Ultrasound Obstet Gynecol 2003; 21: 494497 Published online 4 April 2003 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.

58

Prenatal diagnosis of esophageal atresia with the pouch sign

G. CENTINI, L. ROSIGNOLI, A. KENANIDIS and F. PETRAGLIA
Prenatal Diagnosis Centre, Department of Pediatrics, Obstetrics, and Reproductive Medicine, University of Siena, Siena, Italy

K E Y W O R D S: esophageal atresia; prenatal diagnosis

ABSTRACT A 19-year-old primipara was referred to our center for a routine scan at 33 + 4 weeks gestation. A visible but small stomach bubble (18 mm) was detected in the fetal abdomen, associated with a dilation of the esophagus with a tapering distal part. Direct visualization of lling and emptying of the proximal esophagus suggested the presence of obstruction. A cystic pouch in the region of the esophagus was observed to be full and empty in accordance with fetal swallowing. No other fetal anomalies were detected. Fetal biometric measurements were on the 10th percentile and polyhydramnios was detected. Esophageal atresia type I was suspected. Fetal karyotyping was declined by the parents, no therapy was given and the pregnancy continued until week 37 when a female baby was delivered following spontaneous labor. Esophageal atresia type I was radiographically conrmed. Bougienage was used for 8 weeks and esophageal anastomosis was successfully performed when the infant was 11 months old. No other anomalies were found and the child is currently in excellent health. The present case shows that the upper neck pouch sign may be a delayed manifestation present only in certain types of esophageal atresia. However, when it occurs it should prompt careful fetal examination as it is an important step in the diagnosis of esophageal atresia. Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.

INTRODUCTION
Esophageal atresia is characterized by the absence of a segment of the esophagus and in most cases it is associated with a stula between the gastrointestinal and the respiratory tracts. Normal division of the foregut into trachea and esophagus occurs in the fourth week of gestation. The esophagus and trachea develop from a common diverticulum of the primitive pharyngeal cavity. The diverticulum subsequently partitioned by the tracheoesophageal septum forms the laryngotracheal

tube and the esophagus. The development of the upper respiratory and gastrointestinal tracts takes place between day 21 and week 5 of gestation. In view of their common embryonic origin, it is not surprising that anomalies of the trachea and esophagus are often associated1 . Any interruption of embryonic septation results in an esophageal atresia and tracheoesophageal stula (TEF). Generally these abnormalities occur together and only occasionally separately: the most common form is esophageal atresia with distal TEF (87%). Isolated esophageal atresia and isolated TEF occur in 8% and 5% of cases, respectively. The overall incidence is 1 in 3000 births2 . The etiology of this condition is unknown; there is no demonstrable genetic predisposition, and occurrence seems to be sporadic3 . There is no established sex preponderance. Chromosomal, gastrointestinal, and urogenital anomalies are found in 58% of patients and congenital heart disease in 1539%4 7 . The prognosis of esophageal atresia depends on four factors: associated congenital malformations, respiratory complications, birth weight, and gestational age at delivery. The prenatal detection of esophageal atresia is possible and is based on the ultrasonographic ndings of a small or absent fetal stomach bubble associated with polyhydramnios8,9 . The overabundance of amniotic uid (AF) is related to decreased turnover as a consequence of the esophageal obstruction and occurs in 32% of infants with EA10,11 . Visualization of a uid-lled pouch corresponding to the atretic segment of the esophagus has been proposed as the most reliable sign of esophageal atresia. The use of the pouch sign can increase the positive predictive value of ultrasonography, and recently has been differentiated into two forms: the mediastinal pouch (when the pouch is seen below the clavicle) and the neck pouch (when in the cervical area). The exact location of the pouch seems to be important in providing the prognosis of esophageal atresia, as the presence of a mediastinal pouch has been found to have a better prognosis, lower incidence of associated anomalies,

Correspondence to: Dr G. Centini, Chair of Obstetrics and Gynecology, University of Siena, Policlinico Le Scotte, Viale Bracci 1, 53100 Siena, Italy (e-mail: centini@unisi.it) Accepted: 18 December 2002

Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.

CASE REPORT

Prenatal diagnosis of esophageal atresia and require only a one-stage operation12,13 . Prenatal diagnosis enables parents to be prepared for the birth and treatment of their affected child, permits prompt neonatal management, thereby avoiding potentially hazardous delays in diagnosis, and leads to earlier identication of associated anomalies. Early diagnosis is also important to avoid complications of aspiration and chemical pneumonia. Early management requires gastrostomy and continuous suction of the blind upper pouch. Surgical repair is normally undertaken when the infants condition becomes stable14,15 .

495

CASE REPORT
A 19-year-old primipara was referred to our center for a routine scan at 33 + 4 weeks gestation. Ultrasound examination was performed using an ATL 3000 HDI (Philips Medical Systems, Eindhoven, The Netherlands) with a 3.5-MHz transabdominal probe. A visible, small stomach bubble measuring 18 mm was detected in the fetal abdomen (Figure 1). Visualization of the esophageal region showed dilatation of the esophagus with a tapering distal part (Figure 2). Direct visualization of lling and emptying of the proximal esophagus suggested an obstruction. A cystic pouch in the region of the esophagus was seen to be full and empty in accordance with fetal swallowing. No other fetal anomalies were detected and fetal biometric measurements were on the 10th percentile. AF volume was increased (amniotic uid index (AFI) = 316), and polyhydramnios was diagnosed. Esophageal atresia type I was suspected (Ladds classication16 ). The patient refused cordocentesis for determination of fetal karyotype. Second-trimester biochemical screening for chromosomal anomalies performed at week 16 had been negative. No therapy was given and the pregnancy continued until week 37 when a female infant (2200 g,

Figure 2 Ultrasound examination at 34 weeks gestation. Sagittal view shows the dilated esophagus with a tapering distal part (pouch sign).

Apgar scores 5 and 7 at 1 and 5 min, respectively) was delivered following spontaneous onset of labor. After birth type I atresia (isolated esophageal atresia without stula) was conrmed radiographically. Bougienage was used for 8 weeks, and esophageal anastomosis was successfully carried out when the infant was 11 months old. No cardiovascular, urogenital, or skeletal anomalies were found. The child is now 2 years old and in excellent health.

DISCUSSION
The present case of esophageal atresia had the following characteristics: it was (1) not associated with TEF, (2) was not associated with other malformations, and (3) polyhydramnios was detected. Also in this case the prenatal diagnosis of esophageal atresia was suggested by ultrasonographic evidence of a small stomach bubble and polyhydramnios17 19 . However, these criteria are not denitive, the maximum probability of a correct diagnosis being as low as 42%20 . This is related to the following facts: (1) esophageal atresia in most (90%) cases is accompanied by distal TEF, so that AF occasionally reaches the stomach; (2) in the rare cases without TEF (10%), visualization of the stomach may be possible, owing to the presence of gastric secretion; and (3) polyhydramnios rarely develops before 24 weeks of gestation. Moreover, when the diagnosis of esophageal atresia is based on polyhydramnios and non-detection of the fetal stomach, conrmation is not possible until birth. The differential diagnosis of esophageal atresia includes other conditions that adversely affect fetal swallowing such as nervous system anomalies, facial anomalies, and herniated diaphragm21 . The visualization of the proximally dilated esophagus has been proposed as a direct sign of esophageal atresia,

Figure 1 Ultrasound examination at 34 weeks gestation. Transverse view demonstrates the presence of a small stomach.

Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.

Ultrasound Obstet Gynecol 2003; 21: 494497.

496

Centini et al. TEF 7% of the fetuses will have aneuploidy; and in cases of pure esophageal atresia 2025%, the most likely one being trisomy 18. Esophageal atresia is often associated with other malformations: gastrointestinal (21%), cardiovascular (20%), urogenital (10%), and skeletal (20%). Cardiac and vertebral anomalies may be seen as part of the VATER association28 30 . Karyotyping should be considered, particularly in cases in which associated anomalies are detected. Thorough examination of the fetal neck region in different planes may be an important diagnostic step in all cases of prenatally suspected esophageal atresia. Prenatal diagnosis enables parents to be prepared for the birth and treatment of their affected child, permits prompt neonatal management, thereby avoiding potentially hazardous delays in diagnosis, and leads to earlier identication of associated anomalies.

but few cases have been documented in the literature. Eyheremendy and Pster22 were the rst to identify the upper neck pouch sign. A retrospective study of ten babies with esophageal atresia showed that transient upper neck pouch was observed in all cases. Although onset has been observed as early as 23 weeks by Shulman et al.13 and 26 weeks by Kalache et al.23 , several reports describe later visualization of the pouch sign, which seems to be a late sign. It has been hypothesized that in the rst and early mid-trimesters of pregnancy, the fetus is unable to develop sufcient pressure in swallowing to dilate a blind esophagus. Moreover, the fetus may not swallow during the period of ultrasound examination, necessitating a long examination period (at least 2030 min) performed at repeated short time intervals, as well as multiple ultrasonographic sections for more thorough evaluation of the neck and chest regions. The type of malformations, the size, as well as the location, of the stula could also affect the time of the appearance, since a malformation with a large stula will empty more rapidly than one with a narrow stula, compromising the presence of the pouch sign. Pretorius et al.24 in a retrospective review reported that only 7/22 infants born with TEF had both polyhydramnios and no visible stomach. In a further seven pregnancies, polyhydramnios and a fetal stomach were both detected sonographically. None of these women developed polyhydramnios before week 24 and 12/22 infants had other anomalies. In the rare case of esophageal atresia with proximal TEF, AF probably enters the trachea without dilating the esophagus, and the upper esophagus remains collapsed and is not detected by ultrasound. In the latter half of gestation, the fetus excretes a large volume of urine into the AF and swallows a large volume of AF each day25 . Hence polyhydramnios could be expected to develop if swallowing does not occur but not always develops with esophageal atresia or obstruction. In monkeys, surgical ligation of the fetal esophagus is followed by transient polyhydramnios with AF volume returning to normal over a period of several days26 ; absorption of water and solutes is doubled by esophageal ligation. What mediates this large increase in the ability of the intramembranous pathway to absorb water and solutes is still unclear. In fetal sheep, the ligation or obstruction of the esophagus is followed by increased AF volume after 13 days, and normal AF volume at 5 days and 23 weeks27 . Ovine fetal urinary output remains normal during esophageal obstruction/occlusion. The combined volume of uid from the fetal kidneys and lungs would be approximately 1000 mL/day, which has to be absorbed by the amniotic compartment if the AF volume is to remain within normal limits when swallowing does not occur. We are aware that the upper neck pouch sign may be a delayed manifestation, only occurring in certain types of esophageal atresia. However, when it is observed, it is of fundamental importance as it conrms the diagnosis and is a signal for a careful search for associated anomalies. In the presence of esophageal atresia combined with

REFERENCES
1. Farrant P. The antenatal diagnosis of esophageal atresia by ultrasound. Br J Radiol 1980; 53: 12021203. 2. Beasley SW. Oesophageal atresia without stula. In Oesophageal Atresia, Beasley S, Myers N, Auldist A (eds). London: Chapman & Hall Medical, 1991; Chapter 10, 137159. 3. Depaepe A, Dolk H, Lechat M. The epidemiology of tracheooesophageal stula and oesophageal atresia in Europe. Arch Dis Child 1993; 68: 743748. 4. Nicolaides K, Snijders R, Cheng H. Fetal gastro-intestinal and abdominal wall defects: associated malformations and chromosomal abnormalities. Fetal Diagn Ther 1992; 7: 102115. 5. Damato N, Filly R, Goldstein R. Frequency of fetal anomalies in sonographically detected polyhydramnios. J Ultrasound Med 1993; 12: 1115. 6. Havard A, MacDonald L. Oesophageal atresia and other disorders with a similar antenatal presentation. Br J Radiol 1991; 64: 557558. 7. Ein S, Shandling B, Wesson D. Esophageal atresia with distal tracheoesophageal stula: associated anomalies and prognosis in the 1980s. J Pediatr Surg 1989; 24: 10551059. 8. Pretorius D, Gosink B, Clautice-Engle T. Sonographic evaluation of the fetal stomach: signicance of non visualization. AJR Am J Roentgenol 1988; 151: 987989. 9. Millener P, Anderson N, Chisholm R. Prognostic signicance of nonvisualization of the fetal stomach by sonography. AJR Am J Roentgenol 1993; 160: 827830. 10. Chamberlain P, Manning F, Morrison I, Harman C, Lange I. Ultrasound evaluation of amniotic uid volume. The relationship of increased amniotic uid volume to perinatal outcome. Am J Obstet Gynecol 1984; 150: 250254. 11. Hill L, Breckle R, Thomas M, Fries J. Polyhydramnios: ultrasonographically detected prevalence and neonatal outcome. Obstet Gynecol 1987; 69: 21. 12. Kalache KD, Wauer R, Mau H, Chaoui R, Bollman R. Prognostic signicance of the pouch sign in fetuses with prenatally diagnosed esophageal atresia. Am J Obstet Gynecol 2000; 184: 978981. 13. Shulman A, Mazkereth R, Zalel Y, Kuint J, Lipitz S, Avigad I, Achiron R. Prenatal identication of esophageal atresia: the role of ultrasonography for evaluation of functional anatomy. Prenat Diagn 2002; 22: 669674. 14. Louhimo I, Lindahl H. Esophageal atresia: primary results of 500 consecutively treated patients. J Pediatr Surg 1983; 18: 217229. 15. Jassani M, Gauderer M, Faranoff A. A perinatal approach to the diagnosis and management of gastrointestinal malformations. Obstet Gynecol 1982; 59: 33.

Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.

Ultrasound Obstet Gynecol 2003; 21: 494497.

Prenatal diagnosis of esophageal atresia

16. Ladd WE, Wenson O. Esophageal atresia and tracheoesophageal stula. Ann Surg 1947; 125: 2324. 17. Zemlyn S. Prenatal detection of esophageal atresia. J Clin Ultrasound 1981; 9: 453454. 18. Rahmani M, Zalev A. Antenatal detection of esophageal atresia with distal tracheoesophageal stula. J Clin Ultrasound 1986; 14: 143145. 19. Satoh S. Antenatal sonographic detection of the proximal esophageal segment: specic evidence for congenital esophageal atresia. J Clin Ultrasound 1995; 23: 419423. 20. Robertson F, Crombleholme T, Paidas M, Harris B. Prenatal diagnosis and management of gastrointestinal disorders. Semin Perinatol 1994; 18: 182. 21. Bowie J, Clair M. Fetal swallowing and regurgitation: observation of normal and abnormal activity. Radiology 1982; 144: 877878. 22. Eyheremendy E, Pster M. Antenatal real-time diagnosis of esophageal atresias. J Clin Ultrasound 1983; 11: 395397. 23. Kalache K, Chaoui R, Mau H, Bollmann R. The upper neck pouch sign: a prenatal sonographic marker for esophageal

497
atresia.Ultrasound Obstet Gynecol 1998; 11: 138140. 24. Pretorius D, Meier P, Johnson M. Diagnosis of esophageal atresia in utero. J Ultrasound Med 1983; 2: 475476. 25. Kirshon B. Fetal urine output in hydramnios. Obstet Gynecol 1989; 73: 240242. 26. Minei L, Suzuki K. Role of fetal deglutition and micturition in the production and turnover of amniotic uid in the monkey. Obstet Gynecol 1976; 48: 177181. 27. Fujino Y, Agnew C, Schreyer P, Ervin M, Sherman D, Ross M. Amniotic uid volume response to esophageal occlusion in fetal sheep. Am J Obstet Gynecol 1991; 165: 16201626. 28. Kinoshita M, Nakamura Y, Nakano R. Thirty-one autopsy cases of trisomy 18: clinical features and pathological ndings. Pediatr Pathol 1989; 9: 445457. 29. Claiborne A, Blocker S. Prenatal and postnatal sonographic delineation of gastrointestinal abnormalities and the case of the VATER syndrome. J Ultrasound Med 1986; 5: 45. 30. Torfs CP, Curry CJ, Bateson TF. Population-based study of tracheoesophageal stula and esophageal atresia. Teratology 1995; 52: 220232.

Copyright 2003 ISUOG. Published by John Wiley & Sons, Ltd.

Ultrasound Obstet Gynecol 2003; 21: 494497.