POSTANESTHESIA MONITORING REVISITED - FREQUENCY OF TRUE AND FALSE ALARMS FROM DIFFERENT MONITORING DEVICES

Study Objectives: To determine the frequency of true and false alarms and to determine the frequency of alarm failures for various parameters when using a postanesthesia monitoring system. Design: Open prospective study. Setting: Postanesthesia care unit at a university hospital. Patients: 123 ASA physical status I-IV patients who underwent general or spinal-epidural anesthesia for general, urologic, orthopedic, ophthalmologic, otolaryngologic, or plastic surgery. Measurements and Main Results: Monitoring included ECG, oxygen saturation, and respiratory rate (RR) by either transthoracic impedance or nasal-oral air-flow detection. We undertook careful, continuous observation of patients and monitors, recording of true and false alarms; the latter were defined as those coinciding with a clearly observed, unrelated cause. During a mean observation period of 101 minutes per patient, the average frequency of pure oximetry alarms was once every 8 minutes, with 77% of the alarms being false, caused by sensor displacement, motion artifacts, poor perfusion, or a combination of these factors. Apnea alarms occurred on average once every 37 minutes, with the ''false'' fraction being 28% and 27% for impedance and flow detection, respectively. The impedance sensor failed to detect apnea on at least G occasions; the flow sensor failed on I occasion. The coincidence of pulse oximetry and apnea alarms was small, and ECG exhibited a low alarm rate but a high fraction of false alarms. Patients receiving opioids and neuromuscular relaxants had a higher frequency of ''true'' apneas than patients not receiving these drugs. No clear correlation was found in regard to age distribution, type of surgery, duration of anesthesia, oxygen administration, or ASA physical status between the patient groups that exhibited many apneas or many pulse oximeter alarms and the whole study population. Conclusions: The high frequency of apnea strongly motivates the use of continuous RR monitoring, preferably by flow-sensing techniques, since both central and obstructive apneas are then detected. Further study and development is necessary before pulse oximetry can be unconditionally recommended for postanesthesia monitoring.