Experimental in vitro calibration of a pulse oximeter at low saturations

Abstract

The calibration of a fetal pulse oximeter presents major problems. Pulse oximeters calculate a parameter R, which is correlated with arterial oxygen- saturation. R is defined as the ratio of the optical absorption differences (systolic to diastolic) in two wavelengths. We relate R to the conventionally measured O₂ saturation using a curvilinear plot called the calibration curve. Calibration attempts on adults are obviously only possible in relatively high saturation regions. On the other hand fetuses present much lower saturations, while intermittent measurements of fetal arterial O₂ saturation, in direct comparison to subpartu pulse oximetry data, is impossible for technical and ethical reasons. A derivation of the calibration curve by Lambert-Beer Law is possible over the entire range of O₂ saturation, but is not in agreement with the empirical data (Fig. 1).