Artefactual episodic hypoxaemia during postoperative respiratory monitoring

To determine the reliability of continuous pulse oximetry in the detection of episodic hypoxaemia in total hip joint replacement patients. Episodic hypoxaemia has been described in postoperative patients and is associated with analgesic technique. This study compared the incidence of hypoxaemic episodes identified solely by pulse oximetry and then subsequently where additional monitoring, as indicated for sleep-related breathing disorders, was also utilised. Eight patients were studied on the night before and for three nights after surgery. Pulse oximetry, ECG, chest impedance, nasal and oral airflow and transcutaneous CO2 were recorded. Sudden episodic hypoxaemia was defined as a decrease in oxygen saturation of > or = 5% within two minutes, for > five seconds and with a nadir of < 90%. Artefacts were identified by noise signals on the ECG and impedance recordings and by a motion annotation wave superimposed on the oximetry trace. Using these criteria 172 (79%) of 219 desaturation events were classified as artefactual. The median duration of genuine events was greater (P < 0.001 Mann Whitney U test) than artefactual events; 21 sec (range, 6-443) vs 11 sec (5-63). Genuine desaturations reached a median nadir of SpO2 87% (range 83-89%) compared with 81% (61-88%) for the artefactual. These differences were statistically significant (P < 0.001). Previous studies utilising unobserved pulse oximetry data alone may have overestimated the incidence of episodic hypoxaemic events in postoperative patients.