Clinical Evaluation of the Ohmeda Biox III Pulse Oximeter: A Comparison of Finger and Ear Cuvettes

Abstract

We evaluated the Ohmeda Biox III pulse oximeter for accuracy of intermittent measurement of arterial oxygen saturation (SaO2) in 201 attempted measurements in 109 patients with a wide range of illnesses and severity of illness. Method: Arterial blood was sampled at the same time that two oximeters were operated, one equipped with an ear cuvette and the other with a finger cuvette. The blood SaO2 was measured with an IL282 CO-Oximeter. We obtained acceptable data (noted by absence of a low-perfusion warning) in 184 of 201 attempts (91.5%) with the ear cuvette and in all 201 attempts with the finger cuvette. Results: The mean difference between finger-cuvette SaO2 and blood SaO2 was 0.21%, which was not statistically significant. The mean difference between ear-cuvette SaO2 and blood SaO2 was 0.53%, which was statistically significant (P<0.005) but not clinically important. Both these data sets were highly correlative, 0.871 for the ear cuvette and 0.904 for the finger cuvette (P<0.0005). In the 17 samples when the ear-cuvette oximeter signaled insufficient perfusion to accurately recover data, the finger-cuvette correlation was high and significant (r = 0.969, P < 0.0005). We recorded the infrared-channel raw output from measurements being made on one patient and found a significantly higher amplitude when the finger cuvette was used (3.0 volt) than when the ear cuvette was used (0.5 volt). Conclusion: This study demonstrates that the Ohmeda Biox III pulse oximeter can accurately monitor arterial oxygen saturation in most patients. (Respir Care 1985;30:840-845).