Changes in short-term variation of antenatal cardiotocography to identify intraamniotic infection: a historical cohort study.

Introduction: Intraamniotic infection (IAI) is one of the main possible complications of preterm prelabor rupture of membranes (PPROM) and can lead to severe consequences for the neonate, such as early onset neonatal sepsis (EONS). Available diagnostic tools for IAI have poor diagnostic performance, which may result in both over- and underdiagnoses of IAI. In a search for better diagnostic tools, we have examined short-term variation (STV) in fetal heart rate. We have previously shown that in IAI exposed pregnancies, the STV was more than 20% lower in the last cardiotocography trace before the start of labor, as compared to those not exposed to IAI. The association between IAI and STV needs further evaluation and we therefore continued by examining the longitudinal change in STV in association with IAI.

Material and methods: We performed a historical cohort study on 628 singleton pregnancies with PPROM, delivering between 24 + 0 to 33 + 6 gestational weeks. The main exposure of the study was IAI, using EONS as a proxy as no easily available method exists for confirming IAI antepartum, and IAI and EONS are strongly associated. The main outcome was STV in fetal heart rate. At least two available cardiotocography traces per fetus were required as a minimum, from PPROM or from seven days before birth, whichever came later, until the start of labor or planned cesarean birth. A total of 9 690 cardiotocography traces were analyzed.

Results: Fetuses exposed to IAI had a 26.5% steeper decline in their STV during the last 24 h before the start of labor when compared to fetuses not exposed (95% CI -32.9% to -19.4%; p < 0.001). After adjustment for antenatal corticosteroids, the decline remained significant. The decline became less prominent but the significance remained when also adjusting for the baseline frequency (-12.7% [95% CI -19.3% to -5.5%], p < 0.001). In the IAI-exposed group, the baseline frequency increased by 11.1 bpm during the last 12 h before the start of labor, beyond those who were not exposed (95% CI 8.3 bpm to 13.8 bpm; p < 0.001).

Conclusions: In pregnancies affected by IAI the STV declines steeper in the last 24 h before the start of labor as compared to pregnancies not affected by IAI, even after adjustment for increasing baseline frequency. The association of STV in relation to IAI needs to be further studied in order to evaluate and establish STVs usefulness in monitoring patients for IAI.