It’s Complex: Predicting Gastroschisis Outcomes Using Prenatal Imaging

Abstract

Introduction: Gastroschisis occurs in 1 out of 2,000 births with survival rates partially contingent on intestinal complications and time to establishing feeding. Enhancements in prenatal imaging have given better insight into postnatal outcomes. The goal of this study was to examine the gastroschisis patient population at a single children’s hospital in the modern era and to utilize prenatal ultrasound to develop new prenatal prognostic indicators.

Methods: We performed a retrospective review of gastroschisis patients at a quaternary care referral children’s hospital from 2010 through 2018. We recorded demographics, prenatal data and imaging, early postnatal data, operative data, and patient outcomes. We compared patients within our cohort born with complex gastroschisis (bowel atresia/ perforation) to uncomplicated gastroschisis patients. Second trimester and third trimester prenatal ultrasounds (US) were evaluated for changes in amount of external bowel, bowel dilatation, and bowel wall edema to identify prognostic indicators of the status of the bowel at birth. For categorical variables, Chi-square tests were used to assess for significance. Univariate and multivariable analyses were used to asses significance between categorical and continuous variables using medians and interquartile ranges or means.

Results: 134 patients were included in the study: complex (24), uncomplicated (110). Compared to uncomplicated gastroschisis, complex patients required longer median days to feeding initiation (44 vs 10, p<0.001), full feeding (80 vs 23, p<0.001), length of stay (LOS) (83 vs 33, p<0.001), and TPN at discharge (p=0.004). Full US data was available on 81% of patients, and partial data was identified on 19%. Prenatal US analysis showed significantly more complex patients had polyhydramnios amniotic fluid on third trimester US (4.3% to 23.5%, p=0.018). US analysis between complex and uncomplicated patients showed large amount of external bowel (41.2% vs 22.3%, p=0.129) and prevalence of internal bowel dilation (29.4% vs 10.6%, p=0.053) on third trimester US and increase in bowel edema (29.4% vs 13.8%, p=0.148) and external bowel dilation (64.7% vs 51.1%, p=0.429) from second to third trimester US. Multiple multivariable logistic regression analyses revealed amniotic fluid on third trimester US to be the most significant predictor of complex gastroschisis. However, there were no differences in perioperative or long-term complications in the complex group when compared to the group with uncomplicated gastroschisis.

Conclusions: Markers on prenatal ultrasound can predict intestinal complications at birth. Complex gastroschisis is associated with increased time to feeds and LOS.