Tracheoesophageal anomalies probably arise as a result of events that occur around the 4th week of gestation. The trachea and esophagus first develop as a ventral diverticulum off of the foregut at 22 to 23 days of gestation (Skandalakis et al., 1994). This diverticulum elongates and there is an influx of endodermal cells that form ridges of tissue, which divide the foregut into esophageal and tracheal lumens beginning at the carina and progressing cephalad. By the 26th day of gestation the esophagus and trachea have become completely separated up to the level of the larynx. Interruption in the in growth of ectodermal ridges is thought to result in tracheoesophageal fistula. However, the cause of esophageal atresia when associated with tracheoesophageal fistula is less well understood. One theory suggests that rapid caudal elongation of the trachea, in the presence of a distal tracheoesophageal fistula, produces fixation of the distal esophagus to trachea (Smith, 1957; Moore and Pessaud, 1993). The dorsal wall of the esophagus is drawn forward and downward to be incorporated with the trachea, and esophageal atresia results. Consequently, the esophageal atresia associated with tracheoesophageal fistula occurs as a direct result of the presence of the fistula. In contrast, isolated, or pure, esophageal atresia is thought to arise as a result of a vascular deficiency.

More recently, mutations in the hedgehog (Hh) signaling pathway have been implicated in the development of tracheoesophageal fistula (Lees et al., 2005). Studies in transgenic mouse models examining the effects of homozygous sonic hedgehog (Shh) knockout mutations demonstrate esophageal atresia and tracheoesophageal fistula (Litingtung et al., 1998; Ramalho-Santos et al., 2000). The foregut abnormalities become evident as early as embryonic day 9.5 in the mouse when the tracheal diverticulum is developing. The role of hedgehog signaling in esophageal atresia and tracheoesophageal fistula is further supported by Gli 2–/– and Gli 3+/– double knock out mice. These mutations are downstream from sonic hedgehog and also result in esophageal atresia and tracheoesophageal fistula (Lees et al., 2005). Studies performed on biopsies of proximal and distal esophagus in infants undergoing repair of esophageal atresia (EA) and tracheoesophageal fistula (TEF) demonstrated that Shh expression was present in the proximal esophagus, but absent in the distal esophagus by both immunohistochemistry and RT-PCR (Spilde et al., 2003).

The various forms of esophageal atresia constitute one of the most common gastrointestinal anomalies, occurring in 1 in 3000 livebirths (Holder et al., 1964). The embryologic events at approximately 28 days of gestation that result in esophageal atresia can result in a spectrum of anomalies (Figure 40-1). Isolated esophageal atresia occurs in only 1 in 15,000 livebirths (Figure 40-1A), but esophageal atresia is most commonly seen in association with distal esophageal fistula (Figure 40-1C), in 86% of the observed cases (Holder et al., 1964). Approximately one half of patients with esophageal atresia have anomalies of other organs. Cardiac malformations are the most common, occurring in approximately 25% of cases. They are responsible for most of the mortality and morbidity associated with esophageal atresia (Landing, 1975; Greenwood and Rosenthal, 1976). Atrial and ventricular septal defects are the most common cardiac abnormalities. Genitourinary, additional gastrointestinal, anorectal, and musculoskeletal anomalies occur in approximately 10% of the cases (David and O’Collaghan, 1975; Manning et al., 1986; Spitz et al., 1993) (Table 40-1). These anomalies tend to cluster in groups as part of the “VACTERL” association—vertebral, anorectal, cardiac, tracheal, esophageal, renal, and limb anomalies.

Maternal exposure to the teratogen methimazole is associated with esophageal and tracheal defects (Shaw-Smith, 2006).

It is important to recognize that the diagnosis of esophageal atresia is inferred from the presence of polyhydramnios and the absence of a fetal stomach. This combination of findings has a positive predictive value ranging from 30% to 70% for esophageal atresia (Pretorius et al., 1987; Stringer et al., 1995; Crombleholme et al., 1996). Observation of the dilated proximal pouch is a far more specific finding for esophageal atresia, but this is difficult to see in most fetuses with esophageal atresia (Figure 40-2).

Sonographic observation of fetal breathing movements, swallowing, and general muscular activity should be assessed whenever the stomach bubble is not observed, in order to exclude a neuromuscular cause. Failure to observe the fetal stomach may also occur in normal fetuses because of variations in swallowing and gastric emptying. Serial ultrasound examinations are useful to distinguish this cause of absent fetal stomach. However, visualization of a normal-appearing stomach does not exclude the diagnosis of esophageal atresia. Even in isolated esophageal atresia, gastric secretions alone may be sufficient to distend the stomach and make it visible on prenatal sonography. In esophageal atresia associated with a distal tracheoesophageal fistula, amniotic fluid may be inhaled and passed into the stomach. When the size of the tracheoesophageal fistula is small and limits flow of fetal lung fluid, polyhydramnios develops (Bovicelli et al., 1983).

Prenatal diagnosis of esophageal atresia is suspected when polyhydramnios is seen in association with an absent stomach bubble (Figure 40-3) (Farrant, 1980; Zemlyn, 1981; Bovicelli et al., 1983). Considerations in the differential diagnosis of these findings include congenital diaphragmatic hernia, situs inversus, and musculoskeletal or neurologic anomalies that result in fetal inability to swallow. Some authors argue that prenatal diagnosis of esophageal atresia is possible only in isolated esophageal atresia, which accounts for only 10% to 15% of esophageal anomalies. Rarely, a dilated proximal esophageal pouch may be visualized, which is diagnostic of esophageal atresia (Estroff et al., 1994) (Figure 40-4). A dilated proximal esophageal pouch ending at the point of atresia with regurgitation after swallowing has been observed on prenatal ultrasound examination in cases of pure esophageal atresia (Bowie and Clair, 1982; Eyheremendy and Pfister, 1983). Absence of the stomach on ultrasound examination is commonly seen in isolated esophageal atresia, but may also be observed in esophageal atresia with tracheoesophageal fistula. Prenatal sonographic absence of the fetal stomach may also be seen in neuromuscular disorders due to absence of normal fetal swallowing. Severe central nervous system disorders can diminish fetal swallowing and breathing, resulting in an absent stomach bubble and polyhydramnios (Bowie and Clair, 1982).