False-positive findings during electronic fetal heart rate monitoring may explain the increased rates of cesarean delivery without associated improvements in neonatal outcome. Because of the poor specificity of fetal heart rate monitoring in predicting fetal distress, methods of fetal blood gas monitoring are being investigated as a way to improve the accuracy of assessing the infant's condition during labor. Seelbach-Göbel and associates prospectively investigated the use of fetal arterial oxygen saturation (SaO₂) as a predictor of acidosis caused by hypoxemia. The authors sought to identify the minimum duration of a low SaO₂ value (i.e., 30 percent or less) that would serve as a predictor of a significant decline in fetal pH.

Fetal pulse oximetry was used to monitor fetal SaO₂ during 400 deliveries of infants between 37 and 42 weeks' gestation. To identify the minimum duration of reduced SaO₂ before the onset of acidosis, the authors determined the number of minutes in which SaO₂ was low (30 percent or less), medium (31 to 60 percent) and high (more than 60 percent) during each delivery. Umbilical artery pH and base excess values after delivery, as well as Apgar scores, were then compared with the amount of time the infants spent in low, medium and high SaO₂ during labor and delivery. Umbilical artery pH, base excess values and Apgar scores were categorized as follows: pH, less than 7.15 versus 7.15 or more; base excess, less than 12 mmol per L versus 12 mmol per L or more; and one-minute Apgar score, less than 7 versus 7 or more.

Umbilical artery pH immediately after delivery was less than 7.15 in 54 neonates and 7.15 or more in 346 neonates. Infants with a pH of less than 7.15 had a significantly longer duration of low fetal SaO₂ values than did those with a pH of at least 7.15. In the group with lower pH values, the mean duration of SaO₂ of 30 percent or less was 23.5 minutes, compared with a mean duration of 8.1 minutes in infants with a pH of at least 7.15. Neonates with an umbilical artery pH of 7.15 or more had high values of fetal SaO₂ for a significantly longer time than did neonates with a lower pH.

Among the infants in whom fetal scalp blood sampling was performed during labor and delivery, the duration of a low SaO₂ level proved to be the best predictor of a decline in the scalp pH between two samplings. The pH declined significantly with an increasing duration of low fetal SaO₂. No decrease of more than 0.05 was observed unless the fetal SaO₂ remained at less than 30 percent for more than 10 minutes.

The findings of this study confirm that a fetal SaO2 of 30 percent is the critical threshold value during labor. The pH was found to decrease by 0.02 per 10 minutes of a fetal SaO₂ of 30 percent or less. Thus, when the fetal pH is 7.30 at the beginning of labor, an average of 60 minutes of a low fetal SaO₂ level will reduce the fetal pH to 7.20. The findings suggest that fetal scalp sampling should be conducted at shorter intervals if the fetal SaO₂ is continuously far below the 30 percent threshold.

The authors conclude that even in the presence of a nonreassuring fetal heart rate pattern, one should not expect a significant reduction in the fetal pH if the fetal SaO₂ as measured by pulse oximetry does not drop to less than 30 percent for longer than 10 minutes. Because the specific acid-base status for a fetus would not be known without monitoring, the authors believe that monitoring should ideally begin with a fetal scalp sample to determine the actual fetal pH. Fetal blood sampling should then be repeated only if the fetal SaO₂ is 30 percent or lower for a longer interval. The authors stipulate that the interval for repeat monitoring depends on the original pH, the severity of hypoxemia as determined by pulse oximetry and the quality and continuity of the fetal SaO₂ signal. They suggest that the use of fetal pulse oximetry in practice may reduce the need for fetal blood sampling, an important consideration in hospitals in which fetal blood analysis is a routine part of labor and delivery.