Carbon monoxide (CO) has been called a "silent killer", and those patients who survive CO poisoning are at risk of neurological damage, which may be permanent. CO is a leading cause of unintentional poisoning deaths in the United States, and the odorless gas results in an estimated average of 20,636 emergency department (ED) visits each year. Oxygen is the antidote for CO poisoning, and it acts both by attenuating toxic effects and enhancing elimination. A fractional inspired concentration of oxygen (FiO2) of 0.7 to 0.9 may be achieved by administration of 100% oxygen delivered using a reservoir with a facemask that prevents rebreathing. Hyperbaric oxygen therapy may provide added benefit for patients with CO poisoning, but this therapy is unavailable in many parts of the United States including Vermont. Use of a continuous positive airway pressure (CPAP) mask may achieve an FiO2 of 1.0, but the effects of delivering an FiO2 of 1.0 compared to 0.7 in CO poisoning are unknown. CPAP, by comparison, is inexpensive, portable, and available in most EDs. In this study, the investigators are testing the hypothesis that oxygen delivered by CPAP will improve both CO washout kinetics and functional outcomes, compared to the standard therapy of oxygen delivered by non-rebreathing facemask. Specific Aim 1 will provide toxicokinetic data to support a potential benefit in the use of CPAP for CO poisoning, by comparing CO elimination kinetics in response to oxygen therapy delivered by non-rebreathing facemask versus CPAP. The 20 patients expected in our first year will provide adequate power to detect a 20% fall in half-time of CO elimination. While CPAP may increase CO washout rates, as predicted in Specific Aim 1, demonstration of real functional benefit will be tested in Specific Aim 2. This Aim seeks to determine functional (neuropsychological) outcomes in patients with CO poisoning treated with oxygen therapy delivered by non-rebreathing facemask versus CPAP. Data showing a therapeutic benefit from CPAP in CO poisoning would have clinical implications. Compared to hyperbaric oxygen therapy, CPAP therapy can begin earlier, including the pre-hospital setting, for patients with known exposure. With the frequent nature of CO poisoning and the widespread availability of CPAP, a potential benefit could lead to improved outcomes for the 20,000+ patients who present to EDs annually.

Inclusion Criteria:

• Elevated Carboxyhemoglobin Level (non-smokers >8%, smokers >12%)
• 18 years of age or older
• Able to provide informed consent as assessed by Attending Emergency Physician

Exclusion Criteria:

• Requires daily medication for active lung disease
• Altered mental status
• Hemodynamically unstable
• Requires transfer to ICU or hyperbaric oxygen facility
• Previous enrollment in the study
• No concurrent acute psychiatric illness