Fetal MRI improves diagnostic accuracy in patients referred to a fetal center for suspected esophageal atresia Cecilia G. Ethun a, b , Sara C. Fallon a, b , Christopher I. Cassady a, c , Amy R. Mehollin-Ray a, c , Oluyinka O. Olutoye a, b , Irving J. Zamora a, b , Timothy C. Lee a, b , Stephen E. Welty a, d , Darrell L. Cass a, b, ⁎ a Texas Children’s Fetal Center, Texas Children's Hospital, Houston, Texas b Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas c Department of Radiology, Baylor College of Medicine, Houston, Texas d Department of Pediatrics, Division of Neonatology, Baylor College of Medicine, Houston, Texas abstract article info Article history: Received 2 February 2014 Accepted 13 February 2014 Key words: Esophageal atresia Tracheoesophageal fistula Fetal MRI Prenatal diagnosis Purpose: The purpose of this study was to describe prenatal imaging characteristics and outcomes of fetuses with suspected esophageal atresia (EA) in order to improve prenatal diagnosis, counseling, and management. Methods: The medical records of all patients referred to our multidisciplinary fetal center for suspected EA from January 2003 to April 2013 were reviewed retrospectively. Results: Thirty-three patients were referred with a prenatal diagnosis of possible EA. Following fetal center evaluation with MRI, EA was deemed unlikely in 6 (18%) fetuses. Of 27 fetuses in whom EA could not be excluded, EA was confirmed postnatally in 15 (56%), excluded in 7 (26%), and unconfirmed in 5 (3 fetal losses; 2 lost to follow-up). Imaging characteristics on fetal MRI associated with the highest positive predictive values (PPV) were an esophageal pouch (100%) and a small stomach (75%). The finding of polyhydramnios had high sensitivity (93%) but low specificity (31%) and PPV (61%) for a diagnosis of EA. Conclusion: Prenatal imaging and fetal center evaluation correctly identify the presence or absence of esophageal atresia in 78% of patients referred on suspicion of this condition. The presence of an esophageal pouch on fetal MRI has significant predictive value for EA. These data may assist with evidence-based prenatal family counseling. © 2014 Elsevier Inc. All rights reserved. Accuracy in the prenatal diagnosis of esophageal atresia (EA) is elusive; despite recent improvements in prenatal imaging, the antenatal detection rate remains less than 50% [1–4]. Because the normal esophagus is often not seen on routine ultrasound examina- tion [5], suspicion for EA is raised indirectly in the presence of polyhydramnios or a small or absent stomach bubble [1]. In isolation, these findings are not specific for detecting EA, as both are associated with a variety of different, more commonly encountered condi- tions [2,6,7]. Combining the findings of a small or absent stomach bubble and polyhydramnios might slightly improve the positive predictive value of EA detection [4,6,8]. For those fetuses with multiple anomalies the prenatal diagnosis of EA is complicated even further, and false positive rates are as high as 78% [3] in these patients. In an effort to improve antenatal diagnostic capabilities in this disease, recent focus has been placed on the identification of an esophageal pouch, or a ‘pouch sign’. Caused by dilatation of the blind- ending upper esophageal segment during fetal swallowing, it is considered to be the most consistent indicator of EA on prenatal imaging with a reported positive predictive value as high as 100% [9,10]. However, detection of the pouch sign on ultrasound can be difficult, as it depends on fetal position, mobility, gestational age, and active fetal swallowing during the exam [7,10,11]. The aim of this study was to describe the prenatal imaging characteristics and outcomes of fetuses with suspected esophageal atresia referred to a comprehensive, multidisciplinary fetal center in order to improve prenatal diagnosis, counseling, and management. 1. Methods All patients referred to our center underwent a comprehensive ultrasound examination at the initial visit, and most patients had fetal MRI. Those patients with no other indication for fetal MRI and normal amniotic fluid and stomach bubble on radiology-performed ultra- sound at the time of fetal center consultation had no other imaging. MRI scans were performed without sedation or contrast as primarily T2-weighted single shot and balanced steady-state free precession sequences. In a typical case, serial sagittal images at a single level in the midline were obtained more than 20 second intervals to evaluate fetal swallowing and the transient distension of the proximal esophagus. Imaging assessment focused on identifying polyhydram- nios, the presence and size of the stomach, esophageal anatomy, the presence of a dilated esophageal pouch (Fig. 1), and evidence of other Journal of Pediatric Surgery 49 (2014) 712–715 ⁎ Corresponding author at: 6701 Fannin St., Suite 1210, Houston, TX 77030. Tel.: +1 832 822 3135; fax: +1 832 825 3141. E-mail address: dcass@bcm.edu (D.L. Cass). http://dx.doi.org/10.1016/j.jpedsurg.2014.02.053 0022-3468/© 2014 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg