Introduction: The aim of this study was to measure the performance of short-term variation (STV) in predicting the onset of neonatal acidosis in fetuses at risk due to maternal preeclampsia.
Material and methods: This retrospective study examined data from a series of 159 women with singleton pregnancies, hospitalized for preeclampsia in a level 3 reference maternity hospital in northern France, with an STV measurement in the 24 h preceding cesarean delivery and a measurement of the newborn's arterial cord pH at birth. The main outcome was determined by a correlation between STV and neonatal pH.
Results: The last computerized fetal heart rate analysis took place a mean of 7.9 ± 6.3 h before birth, and neonatal acidosis was diagnosed in 38 newborns (23.9%). Although STV and umbilical artery pH at birth were significantly correlated (x03C1; = 0.16, p < 0.05), the performance of STV in predicting neonatal acidosis was poor, with an area under the ROC curve of 0.63. The sensitivity reached only 50.0% and the specificity 71.9% at the best STV threshold for predicting acidosis.
Conclusion: The performance of STV for screening for neonatal acidosis is poor in women with preeclampsia. The divergent results between studies are probably due to the variable intervals between STV measurement and birth.
© 2015 S. Karger AG, Basel.