Don't miss carbon monoxide poisoning: Screen patients

Misdiagnoses are common

When a group of guests at a motel in Ocean City, MD, were brought to an ED with headaches and nausea, they were treated for food poisoning — but a carbon monoxide (CO) leak was the actual cause of their symptoms. The leak went undetected for several hours, and two deaths occurred.

A recently published study found that only 44% of acute care hospitals in the Pacific Northwest had an arterial blood gas machine with a CO-Oximeter on site to measure carbon monoxide levels in the blood.1 The equipment costs approximately $40,000 and can be obtained from Westlake, OH-based Radiometer America, Tarrytown, NY-based Bayer HealthCare, East Walpole, MA-based Ciba-Corning Diagnostics, and Alameda, CA-based Roche Diagnostics. There are more than 40,000 ED visits for carbon monoxide poisoning annually, but many cases go undetected because symptoms often are mistaken for influenza or a migraine headache.2

To quickly screen patients, a small but growing number of ED nurses are using handheld CO detectors. The Rad-57 Pulse CO-Oximeter, manufactured by Irvine, CA-based Masimo Corp., costs approximately $3,000 and noninvasively monitors the level of carbon monoxide and methemoglobin in the blood for early detection and treatment.1 The tool can be used to avoid drawing arterial blood gas levels for some patients. When two boys aged 3 and 7 were brought to the ED at Presbyterian Intercommunity Hospital in Whittier, CA, with smoke inhalation, the tool showed that one child had a carbon monoxide level of 8 that exactly matched the blood gas level. "So we decided not to get a blood gas on the 3-year-old because of that," says Joe Ball, RRT, the respiratory therapist who cared for the boys.

Since ED nurses knew the result was accurate because it had been confirmed with the first boy's blood gas level, the blood gas didn't need to be done for the second boy, which saved him a painful procedure.

At Rhode Island Hospital in Providence, ED nurses screen all patients at triage with the RAD-57. "We use it as a screening tool," says Bruce Gillard, RN, assistant clinical manager of the ED. "It's just like a pulse oximeter: We're talking about seconds, and you get a level."

Nurses document the patient's carbon monoxide level on the chart for the physician to review. "If somebody has a level of 1, that's fine; and if someone is a smoker, the level may be a little higher. But if somebody comes back with a 10 or 13, there may be another reason for that, and we will need to ask more questions," says Gillard.

For example, patients with high levels are asked whether they have been working on a car in a confined space or using a space heater in the home. (See checklist of questions used by ED nurses to assess carbon monoxide toxicity, below.)

Ask these Questions if You Suspect CO Poisoning

Patients with CoHb levels from 2%-5%:
• Do you smoke?
• How much do you smoke?
• When did you last smoke?

Patients who do not smoke and have COHb over 3%:
• What is your occupation?
• Do you work with industrial solvents or paint removers?
• How do you heat your home?
• Do you use a space heater at home?
• Have you been using a charcoal grill at home?
• Do you use gasoline-powered tools such as mowers, snow   blowers, chainsaws, or weed trimmers?
• Have you had any recent exposure to exhaust fumes?
• Have you had any recent exposure to a fire?
• Does your home have a chimney?
• Have you been on a motorboat?
• Have you recently gone camping and used a gas stove?
• Are there any other family members, roommates, or co-workers   with the same symptoms?

Source: Rhode Island Hospital, Providence.

You might suspect carbon monoxide exposure if patients report symptoms such as headache and nausea, but the condition can be overlooked easily if patients don't have these symptoms, Gillard explains. "Previously, we would draw a blood sample to acquire the carbon monoxide level on someone only when we were clinically looking at it to be elevated, such as if somebody was in a fire or smoke inhalation," he says.

Several patients have come in with completely unrelated complaints and had elevated carbon monoxide levels detected by the screening, Gillard reports.

"We are finding an aberrant reading from time to time," he says. In one case, a man came to the ED after a bicycle accident and ended up having a level greater than 10, which was confirmed by blood gas. "We found out the patient had been using a heater in the home that wasn't ventilated appropriately," says Gillard.


1. Kao LW, Nanagas KA. Carbon monoxide poisoning. Emerg Med Clin N Am 2004; 985-1,018.

2. Hampson NB, Scott KL, Zmaeff JL. Carboxyhemoglobin measurement by hospitals: Implications for the diagnosis of carbon monoxide poisoning. J Emerg Med 2006; 31:13-16.


For more information about screening for carbon monoxide levels, contact:

  • Joe Ball, RRT, Presbyterian Intercommunity Hospital, 12401 Washington Blvd., Whittier, CA 90602-1006. Telephone: (562) 698-0811, ext. 2545. Fax: (562) 698-6865. E-mail:
  • Bruce Gillard, RN, Assistant Clinical Manager, Emergency Department, Rhode Island Hospital, 593 Eddy St., Providence, RI 02903. Telephone: (401) 640-5166. E-mail:

For more information about the RAD-57, contact: