Exposure to Elevated Carbon Monoxide Levels at an Indoor Ice Arena — Wisconsin, 2014

Paul D. Creswell, PhD1; Jon G. Meiman, MD1,2; Henry Nehls-Lowe, MPH1; Christy Vogt, MPH1; Ryan J. Wozniak, PhD1; Mark A. Werner, PhD1; Henry Anderson, MD1

On December 13, 2014, the emergency management system in Lake Delton, Wisconsin, was notified when a male hockey player aged 20 years lost consciousness after participation in an indoor hockey tournament that included approximately 50 hockey players and 100 other attendees. Elevated levels of carbon monoxide (CO) (range = 45 ppm–165 ppm) were detected by the fire department inside the arena. The emergency management system encouraged all players and attendees to seek medical evaluation for possible CO poisoning. The Wisconsin Department of Health Services (WDHS) conducted an epidemiologic investigation to determine what caused the exposure and to recommend preventive strategies. Investigators abstracted medical records from area emergency departments (EDs) for patients who sought care for CO exposure during December 13–14, 2014, conducted a follow-up survey of ED patients approximately 2 months after the event, and conducted informant interviews. Ninety-two persons sought ED evaluation for possible CO exposure, all of whom were tested for CO poisoning. Seventy-four (80%) patients had blood carboxyhemoglobin (COHb) levels consistent with CO poisoning (1); 32 (43%) CO poisoning cases were among hockey players. On December 15, the CO emissions from the propane-fueled ice resurfacer were demonstrated to be 4.8% of total emissions when actively resurfacing and 2.3% when idling, both above the optimal range of 0.5%–1.0% (2,3). Incomplete fuel combustion by the ice resurfacer was the most likely source of elevated CO. CO poisonings in ice arenas can be prevented through regular maintenance of ice resurfacers, installation of CO detectors, and provision of adequate ventilation.

WDHS abstracted patient information for persons who were present at the event and subsequently sought care at four area EDs. Information collected included demographics, smoking status, underlying medical conditions, participant type (player, coach, or spectator), symptoms, COHb levels, and any treatment provided. A case of CO poisoning was defined as a blood COHb level >5.0% for nonsmokers or >10.0% for smokers (1) in a person who attended the tournament and was subsequently evaluated at an ED during December 13–14, 2014. Data on smoking behavior from both ED records and a follow-up survey conducted in February 2015 were used to classify cases of CO poisoning on the basis of COHb levels. COHb levels were measured using one of three methods, depending on the test available at the facility: 1) COHb saturation tests from venous blood samples; 2) noninvasive pulse CO-oximetry tests, using handheld devices; and 3) exhaled CO, using breath CO monitors. When multiple methods were used, the order of preference was venous blood sample measures, CO-oximetry, and breath CO monitors (4), an order which reflected the accuracy of the different tests.

In February 2015, WDHS mailed a survey to ED patients whose records had been abstracted to verify smoking status and to gather additional information on symptoms and sequelae. Because incomplete fuel combustion poses a risk for nitrogen dioxide (NO2) exposure (5), respondents were also asked about symptoms associated with NO2 exposure, including cough, coughing up blood, chest pain, chest tightness, sore throat, choking, shortness of breath, and eye irritation. If the survey was not returned, a follow-up postcard was mailed. In April 2015, WDHS personnel contacted persons who had not responded to conduct a telephone interview. Persons who did not return a survey or participate in the telephone interview were considered lost to follow-up. A t-test was used to assess whether persons who reported more symptoms during their ED visit were more likely to respond to the follow-up survey than those who reported fewer symptoms. On February 26, 2015, WDHS personnel conducted unstructured interviews with the ice arena manager and the Lake Delton Fire Chief, toured the facility, and obtained records of CO measurements and emissions test data from the ice resurfacer.

Among the 92 persons who sought ED care during December 13–14, 2014, 57% were aged 16–20 years, and 70% were male. Only one person reported being a current smoker, and two were hospitalized (Table). Symptoms reported included headache (84%), nausea (66%), vomiting (26%), dizziness (19%), and shortness of breath (10%). COHb levels were determined by venous blood for 77 (84%) persons, by CO-oximetry for 10 (11%), and by exhaled CO breath monitoring for five (5%). COHb levels ranged from 0%–21.7% (mean = 9.7%; standard deviation [SD] = 4.4%). Case status was determined using both ED and survey data sources; two persons who did not report smoking during the ED visit but reported it later in the survey were reclassified as non-CO poisonings for this analysis. Overall, 74 (80%) of evaluated patients, including 32 hockey players, met the criteria for CO poisoning (1). Among persons with CO poisoning, COHb levels ranged from 5.1%–21.7% (mean = 11.1%; SD = 3.6%). The 32 hockey players had higher COHb levels (mean = 13.0%; SD = 3.6%), compared with 42 spectators (mean = 9.7%; SD = 3.0%; p<0.001). Two patients, the hockey player who lost consciousness at the ice arena and a woman aged 25 years who was 19 weeks pregnant, received hyperbaric oxygen treatment as inpatients. The hockey player who lost consciousness had the highest measured COHb blood level (21.7%). Supplemental oxygen was administered to 91 patients. Patients were also treated with acetaminophen (n = 14), ibuprofen (n = 1), and meperidine (n = 1) for pain, and ondansetron (n = 18) and promethazine (n = 2) for nausea, vomiting, and dizziness.

Approximately 2 months after the event, surveys were mailed to 91 persons for whom contact information was available; overall, 47 (52%) responded. Persons who reported more symptoms in the ED were not more likely to respond to the survey than those who reported fewer symptoms (t = 1.31, p = 0.194). The symptoms most commonly reported among the 36 survey respondents who had CO poisoning were lightheadedness (70%), shortness of breath (33%), and chest tightness (22%). Six (17%) of these respondents reported persistent symptoms ≥7 weeks after the incident, including memory problems (n = 4) or difficulty concentrating (n = 4). Other persistent symptoms included cough (n = 1), shortness of breath (n = 1), and mood changes (n = 1). Symptoms associated with NO2 exposure were infrequently reported. Cough was reported by five (13.8%) persons, eye irritation by four (11.1%), sore throat by two (5.6%), chest pain by two (5.6%), and choking by two (5.6%); no one reported coughing up blood.

The informant interviews revealed that CO levels of 45 ppm–165 ppm had been detected by the local fire department on the night of the event. The measurement taken near the locker room where the hockey player lost consciousness was 140 ppm. These measurements were higher than the acceptable air quality standard for CO levels at an ice arena, which is ≤20 ppm (6). On December 15, 2014, the CO output of the ice resurfacer used during the weekend was measured at 4.8% of total emissions when actively resurfacing (i.e., loaded) and 2.3% when idling; both of these are above the optimal range of 0.5%–1.0% (2,3). No other mechanical CO sources were identified.

Discussion

CO is a colorless, odorless, and tasteless gas that is highly toxic. When CO is inhaled, it binds to hemoglobin to produce COHb, which appropriates the space that normally carries oxygen, thereby depriving the tissues of sufficient oxygen (7). Exposure to CO can cause various negative health effects ranging from headache and fatigue to coma and death; lasting neurologic problems can also occur (2,7). CO exposure episodes during indoor sporting events have been associated with substantial morbidity among exposed persons (5,8,9), including psychological sequelae (10). Exposure to CO is of particular risk to pregnant women and has been reported to result in miscarriage and developmental problems for the fetus (5).

This CO poisoning event was the largest reported in Wisconsin and steps have been taken to prevent future exposures at ice arena events. At the arena, multiple CO detectors were installed, and maintenance was conducted on the ice resurfacer. WDHS updated outreach materials with recommendations for trade groups, hockey organizations, coaches, and parents. These materials include detailed information on topics such as proper ventilation and symptoms of CO poisoning. WDHS has also improved surveillance for CO poisonings by using real-time alerts generated from Wisconsin Poison Center data.

The findings in this report are subject to at least two limitations. First, discrepancies in smoking status were identified between ED records and surveys. ED visit data only identified one smoker, whereas four survey respondents reported they smoked on some days. Because CO poisoning case definition was dependent upon smoking status, inadequate collection of smoking data could affect case ascertainment. Only half of persons who were evaluated in EDs responded to the survey; therefore, smokers might have been incompletely ascertained, leading to incomplete identification of cases of CO poisoning. Second, the reporting of symptoms differed between medical records and survey responses, with slightly more symptoms reported in the survey. The survey is likely less reliable as it lacks confirmatory assessments (i.e., biologic measurements) available to medical personnel in the clinical setting. Moreover, considering the delay in the administration of the survey, the possibility for recall or participation bias exists.

Despite awareness of the risk for CO poisoning from ice resurfacers with high CO emissions, large-scale CO poisoning events still occur. Increased risk communication and outreach efforts might help to prevent future events. Certain states, including Connecticut, Massachusetts, Minnesota,* Pennsylvania, and Rhode Island have enacted indoor CO monitoring requirements to prevent large-scale CO poisoning events; other public health stakeholders with multiple ice rink venues might consider this approach as an avenue for primary prevention.

Acknowledgments

Kristine M. Bisgard, Fuyuen Yip, Joshua Mott, U.S. Public Health Service; Jenny Camponeschi, Roy Irving, Wisconsin Bureau of Environmental and Occupational Health; Darren Jorgenson, Lake Delton Fire Department, Baraboo, Wisconsin; Cindy Bodendein, Sauk County Health Department, Baraboo, Wisconsin; Poppy Waterman Ice Arena personnel, Lake Delton, Wisconsin; Wisconsin Poison Center; Wisconsin Environmental Public Health Tracking Program; Michael Paul Reid, Fátima Coronado, Division of Scientific Education and Professional Development, Center for Surveillance, Epidemiology, and Laboratory Services, CDC.

1Division of Public Health, Wisconsin Department of Health Services; 2Epidemic Intelligence Service, CDC.

Corresponding author: Paul D. Creswell, paul.creswell@dhs.wisconsin.gov, 608-267-9752.

References

1. Council of State and Territorial Epidemiologists. Public health reporting and national notification for carbon monoxide poisoning. Atlanta, GA: Council of State and Territorial Epidemiologists; 2015. Available at http://c.ymcdn.com/sites/www.cste.org/resource/resmgr/PS/13-EH-01.pdf.
2. Masters GM. Introduction to environmental engineering and science. 2nd ed. Upper Saddle River, NJ: Prentice Hall; 1998.
3. Pelham TW, Holt LE, Moss MA. Exposure to carbon monoxide and nitrogen dioxide in enclosed ice arenas. Occup Environ Med 2002;59:224–33.
4. Agency for Toxic Substances and Disease Registry. Toxicological profile for carbon monoxide. Atlanta, GA: US Department of Health and Human Services, Agency for Toxic Substances and Disease Registry; 2015. Available at http://www.atsdr.cdc.gov/toxprofiles/tp201.pdf.
5. CDC. Nitrogen dioxide and carbon monoxide intoxication in an indoor ice arena—Wisconsin, 1992. MMWR Morb Mortal Wkly Rep 1992;41:383–5.
6. Minnesota Department of Health. Summary of ice arena rules. Minneapolis, MN: Minnesota Department of Health; 2013. Available at http://www.health.state.mn.us/divs/eh/indoorair/arenas/forms/keychangestorule.pdf.
7. Sircar K, Clower J, Shin MK, Bailey C, King M, Yip F. Carbon monoxide poisoning deaths in the United States, 1999 to 2012. Am J Emerg Med 2015;33:1140–5.
8. CDC. Carbon monoxide poisoning at an indoor ice arena and bingo hall—Seattle, 1996. MMWR Morb Mortal Wkly Rep 1996;45:265–7.
9. Paulozzi LJ, Satink F, Spengler RF. A carbon monoxide mass poisoning in an ice arena in Vermont. Am J Public Health 1991;81:222.
10. Chambers CA, Hopkins RO, Weaver LK, Key C. Cognitive and affective outcomes of more severe compared to less severe carbon monoxide poisoning. Brain Inj 2008;22:387–95.