The purpose of this study is to compare the accuracy of EtPCO2 measurements taken with a nasal cannula using the Microcap (Oridion Capnography Inc., Needham, MA) micro-stream device and the TG-920 main-stream device (Nihon Kohden, Tokyo, Japan). Three groups of non-intubated, spontaneously breathing patients will be included: 1) lean patients without obstructive sleep apnea (OSA); 2) obese patients without OSA; and 3) obese patients diagnosed with OSA. This latter group has a higher prevalence of oral breathing.

Additionally, to test the efficacy of oral guides, we propose to compare the accuracy of EtPCO2 measurements from the Microcap with two different nasal cannulas, one with (Smart CapnoLine PlusTM, Oridion Medical, Jerusalem, Israel) and one without (CapnoLineTM H, Oridion Medical) an oral guide to trap gas expired from the mouth. The TG-920 main-stream system requires its purpose-designed YG-122T cannula so that is the only one that will be tested with the Nihon-Kohden product.

After institutional review board approval and informed consent, we will recruit 60 patients who are scheduled for general anesthesia: 20 normal weight patients (defined as BMI < 30 kg/m2) without a diagnosis of OSA, 20 obese patients (BMI > 35 kg/m2) without a diagnosis of OSA, and 20 obese patients with polysomnography-diagnosed OSA. Patients with known severe pulmonary or cardiac disease will be excluded.

Protocol Patients will be given general anesthesia with endotracheal intubation or a laryngeal mask airway. In the post-anesthesia care unit, they will be given oxygen through a nasal cannula. They will be randomly assigned to one of two capnometers: Microcap (Oridion Capnography Inc., Needham, MA) or TG-920 (Nihon Kohden, Tokyo, Japan) with O2 administered at 3 L/min. The order of application of the two capnometers will be randomized according to computer-generated codes kept in opaque envelopes until the study begins. The order of the two nasal cannulas for the Microcap (Oridion Capnography Inc., Needham, MA) will also be randomized.

Thus, each patient will have their EtCO2 values measured under three conditions: with the Microcap capnometer via the Smart CapnoLine PlusTM cannula at 3 L/min O2 flow; with the Microcap capnometer via the CapnoLineTM H cannula at 3 L/min O2 flow; and with the TG-920 capnometer with 3 L/min O2 flow. EtCO2 will be measured for 5 minutes with each device and cannula. At the end of the 5-min sampling period (5 minutes after start of each capnometer and nasal cannula combination), 2-3 ml arterial blood will be drawn to obtain blood gas measurements. This blood sample will be obtained from an indwelling radial arterial catheter that will be inserted pre-, intra-, or postoperatively by the attending anesthesiologist, the resident in charge of the case, or one of our investigators. After the blood gas samples are collected, another capnometer will be applied and the same procedure will be repeated. When the capnometers or cannulas are changed, the patients will be asked to rate their comfort level.

At the beginning of each sampling period for each capnometer and nasal cannula combination, an Angiocath, which will be connected to the capnometer, will be applied to nostrils and presence or absence of nasal breathing will be recorded. The capnometers and arterial blood gas analyzer will be calibrated before each experiment according to the manufacturer's instructions.

Measurements Morphometric and demographic characteristics of the participating patients will be recorded. Our major outcomes will be accuracy of EtCO2 values with each capnometer during administration of 3 L/min O2 in non-obese patients without OSA, obese patients without OSA, and obese patients with OSA. In each patient, during the 5-min application of each combination of capnometer and nasal cannula, EtCO2 values will be measured for 5 minutes and averaged. These averaged EtCO2 values for each set will be subtracted from the arterial partial pressure of carbon dioxide (PaCO2) value measured simultaneously ( GEM Premier 3000, Instrumentation Laboratory, Lexington, MA). At nominal value of PaCO2 at 34 mmHg, precision and accuracy of PaCO2 measurement by GEM Premier 3000are 1 mmHg and 0.15 mmHg, respectively. Our minor outcomes will be oxygenation efficacy and patient's comfort with the three nasal cannulas. The arterial partial oxygen pressure (PaO2) will be measured ( GEM Premier 3000, Instrumentation Laboratory, Lexington, MA) for each nasal cannula. Patients will rate their comfort level on a 10-cm visual analogue scale (VAS) for each nasal cannula.

Inclusion Criteria:

• Normal weight patients (defined as BMI < 30 kg/m2) without a diagnosis of OSA
• 20 obese patients (BMI > 35 kg/m2) without a diagnosis of OSA
• 20 obese patients with polysomnography-diagnosed OSA

Exclusion Criteria:

• Severe pulmonary disease
• Cardiac disease