Sage Journals: Discover world-class research

Abstract

Introduction

Carbon monoxide (CO) poisoning is a public health concern. Devices relying on internal combustion represent a common source of contamination. The colorless and odorless nature of the gas reinforces the importance of active monitoring. Intoxication may be missed, with the potential of life-threatening compromise. The impact may be accentuated at high altitude, where lower barometric pressure reduces oxygen availability. CO hazards are a known and likely underreported issue in remote high altitude facilities. We evaluated CO levels produced by internal combustion sources in various field camp environments.

Methods

CO levels in kitchens, dining areas, shower spaces, cooking tents, and seasonal ice-fishing tents were opportunistically measured using portable CO detectors (Reed Instruments R9400). Normally distributed data are reported as mean±SD or, with small samples, as range values.

Results

Measurements were taken in Nepal (n=24; altitude 3606±520 m), Argentina (n=3; altitude 3240–4950 m), and Canada (n=3; altitude 198 m). The highest CO levels were found in shower spaces. Half the shower measures (7 of 13) exceeded World Health Organization recommendations for a 15-min CO exposure (86 ppm), with a peak measure of 1036 ppm.

Conclusions

Our sample size was small, but CO levels were seen to vary widely across field facilities, peaking in enclosed shower spaces, where World Health Organization recommendations were frequently exceeded. Elevated levels may pose safety risks, highlighting the need for better monitoring and public education.

Keywords

altitude combustion source domestic exposure inhalation intoxication poisoning remote

Introduction

Carbon monoxide (CO) poisoning resulting from the use of internal combustion devices is a public health issue affecting many populations globally. Increased levels of CO often result from common practices such as the operation of internal combustion engines, fossil fuel furnaces, and fires.¹ Global mortality from unintentional CO poisoning approaches 30,000 cases annually.² Contamination may occur in a wide array of spaces, from permanent through temporary field constructs.³

CO is a by-product of incomplete combustion of fossil fuels. When the gas is inhaled, it binds to hemoglobin with an affinity 200 to 250 times greater than oxygen to create carboxyhemoglobin (COHb). The competitive binding of CO compromises oxygen transport through the body.⁴ Toxicity results from hypoxia secondary to COHb and myoglobin binding, disruption of electron transport system reactions, and inflammation secondary to increased reactive oxygen species production. The electron transport system disruption compromises aerobic metabolism. CO-myoglobin formation alters myocardial contractility and promotes cardiovascular dysfunction. CO-induced oxidative stress can damage cell membranes through lipid peroxidation.

CO poisoning symptoms are often nonspecific, with intoxication presenting with headache, dizziness, confusion, or loss of consciousness. Acute exposure commonly leads to cognitive impairment with a possibility of long-term cognitive sequelae.^5,6 Prolonged exposure can exacerbate chronic illness, such as cardiovascular and pulmonary diseases. COHb levels of 30 to 70% can lead to a loss of consciousness and death.⁴

The World Health Organization (WHO) has indoor exposure guidelines for CO. Individuals should not be exposed to CO levels >100 mg·m^–3 (87 ppm) for >15 min, >35 mg·m^–3 (31 ppm) for >1 h, >10 mg·m^–3 (9 ppm) for >8 h, or >4 mg·m^–3 (3 ppm) for >24 h.⁷ The guidelines were based on available scientific and clinical data. Any exposure exceeding these limits can pose significant health risks.

Previous studies have evaluated CO levels in various living environments using parameters of COHb and atmospheric concentrations. Cooking equipment, more specifically open-burning stoves, were noted as the main culprit of increased household levels.⁸ Studies that measured CO levels throughout the day found that CO levels were higher during periods that coincided with mealtimes.⁹ Although cooking may be a principal concern, CO hazards also can arise from other common activities. For example, one case report described an otherwise healthy 34-y-old female who succumbed to CO poisoning after taking a gas-heated bath in the Kathmandu Valley region.¹⁰ Other unpublished reports have suggested that the use of heaters to warm water and shower spaces may represent an underappreciated risk. Public awareness is important to ensure mindfulness of this preventable issue.

The objective of this research was to evaluate the levels of CO contamination associated with the use of internal combustion sources in a variety of heated spaces in remote facilities.

Methods

We studied environmental levels of CO in remote environments that employed permanent or temporary structures and combustion-based heating systems. Human research ethics authorization was not required because no measures were collected on or from humans.

Measures were captured opportunistically during treks when access and logistics allowed. The research plan was communicated with local residents. They were informed of the findings and told that aggregate data would be published without linking high values to any particular site.

A portable CO detector was used to measure CO concentration in targeted shared-space areas (Reed Instruments Model R9400, Wilmington, NC; range 1–1000 ppm with an accuracy of ±5% or ±5 ppm using a stabilized electrochemical sensor). A multifunction altimeter was used to measure date, time, altitude, ambient pressure, temperature, and relative humidity of each assessed space (Sunroad FR500 Multifunction Digital Altimeter, Sunroad Technology, San Diego, CA; altimeter range −700–9000 m [–2297–29,528 ft] with resolution of 1 m [3.28 ft], pressure range 300–1100 kPa with resolution of 0.1 kPa, temperature range −20–60 °C, and relative humidity range 20–95%). iPhone cameras (iPhone X and iPhone 12, Apple Inc. Cupertino, CA), and a measuring mobile app (Measure app by Apple, iOS 18, 2024) were used to document the general layout of the studied areas.

CO detectors were auto-calibrated before use and positioned for 60 s at each height prior to recording. The digital screen displayed the highest level of CO recorded during the sample period. Each CO measure was accompanied by simultaneous capture of all secondary measures.

Spaces to be assessed included kitchens, dining areas, and showers. When feasible, 2 CO detectors were used to capture levels within a given space. For non-shower space locations, CO levels were recorded at 3 different heights, where possible, to account for vertical air dispersion: ground level, head-level (∼160 cm), and the height of arms extended above the technician's head (∼200 cm). The peak value from multiple height measures was accepted as the measurement peak for the cycle.

Measures were taken in kitchen and dining areas without modifying the ventilation parameters of the room beforehand. For example, if a window was open, it was left open. All windows and doors were closed in shower spaces to obtain results closest to real-life behaviors. CO levels in shower spaces were measured after a controlled shower of 8 to 10 min duration with no individuals inside. Depending on the circumstances, 1 or 2 CO detectors were installed in showers, both at approximately the same distance from the heating system and approximately at head level.

Nonnormally distributed data are presented as median (first quartile–third quartile) or peak values as appropriate. Normally distributed data are presented as mean±SD (range). CO data are reported descriptively, with summarization and analysis completed with peak values. The agreement between simultaneous CO readings taken with 2 devices was evaluated with a Pearson product-moment correlation.

Results

Kitchens, dining areas, and shower spaces were assessed in Nepal, cooking tents and a shower space in Argentina, and seasonal ice fishing huts in Quebec, Canada. The trekking lodges in Nepal shared a similar configuration: permanent structures with kitchen, shared dining and living areas, bedrooms, and shower spaces (Figures 1–4). The dining areas were generally large enough to accommodate 20 to 30 people. The living areas had a central fireplace with chimney and seating and tables arranged around the perimeter of the room. Bedrooms were usually in separate unheated buildings, and shower spaces were usually detached from the main buildings. Construction materials were primarily wood for the interior structure, including uninsulated wooden doors, and a combination of wood or wood and stone facades. Heat was typically provided by wood-burning stoves and wall-mounted gas water heaters connected to fixed shower outlets. The cooking tents in Argentina were temporary Quonset hut style structures (semicylindrical arch form) with heavy-duty reinforced tarpaulin covering a metal frame (Figures 5 and 6). The shower block was constructed using a wooden floor, thin metal walls, and a corrugated metal roof. The water for the showers was heated by propane boilers. The ice fishing shelters in Quebec were 3-layer bonded and insulated fabric top and sides (Eskimo Outbreak 850XDP; Cumberland, WI) heated with portable radiant propane heaters (Figure 7).

Figure 1.

Example of a shower space in Nepal with deposits on the wall and ceiling above the heater unit.

Figure 2.

Example of a shower space in Nepal with deposits on the wall above the heater unit.

Figure 3.

Example of a temporary cooking space in a high altitude field facility in Nepal.

Figure 4.

Examples of living spaces in long-standing high altitude facilities in Nepal.

Figure 5.

Example of a temporary cooking space in a high altitude field facility in Argentina.

Figure 6.

Example of a shower space in Argentina.

Figure 7.

Examples of temporary ice fishing tents and a propane heating system in Canada.

Data were collected over a 15-d period in Nepal (n=24; altitude 3606±520 m), over a 7-d period in Argentina (n=3; altitude 3240–4950 m), and on a single day in Canada (n=3; altitude 198 m) (Table 1). The correlation between simultaneous measures recorded with the 2 CO detectors was very strong (r=0.97, n=33). The highest CO levels were found in shower spaces in Nepal (n=13) at 111 (98–283) ppm, with a peak reading of 1036 ppm. Nepalese kitchen (n=8) levels were 17 (8–29) ppm, with a peak reading of 99 ppm. Nepalese dining area (n=3) levels were 0, 0, and 8 ppm. Cooking tents in Argentina (n=2) had CO levels of 13 and 206 ppm. A shower in Argentina (n=1) had a CO level of 6 ppm. Ice fishing tents in Canada (n=3) had CO levels of 0, 0, and 6 ppm.

Table 1.

Maximum Carbon Monoxide Levels and Associated Ambient Conditions in Different Field Sites.

Measurement site	Maximum CO level (ppm)	Altitude (m)	Inside temperature (°C)
Shower	1036	3440	—
Shower	926	3650	14.0
Shower	509	2860	15.4
Shower	283	3675	17.5
Cooking tent	206	4950	—
Shower	192	3440	18.4
Shower	156	3963	18.3
Shower	111	3346	21.7
Cooking tent	99	4855	12.7
Shower	69	3440	17.3
Kitchen	48	3598	16.9
Kitchen	23	2564	18.8
Kitchen	19	3139	14.8
Shower	17	3137	9.5
Kitchen	14	4047	4.4
Cooking tent	13	3240	—
Shower	8	3599	14.3
Kitchen	8	3671	21.1
Kitchen	7	3300	13.5
Ice fishing tent	6	190	—
Shower	6	4038	15.3
Shower	6	4100	—
Shower	5	4525	12.9
Ice fishing tent	0	190	—
Ice fishing tent	0	190	—
Shower	0	3133	18.7
Kitchen	0	3671	12.9

All peak CO levels were found at the highest vertical measurement points. This was at ∼200 cm above ground level in shower spaces and ∼160 cm for more vertically constrained cooking tents and ice fishing tents.

Discussion

Previous field studies have reported elevated CO levels associated with cooking in temporary tent and snow cave shelters.^3,11,12 CO intoxication from field use of combustion sources is a widespread and likely underreported issue. This is a concern in Nepal, where local news outlets have reported suspected fatalities.¹³ We focused on established but sometimes dated or temporary field facilities in this effort.

The main finding of this study is that CO levels can be elevated in a variety of spaces heated with internal combustion sources. The levels can exceed, sometimes substantially, the established WHO indoor exposure guidelines.⁷ The highest levels were found in shower spaces. This is not surprising because users can be highly motivated to seal the space against drafts and to maximize active heating. The risk in poorly ventilated spaces clearly increases if CO-producing sources are not fully isolated from them.

The finding that the peak CO levels were at the highest of multiheight samples may reflect the fact that carbon monoxide is slightly less dense than air (1.25 vs 1.29 kg·m⁻³, respectively). Although complete mixing with air has been reported in an airtight plexiglass container into which CO was infused,¹⁴ we saw more vertical stratification in our dynamic environments. It may be that the risk is increased for persons standing in heated spaces with low ceilings.

The risk of CO accumulation should be appreciated by all individuals using internal combustion sources because intoxication and death are possible. Actively heated spaces that are more tightly closed are expected to have the highest environmental levels of CO. Our study did not assess COHb or physiologic impact, but some of the measures were high enough to generate substantial concern.

There is generally a direct linear correlation between ambient CO levels and COHb percentage. The predicted COHb levels are 7% at a CO concentration of 50 ppm, 15% at 100 ppm, 30% at 300 ppm, and 40% at 500 ppm.¹⁵ It is noteworthy that half (7 of 13) of the shower measures in this study exceeded the WHO guidelines for 15 min of exposure, sometimes dramatically. Two of 3 of the cooking tent measures substantially exceeded the WHO guidelines for 1 h of exposure.

The rate of COHb formation is largely independent of altitude,¹⁶ but the risk of impairment increases with altitude. Decreased barometric pressure lowers the oxygen partial pressure. This reduces hemoglobin saturation with oxygen (SaO₂) and the amount of oxygen that will be carried to the tissues.¹⁷ A study conducted in Alaska demonstrated that the normal SaO₂ of a climber after 2 d of acclimatization at 4400 m was 80%.¹¹ The altitude of the 7 shower measures in this study that exceeded the 15-min WHO exposure guidelines was 3482±343 (2860–3963) m (11,424±1125 [9383–13002] ft). The barometric pressure at these altitudes, according to the US standard atmosphere, ranges from 71.4 to 61.9 kPa (535–465 mm Hg), in contrast with the 101.3 kPa (760 mm Hg; 1.0 ATA) barometric pressure found at sea level.¹⁸ The ambient partial pressure of oxygen would range from 14.9 to 13.0 kPa (112–97 mm Hg), respectively, in contrast with the 21.2 kPa (159 mm Hg) found at sea level.

CO levels are likely to be influenced by a host of factors, including environmental conditions (eg, ambient temperature and wind), the porosity of barrier materials to CO, the amount of ventilation maintained, the heating source used, and the amount of heating maintained. Our on-site observations reinforced the expectation that colder ambient temperatures would drive an increased reliance on internal combustion sources and active reduction in ventilation. The effect was most noticeable in shower spaces, where people do not have the protection of clothing and equipment typical elsewhere. Behavior-linked patterns have been documented in residential settings where an increased reliance on fuel-burning appliances and decreased ventilation such as closing windows led to a 43% increase in peak hourly CO levels during Canadian winter months.¹⁹

Other factors can contribute to elevated CO levels. A recent field study in snow caves found that repeated use of combustion sources could produce a cumulative rise in CO levels.¹² A similar effect may occur in shower spaces, where consecutive showers and a desire to keep the space closed could promote similar patterns.

Recognition of CO-related issues is complicated by the overlap between symptoms of CO intoxication and those of high altitude sickness, including headache, dizziness, nausea, and altered mental status.^5,20 Both conditions frequently present with intense headache, which may be the result of hypoxia-driven cerebral changes that can lead to cerebral edema and increased intracranial pressure.²¹ The similarity in clinical presentation may delay recognition of CO intoxication. It also could lead to underreporting of CO-related problems and decreased mindfulness of the risk.

Active monitoring of CO levels can protect and inform users and local authorities about this ongoing public health concern. Residential monitoring is becoming more common in many countries, but guidance is less consistent concerning retrofitting existing properties and less formal facilities. The colorless and tasteless nature of CO make it easy to miss, as can many of the nonspecific symptoms that may arise initially or develop with low-level exposures.

The price and availability of reliable and compact CO detectors have improved in recent years, even in highly portable units marketed to travelers.²² Fixed units would be preferred for the protection of all persons, but providing this equipment represents another demand on resources that may be limited. Both the risk and the importance of avoidance must be understood for action to be prioritized. Some lodge owners met in the course of this project expressed a lack of familiarity with CO poisoning. After a brief explanation, they indicated a willingness to adjust the configuration of spaces to allow better ventilation and to obtain CO detectors. This suggests both an interest and an ongoing need for public education and facility review.

Strategies to manage persons suffering from CO exposure include removal of or isolation from CO sources, administration of oxygen at the highest fraction possible, and medical evaluation or guidance. Options for the care of multiple casualties and severe cases should be considered, including evacuation, if needed. Severe CO intoxication is a well-established indication for hyperbaric oxygen (HBO)²³ because CO is cleared much faster than with ground-level oxygen and inflammatory processes resulting from the poisoning are modulated. Despite the merits, HBO almost certainly will not be available in remote or austere environments. It is important to understand that Gamow bags are not a substitute for HBO. HBO involves oxygen breathing during pressurization to the 203 to 280 kPa (29–41 lb·in⁻²) range. There can be confusion because 1 report has described accelerated CO clearance in smokers using a “modified Gamow bag” and oxygen breathing.²⁴ In this case, the device was pressurized to 1.58 ATA (59 kPa [9 lb·in⁻²] above ambient pressure) with compressed gas. This is far beyond the ∼14 kPa (2.0 lb·in⁻²) pressurization above ambient typically achieved with a standard foot-pump-regulated Gamow bag. Any benefit of the minimal pressurization would be further reduced by the loss of ability to closely monitor and manage patients in a Gamow bag.

Limitations

This work has several limitations. The small sample size restricts statistical power and generalizability. Details on heating systems and heating system performance were not captured. The internal volume of measured spaces was not calculated. Multiple vertical measures were not always captured. The 1000-ppm calibration range of the CO detectors was exceeded in 1 case. The presampling use of shower spaces was not controlled. Testing was not conducted in personal tents that may be commonly used in some environments. The study also did not include measures of COHb in any persons exposed to the assessed spaces, thus restricting any discussion of physiologic impact.

This work prompts several recommendations:

Increase commitment to CO monitoring. CO detectors should be installed in enclosed spaces to protect users and help ensure safe practice. The placement of monitors at head level could provide the most timely warning. Review and potentially strengthening or developing monitoring standards is warranted.

Increase outreach efforts to promote public awareness. Enhanced public education could promote a better understanding of risk factors, quicker recognition of CO intoxication symptoms, and appropriate mitigation and management strategies. Preparedness for multicasualty situations and severe cases of intoxication should be encouraged.

Expand related research. Measuring COHb levels in a variety of activities and environments would help to determine the physiologic impact of real-world field activities and more fully define the scope of the problem. While documentation is important, efforts to minimize risk should not be delayed.

Develop, implement, and evaluate public health education programming to improve community awareness and preparedness.

Develop, evaluate, and promote heating systems that do not produce harmful by-products to benefit public health. This could result in a reduction of risk in remote or underserved environments where medical care and evacuation options are likely to be limited. Initiatives to offset cost would be desirable where this could be a substantial barrier to procurement.

Conclusions

The study found variable environmental levels of CO in human-occupied field facilities. The highest CO levels were observed in enclosed shower spaces, often substantially exceeding WHO-recommended exposure limits. Although COHb levels were not measured, the high environmental CO levels could pose a meaningful threat to safety. Efforts to increase monitoring and enhance public education could help to reduce the risk to public health.

Footnotes

Author Note

Neal W. Pollock is a member of the editorial board of Wilderness & Environmental Medicine but was not involved in the peer review or publication decision-making process for this article.

ORCID iD

Neal W. Pollock

Author Contribution(s)

Léa Perron: Data curation; Investigation; Visualization; Writing – original draft; Writing – review & editing.

Rose Plourde: Data curation; Investigation; Visualization; Writing – original draft; Writing – review & editing.

Marie-Kristelle Ross: Conceptualization; Formal analysis; Investigation; Methodology; Resources; Supervision; Validation; Writing – original draft; Writing – review & editing.

Neal W. Pollock: Conceptualization; Data curation; Formal analysis; Methodology; Project administration; Supervision; Visualization; Writing – original draft; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Test equipment was provided by the Research Chair in Hyperbaric and Diving Medicine, Université Laval, Quebec, Canada.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Piantadosi

. Carbon monoxide poisoning. N Engl J Med. 2002;347(14):1054-1055.

GBD 2021 Carbon Monoxide Poisoning Collaborators. Global, regional, and national mortality due to unintentional carbon monoxide poisoning, 2000–2021: results from the Global Burden of Disease Study 2021. Lancet Public Health. 2023;8:e839-e849.

Thurman

Leavitt

Buchanan

Shreffler

Huecker

Early

. Carbon monoxide levels produced by propane/isobutane canister stoves inside a tent. Wilderness Environ Med. 2023;34(4):513-516.

Bleecker

. Carbon monoxide intoxication. Handb Clin Neurol. 2015;131:191-203.

Weaver

. Carbon monoxide poisoning. Undersea Hyperb Med. 2020;47(1):151-169.

Weaver

Hopkins

Chan

, Churchill S, Gregory Elliott C, Clemmer TP, et al. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med. 2002;347(14):1057-1067.

World Health Organization.

WHO Global Air Quality Guidelines . Particulate Matter (PM_2.5 and PM₁₀), Ozone, Nitrogen Dioxide, Sulfur Dioxide and Carbon Monoxide. WHO; 2021.

Chen

Zeger

Breysse

, Katz J, Checkley W, Curriero FC, et al. Estimating indoor PM_2.5 and CO concentrations in households in southern Nepal: the Nepal cookstove intervention trials. PLoS One. 2016;11(7):e0157984.

Powers

Ezzati

Butler

Zigler

Splenger

. Chronic household air pollution and exposure patterns among Himalayan nomads. J Expo Sci Environ Epidemiol. 2024;34(7):973-980.

10.

Bista

Manandhar

Mishra

Shrestha

Dhungel

. Carbon monoxide poisoning due to gas water heater. Janaki Med Coll J Med Sci. 2018;5(2):56.

11.

Turner

Cohen

Moore

Spengler

Hackett

. Carbon monoxide exposure in mountaineers on Denali. Alaska Med. 1988;30(3):85-90.

12.

Nielsen

Karlsen

. A field study of carbon monoxide levels in snow caves during short-term stove use. Wilderness Environ Med. 2024;35(2):119-137.

13.

Acharya

Purpura

Kao

House

. Incidence and risk factors for carbon monoxide poisoning in an emergency department in Nepal. Toxicol Comm. 2022;6(1):13-19.

14.

Hampson

Courtney

Holm

. Should the placement of carbon monoxide (CO) detectors be influenced by CO’s weight relative to air? J Emerg Med. 2012;42(4):478-482.

15.

Leigh-Smith

. Carbon monoxide poisoning in tents—a review. Wilderness Environ Med. 2004;15(3):157-166.

16.

Horvath

Bedi

. Alteration in carboxyhemoglobin concentrations during exposure to 9 ppm carbon monoxide for 8 h at sea level and 2134 m altitude in a hypobaric chamber. JAPCA. 1989;39(10):1323-1327.

17.

McGrath

JJ.

The interacting effects of altitude and carbon monoxide. In: Penney

, ed. Carbon Monoxide Toxicity. CRC Press; 2000:135-156.

18.

National Oceanic and Atmospheric Administration; National Aeronautics and Space Administration; United States Air Force. US Standard Atmosphere, 1976 (NOAA-S/T 76-1562). US Government; 1976.

19.

Sun

Héroux

M-È

, Wheeler AJ. Associations between residential fossil fuel combustion and indoor concentrations of nitrogen dioxide, carbon monoxide, and aldehydes in Canadian homes. J Expo Sci Environ Epidemiol. 2025;35(4):649-660.

20.

Luks

Swenson

Bärtsch

. Acute high-altitude sickness. Eur Respir Rev. 2017;26(143):160096.

21.

Foutch

Henrichs

. Carbon monoxide poisoning at high altitudes. Am J Emerg Med. 1988;6(6):596-598.

22.

Hampson

Holm

. Comparison of four low-level carbon monoxide alarms suitable for home use or when traveling. Undersea Hyperb Med. 2022;49(3):307-313.

23.

Huang

ET,

ed. Undersea and Hyperbaric Medical Society Hyperbaric Medicine Indications Manual. 15th ed. Best Publishing; 2023.

24.

Jay

Tetz

Hartigan

Lane

Aghababian

. Portable hyperbaric oxygen therapy in the emergency department with the modified Gamow bag. Ann Emerg Med. 1995;26(6):707-711.

Carbon Monoxide Contamination Associated with Field Use of Internal Combustion Sources

Abstract

Introduction

Methods

Results

Conclusions

Keywords

Introduction

Methods

Results

Discussion

Limitations

Conclusions

Footnotes

Author Note

ORCID iD

Author Contribution(s)

Funding

Declaration of Conflicting Interests

References