The purpose of this pilot study is to determine the feasibility of using a novel treatment for sleep apnea in which the patient's own exhaled CO2 is tightly controlled and used in a rebreathe system to reduce sleep disordered breathing.

Obstructive sleep apnea (OSA) needs to be treated with devices that will be utilized with greater compliance by patients than is currently the case with positive airway pressure. To this end, the investigators recently showed that increased inspired CO2 via rebreathing was effective in reducing most obstructive and central apneas in OSA patients. Then, the investigator's University of Wisconsin (UW) team of bioengineers, physiologists, and sleep physicians built a novel variable dead space rebreathe device-with no added positive pressure-which monitors breath by breath ventilation and automatically adjusts the rebreathe dead space volume to add or subtract the level of inspired CO2 depending on the degree of sleep disordered breathing. This approach provides the minimum effective CO2 dose needed in individual OSA patients to stabilize central respiratory motor output and to recruit upper airway dilator muscles, thereby treating obstructive and central apneas. The investigators propose to determine the effectiveness of this "Smart CO2" treatment as well as its effects on sleep state stability, sleep quality, and blood pressure in 10 moderate to severe OSA patients studied over several nights. The investigators expect the findings from the proposed study to be sufficient to determine if the "Smart CO2" treatment is a viable approach to OSA treatment worthy of testing in a clinical trial.

Inclusion Criteria:

1. Age 18-75 years old, inclusive
2. Moderate to severe OSA (Apnea hypopnea index [AHI] >15 events/hr of sleep)
3. Newly diagnosed OSA, or have either refused to or have not tolerated PAP treatment, or are currently using Positive Airway Pressure (PAP) and are willing to hold it during the night of sleep study visits.
4. Willing and able to provide consent to 5 laboratory visits with 4 full nights of sleep study.
5. Self report of still having sleep apnea since the last sleep test (if diagnosis was greater than a year prior).

Exclusion Criteria:

1. BMI >45 kg/m2
2. AHI<15 events/hour on the first night of sleep testing in our research laboratory (Visit 2)
3. Cigarette smoking of 1 pack per day or more within 6 months of screening;
4. Diagnosed heart failure or coronary artery disease.
5. Diagnosed cerebrovascular disease, Transient Ischemic Attack (TIA) or stroke.
6. Diagnosed asthma or chronic obstructive pulmonary disease with significant airways obstruction (FEVI/Forced Vital Capacity <65%)
7. End-stage hepatic or renal disease
8. Significant daytime sleepiness in subjects who have risky occupations or life-style (for example, working in the transportation industry, history of motor vehicle crash, occupation that requires vigilance for safety or performance)
9. Pregnancy (assessed on urine test at V1 in females with childbearing potential)
10. Clinically significant arrhythmia/dysrhythmias on the clinical diagnostic polysomnography (per medical record) or on research polysomnography conducted at V2
11. Established diagnosis of neuromuscular disease (e.g., Parkinsonism, multiple sclerosis, syringomyelia, transverse myelitis, amyotrophic lateral sclerosis, poliomyelitis, Eaton-Lambert, Guillain-Barre, myasthenia gravis, myotonic dystrophy, mononeuritis multiplex, in the setting of polymyositis/dermatomyositis or severe cervical spine disease)
12. Any current use of benzodiazepines, opioids, or barbiturates
13. Any medical or psychiatric condition which in the opinion of the medical director interferes with the subject's ability to enroll or to continue in the study (once enrolled)
14. Veteran, cared for in the VA system