RESUMEN
OBJECTIVE: Evaluate accuracy of prenatal ultrasound findings in predicting the risk of bowel atresia in patients with gastroschisis. METHODS: A retrospective study was conducted on 18 fetuses with a prenatal diagnostic of gastroschisis treated at University hospital of Saint Etienne France between 2002 and 2012. Ultrasound abnormalities were used to classify them into three groups: no ultrasound abnormality (n=4), oligohydramnios (n=9), intra-abdominal bowel dilatation ≥20.5mm (n=5). Postnatal outcomes were compared between groups. The threshold value of 20.5mm for the prediction of atresia was determined through the receiver operator characteristics curve. RESULTS: In the group with oligohydramnios, intra uterine growth restriction were significantly more frequent (p=0.015) and three newborns had serositis including two with secondary complications after the initial surgery. In the group with major intra-abdominal bowel dilatation, all had a narrow defect <10mm significantly more than other fetuses (p=0.002). Intra-abdominal bowel dilatation reaching 20.5mm started at a mean gestational age significantly lower than that of the other fetuses (23.3 versus 29.7 weeks p=0.02). On the five fetuses presented intra-abdominal bowel dilatation ≥20.5mm, four showed atresia and no other newborn has this complication (p=0.0016). The threshold value of 20.5mm has a sensitivity of 100% and a specificity of 92.9%. The area under the curve was equal to 96.4%. CONCLUSION: Intra-abdominal bowel dilatation ≥20.5mm seems to be associated with the risk of postnatal atresia. MRI could help to clarify a complicated or uncertain ultrasound aspect.