Objective Oesophageal atresia (OA) is a major anomaly of varying severity. The complexity of surgical correction highly depends on the gap length of missing oesophagus and the presence of a distal fistula. The aim of this study was to identify antenatal sonographic findings associated with presence of a distal fistula and type of surgical repair

Methods Prenatal medical records of neonates postnatally diagnosed with OA were reviewed. Sonographic signs of OA (small/absent stomach, polyhydramnios, oesophageal pouch) and the trimester at sign detection were recorded and compared between (1) OA with and without a distal fistula and (2) early one-step versus delayed two-step anastomosis. Multivariate analysis was performed.

Results Overall, 80 cases of OA were included. Absence of a distal fistula was significantly associated with higher rates of small/absent stomach (100% vs 28.6%, P<0.0001), oesophageal pouch (100% vs 24.3%, P<0.0001) and severe polyhydramnios (66.7% vs 22.9%, P=0.006), compared with OA with a distal fistula.

Cases requiring a delayed two-step repair had higher rates of small/absent stomach (84.2% vs 16.7%, P>0.0001), severe polyhydramnios (47.4% vs 16.7%, P=0.008) and oesophageal pouch (73.7% vs 18.5%, P<0.0001), compared with those corrected in an early one-step anastomosis.

Multivariate logistic regression found small/absent stomach and pouch to be significantly and independently associated with a delayed two-step anastomosis.

Conclusion OA without a distal fistula is associated with higher rates of prenatal sonographic signs. Both small/absent stomach and a pouch are independently associated with a delayed two-step anastomosis. These findings may help improve antenatal parental counselling regarding the anticipated surgical repair.