Objective: Our purpose was to assess the level of agreement between oxygen saturation values obtained from two identical sensors used on different sides of the face of the same human fetus during labor.
Study design: Two identical fetal pulse oximeter sensors were placed on 12 fetuses during uncomplicated labor at < or = 38 weeks' gestation. Oxygen saturation, fetal heart rate, and uterine activity were recorded. The agreement between synchronous values of oxygen saturation was assessed by calculating the mean difference and SD of the difference. The SD of a single sensor was estimated as the SD of the difference divided by the square root of 2.
Results: The mean oxygen saturation value returned from one sensor was 49.2% and from the other 49.9%; the mean difference was -0.7%. The SD of the difference was 7.5%, and the 95% limits (+/- 2 SDs of the difference) were -15.5% to 14.1%. The SD from a single sensor was estimated as 5.3%.
Conclusions: There was no clinically significant difference between the oxygen saturation values returned from two identical sensors on the one fetus. The magnitude of the SD from a single sensor must be taken into account when an arbitrary "cutoff" or "action" oxygen saturation value in a clinical setting or trial is defined.