Apnea and airway obstruction are common during monitored anesthesia care (MAC). Because their early detection is essential, we sought to measure the efficacy of capnography as an indicator of apnea during MAC at a variety of oxygen flow rates compared with thoracic impedance. Anesthesia care providers using standard American Society of Anesthesiologists monitors were blinded to capnography and thoracic impedance monitoring. Ten (26%) of the 39 patients studied developed 20 s of apnea; none was detected by the anesthesia provider, but all were detected by capnography and impedance monitoring. There was no difference in detection rates between the two methods. Higher oxygen flow rates decreased the amplitude of the capnograph but did not interfere with apnea detection. This pilot study revealed that apnea of at least 20 s in duration may occur in every fourth patient undergoing MAC. Although these episodes were undetected by the anesthesia provider, they were reliably detected by both capnography and respiratory plethysmography. Monitoring of nasal end-tidal CO(2) is an important way to improve safety in patients undergoing MAC.