Characterization of respiratory compromise and the potential clinical utility of capnography in the post-anesthesia care unit: a blinded observational trial

The utility of capnography to detect early respiratory compromise in surgical patients after anesthesia is unclear due to limited prospective data. The purpose of this trial was to determine the frequency and duration of capnography-detected respiratory adverse events in the post-anesthesia care unit (PACU). In this prospective observational trial, 250 consenting patients undergoing elective surgery with general anesthesia were monitored by standard monitoring together with blinded capnography and pulse oximetry monitoring. The capnography notification settings were adjusted to match nursing (Level II) and physician (Level I) alarm thresholds. 163 (95%) patients had a Level II notification and 135 (78%) had a Level I notification during standard monitoring. The most common events detected by the capnography monitor included hypocapnia, apnea, tachypnea, bradypnea and hypoxemia, with silent notification duration for these events ranging from 17 ± 13 to 189 ± 127 s. During standard monitoring, 15 respiratory adverse events were reported, with 8 events occurring when valid blinded/silenced capnography and pulse oximetry data was collected simultaneously. Capnography and the Integrated Pulmonary Index™ algorithm (IPI) detected respiratory adverse events earlier than standard monitoring in 75% and 88% of cases, respectively, with an average early warning time of 8 ± 11 min. Three patients’ blinded capnography was unblinded to facilitate clinical care. Respiratory adverse events are frequent in the PACU, and the addition of capnography and IPI to current standard monitoring provides potentially clinically relevant information on respiratory status, including early warning of some respiratory adverse events. Trial registration ClinialTrials.gov Identifier NCT02707003 (https://clinicaltrials.gov/ct2/show/NCT02707003). ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s10877-019-00333-9) contains supplementary material, which is available to authorized users.