Abstract
Pulse oximeter (Nellcor 100) was used to continuously monitor (via skin surface) O2 saturation (TcSO2) in 26 neonates (GA 25-40 wks, BW 700-3500 gm) with cardiac or respiratory distress. First, we studied whether TcSO2 reading would correlate with actual measurement of arterial O2 saturation (SaO2). Simultaneous deteriminations of SaO2 by OSM3 hemoximeter (Radiometer) revealed a linear correlation of TcSO2 and SaO2 with r=0.969 (p<106). Secondly, we obtained a correlation equation of TcSO2 and PaO2 through simultaneous blood samplings (N=394): log SaO2/100-SaO2=-2.951+2.262 log PaO2. (r=0.871, p<0.0001). This equation was almost identical with fetal Hb O2 dissociation curve. We derived from these data the range of TcSO2 (80% to 95%) to maintain PaO2 in the range of 40 to 80 torr. Finally using these criteria we studied the sensitivity and specificity of pulse oximetry by analyzing 394 paired TcSO2 and PaO2 samples.
Confining TcSO2 in the range of 80 to 95% in order to restrict PaO2 within the limit between 40 to 80 torr revealed a sensitivity of 88% and specificity of 77%. Keeping O2 saturation in this range would overestimate PO2 in 6.5% (18/278 PaO2 <40 torr) and underestimate PaO2 in 2.1% (6/278 PaO2 >80 torr) of determinations. Understanding its optimal range will enhance effectiveness in clinical practice.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Wei, T., Bautista, A., Ko, S. et al. PULSE OXIMETRY-ITS RELIABILITY IN PREDICTING ARTERIAL OXYGENATION. Pediatr Res 21 (Suppl 4), 208 (1987). https://doi.org/10.1203/00006450-198704010-00249
Issue Date:
DOI: https://doi.org/10.1203/00006450-198704010-00249