Pollack Jessica C, Mathew Leny, Williams Clara, Hornyak Jake, Flohr Sabrina, Reynolds Tom, Hedrick Holly L, Oliver Edward R, Manfredi Michael A, Hamilton Thomas E, Dao Duy T
Division of Pediatric General, Thoracic, and Fetal Surgery, Children's Hospital of Philadelphia, Philadelphia, USA.
The Richard D. Wood Jr. Center for Fetal Diagnosis and Treatment, Children's Hospital of Philadelphia, Philadelphia, USA.
J Pediatr Surg. 2025 Aug 12:162530. doi: 10.1016/j.jpedsurg.2025.162530.
We aimed to determine the performance of fetal ultrasound in prenatal detection of esophageal atresia/tracheoesophageal fistula (EA/TEF) and evaluate the impact of prenatal diagnosis on postnatal outcomes at a major quaternary care center.
We conducted a retrospective review of patients who underwent prenatal screening for suspected congenital anomalies at our institution from 2013 to 2024 and included those with prenatal suspicion on ultrasound and/or postnatal diagnosis of EA/TEF (N = 70). We then performed a retrospective cohort analysis of all patients who underwent repair of EA/TEF at our institution, comparing outcomes between those with correct prenatal diagnosis (N = 28) and those with postnatal diagnosis only (N = 168).
The sensitivity of fetal ultrasound for prenatal diagnosis of EA/TEF was 48 %, with a positive predictive value (PPV) of 78 %. Sensitivity was 36 % in patients with type C and 100 % in patients with type A. Later gestational age (GA) at ultrasound and presence of fluid-filled esophageal pouch, under-distended/collapsed stomach, and polyhydramnios increased the odds of prenatal diagnosis among patients with type C EA/TEF. There was high concordance between fetal ultrasound and MRI in prenatal detection of EA/TEF. After adjusting for type of EA/TEF, GA at birth, and presence of major cardiac anomalies, there was no difference in age at initial operation, age at discharge, or rate of reoperation.
Fetal ultrasound has high PPV for prenatal diagnosis of EA/TEF and sensitivity is highly dependent on type of EA/TEF. There are no differences in outcomes between prenatally and postnatally diagnosed patients after adjusting for confounders.
