Comparison of two new generation pulse oximeters with arterial oxygen saturation in critically ill children.

OBJECTIVES: To compare the performance of two new generation pulse oximeters, one with enhanced signal extraction technology (SET) and other without enhanced SET in detecting hypoxemia and to correlate it with arterial blood gas analysis.
METHODS: Forty-eight patients, admitted to pediatric intensive care unit (PICU) of a teritiary care teaching hospital in India for critical care and support during the study period, who had an arterial catheter in situ were included. Children with those disease conditions known to interfere with pulse oximetry and blood gas analysis were excluded.184 set of observations were made during the study period. Each set had oxygen saturation (SpO2) measured from both the pulse oximeters and the corresponding arterial oxygen saturation (SaO2). The values were compared for occurrence of true and false alarms during periods of normal BP, hypotension and varying degrees of hypoxia.
RESULTS: The mean arterial SaO2 in the study was 94.4 % ± 4.9. The mean SpO2 recorded in conventional and enhanced signal extraction technology (SET) pulse oximeters were 94.9 % ± 4.5 and 97.2 % ± 4.7 respectively. Enhanced signal extraction technology pulse oximeter detected 4/27 (15 %) of true hypoxemic events and 1 event was a false alarm. Conventional pulse oximeter detected 11/27 (41 %) true hypoxemic events but recorded 6 false alarms.
CONCLUSIONS: Both pulse oximeters were not found to be performing satisfactorily in picking up hypoxemia in the study. There was good correlation with mean SpO2 from pulse oximeters and arterial SaO2. The reliability of pulse oximetry decreases with worsening hypoxemia and hypotension, and the sensitivity for picking up hypoxemia can be as low as 15 %.