Carbon Monoxide Poisoning — Comprehensive Research Report

Prepared for: dismech disease knowledge base entry Category: Environmental (toxic gas exposure) Report date: 2026-07-09

Scope note: Carbon monoxide (CO) poisoning is a toxic-exposure syndrome, not a Mendelian or genetically-defined disease. Sections framed around causal genes, inheritance, penetrance, founder effects, and germline variants are largely Not Applicable; they are retained below with explicit "N/A" annotations and repurposed toward host susceptibility modifiers where evidence exists. The mechanistic, phenotypic, diagnostic, treatment, and prevention sections are the substantive core.

1. Disease Information

Overview. Carbon monoxide poisoning is the clinical syndrome resulting from inhalation of carbon monoxide, a colorless, odorless, tasteless, non-irritating gas produced by incomplete combustion of carbon-containing (hydrocarbon) fuels. It is one of the most common causes of fatal poisoning worldwide. CO produces injury through two convergent mechanisms: (1) hypoxic stress from formation of carboxyhemoglobin (COHb), which reduces oxygen-carrying capacity and shifts the oxyhemoglobin dissociation curve leftward (impairing tissue oxygen offloading); and (2) direct cellular toxicity independent of COHb — binding to mitochondrial cytochrome c oxidase, myoglobin, and other heme proteins, plus oxidative/nitrosative stress and an immune-mediated delayed neuroinflammatory cascade (StatPearls Carboxyhemoglobin Toxicity, NBK557888; Weaver, NEJM 2002, PMID 12362006).

Key identifiers. - MONDO: MONDO:0021113 (carbon monoxide poisoning) — verify against the local sqlite:obo:mondo adapter before committing (runoak -i sqlite:obo:mondo info MONDO:0021113 -O obo); OLS did not render the record in this session, so treat the ID as provisional. A closely related concept is carboxyhemoglobinemia. - ICD-10-CM: T58.- "Toxic effect of carbon monoxide," with source/intent subcodes — e.g., T58.01XA (motor-vehicle exhaust, accidental, initial encounter), T58.11XA (utility gas, accidental), T58.91XA (unspecified source, accidental), and intentional-self-harm variants (T58.-2-). - ICD-11: NE61 / stem for toxic effect of carbon monoxide (foundation "Toxic effect of carbon monoxide"). - MeSH: D002249 "Carbon Monoxide Poisoning." - OMIM/Orphanet: No dedicated Mendelian OMIM entry (not a genetic disease). Not a designated rare disease in Orphanet. - SNOMED CT: 284196006 "Carbon monoxide poisoning" (concept present; verify current ID).

Synonyms / alternative names. CO poisoning; carbon monoxide toxicity; carbon monoxide intoxication; carboxyhemoglobinemia (the biochemical state); "silent killer" (lay). Historically overlapping legacy terms: "coal gas poisoning," "flue gas / smoke inhalation CO component."

Data derivation. Disease-level knowledge here derives from aggregated resources (CDC/WHO surveillance, poison-center data, UHMS/society guidelines, and clinical-trial cohorts) rather than individual EHR records, though large administrative/EHR and forensic cohorts underpin the epidemiology and prognosis figures.

Sources: StatPearls Carboxyhemoglobin Toxicity; MSD/Merck Manual; ICD10Data T58.

2. Etiology

Primary cause (environmental/toxic). Inhalation of CO gas. There is no genetic or infectious primary etiology — this is a pure environmental/toxicological exposure disease. CHEBI: carbon monoxide = CHEBI:17245.

Common exposure sources. - Faulty or unvented fuel-burning appliances: furnaces, gas/oil boilers, water heaters, gas ranges/ovens, space heaters. - Motor-vehicle exhaust (running engines in attached/closed garages; a frequent suicide method). - Portable generators, a major cause of poisoning and death during storm-related power outages (CDC disaster guidance). - Charcoal grills / hibachis used indoors; camping stoves and lanterns in tents. - House fires / smoke inhalation (CO is a leading cause of fire-related death). - Methylene chloride (dichloromethane, CHEBI:15767) — paint strippers and solvents; hepatically metabolized to CO, causing delayed, prolonged COHb elevation. - Indoor use of gasoline-powered tools; ice-rink resurfacers; boating "houseboat/exhaust" exposures. - Tobacco/hookah smoke (chronic low-level source; baseline COHb 3–10% in smokers).

Risk factors (environmental / demographic). - Season: winter months (heating use); poisonings and deaths peak in cold months (CDC/NVSS). - Setting: most exposures occur at home; disasters and power outages (generators). - Sex: male death rate ~0.22/100,000 vs female ~0.07/100,000 (>3× higher in males, 1999–2010 CDC data) — partly reflecting occupational exposure and intentional poisoning. - Age: highest death rates in adults ≥65 years (males 0.42, females 0.18 per 100,000). - Occupation: firefighters, garage/toll workers, welders, mechanics, foundry/blast-furnace workers, propane forklift operators. - Pregnancy: enhanced fetal vulnerability (see §9, §12). - Comorbidity: pre-existing coronary artery disease, anemia, chronic respiratory disease, and cerebrovascular disease increase susceptibility to a given COHb level. - Absent CO detectors in the home is a strong, modifiable risk factor for fatal poisoning (Yoon, JAMA 1998, PMID 9496987).

Genetic / host-susceptibility modifiers (not causal). True Mendelian causation is N/A. Reported modifiers of susceptibility or of delayed neurological sequelae (DNS) are limited and largely candidate-gene/observational: haptoglobin phenotype, HMOX1 (heme oxygenase-1) promoter (GT)n repeat polymorphisms, apolipoprotein E (APOE) genotype, and inflammatory-cytokine polymorphisms have been proposed to modulate outcome, but none are established. Individuals with higher baseline COHb (smokers), anemia (lower total O₂-carrying reserve), or cardiac/cerebrovascular disease tolerate less CO.

Protective factors. - CO detectors/alarms (primary environmental protection; legislative mandates reduce deaths). - Appliance maintenance, adequate ventilation, and never running engines/generators indoors. - No established genetic protective variant.

Gene–environment interaction. The dose–response to a fixed CO exposure is modulated by host oxygen-carrying reserve and antioxidant/inflammatory genotype (e.g., HMOX1, antioxidant enzymes), but this remains hypothesis-level (CTD lists CO–gene interactions primarily from toxicogenomic models, not clinical GxE). Treat as an open knowledge gap (KNOWLEDGE_GAP).

Sources: CDC QuickStats 1999–2010; Yoon 1998 PMID 9496987; CDC disaster clinical guidance.

3. Phenotypes

CO poisoning is a multisystem syndrome with nonspecific early symptoms (often misdiagnosed as viral illness or food poisoning). Manifestations correlate imperfectly with COHb level; clinical severity, duration of exposure, and host factors matter more than a single COHb number.

Neurologic / neuropsychiatric (dominant and prognostically important). - Headache — the most common early symptom (often "dull, frontal"). HPO: Headache HP:0002315. Frequency: very frequent/most common. - Dizziness / lightheadedness — HPO: Vertigo HP:0002321 / Dizziness. Frequent. - Confusion, impaired cognition, difficulty concentrating — HPO: Confusion HP:0001289; Cognitive impairment HP:0100543. Frequent-to-severe cases. - Syncope / loss of consciousness — HPO: Syncope HP:0001279; Loss of consciousness HP:0007185. A key severity marker and HBO indication. - Seizures — HPO: Seizure HP:0001250. Severe poisoning. - Coma — HPO: Reduced consciousness/confusion → Coma HP:0001259. Severe. - Delayed neurological/neuropsychiatric sequelae (DNS / DEACMP) — a hallmark: after apparent recovery, a lucid interval of days–weeks precedes cognitive decline, memory loss, parkinsonism, gait/movement disorders, personality/affective change, incontinence, and akinetic mutism. HPO overlaps: Parkinsonism HP:0001300, Memory impairment HP:0002354, Gait disturbance HP:0001288, Personality changes HP:0000751, Dystonia HP:0001332. Onset typically 2–40 days post-recovery; occurs in ~3–40% depending on severity/definition. - Peripheral neuropathy, hearing loss, vestibular dysfunction (less common).

Cardiovascular. - Myocardial ischemia / injury — chest pain, ECG ischemia, troponin rise, arrhythmia; myocardial infarction with angiographically normal coronaries possible. HPO: Myocardial infarction HP:0001658; Angina pectoris HP:0001681; Arrhythmia HP:0011675. Common in moderate–severe poisoning; associated with increased long-term mortality. - Hypotension, tachycardia.

Respiratory. - Dyspnea (HPO: Dyspnea HP:0002094), tachypnea; noncardiogenic pulmonary edema in severe cases.

General / constitutional. - Nausea and vomiting — HPO: Nausea and vomiting HP:0002017 (or Nausea HP:0002018 / Vomiting HP:0002013). Frequent; often mistaken for gastroenteritis. - Fatigue / weakness / malaise — HPO: Fatigue HP:0012378; Muscle weakness HP:0001324. - Visual disturbance / blurred vision — HPO: Blurred vision HP:0000622; rarely retinal hemorrhages, cortical blindness.

Dermatologic. - The classically taught "cherry-red" skin/lips is rare and unreliable (usually a postmortem finding); cyanosis or normal color is more typical. Bullous skin lesions/pressure necrosis can occur in comatose patients.

Laboratory abnormalities (phenotype: laboratory). - Elevated carboxyhemoglobin (COHb) — the defining lab abnormality. Normal <3% (nonsmokers), up to ~10–15% in heavy smokers. Symptomatic poisoning usually >10–20%. LOINC: COHb/Hb.total — LOINC 20563-3 (Carboxyhemoglobin/Hemoglobin.total in Blood). - Metabolic acidosis with elevated lactate (anaerobic metabolism) — LOINC: Lactate 2524-7; a severity marker. - Elevated cardiac troponin / CK-MB (myocardial injury). - Elevated creatine kinase ± rhabdomyolysis/AKI in immobilized/comatose patients. - Falsely normal SpO₂ on standard pulse oximetry — conventional oximeters cannot distinguish COHb from oxyhemoglobin ("saturation gap"); requires CO-oximetry (multi-wavelength). A critical diagnostic pitfall.

Quality-of-life impact. Survivors of moderate–severe poisoning, especially those developing DNS, may have persistent cognitive deficits, mood/anxiety disorders, chronic headache, and impaired executive function, with measurable declines on neuropsychological batteries and quality-of-life instruments; some do not fully recover by 12 months (Weaver, NEJM 2002). Chronic low-level exposure produces persistent headache, fatigue, and cognitive complaints.

Onset/severity/progression summary. Onset acute (minutes–hours of exposure); severity mild → severe/fatal and variable; a distinct biphasic course is possible (acute illness → apparent recovery → DNS). Symptom–COHb correlation is weak.

Sources: Merck Manual; MedLink Neurology; Weaver NEJM 2002 PMID 12362006.

4. Genetic / Molecular Information

Causal genes: None — Not Applicable. CO poisoning is a toxic exposure, not a Mendelian disorder. There are no causal genes, pathogenic variants (ACMG/AMP classes), allele frequencies, somatic/germline distinctions, or chromosomal abnormalities to report.

Molecular target of the toxicant (the meaningful "molecular" content). The pathogenic ligand is CO gas (CHEBI:17245), which binds ferrous (Fe²⁺) heme iron in multiple hemoproteins: - Hemoglobin (HBB/HBA) → carboxyhemoglobin. CO affinity ~200–250× that of O₂ (StatPearls cites ~240×). - Myoglobin (MB, HGNC gene MB) → carboxymyoglobin, impairing cardiac and skeletal muscle O₂ storage/utilization; contributes to myocardial dysfunction. - Cytochrome c oxidase (Complex IV, MT-CO1/2/3 + nuclear COX subunits) → inhibition of mitochondrial electron transport and oxidative phosphorylation (direct histotoxic hypoxia). - NADPH oxidase, cytochrome P450, guanylate cyclase, and NOS-associated hemes are additional CO targets influencing signaling and reactive-species production.

Modifier genes / epigenetics / chromosomal: No established modifiers (see §2 for candidate host modifiers such as HMOX1, APOE). Epigenetic and chromosomal sections are N/A for causation. (Note that CO is itself an endogenous signaling molecule generated by heme oxygenase-1, HMOX1, HGNC:5013 — biologically relevant context, not a disease gene.)

Sources: StatPearls NBK557888; ROS/oxidative-stress review PMID 24773392.

5. Environmental Information

• Environmental factors: ambient CO from incomplete combustion (see §2 source list). Poorly ventilated enclosed spaces concentrate CO; concentration (ppm) × exposure duration determines dose. Regulatory context: OSHA PEL 50 ppm (8-h TWA); NIOSH IDLH 1,200 ppm; ambient CO tracked by EPA.
• Lifestyle factors: tobacco smoking (chronic endogenous/exogenous CO load, baseline COHb elevation); indoor charcoal/generator use; occupational exposure.
• Infectious agents: None — Not Applicable.

Sources: CDC CO topic; Merck Manual.

6. Mechanism / Pathophysiology

CO injures tissue through two integrated arms — impaired oxygen delivery/utilization (hypoxic-ischemic) and direct cellular toxicity with oxidative/nitrosative stress and delayed immune-mediated neuroinflammation. The brain and heart, with the highest oxygen demand, are most vulnerable.

Causal chain (upstream → downstream)

A. Hypoxic-ischemic arm. 1. Inhaled CO diffuses across the alveolar–capillary membrane and binds hemoglobin Fe²⁺ → carboxyhemoglobin, reducing O₂-carrying capacity. 2. CO binding to one heme shifts the oxyhemoglobin dissociation curve leftward, further impairing O₂ release to tissue (functional anemia worse than simple loss of capacity). 3. → Tissue hypoxia, anaerobic metabolism, lactic acidosis (GO: cellular response to hypoxia GO:0071456; anaerobic respiration).

B. Direct cytotoxic / histotoxic arm (COHb-independent). 4. CO binds cytochrome c oxidase (Complex IV), inhibiting mitochondrial electron transport and ATP synthesis → histotoxic hypoxia even where O₂ is available (GO: mitochondrial electron transport, cytochrome c to oxygen GO:0006123; oxidative phosphorylation GO:0006119). CO also binds myoglobin → impaired cardiac oxygen utilization → myocardial depression, hypotension, and secondary global ischemia/reperfusion.

C. Oxidative/nitrosative stress and vascular arm. 5. CO displaces nitric oxide (NO) from platelets and hemoproteins → excess NO → peroxynitrite formation; NO/peroxynitrite drive leukocyte (neutrophil) adhesion to injured cerebral microvascular endothelium (β2-integrin–mediated). 6. Adherent neutrophils release myeloperoxidase, generating reactive oxygen species (ROS). ROS are produced from three temporally distinct sources: mitochondria (first minutes of exposure), xanthine oxidase (~20 min, from energy deprivation/purine catabolism), and NADPH oxidase during the post-exposure reoxygenation period — i.e., an ischemia–reperfusion–like injury (Chang et al., ROS review PMID 24773392; PMC9852609). GO: reactive oxygen species metabolic process GO:0072593; response to oxidative stress GO:0006979. 7. → Lipid peroxidation of membrane and myelin lipids; glutathione depletion (GO: lipid oxidation; lipid peroxidation).

D. Delayed immune-mediated demyelination (the DNS engine). 8. Lipid peroxidation generates malondialdehyde (MDA), which forms adducts with myelin basic protein (MBP) in the CNS. Modified MBP loses its normal cationic charge and antibody-recognition profile. 9. Chemically altered MBP is immunogenic: over days, degraded MBP appears in brain with influx of macrophages and CD4⁺ T-lymphocytes, and autoreactive lymphocyte proliferation to MBP develops, with microglial activation → adaptive autoimmune demyelination and delayed neuropathology. Rats made immunologically tolerant to MBP before CO poisoning show the acute biochemical MBP change but no lymphocyte response and no learning deficit, establishing causation (Thom SR et al., PNAS 2004;101(37):13660–13665, PMID 15342916). GO: adaptive immune response GO:0002250; inflammatory response GO:0006954; demyelination. 10. → Delayed neuronal apoptosis (hippocampus, basal ganglia), white-matter demyelination, and DNS.

Cell types involved (CL terms). Neurons (CL:0000540), especially hippocampal neurons and basal-ganglia/globus-pallidus neurons; oligodendrocytes (CL:0000128, myelin) as demyelination targets; microglial cells (CL:0000129); cardiac myocytes (CL:0000746); vascular endothelial cells (CL:0000115); neutrophils (CL:0000775); CD4⁺ T cells (CL:0000624); macrophages (CL:0000235); erythrocytes (CL:0000232, the COHb site).

Subcellular compartments (GO Cellular Component). Mitochondrion (GO:0005739), specifically the mitochondrial respiratory chain complex IV (GO:0005751); myelin sheath (GO:0043209); plasma membrane / cytosol.

Metabolic changes. Shift to anaerobic glycolysis; lactic acidosis; ATP depletion; glutathione depletion; disrupted heme-protein oxygen handling.

Immune involvement. Innate (neutrophil/microglia/macrophage, ROS) acutely; adaptive autoimmune (anti-MBP CD4⁺ T-cell) response driving delayed demyelination — a rare example of a toxic exposure triggering an autoimmune neurologic sequela.

Tissue-damage mechanisms. Oxidative stress, ischemia–reperfusion injury, lipid peroxidation, apoptosis/necrosis, and immune-mediated demyelination converge on the globus pallidus (watershed, high metabolic demand, vulnerable) and deep white matter.

Molecular profiling / advanced tech. Transcriptomic and proteomic studies in rodent DEACMP models show upregulated inflammatory and apoptotic pathways and MBP degradation; candidate serum biomarkers (S100B, NSE, GFAP, myelin-related autoantibodies) have been studied for DNS prediction. Human single-cell/spatial and CRISPR-screen data are not established for this exposure — an open knowledge gap.

Sources: ROS review PMID 24773392; Thom PNAS 2004 PMID 15342916; Mechanism of delayed encephalopathy PMID 32594050; MBP degradation rat PMID 20633582.

7. Anatomical Structures Affected

Organ level. - Primary: Brain (UBERON:0000955) and heart (UBERON:0000948) — highest O₂ demand. - Secondary / systemic: skeletal muscle (rhabdomyolysis), kidney (UBERON:0002113; AKI from rhabdomyolysis/hypoperfusion), lungs (UBERON:0002048; pulmonary edema), skin, retina/eye, peripheral nerves. - Body systems: nervous, cardiovascular, respiratory, musculoskeletal, and (fetal) reproductive/placental.

Tissue and cell level. - Basal ganglia — especially globus pallidus (UBERON:0001875), the signature CO lesion — and substantia nigra; deep cerebral white matter (UBERON:0002316); hippocampus (UBERON:0002421); cerebral cortex; occasionally putamen, caudate, thalamus (imaging: bilateral, symmetric). - Cell populations: pallidal neurons, hippocampal neurons, oligodendrocytes/myelin, cardiac myocytes, microvascular endothelium.

Subcellular level. Mitochondria (Complex IV) and myelin sheath are the principal molecular battlegrounds.

Localization / laterality. CNS lesions are characteristically bilateral and symmetric (globus pallidus > white matter). On MRI: globus pallidus shows T1 hypo-/T2-FLAIR hyperintensity with restricted diffusion (DWI) acutely; late subacute period shows diffuse white-matter demyelination.

Sources: AJNR Pallidoreticular DWI; Advanced neuroimaging of CO poisoning PMC5602327.

8. Temporal Development

• Onset: acute — minutes to hours of exposure; any age (congenital exposure via maternal poisoning to geriatric).
• Acute course: symptoms during/after exposure; severity tracks cumulative dose (concentration × time) more than a single COHb.
• Biphasic pattern / DNS: after treatment and apparent recovery, a lucid interval of ~2–40 days may precede delayed neuropsychiatric sequelae (cognitive decline, parkinsonism, affective/behavioral change). This is the defining temporal feature.
• Progression / recovery: many mild cases resolve fully within hours–days on oxygen. Moderate–severe cases risk persistent or delayed deficits; DNS may partially recover over 6–12 months, but a subset has permanent impairment.
• Critical intervention window: benefit of hyperbaric oxygen (HBO) is greatest when started within 6 hours and not later than ~24 hours (UHMS).

Sources: Weaver NEJM 2002; Predictors of delayed encephalopathy PMC11979149.

9. Inheritance and Population (Epidemiology)

Inheritance: Not Applicable (non-genetic exposure). Penetrance, expressivity, anticipation, mosaicism, founder effects, consanguinity, and carrier frequency are all N/A.

Epidemiology (aggregate). - US unintentional mortality: ~430 deaths/year average (1999–2010, total 5,149; CDC); ~2,244 deaths over 2010–2015. In 2022, CDC provisional data recorded 1,244 total CO deaths (624 accidental + 579 suicides). - Morbidity: >100,000 US emergency-department visits/year for accidental CO poisoning, with >14,000 hospitalizations/year (CDC/USAFacts). - Sex ratio: male death rate ~0.22/100,000 vs female ~0.07/100,000 (males >3× higher, 1999–2010). - Age: highest death rates in adults ≥65 years (males 0.42, females 0.18 per 100,000). - Seasonality: strong winter predominance (heating season, power outages). - Global: CO poisoning is a leading cause of poisoning death worldwide; incidence and trends analyzed via GBD 1990–2021 (joinpoint/ARIMA; PMC12373207). Rates are higher where unvented biomass/coal heating and gas water heaters are common.

Prevalence framing for the KB (PrevalenceMeasureEnum): best modeled as ANNUAL_INCIDENCE (ED visits ~30/100,000/year in the US; deaths ~0.1–0.4/100,000/year), not a chronic point prevalence. prevalence_class qualitatively COMMON among acute poisonings.

Population demographics. Higher burden in lower-income households (older/unmaintained appliances), disaster-affected populations (generators), and in regions with indoor combustion heating/cooking. Intentional CO poisoning (suicide) skews male and adult.

Sources: CDC QuickStats 1999–2010; CDC NVSS 2010–2015; USAFacts 2022; GBD 1990–2021 PMC12373207.

10. Diagnostics

Core principle: diagnosis requires clinical suspicion (nonspecific symptoms + compatible exposure history, often multiple household members or a pet affected) plus direct COHb measurement. Standard pulse oximetry is falsely reassuring.

Laboratory / functional tests. - Carboxyhemoglobin (COHb) by CO-oximetry on arterial or venous blood (venous adequate for the level) — the confirmatory test. LOINC 20563-3. Interpret against baseline: nonsmoker <3%; smoker up to ~10–15%. Elevation confirms exposure; magnitude does not reliably grade severity. - Non-invasive pulse CO-oximetry (multi-wavelength, e.g., SpCO) — screening/triage; correlates imperfectly with blood COHb (PMC10890311). - ABG/VBG with lactate: metabolic acidosis, elevated lactate = severity markers. Measured (not calculated) SaO₂ needed. - Cardiac troponin, ECG — screen all moderate–severe poisonings for myocardial injury/ischemia (prognostic). - CK, renal function, urinalysis — rhabdomyolysis/AKI. - Pregnancy test in women of childbearing age (alters HBO threshold).

Imaging. - CT/MRI brain: bilateral symmetric globus-pallidus lesions; DWI restricted diffusion acutely; diffuse white-matter demyelination subacutely — supports diagnosis and prognosticates DNS risk. Advanced MRI (DTI, MRS, CEST-glutamate) is research-grade.

Biomarkers (investigational for DNS prediction). S100B, neuron-specific enolase (NSE), GFAP, and myelin-related autoantibodies have been studied as predictors of delayed encephalopathy but are not standard of care.

Genetic/omics testing: Not Applicable for diagnosis.

Clinical criteria / differential diagnosis. No formal DSM/ICD diagnostic criteria set beyond exposure + elevated COHb + compatible illness. Differential: viral illness/influenza, gastroenteritis/food poisoning, migraine, acute coronary syndrome, stroke, other toxic gas/cyanide exposure (concurrent in fires — consider cyanide co-toxicity), and psychiatric/functional disorders. Clues favoring CO: multiple people/pets ill in the same environment, symptoms improving away from home, winter/heating context.

Screening. Not a population lab-screening target; home CO alarms are the practical "screening" tool for asymptomatic detection of dangerous ambient levels.

Sources: StatPearls NBK557888; The Diagnosis and Treatment of CO Poisoning PMC6381775; CO-oximetry correlation PMC10890311.

11. Outcome / Prognosis

• Acute mortality: most treated symptomatic patients survive; case fatality is driven by severe exposure, coma, cardiac arrest, and delayed presentation. Severe poisoning with coma or myocardial injury carries substantially higher mortality.
• Cardiac injury as a prognostic marker: myocardial injury (elevated troponin) during acute poisoning is associated with increased long-term mortality independent of acute severity.
• Delayed neurological sequelae (DNS/DEACMP): the principal morbidity — reported in roughly 3–40% of moderate–severe cases depending on definition and follow-up; risk factors include older age, loss of consciousness, longer exposure, severe acidosis, and abnormal early neuroimaging. In Weaver's RCT, cognitive sequelae at 6 weeks occurred in 25% (HBO group) vs 46% (normobaric group).
• Recovery: many with DNS improve over 6–12 months; a subset has permanent cognitive, extrapyramidal (parkinsonism/dystonia), or affective deficits.
• Morbidity/QoL: persistent neurocognitive impairment, mood disorders, chronic headache, functional disability; measured with neuropsychological batteries and QoL tools.
• Fetal outcome: maternal poisoning can cause fetal death, CNS malformation, or neurodevelopmental injury; fetal COHb runs higher and clears more slowly than maternal (see §12).

Sources: Weaver NEJM 2002 PMID 12362006; Predictors of delayed encephalopathy PMC11979149.

12. Treatment

Immediate / supportive (MAXO: supportive care MAXO:0000950). - Remove from exposure; secure airway, breathing, circulation; treat seizures, hypotension, arrhythmia; cardiac monitoring; correct acidosis by restoring oxygenation/perfusion.

Normobaric oxygen — first-line (MAXO: oxygen therapy MAXO:0035013; therapeutic agent dioxygen CHEBI:15379). - High-flow 100% O₂ via non-rebreather mask (or ETT) for all suspected/confirmed cases. Rationale: accelerates COHb elimination. CO half-life ~300 min (4–5 h) on room air → ~60–90 min on 100% normobaric O₂ → ~20–30 min on hyperbaric O₂. Continue until asymptomatic and COHb near-normal (typically <3–5%).

Hyperbaric oxygen (HBO) therapy (MAXO: hyperbaric oxygen therapy MAXO:0000257). - Mechanism/rationale: 100% O₂ at 2.5–3.0 ATA dramatically shortens COHb half-life, rapidly dissociates CO from cytochrome c oxidase and myoglobin, and is proposed to reduce lipid peroxidation, leukocyte–endothelial adhesion, and the immune-mediated demyelination cascade — potentially preventing delayed neurological sequelae, not just clearing COHb. - Indications (UHMS 2020 / clinical consensus): consider HBO — ideally within 6 h, no later than ~24 h — for patients with loss of consciousness (any duration), neurologic deficits/abnormal cognition, cardiac ischemia/arrhythmia, severe metabolic acidosis, COHb >25% (adults), and pregnancy (lower threshold, generally symptomatic exposure or COHb >15–20%, because fetal COHb is higher and clears slowly). Prolonged HBO/oxygen may be warranted after methylene-chloride exposure (ongoing endogenous CO production). - Evidence base (conflicting): - Thom et al., 1995 — prospective RCT: HBO reduced incidence of DNS in mild–moderate CO poisoning presenting within 6 h (PMID 7710151). - Weaver et al., NEJM 2002 — quadruple-blinded RCT: three HBO sessions within 24 h reduced cognitive sequelae at 6 weeks (25% vs 46%) and at 12 months (PMID 12362006). - Scheinkestel et al., MJA 1999 — RCT found no benefit (methodological differences; delayed/varied protocols) (PMID 10092916). - Cochrane review (Buckley et al., 2011, CD002041) concluded evidence is insufficient/conflicting to define which patients benefit — HBO remains standard practice for severe poisoning at many centers despite equipoise.

Pharmacogenomics / targeted / gene / cell / RNA therapies: Not Applicable (no molecular-target drug therapy). Experimental adjuncts studied mainly in models: erythropoietin (neuroprotection; rat serum-biomarker study PMC3586885), N-acetylcysteine, hypothermia, and — as an experimental antidote concept — engineered high-affinity CO-scavenger molecules (e.g., recombinant neuroglobin/"CO-scavenger" therapeutics in preclinical development to accelerate CO removal). None are approved.

Treatment strategy summary. All symptomatic patients → immediate high-flow 100% O₂ + supportive care; risk-stratify (LOC, neuro deficit, cardiac ischemia, acidosis, COHb, pregnancy) → refer for HBO if indicated and available within the therapeutic window. Arrange neuropsychological follow-up to detect DNS.

Adverse events. HBO risks: barotrauma (middle ear/sinus, rare pulmonary), oxygen toxicity seizures, confinement anxiety, transient myopia. Normobaric high-FiO₂ is generally safe over the short treatment course.

Sources: UHMS HBO indications 2020; Weaver NEJM 2002; Thom 1995 PMID 7710151; Scheinkestel 1999 PMID 10092916; Cochrane CD002041.

13. Prevention

Primary prevention (the highest-yield domain). - CO alarms/detectors in homes near sleeping areas and on every level (MAXO/behavioral: environmental intervention). Legislative mandates for residential CO alarms reduce fatalities; battery replacement and testing emphasized. - Appliance safety: annual professional inspection/maintenance of furnaces, water heaters, and vents; never use ovens/ranges for heating. - Generator safety: operate portable generators outdoors, ≥20 ft from windows/doors/vents, never indoors/garages (key message during storms/outages). - Vehicle/engine safety: never run engines in attached/closed garages; check exhaust systems. - No indoor charcoal grills / camp stoves. - Public health education, especially before winter and after disasters; CDC/EPA campaigns.

Secondary prevention (early detection). - CO alarms detecting dangerous ambient levels before symptoms; prompt evaluation of clustered household symptoms; consider CO in nonspecific winter illness ("influenza-like" clusters).

Tertiary prevention (limiting sequelae in the poisoned). - Prompt high-flow/hyperbaric oxygen within the therapeutic window; neuropsychological follow-up to identify and rehabilitate DNS; removal/remediation of the CO source before discharge to prevent re-exposure.

Prophylaxis / immunization / genetic screening / counseling: vaccination and genetic counseling are Not Applicable. "Prophylaxis" = engineering/behavioral source control + alarms.

Sources: CDC clinical/disaster guidance; Yoon 1998 PMID 9496987.

14. Other Species / Natural Disease

• Taxonomy / cross-species susceptibility: CO is toxic to all aerobic, hemoglobin-bearing animals — mechanism (heme binding) is evolutionarily conserved. Species commonly poisoned alongside humans include domestic dog (Canis lupus familiaris, NCBITaxon:9615) and cat (Felis catus, NCBITaxon:9685) — pets are a classic "sentinel" for household CO exposure. Also documented in birds (historically the "canary in the coal mine," Serinus canaria), which are especially sensitive due to high metabolic/respiratory rates.
• Natural/veterinary disease: accidental CO poisoning occurs in companion animals (house fires, faulty heaters, vehicle transport) and livestock in poorly ventilated heated barns; veterinary toxicology recognizes it with a pathophysiology mirroring humans. Not a Mendelian OMIA entry (it is a toxic exposure).
• Comparative biology: rodents show the same globus-pallidus/hippocampal vulnerability, MBP-adduct autoimmunity, and DNS-like learning deficits, validating cross-species conservation of both the hypoxic and immune arms.
• Zoonotic potential: None — Not Applicable (non-infectious).

Sources: Thom PNAS 2004 (rat); veterinary toxicology consensus (background).

15. Model Organisms

• Rat (Rattus norvegicus, NCBITaxon:10116) — the dominant model. Standard induced-exposure protocols (e.g., 1,000 ppm 40 min then 3,000 ppm 20 min) reproduce: transient hippocampal MBP degradation (Hara et al., PMID 20633582), immune-mediated delayed neuropathology and learning deficits (Thom et al., PMID 15342916), decreased hippocampal neural precursor cells in DEACMP models (Nat Sci Rep, s41598-021-85860-9), altered nicotinic cholinergic signaling (PMID 24704181), and neuroprotection studies (erythropoietin, PMC3586885). Model type: induced (inhalational exposure), not genetic.
• Mouse (Mus musculus, NCBITaxon:10090) — used for oxidative-stress/ROS-source dissection (mitochondria/xanthine oxidase/NADPH oxidase) and inflammatory-pathway studies; transgenic/knockout mice (e.g., NOS, NADPH oxidase subunits) dissect specific mechanistic arms.
• In vitro / cellular: neuronal and oligodendrocyte cultures, endothelial–neutrophil adhesion assays, and mitochondrial respiration assays (Complex IV inhibition).

Phenotype recapitulation. Rodent models reproduce the biphasic course (acute exposure → delayed neuropathology), the globus-pallidus/hippocampal vulnerability, demyelination, and the autoimmune anti-MBP mechanism — strong construct/face validity, including the pivotal MBP-tolerance experiment establishing causation.

Limitations. Species differences in CO tolerance and COHb kinetics; some protocols produce MBP degradation without measurable cognitive deficit (dose-dependence), highlighting a human–model translational caveat (HUMAN_MODEL_MISMATCH): the immune-mediated DNS mechanism is best characterized in rodents, and its quantitative fidelity to human DNS remains an open translational question.

Resources: MGI/RGD for mouse/rat strains; primary literature via PubMed. No dedicated CO-poisoning model repository (it is an induced-exposure paradigm, not a genetic line).

Sources: Thom PNAS 2004 PMID 15342916; MBP degradation PMID 20633582; Neural precursor cells Sci Rep; Nicotinic cholinergic PMID 24704181.

Ontology Term Quick-Reference (for KB population)

Curation Caveats (dismech-specific)

1. Verify every ontology ID before committing — MONDO:0021113 and the MAXO oxygen/HBO IDs are provisional in this report (OLS/OAK not confirmed in-session). Run just validate-terms-file and runoak … info per the anti-hallucination SOP.
2. All snippets must be exact abstract quotes — this report paraphrases sources; before entering any evidence: snippet, run just fetch-reference PMID:XXXX and confirm the exact substring (e.g., PMIDs 12362006, 7710151, 15342916, 24773392, 20633582, 32594050, 24704181, 10092916, 9496987).
3. evidence_source tagging: Weaver/Thom-1995/Scheinkestel = HUMAN_CLINICAL; Thom PNAS 2004, MBP-degradation, EPO, nicotinic-cholinergic studies = MODEL_ORGANISM; ROS-source dissection = mix of IN_VITRO/MODEL_ORGANISM (split items accordingly).
4. Genetic/inheritance sections should be curated as explicitly N/A or omitted — this is an environmental exposure, not a Mendelian disease.
5. Key KNOWLEDGE_GAP / HUMAN_MODEL_MISMATCH candidates: (a) whether the rodent anti-MBP autoimmune DNS mechanism quantitatively explains human delayed encephalopathy; (b) which patients benefit from HBO (unresolved per Cochrane); (c) validity of serum biomarkers (S100B/NSE/GFAP) for DNS prediction.

Primary sources cited

• Weaver LK et al., NEJM 2002 — PMID 12362006
• Thom SR et al., DNS prevention by HBO, 1995 — PMID 7710151
• Scheinkestel CD et al., MJA 1999 — PMID 10092916
• Thom SR et al., "Delayed neuropathology…is immune-mediated," PNAS 2004;101:13660 — PMID 15342916
• ROS & oxidative stress in CO toxicity review — PMID 24773392
• Transient MBP degradation in rat hippocampus — PMID 20633582
• Mechanism of delayed encephalopathy — PMID 32594050
• Yoon SS et al., CO deaths & detectors, 1998 — PMID 9496987
• StatPearls Carboxyhemoglobin Toxicity — NBK557888
• UHMS HBO Indications — CO Poisoning (2020)
• CDC CO mortality QuickStats & clinical guidance
• AJNR — Pallidoreticular DWI in acute CO poisoning
• Cochrane — HBO for CO poisoning (Buckley 2011, CD002041)

Bottom line: Carbon monoxide poisoning is an acute environmental toxic-inhalation syndrome (MONDO:0021113, ICD-10 T58, MeSH D002249) with no genetic etiology. Its pathophysiology is a dual insult — carboxyhemoglobin-mediated hypoxia plus direct mitochondrial/heme-protein toxicity, oxidative/ischemia-reperfusion injury, and a distinctive immune-mediated (anti-MBP) delayed demyelinating cascade that produces delayed neuropsychiatric sequelae. The brain (globus pallidus, white matter, hippocampus) and heart bear the injury. Management is 100% oxygen ± hyperbaric oxygen (benefit best-supported by Weaver 2002 but contested by Scheinkestel/Cochrane), and the disease is highly preventable through CO alarms and combustion-source safety. Curate the genetic/inheritance sections as Not Applicable, verify all ontology IDs and PMID snippets against the dismech validation stack before committing, and flag the HBO-benefit question and rodent-DNS translational fidelity as knowledge gaps.