P0710 - Association between antenatal ultrasound findings and neonatal outcomes in rural Uganda

Background: Although the use of antenatal ultrasound (US) services has increased in low and middle income countries (LMICs), there has not been a concurrent significant improvement in perinatal mortality. It remains unknown if individual US findings in this setting are associated with neonatal death or the need for resuscitation at delivery. If such an association were identified it could be used to inform the birth attendant and counsel the family regarding risk, potentially altering the location of delivery and preparedness of the team responsible for the neonate in order to reduce neonatal mortality.

Methods: This was a secondary analysis of data collected from a prospective cohort. This study was conducted with the support of the American Academy of Pediatrics Marshall Klaus Neonatal-Perinatal Research Award. Data had been gathered at Nawanyago Health Centre III (HCIII) in Kamuli District, Uganda. Participants included pregnant women who received prenatal US scans and also delivered at that center between July 2010 - August 2018, and their neonates. All US were performed and deliveries were attended solely by midwives. Predictor variables included the following US findings: number of fetuses, fetal position, and amniotic fluid volume. The primary outcome was bag-mask ventilation (BMV) of the neonate at delivery. The secondary outcome was stillbirth or neonatal death at delivery. After data analysis was complete, study investigators traveled to Uganda to provide education around the results to providers at HCIIIs, as well as to provide Helping Babies Breathe training in order to facilitate preparedness for resuscitation at delivery.

Results: Primary outcome data was available for 1105 infants and secondary outcome data was available for 1098 infants. A total of 33 infants received BMV at delivery. The odds of receiving BMV at delivery was significantly increased if amniotic fluid volume was abnormal (OR 4.2, CI 1.2-14.9) and was also increased for multiples (OR 1.9, CI 0.7-5.4) and for non-vertex fetal position (OR 1.4, CI 0.6-3.2). Stillbirth or neonatal death was diagnosed for 20 infants. Multiple gestation (OR 4.7, CI 1.6-14.2) and abnormal amniotic fluid volume (OR 4.8, CI 1.0-22.1) significantly increased the odds of stillbirth or neonatal death.

Conclusion: Some findings that are identifiable on US in a LMIC are associated with adverse neonatal outcomes. Education around this could lead to improved delivery preparedness, with the potential to reduce neonatal mortality. This was a preliminary retrospective study, and was affected by limitations with regards to the data that was collected. Larger prospective studies are needed to adequately assess the strength of these associations, as well as to analyze potential confounders and examine other prenatal US findings.