Acute Carbon Dioxide Poisoning Reveals Issues in Patient Rescue and Management

Two employees of a local restaurant go down to the basement to investigate a leak in the pressurization system for the carbonated beverage dispenser, which is producing an audible hiss. On entering the basement, one of the employees, a healthy 64-year-old male, notices the hissing sound is coming from a compressed gas cylinder of carbon dioxide with some frost on it. (See Figure 1 below.)

He turns to pick up a crescent wrench to close the valve and falls to the ground unresponsive, with seizure-like activity. His colleague and others at the restaurant attempt a rescue but are unable due to a sensation of suffocation. Police officers arriving on scene recognize the potentially hazardous atmosphere and prevent any further attempts at entry.

Figure 1: Compressed gas cylinder containing carbon dioxide

Simple asphyxiants displace oxygen from ambient air. Figures couresy Matthew Sztajnkrycer


Responding fire department personnel make entry using self-contained breathing apparatus (SCBA) equipment and effect a rescue. Initial environmental readings include an ambient oxygen level of 14.7% with no explosive gases, carbon monoxide, or hydrogen sulfide detected.

On initial assessment, the patient is hypoxic (84%) and unresponsive. Bag-valve mask respirations are delivered with 100% oxygen, with rapid improvement. Upon awakening, the man is initially combative, and is  administered midazolam 2 mg IV. He’s subsequently transferred to the ED of the regional tertiary care hospital and Level 1 trauma center.

Hospital personnel are informed of an unresponsive individual with seizure activity. Assuming the exposure to be related to carbon monoxide, they make plans to activate the hyperbaric oxygen chamber. It’s only upon arrival that the actual nature of the exposure becomes clear.

Upon arrival in the ED, the patient remains somnolent and intermittently combative, but is maintaining a patent airway. Initial vital signs include heart rate 130, blood pressure 95/69, respiratory rate 24, oxygen saturation 97% on 15 Lpm non-rebreather. Additional physical exam findings include tachycardia and bilateral rales on pulmonary exam, left more so than right. Initial arterial blood gas analysis was significant at pH 7.18 and pO2 at 109 mmHg on 100% oxygen (typically greater than 500 mmHg). Point-of-care lactate is 6.9 mmol/L. Chest X-ray demonstrates evidence of pulmonary edema. (See Figure 2 below.) ECG shows sinus tachycardia, nonspecific ST T-wave changes, but no evidence of acute ischemia. (See Figure 3 below.) Initial troponin T is 0.02 ng/mL, with three-hour troponin increasing to 0.78 ng/mL. A CT scan of the brain demonstrates neither acute ischemic events nor any evidence of traumatic bleed.

Figure 2: Single view chest X-ray

This chest X-ray demonstrates bilateral interstitial infiltrates, left greater than right. Regions of bronchial wall cuffing are noted in both lungs. These findings are consistent with parenchymal infection or edema.


The patient is admitted to the ICU for further monitoring of his acute hypoxic event with evidence of a non-ST segment elevation myocardial infarction and pulmonary edema. An echocardiogram demonstrates preserved ejection fraction (62%) but new anterior regional wall motion abnormalities. (See Figure 4 below.) Coronary angiography demonstrates only mild coronary artery atherosclerosis. He’s discharged from the hospital in stable condition and without sequelae on hospital day three.

Simple Asphyxiants

Oxygen is essential for life, serving as the electron receptor for mitochondrial oxidative phosphorylation, with resultant energy production. At sea level (760 mmHg), oxygen accounts for 20.9% of atmospheric gases, and results in a partial pressure of 150 mmHg in the lungs of healthy individuals.

An asphyxiant is any substance or environment that interferes with the body’s ability to use oxygen for energy production. Most asphyxiants encountered in the hazardous materials (hazmat) environment are classified as cellular asphyxiants. These agents, which include cyanide, hydrogen sulfide, carbon monoxide and sodium azide, interfere with the utilization of oxygen at the cellular level, including oxygen transport by hemoglobin or mitochondrial oxidation and energy production.1—6

In contrast, a simple asphyxiant causes end-organ toxicity by displacing oxygen from ambient air, reducing the partial pressure of oxygen and creating an oxygen deficient atmosphere. Typically inert, these gases include nitrogen, carbon dioxide, noble gases and short chain aliphatic carbon gases. (See Table 1 below.)7—9

Figure 3: Initial ECG

ECG shows sinus tachycardia (104 beats per minute), QRS duration of 86 ms, QTc interval 460 ms and nonspecific ST-T wave abnormalities.


An oxygen-deficient atmosphere is any atmosphere containing less than 19.5% ambient oxygen.10 According to United States Occupational Safety and Health Administration standards, an oxygen-deficient atmosphere is considered immediately dangerous to life and health.11

Simple asphyxiants cause physical harm through reduction in ambient oxygen concentration, thereby depriving the body of oxygen. The accidental release of inert gas from pressurized cylinders in enclosed spaces has contributed to many cases.7 In contrast to physical suffocation, this form of asphyxia hasn’t typically been associated with a sensation of breathing difficulty.12

While ambient oxygen concentrations less than 19.5% are considered immediately dangerous to life and health, significant signs and symptoms of hypoxia don’t occur until concentrations fall below 16%, especially in otherwise healthy individuals.7,10 Oxygen concentrations less than 10% are associated with coma.

As a consequence, this case is unusual both because of the knockdown-type effect noted in the victim despite an oxygen concentration of 14.7%, and because of the sensation of breathing difficulty noted during the initial rescue attempts. Although it may occur with any agent if the levels are high enough, knockdown effect is classically associated with the cellular asphyxiant hydrogen sulfide, which wasn’t detected in this case.4,5 The mechanism of knockdown in hydrogen sulfide exposure is believed to be rapid uptake of lipophilic hydrogen sulfide by the lipid-rich respiratory center of the brainstem.13

Table 1: Examples of simple asphyxiant gases


In contrast to other simple asphyxiants, which have no pharmacological activity, carbon dioxide is an end product of cellular respiration and the biologic marker of ventilation. (See Table 1.) Elevated carbon dioxide is buffered in the serum as carbonic acid; the subsequent acidemia stimulates ventilation to rapidly equilibrate acid-base homeostasis. Unfortunately, in this situation, increased respiratory rate results in increased carbon dioxide inhalation. The sensation of suffocation occurs more rapidly in the setting of hypercapnea than hypoxia.12

Elevated carbon dioxide levels are also associated with narcosis and central nervous system depression.14 It’s likely these effects contributed to the findings in this case. In 1986, a massive carbon dioxide release from the volcanic Lake Nyos in Cameroon killed nearly 2,000 people.15 The death was described as so rapid that victims appeared to “die in their tracks.”

Assumptions & Preconceptions

This case highlights the perils of assumptions and preconceptions, as well as issues with communications. Initial responders reported to the dispatch center that a victim was unresponsive and seizing due to gas exposure. This was interpreted by the responding ED staff to represent a carbon monoxide exposure, as this was the most commonly encountered gas exposure in the region and is associated with both coma and seizures. Although later amended to specifically identify the gas as carbon dioxide, this was felt by ED personnel to reflect a communication error on the part of the dispatch center. On multiple previous occasions, dispatch had reported “carbon dioxide” rather than “carbon monoxide” exposure despite direct communication to the contrary.

Confusion regarding the nature of chemical exposure isn’t uncommon. In the initial minutes after the 1995 Tokyo Sarin chemical weapon attack, the primary receiving hospital was alerted by the Tokyo Metropolitan Ambulance Communications Center that an explosion and toxic gas release had occurred within the subway system. Based upon this information, the ED began to prepare for patients suffering from burns, inhalation injuries and carbon monoxide poisoning.16

Figure 4: Bedside echocardiogram results

Emergent bedside echocardiogram in the ICU demonstrates areas of hypokinesis and akinesis in the anterior segments of the heart.


No direct communication occurred between the on-scene responders and the ED, and ED information regarding environmental sampling measurements was only received after patient arrival by directly contacting fire dispatch. Ideally, both direct communications from the scene to ED medical control, as well as transport of specific field data with the patient, should occur to avoid miscommunication and agent misidentification.

Although only a single victim occurred in this event, there was a significant potential for a hazmat mass casualty event.17 Two people initially entered the basement, and both became rapidly symptomatic. The individual closest to the door was able to self-evacuate to safety, but then he and others attempted a second rescue. Further attempts were prevented by the recognition of a potential hazmat event by initial law enforcement responders. Multiple previous studies have demonstrated the potential for bystanders and first responders to become incapacitated during rescue attempts at hazmat events, with resultant mass casualty and mass fatality incidents.18,19 This finding highlights the importance of all first responders to be trained to recognize potential hazmat environments, and avoid unprotected entry.


Although uncommon, simple asphyxiants have the potential to cause rapid and significant end-organ toxicity. The case above highlights several key issues in patient rescue and management: the potential for rapid incapacitation by simple inert gases, the potential for mass casualty events, and issues of miscommunications between the responding services and the receiving hospital. Responders should be aware of all these issues when attending these incidents. 


1. Hampson NB, Piantadosi CA, Thom SR, et al. Practice recommendations in the diagnosis, management, and prevention of carbon monoxide poisoning. Am J Respir Crit Care Med. 2012;186(11):1095—1101.
2. Guzman JA. Carbon monoxide poisoning. Crit Care Clin. 2012;28(4):537—548.
3. Weaver LK. Clinical practice: Carbon monoxide poisoning. N Engl J Med. 2009;360(12):1217—1225.
4. Guidotti TL. Hydrogen sulfide: Advances in understanding human toxicity. Int J Toxicol. 2010;29(6):569—581.
5. Novotny-Baumann M, Baud FJ, Descatha A. Can the initial clinical signs be used for triage of patients with acute H2S poisoning? J Emerg Med. 2011;41(4):403—404.
6. Reade MC, Davies SR, Morley PT, et al. Review article: Management of cyanide poisoning. Emerg Med Australas. 2012;24(3):225—238.
7. Miller TM, Mazur PO. Oxygen deficiency hazards associated with liquefied gas systems: Derivation of a program of controls. Am Indust Hyg Assoc J. 1984;45(5):293—298.
8. Tan KH, Wang TL. Asphyxiants: Simple and chemical. Ann Disaster Med. 2005;4(suppl 1):S35—S40.
9. Dorevitch S, Forst L, Conroy L, et al. Toxic inhalation fatalities of US construction workers, 1990—1999. J Occup Environ Med. 2002;44(7):657—662.
10. National Institute for Occupational Safety and Health: Criteria for a recommended standard: Working in a confined space. U.S. Government Printing Office: Washington, D.C., 1979.
11. Major requirements of OSHA’s Respiratory Protection Standard 29 CFR 1910.134. OSHA Office of Training and Education: Washington, D.C., 2006.
12. Manning HL, Schwartzstein RM. Pathophysiology of dyspnea. N Engl J Med. 1995;333(23):1547—1553.
13. Haouzi P. Ventilatory and metabolic effects of exogenous hydrogen sulfide. Respir Physiol Neurobiol. 2012;184(2):170—177.
14. Leikin JB, Mitton JF, Freedom T. Carbon dioxide-induced narcosis due to dry ice exposure in a patient with sleep apnea. Ann Intern Med. 2009;150(5):361—362.
15. Kling GW, Clark MA, Wagner GN, et al. The 1986 Lake Nyos gas disaster in Cameroon, West Africa. Science. 1987;236(4798):169—175.
16. Okumura T, Takasu N, Ishimatsu S, et al. Report on 640 victims of the Tokyo subway sarin attack. Ann Emerg Med. 1996;28(2):129—135.
17. Halpern P, Raskin Y, Sorkine P, et al. Exposure to extremely high concentrations of carbon dioxide: A clinical description of a mass casualty incident. Ann Emerg Med. 2004;43(2):196—199.
18. Suruda A, Agnew J. Deaths from asphyxiation and poisoning at work in the United States, 1984—1986. Br J Ind Med. 1989;46(8):541—546.
19. Gabbay DS, De Roos F, Perrone J. Twenty-foot fall averts fatality from massive hydrogen sulfide exposure. J Emerg Med. 2001;20(2):141—144.

No posts to display