a 1-month period) and continued supportive care, including sedation with quetiapine, were required, resulting in a gradual return to independence in her activities of daily living and almost complete normalization in cognitive function (Mizuno et al, 2014).

M) METHEMOGLOBINEMIA

1) Fire victims may be suffering from both CO and cyanide poisoning. Conventional therapy of cyanide poisoning incorporates the use of methemoglobin producers (nitrites) with sodium thiosulfate. The production of methemoglobinemia may increase the toxicity of CO by further compromising tissue oxygenation (Moore et al, 1987).

Abstracts

Case Reports

1) SUBACUTE POISONING: Eleven members of 1 family (aged several months to 73 years) presented with vomiting, headache, and general discomfort. Measured carboxyhemoglobin levels were in the range of 25% to 35%. All patients appeared to recover following the administration of oxygen. No follow-up was available (Kunkel, 1987).
2) INDOOR BARBECUE USE: Five members of a household had carboxyhemoglobin concentrations measured at 38.3%, 31.9%, 25.5%, 21.2%, and 18.5% following carbon monoxide (CO) exposure from preparing food with an indoor barbecue (Sternbach & Varon, 1991).
3) INFLUENZA-LIKE ILLNESS: Thirteen out of 55 patients (23.6%) who presented with flu-like symptoms during 1 winter month had carboxyhemoglobin levels ranging from 10% to 21%. The diagnosis of CO exposure was missed in all patients (Dolan et al, 1987).
4) TRANSIENT UNILATERAL DIAPHRAGMATIC PARALYSIS: A case was associated with acute CO poisoning (carboxyhemoglobin 10.2 g% 2 hours postexposure) in a 56-year-old man (Joiner et al, 1990).
5) BOWEL NECROSIS: One case has been reported, manifested by abdominal pain and loose bloody stools 12 hours after exposure (Watson & Williams, 1984).
6) HEPATIC NECROSIS: One case reported with prolonged prothrombin time noted 12 hours postexposure and jaundice and elevated transaminases 48 hours after exposure (Watson & Williams, 1984).
7) NEUROLOGICAL IMPAIRMENT: In a study of 79 patients who had severe CO toxicity and were treated in an intensive care unit, 14% of the survivors had permanent neurological impairment. All patients were treated with 100% oxygen, and none received hyperbaric oxygen therapy (Krantz et al, 1988).

a) DELAYED NEUROLOGICAL EFFECTS

1) An 82-year-old man who lived independently was found collapsed after being exposed to gas from an unlit fireplace. Initial carboxyhemoglobin level was 42%; he was treated with oxygen by mask only, and within 48 hours neurological status improved. However, within 6 days of exposure, the patient's neurological status deteriorated and symptoms of Parkinson disease were present along with incontinence, difficulty ambulating, and inability to complete activities of daily living (Gillespie et al, 1999).

a) A CT scan revealed mild atrophic changes, and MRI indicated ischemic demyelination, which was diagnosed as CO-induced parkinsonism. Despite receiving a 4-week course of hyperbaric oxygen therapy (1 hour daily) for 6 days a week, the patient had only transient clinical improvement; severe parkinsonism required supportive care (Gillespie et al, 1999).

1) Respiratory alkalosis was reported in an elderly woman who was taking diuretics and a beta-blocker for hypertension The patient presented 3 times over a 3-week period complaining of nausea, vomiting, diarrhea, weakness, and shortness of breath before a carboxyhemoglobin level of 10.5% was obtained (Skorodin et al, 1986).
2) A case of Kluver-Bucy encephalopathy has been reported in a 32-year-old man 8 days after an acute exposure to CO (Sandson et al, 1988).
3) A 60-year-old man who had been found comatose after a 3-day CO exposure developed delayed effects (affective disorder, akinetic mutism, and tremor). CT scan and MRI findings were normal, but cerebral blood flow was decreased (Bedry et al, 1992).
4) PREGNANCY: Exposure to the fetus can result in permanent brain damage, including mental retardation, microcephalus, seizures, and stillbirth (Farrow et al, 1990; Caravati et al, 1987; Barlow & Sullivan, 1982).

a) In a report of 6 consecutive cases of CO poisoning during pregnancy, with carboxyhemoglobin levels of 2.8% to 39%, 3 resulted in fetal death. Two of these were stillborn within 36 hours of exposure. One fetus, exposed at 13 weeks' gestation, was delivered at 33 weeks, had multiple anomalies, and was nonviable (Caravati et al, 1988).

5) PREGNANCY: A 20-year-old woman who became unconscious and acrocyanotic, with a carboxyhemoglobin level of 7%, delivered a stillborn female fetus approximately 7 months in gestation, with a carboxyhemoglobin saturation of 61% (Farrow et al, 1990).
6) PREGNANCY: Hyperbaric oxygen therapy was used successfully in a woman who was 32 weeks pregnant and had been fighting fires. She reported headache, vomiting, and unconsciousness. Initial carboxyhemoglobin level was 22.4% (time unknown). The patient was dived twice and recovered. Two months later a normal child was born (Brown et al, 1992).

Range Of Toxicity

Summary

Minimum Lethal Exposure

1) A carbon monoxide (CO) concentration of 5000 parts per million (ppm) in air is lethal to humans after 5 minutes of exposure (RTECS , 2001).
2) LCLo - (INHL) MAN: 4000 ppm for 30M (RTECS , 2001).

Maximum Tolerated Exposure

1) A person with carboxyhemoglobin (blood saturation) above 65% to 70% will die if not treated with oxygen (Bingham et al, 2001).
2) It has been noted that carboxyhemoglobin levels over 60% usually result in death (Hathaway et al, 1996).

1) LD50 values for animals can vary greatly. However, in most studies, carboxyhemoglobin (carbon monoxide bound to hemoglobin molecules) reaches a 60% level before a majority of animals die. Most die at COHb levels between 65% and 75% (Bingham et al, 2001).

Serum Plasma Blood Concentrations

7.5.2)

A) TOXIC CONCENTRATION LEVELS

1) CONCENTRATION LEVEL

a) The range of toxicity for carbon monoxide (CO) poisoning has been expressed in tabular form (Finck, 1966; Schulte, 1963). However, levels do not correlate with clinical symptoms, particularly if they have been obtained after initiation of oxygen therapy or if there has been a significant time delay between obtaining the sample and exposure (Sanchez et al, 1988; Norkool & Kirkpatrick, 1985). NOTE: Patient prognosis and therapy should NOT be based solely on these correlations.

CO IN ATMOSPHERE	DURATION OF EXPOSURE	BLOOD SATURATION (80% EQUILIBRIUM)	POSSIBLE SYMPTOMS
.			MILD TOXICITY
up to 0.01%	indefinite	1% to 10% (0.01 to 0.1)	Tightness across forehead; slight headache; dilation of cutaneous vessels
0.01%-0.02%	Indefinite	10% to 20%	.
0.02%-0.03%	5 to 6 hr	20% to 30% (0.1 to 0.2)	Headache; throbbing in temples
.			MODERATE TOXICITY
0.04%-0.06%	4 to 5 hr	30% to 40% (0.3 to 0.4)	Severe headache; weakness and dizziness; nausea and vomiting; collapse; leukocytosis
0.07%-0.1%	3 to 4 hr	40% to 50% (0.4 to 0.5)	As above, plus: increased tendency to collapse and syncope; increased pulse and respiratory rate
.			SEVERE TOXICITY
0.11%-0.15%	1.5 to 3 hr	50% to 60% (0.5 to 0.6)	Increased pulse and respiratory rate; syncope; Cheyne-Stokes respiration; coma with intermittent seizures
0.16%-0.3%	1 to 1.5 hr	60% to 70% (0.6 to 0.7)	Coma with intermittent seizures; depressed heart action and respirations; death possible
0.5%-1%	1 to 2 min	70% to 80% (0.7 to 0.8)	Weak pulse; depressed respirations; respiratory failure and death

b) Factors which might affect the above carboxyhemoglobin level-clinical response relationship include age, metabolic rate, physical activity, pulmonary function, and the presence of pulmonary-cardiovascular or cardiovascular disease.
c) In smoke inhalation victims, altered response may be due to the presence of airway injury or exposure to other toxic gases, including hydrogen cyanide.
d) Sudden short-time exposure to a high atmosphere concentration of CO may result in impairment without greatly elevated carboxyhemoglobin concentration.
e) An 8- to 10-hour exposure to exhaust fumes from an automobile with a catalytic converter resulted in a low carboxyhemoglobin concentration of 4.8% and severe symptoms of rhabdomyolysis and compartment syndrome requiring urgent fasciotomy. Following surgery, the patient's neurological condition worsened; he developed sepsis and died on the fifth hospital day (Vossberg & Skolnick, 1999).
f) A slow build-up of atmospheric CO may result in elevated carboxyhemoglobin concentration without symptoms (Back & Dominguez, 1968).

Workplace Standards

1) Editor's Note: The listed values are recommendations or guidelines developed by ACGIH(R) to assist in the control of health hazards. They should only be used, interpreted and applied by individuals trained in industrial hygiene. Before applying these values, it is imperative to read the introduction to each section in the current TLVs(R) and BEI(R) Book and become familiar with the constraints and limitations to their use. Always consult the Documentation of the TLVs(R) and BEIs(R) before applying these recommendations and guidelines.

a) Adopted Value

1) Carbon monoxide

a) TLV:

1) TLV-TWA: 25 ppm
2) TLV-STEL:
3) TLV-Ceiling:

b) Notations and Endnotes:

1) Carcinogenicity Category: Not Listed
2) Codes: BEI
3) Definitions:

a) BEI: The BEI notation is listed when a BEI is also recommended for the substance listed. Biological monitoring should be instituted for such substances to evaluate the total exposure from all sources, including dermal, ingestion, or non-occupational.

c) TLV Basis - Critical Effect(s): COHb-emia
d) Molecular Weight: 28.01

1) For gases and vapors, to convert the TLV from ppm to mg/m(3):

a) [(TLV in ppm)(gram molecular weight of substance)]/24.45

2) For gases and vapors, to convert the TLV from mg/m(3) to ppm:

a) [(TLV in mg/m(3))(24.45)]/gram molecular weight of substance

e) Additional information:

B) NIOSH REL and IDLH Values for CAS630-08-0 (National Institute for Occupational Safety and Health, 2007):

1) Listed as: Carbon monoxide
2) REL:

a) TWA: 35 ppm (40 mg/m(3))
b) STEL:
c) Ceiling: 200 ppm (229 mg/m(3))
d) Carcinogen Listing: (Not Listed) Not Listed
e) Skin Designation: Not Listed
f) Note(s):

3) IDLH:

a) IDLH: 1200 ppm
b) Note(s): Not Listed

C) Carcinogenicity Ratings for CAS630-08-0 :

1) ACGIH (American Conference of Governmental Industrial Hygienists, 2010): Not Listed ; Listed as: Carbon monoxide
2) EPA (U.S. Environmental Protection Agency, 2011): Not Listed
3) IARC (International Agency for Research on Cancer (IARC), 2016; International Agency for Research on Cancer, 2015; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2010; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2010a; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2008; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2007; IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2006; IARC, 2004): Not Listed
4) NIOSH (National Institute for Occupational Safety and Health, 2007): Not Listed ; Listed as: Carbon monoxide
5) MAK (DFG, 2002): Not Listed
6) NTP (U.S. Department of Health and Human Services, Public Health Service, National Toxicology Project ): Not Listed

D) OSHA PEL Values for CAS630-08-0 (U.S. Occupational Safety, and Health Administration (OSHA), 2010):

1) Listed as: Carbon monoxide
2) Table Z-1 for Carbon monoxide:

a) 8-hour TWA:

1) ppm: 50

a) Parts of vapor or gas per million parts of contaminated air by volume at 25 degrees C and 760 torr.

2) mg/m3: 55

a) Milligrams of substances per cubic meter of air. When entry is in this column only, the value is exact; when listed with a ppm entry, it is approximate.

3) Ceiling Value:
4) Skin Designation: No
5) Notation(s): Not Listed

Toxicity Information

7.7.1)

A) References: ITI, 1995 Lewis, 2000 RTECS, 2001

1) TCLo- (INHALATION)HUMAN:

a) 4 mg/m(3) for 12H -- coma, methemoglobinemia-carboxyhemoglobin
b) 600 mg/m(3) for 10M -- headache
c) 650 ppm for 45M -- methemoglobinemia-carboxyhemoglobin

2) TCLo- (INHALATION)MOUSE:

a) 50 ppm for 30D - Intermittent -- changes in trachea or bronchi
b) Female, 65 ppm for 24H at 7-18D of pregnancy -- behavioral effects in offspring
c) Female, 250 ppm for 7H at 6-15D of pregnancy -- post-implantation mortality, developmental abnormalities of the musculoskeletal system
d) Female, 8 pph for 1H at 8D of pregnancy -- fetotoxicity, fetal death
e) Female, 8 pph for 1H at 8D of pregnancy -- CNS developmental abnormalities
f) Female, 125 ppm 24H at 7-18D of pregnancy -- fetotoxicity

3) TCLo- (INHALATION)RAT:

a) Female, 75 ppm for 24H at 0-22D of pregnancy -- behavioral, developmental effects in offspring
b) 96 ppm for 24H for 90D -- pigmented or nucleated red blood cells
c) 150 ppm for 24H
d) Female, 150 ppm for 24H at 1-22D of pregnancy -- behavioral, developmental effects in offspring
e) 250 ppm for 5H for 20D - Intermittent -- pigmented or nucleated red blood cells
f) 1800 ppm for 1H for 14D - Intermittent -- cardiac changes
g) Female, 1 mg/m(3) for 24H at 72D prior to mating -- menstrual cycle disorders, parturition
h) 30 mg/m(3) for 8H for 10W - Intermittent -- degenerative changes, muscle spasticity

Pharmacology Toxicology

Toxicologic Mechanism

1) CO reduces the oxygen-carrying capacity of hemoglobin by binding to hemoglobin and preventing the binding of oxygen. The affinity with which CO binds to hemoglobin is more than 200 times (approximately 210 times greater) that of oxygen, which can result in toxic levels of carboxyhemoglobin even after a relatively small environmental exposure to CO.
2) CO reduces oxygen delivery to tissues by shifting the oxygen-dissociation curve to the left.
3) The affinity of CO for myoglobin is even greater than for hemoglobin. Impaired tissue perfusion, hypotension induced by hypoxia and myocardial depression, and subsequent hypoxic ischemic injury to the brain may all play a role in CO sequelae (Seger & Welch, 1994; Blumenthal, 2001).

1) CO binds to the cytochrome oxidase system to reduce cellular utilization of oxygen, and it is possible that this is a major site of CO action since tissues most sensitive to oxygen deprivation (CNS, myocardium) are most sensitive to CO toxicity. The clinical effects of such binding may contribute little to the overall clinical picture of CO poisoning (Olsen, 1984).

1) CO poisoning in experimental animals has been associated with brain lipid peroxidation. A free radical, peroxynitrate, is produced, which causes cellular toxicity (Thom, 1988). In the brain this can cause further mitochondrial dysfunction, capillary leakage, leukocyte sequestration, and apoptosis. This change primarily occurs during the recovery phase when lipid peroxidation occurs, which produces an overall reversible demyelination in the brain. Common sites for CO-induced brain injury are the basal ganglia, cerebral white matter, hippocampus, and cerebellum (Blumenthal, 2001).
2) Animal studies have shown increased conjugated diene and thiobarbituric acid activity, both markers for lipid peroxidation. However, this was true only if CO exposure was limited to 90 minutes and the rats were returned to CO-free air. These plus the lack of correlation with carboxyhemoglobin levels emphasize the need for further studies (Thom, 1990).
3) In a rat model of CO poisoning, CO induced increased CNS levels of adenosine, and this increase was eliminated by hyperbaric oxygen treatment (Tomaszwski et al, 1995).
4) ANIMAL DATA/OXIDATIVE STRESS: In a study of Wistar male rats, the delayed effects of neuropathology following CO poisoning were studied. The authors hypothesized that acute CO-mediated oxidative stress can cause alterations in myelin basic protein (a major myelin protein of the CNS), and that the immune response to these modified proteins can precipitate delayed neurological dysfunction. The results suggested that following CO poisoning adduct formation between MBP and malonylaldehyde, a reactive product of lipid peroxidation, causes an immunological cascade resulting in part in a loss of antibody recognition of MBP. Thus, the neuropathology observed following acute CO exposure may be linked to an adaptive immunological response to chemically modified MBP. The authors suggested that these findings may have clinical application in the treatment of delayed neurotoxicity with anti-inflammatory agents (Thom et al, 2004).

Physicochemical

Physical Characteristics

Molecular Weight

Other

1) Odorless (CHRIS , 2001)

Animal Toxicology

Clinical Effects

11.1.13)

A) OTHER

1) SIGNS: Drowsiness, disorientation, incoordination, dyspnea, coma, and carboxyhemoglobinemia.
2) Household pets may be more rapidly and obviously affected by carbon monoxide (CO) than their owners (Ilano & Raffin, 1990).

Treatment

11.2.2)

A) GENERAL

1) MAINTAIN VITAL FUNCTIONS: Secure airway, supply oxygen, and begin supportive fluid therapy if necessary.

11.2.5)

A) GENERAL TREATMENT

1) Provide oxygen 95% with 5% carbon dioxide, normal or high-pressure administration. Mechanical ventilation. Blood transfusion.

Range Of Toxicity

11.3.2)

A) GENERAL TREATMENT

1) Household pets may become ill concurrent with or just preceding the onset of a patient's own illness. This is generally due to their smaller size and higher rate of metabolism (Ilano & Raffin, 1990)
2) DOG: Inhalation of 13% CO by 5 dogs was fatal within 15 minutes to 1 hour. Measured carboxyhemoglobin concentrations were from 54% to 90% in these dogs (Martindale, 1989).

Other

1) GENERAL

a) LABORATORY

1) CARBOXYHEMOGLOBIN: Absorption maxima occurs at 539 and 568 nm for canine carboxyhemoglobin (Smith, 1990). A slight minimum occurs at 554 nm.
2) HEMOGLOBIN: Absorption maxima for hemoglobin occurs at 541 and 576 nm, with a pronounced minimum at 561 nm (Smith, 1990).
3) The spectrum will become more like that of carboxyhemoglobin as carboxyhemoglobin concentration in diluted blood increases (Smith, 1990).

References

General Bibliography

10)

11)

12)

13)

14)

15)

16)

17)

18)

19)

20)

21)

22)

23)

24)

25)

26)

27)

28)

29)

30)

31)

32)

33)

34)

35)

36)

37)

38)

39)

40)

41)

42)

43)

44)

45)

46)

47)

48)

49)

50)

51)

52)

53)

54)

55)

56)

57)

58)

59)

60)

61)

62)

63)

64)

65)

66)

67)

68)

69)

70)

71)

72)

73)

74)

75)

76)

77)

78)

79)

80)

81)

82)

83)

84)

85)

86)

87)

88)

89)

90)

91)

92)

93)

94)

95)

96)

97)

98)

99)

100)

101)

102)

103)

104)

105)

106)

107)

108)

109)

110)

111)

112)

113)

114)

115)

116)

117)

118)

119)

120)

121)

122)

123)

124)

125)

126)

127)

128)

129)

130)

131)

132)

133)

134)

135)

136)

137)

138)

139)

140)

141)

142)

143)

144)

145)

146)

147)

148)

149)

150)

151)

152)

153)

154)

155)

156)

157)

158)

159)

160)

161)

162)

163)

164)

165)

166)

167)

168)

169)

170)

171)

172)

173)

174)

175)

176)

177)

178)

179)

180)

181)

182)

183)

184)

185)

186)

187)

188)

189)

190)

191)

192)

193)

194)

195)

196)

197)

198)

199)

200)

201)

202)

203)

204)

205)

206)

207)

208)

209)

210)

211)

212)

213)

214)

215)

216)

217)

218)

219)

220)

221)

222)

223)

224)

225)

226)

227)

228)

229)

230)

231)

232)

233)

234)

235)

236)

237)

238)

239)

240)

241)

242)

243)

244)

245)

246)

247)

248)

249)

250)

251)

252)

253)

254)

255)

256)

257)

258)

259)

260)

261)

262)

263)

264)

265)

266)

267)

268)

269)

270)

271)

272)

273)

274)

275)

276)

277)

278)

279)

280)

281)

282)

283)

284)

285)

286)

287)

288)

289)

290)

291)

292)

293)

294)

295)

296)

297)

298)

299)

300)

301)

302)

303)

304)

305)

306)

307)

308)

309)

310)

311)

312)

313)

314)

315)

316)

317)

318)

319)

320)

321)

322)

323)

324)

325)

326)

327)

328)

329)

330)

331)

332)

333)

334)

335)

336)

337)

338)

339)

340)

341)

342)

343)

344)

345)

346)

347)

348)

349)

350)

351)

352)

353)

354)

355)

356)

357)

358)

359)

360)

361)

362)

363)

364)

365)

366)

367)

368)

369)

370)

371)

372)

373)

374)

375)

376)

377)

378)

379)

380)

381)

382)

383)

384)

385)

386)

387)

388)

389)

390)

391)

392)

393)

394)

395)

396)

397)

398)

399)

400)

401)

402)

403)

404)

405)

406)

407)

408)

409)

410)

411)

412)

413)

414)

415)

416)

417)

418)

419)

420)

421)

422)

423)

424)

425)

426)

427)

428)

429)

430)

431)

432)

433)

434)

435)

436)

437)

438)

439)

440)

441)

442)

443)

444)

445)

446)

447)

448)

449)

450)

451)

452)

453)

454)

455)

456)

457)

458)

459)

460)

461)

462)

463)

464)

465)

466)

467)

468)

469)

470)

471)

472)

473)

474)

475)

476)

477)

478)

479)

480)

481)

482)

483)

484)

485)

486)

487)

488)

489)

490)

491)

492)

493)

494)

495)

496)

497)

498)

499)

500)

501)

502)

503)

504)

505)

506)

507)

508)

509)

510)

511)

512)

513)

514)

515)

516)

517)

518)

519)

520)

521)

522)

523)

524)

525)

526)

527)

528)

529)

530)

531)

532)

533)

534)

535)

536)

537)

538)

539)

540)

541)

542)

543)

544)

545)

546)

547)

548)

549)

550)

551)

552)

553)

554)

555)

556)

557)

558)

559)

560)

561)

562)

563)

564)

565)

566)

567)

568)

569)

570)

571)

572)

573)

574)

575)

576)

577)

578)

579)

580)

581)

582)

583)

584)

585)

586)

587)

588)

589)

590)

591)

592)

FeedBack

CARBON MONOXIDE

Classification | Detailed evidence-based information

Therapeutic Toxic Class

Specific Substances

Available Forms Sources

Life Support

Clinical Effects

Laboratory Monitoring

Treatment Overview

Range Of Toxicity

Summary Of Exposure

Vital Signs

Heent

Cardiovascular

Respiratory

Neurologic

Gastrointestinal

Hepatic

Genitourinary

Acid-Base

Hematologic

Dermatologic

Musculoskeletal

Endocrine

Reproductive

Carcinogenicity

Monitoring Parameters Levels

Radiographic Studies

Methods

Life Support

Patient Disposition

Monitoring

Inhalation Exposure

Case Reports

Summary

Minimum Lethal Exposure

Maximum Tolerated Exposure

Serum Plasma Blood Concentrations

Workplace Standards

Toxicity Information

Toxicologic Mechanism

Physical Characteristics

Molecular Weight

Other

Clinical Effects

Treatment

Range Of Toxicity

Other

General Bibliography