Fetal oxygen saturation and fractional extraction at birth and the relationship to measures of acidosis

OBJECTIVE: We sought to determine umbilical cord oxygen saturation and fractional oxygen extraction values as measured at birth for a large tertiary hospital population and their predictive value for measures of acidosis. STUDY DESIGN: The computerized perinatal database of St. Joseph's Health Centre, London, Ontario, was used to obtain the umbilical cord gases, pH, mode of delivery, gestational age at delivery, and nuchal cord status for all live-born infants >500 gm between January 1991 and December 1995 (n = 22,134), Oxygen saturation values were calculated from the umbilical cord PO2 and pH data with a previously derived empirical equation, the accuracy of which was rechecked with 100 consecutive cord blood samples where oxygen saturation values were both calculated and measured with a hemoximeter (r = 0.99, p = 0.001). Fractional oxygen extraction values were calculated from the umbilical cord oxygen saturation data. RESULTS: There were 18,250 "validated" paired umbilical vein and artery blood gas and pH results available for analysis after patient case exclusions for missing, unreliable, or "unphysiologic" data. For all validated patient cases, mean umbilical vein oxygen saturation was 63% +/- 16% (SD), mean umbilical artery oxygen saturation was 24% +/- 15%, and mean fractional oxygen extraction was 0.62 +/- 0.20, with all three of these parameters significantly affected by mode of delivery, gestational age at delivery, and nuchal cord status, Umbilical vein and artery oxygen saturation and fractional oxygen extraction values showed significant relationships with umbilical artery base excess, albeit weak (r = 0.18 to 0.22), and pH (r = 0.46), which were best described using cubic regression models. Receiver-operator characteristic curve statistics for the prediction of acidosis at birth were also significant for all three of these parameters but lower when predicting metabolic versus mixed acidosis. However, all showed a poor positive predictive value for significant acidosis at birth, whether metabolic or mixed and regardless of the cutoff values used, CONCLUSION: Umbilical cord oxygen saturation and fractional oxygen extraction values as measured at birth for a large tertiary hospital population indicate a decreased oxygen margin of safety for infants born postterm, by cesarean section after labor, and with a nuchal cord. However, these values have a limited relationship to measures of acidosis, which may have clinical implications for the usefulness of intrapartum pulse oximetry.