771 Impact of antenatal fetal surveillance on obstetric and neonatal outcomes in women with diabetes

Abstract

To explore the impact of common antenatal fetal surveillance (AFS) methods, including non-stress tests and biophysical profiles, on obstetric and neonatal outcomes in women with diabetes. A chart review was conducted on a cohort of pregnant women with diabetes (type 1, type 2, or gestational) who received prenatal care at a tertiary care academic center between 2012 and 2019. We identified interventions performed after abnormal AFS testing, either further testing or delivery, and then examined pregnancy outcomes. Descriptive statistics were reported in SPSS (IBM, Armonk, NY). We identified 586 diabetic pregnancies and AFS was performed on 449 (76.6%). A total of 2367 tests (1424 NSTs and 943 BPPs) were performed. Overall, 70% (n=314) had normal AFS, while 30% (n=135) had at least one abnormal AFS. The total number of AFS tests that resulted in delivery was 33/2387 (1.4%), or 33/449 (7.3%) total pregnancies. The positive predictive value of abnormal AFS was 24.4. Of AFS resulting in delivery, 13/33 (39.4%) were preterm (prior to 37 weeks) and 3 had T1DM, 7 had T2DM, and 3 had GDM. Of these preterm deliveries, 2 (15.4%) had APGAR scores < 7 and 12 (92.3%) had significant maternal comorbidities that warranted further testing and consideration of delivery. One fetal death occurred with no AFS, one neonatal death occurred despite normal AFS, and one neonatal death occurred after abnormal AFS. One woman with abnormal AFS had no interventions and resulted in an uncomplicated scheduled repeat cesarean section. AFS rarely led to delivery as an obstetric intervention, and perinatal outcomes were overall favorable. There is no clear correlation between an abnormal AFS test and perinatal mortality due to its low incidence, but a normal AFS test is overall reassuring. Most preterm deliveries associated with abnormal AFS were complicated by other comorbidities that necessitated delivery. New AFS techniques with improved ability to identify the fetus most at risk are needed to improve perinatal outcomes and decrease unnecessary interventions.