Clinical implications of umbilical artery Doppler changes after betamethasone administration^†

Background: Betamethasone (BMZ) is commonly administered to patients with fetal growth restriction (FGR) and abnormal umbilical artery Doppler (UAD) velocimetry due to the increased risk of preterm delivery; however, the clinical impact of UAD changes after BMZ exposure is unknown.Objective: To test the hypothesis that lack of UAD improvement after BMZ administration is associated with shorter latency and greater neonatal morbidity in patients with FGR.Study design: This was a retrospective cohort study of pregnancies complicated by FGR and abnormal UAD between 24⁰ and 33⁶ weeks gestation. Abnormal UAD included the following categories of increasing severity: elevated (pulsatility index >95%), absent end diastolic flow (EDF), or reversed EDF improvement was defined as any improvement in category of UAD within two weeks of BMZ. Sustained improvement was defined as improvement until the last ultrasound before delivery, whereas transient improvement was considered as unsustained. The primary outcome was latency, defined as interval from betamethasone administration to delivery. Secondary outcomes were gestational age at delivery, umbilical artery pH, and a composite of neonatal morbidity (intubation, necrotizing enterocolitis, ionotropic support, intraventricular hemorrhage, total parenteral nutrition, neonatal death). Outcomes were compared between (a) patients with and without UAD improvement and (b) patients with sustained and unsustained improvement, using univariable, multivariable and time-to-event analyses.Results: Of the 222 FGR pregnancies with abnormal UAD, 94 received BMZ and had follow-up ultrasounds. UAD improved in 48 (51.1%), with 27 (56.3%) having sustained improvement. Patients with hypertension and drug use were less likely to have UAD improvement. Patients without UAD improvement had shorter latency (21.5 days [interquartile range (IQR) 8,45] versus 35 [IQR 22,61], p = .02) and delivered at an earlier gestational age (34 weeks [IQR 31,36] versus 37 [IQR 33,37], p < .01) than those with improvement. There were no differences in umbilical artery pH between groups. Composite neonatal morbidity was higher in patients without UAD improvement, but this was not statistically significant after adjusting for confounders (aOR 2.0; 95% CI 0.08-5.1). There were no differences in outcomes between patients with sustained versus unsustained improvement.Conclusions: UAD improved in half of patients following BMZ. Lack of UAD improvement was associated with shorter latency and earlier gestational age at delivery, but no difference in composite neonatal morbidity. UAD response to BMZ may be useful to further risk stratify FGR pregnancies.