Critical incidents detected by pulse oximetry during anaesthesia.

The Critical Incident Technique was used to study anaesthetics given in a major tertiary care teaching hospital in order to define indications for monitoring with a pulse oximeter during anaesthesia. Anaesthetists were asked to use a pulse oximeter in every case and trained to report Critical Incidents in order to determine if the oximeter can shorten the time to detection of these events. Four thousand seven hundred and ninety-seven anaesthetics were given during a four-month period in 1986-87. A Critical Incident was recorded when an unexpected physiologic deterioration requiring intervention by the anaesthetist to prevent a likely bad outcome was signalled first by the pulse oximeter. Critical Incidents were classified by patient characteristics, physiologic change, type of anaesthetic, and the type, length, and place of surgery. Reports were received in 65 per cent of cases, and a Critical Incident occurred in 191 (six per cent) of these. Desaturation was the commonest physiologic change (151) and was further classified as to severity (mild 85-94 per cent saturation, moderate 75-84 per cent, and severe 75 per cent). Desaturations during the maintenance phase of anaesthesia were milder but more frequent than those in the induction or emergence phases. There were no severe desaturations in elderly patients or those receiving regional anaesthesia. No group was free of Critical Incidents. Since undetected hypoxaemia may lead to disastrous complications we recommend that a pulse oximeter be used for every anaesthetic.