Esophageal duplication and congenital esophageal stenosis

Author’s Accepted Manuscript Esophageal duplication and congenital esophageal stenosis A. Francois Trappey, Shinjiro Hirose

www.elsevier.com/locate/sempedsurg

PII: DOI: Reference:

S1055-8586(17)30012-4 http://dx.doi.org/10.1053/j.sempedsurg.2017.02.003 YSPSU50667

To appear in: Seminars in Pediatric Surgery Cite this article as: A. Francois Trappey and Shinjiro Hirose, Esophageal duplication and congenital esophageal stenosis, Seminars in Pediatric Surgery, http://dx.doi.org/10.1053/j.sempedsurg.2017.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title: Esophageal Duplication and Congenital Esophageal Stenosis Authors: A. Francois Trappey, III and Shinjiro Hirose

A. Francois Trappey, III, MD Major, United States Air Force, Medical Corps General Surgeon, David Grant Medical Center, Travis Air Force Base, CA Assistant Clinical Professor of Surgery, Trauma and Acute Care Surgery, UC Davis Medical Center Email: [email protected] Phone: 225-933-1095

Shinjiro Hirose, MD – Corresponding Author Associate Professor of Surgery, UC Davis Vice Chair, Department of Surgery Chief, Division of Pediatric General, Thoracic, and Fetal Surgery, UC Davis Medical Center Director of Pediatric Surgery, Shriners Hospitals for Children – Northern California Email: [email protected] Phone: 916-453-2035

General Correspondence: Shriners Hospital for Children 2425 Stockton Blvd. Sacramento, CA 95817 Phone: 916-453-2080 Fax: 916-453-2035

Abstract Esophageal duplication and congenital esophageal stenosis (CES) may represent diseases with common embryologic etiologies - namely, they are examples of faulty tracheoesophageal separation and differentiation. Here we will re-enforce definitions for these diseases as well as review their embryology, diagnosis, and treatment.

Key Words Esophagus, Esophageal duplication, Congenital esophageal stenosis, Tracheobronchial remnants

Introduction Congenital foregut malformations present as a wide variety of surgically important entities. Of these, congenital malformations of the esophagus are particularly relevant to pediatric surgical practice. This article will examine two of these disease processes in detail: esophageal duplication and congenital esophageal stenosis (CES). While often discussed as distinct clinical entities, esophageal duplication and CES may have similar, but as of yet incompletely understood embryologic origins (1). Both are products of abnormal foregut embryogenesis and may represent points along a spectrum of disease that include esophageal atresia with or without tracheoesophageal fistula (EA/TEF) as well as congenital bronchopulmonary malformations (2). There has historically been some confusion regarding the definitions of esophageal duplication and congenital esophageal stenosis in the literature due to the wide range of diagnoses, common origins, and incomplete understanding of embryogenesis.

The term, “esophageal duplication,” as it is commonly used, describes three different morphologic variants of the more inclusive term, “foregut duplication.” The morphologic variants are as follows: 1) cystic, 2) tubular, and 3) diverticular (3). Classically, esophageal duplication had to have the following three characteristics: 1) a well-developed coat of smooth muscle, 2) an epithelial lining representing some portion of the alimentary tract (though some respiratory components may be present), and 3) an attachment to the esophagus (4). We prefer this classical definition but realize that some authors have used the term, “esophageal duplication” to describe any thoracic duplication containing purely gastrointestinal epithelium even if the duplication is remote from the esophagus (4). When containing respiratory epithelium and having close anatomic relation to the trachea or bronchi, foregut duplication cysts are more aptly named, “bronchogenic cysts” (1). “Neurenteric cysts” are another important subset of foregut duplication and represent foregut duplications that have extension into the spinal canal. These should be recognized as a distinct clinical entity given the complexity of their disease associations, diagnosis, and treatment (5-7). We will limit our discussion here to esophageal duplication as defined above. CES as discussed here will represent three distinct lesions: 1) tracheobronchial remnants, 2) fibromuscular stenosis, and 3) membranous webs (8). The stenosis in these cases represents a congenital narrowing due to a mural process rather than from external compression from mass effect, hence their inclusion under this umbrella term though they may prove to have different etiologies in the future. These must also be distinguished from causes of acquired esophageal stenosis, primarily strictures from gastroesophageal reflux.

Embryology From a conceptual standpoint, the esophagus’ function is simple; its only purpose is antegrade propulsion of liquid and masticated food from the mouth to the stomach. This simplicity belies the complexity of foregut development. The separation of the primitive foregut tube into the esophagus and tracheobronchial tree is key to the normal development of the esophagus and respiratory system. Aberrations in normal development can lead to a wide variety of disease states. The mechanisms by which the endodermal derived foregut tube separates into the esophagus and trachea are still being elucidated. No consensus exists regarding not only the morphological changes that occur during this critical period but also regarding the complex array of cell signaling interactions that drive them (1, 9, 10). What is clear is as follows: tracheoesophageal separation occurs between the fourth and fifth week of life (11). Many mechanisms have been proposed regarding this process. The origin of the socalled tracheoesophageal septum continues to be unclear. Historical descriptions of key structures are credited to His in 1887 and Rosenthal in 1931 (12-14). In these classic descriptions, endodermal ridges form in the lateral walls of the primitive foregut, fusing in a caudal to cranial direction like a “zipper” separating the primitive esophagus and trachea (1). Three contemporary models have been proposed, the first two as a challenge to the latter and the third in an attempt to make it more robust (15, 16)(Figure 1). The first contemporary theory, the outgrowth model, directly challenges the appearance of lateral ridges and instead proposes budding and growth of the trachea outward from the common foregut tube, which becomes the esophagus (17, 18). The second contemporary theory, the watershed model, proposes that a mesenchymal wedge exists to prevent growth and extension of the lateral walls of the common foregut tube while the trachea and esophagus/pharynx elongate (19). The third model, the septation model, builds upon the classical description mentioned above, proposing that lateral

edges of the primitive foregut fuse to form the tracheal and esophageal compartments (20). Lack of experimental evidence and correlation with disease states is commonly cited as the reason any one of these models is not favored over the others (15, 16). In the septation model, the lack of evidence by imaging (electron microscopy of chick embryos) of developing lateral folds is at odds with this model’s accuracy. However, Que and Metzger have recently described a saddle-like structure that has been experimentally observed and is supportive of a variation of the septation model. Que has deemed this new model the, “splitting and extension model” (10, 16). The above models describe morphogenic behavior irrespective of the molecular signaling occurring at the point of differentiation. Much progress has been made in the characterization of these signals (Figure 2). It is clear that reciprocal signaling occurs between the endodermal foregut derivative and surrounding structures, notably the mesenchyme and notochord. The notochord’s influence, especially signaling in the Hedgehog pathway (Sonic Hedgehog [Shh] is the most well described) is important to understand as defects in this signaling pathway in experimental models have led to diseases of not only the foregut, but other structures, especially the spine (1, 21-25). In any case, reciprocal signaling is necessary to form a dorsal-ventral pattern of expression that influences the compartmentalization of the primitive foregut (9). Two additional key players in this are Sox2 and Nkx2.1. Sox2 is primarily seen in the dorsal foregut (an esophageal/GI marker) with Nkx2.1 in the ventral foregut (a tracheal marker). Deficiencies in these arrangements have led to EA/TEF in multiple knockout models (9). While no model exists to specifically study foregut duplication or CES, a robust model for induction of the VACTERL association in rats has been described. This model has been used for most of the embryologic studies of foregut disease. Induction of the disease state is accomplished by exposure of fetal rats to Adriamycin (26, 27). The model is reported to be able to induce foregut anomalies in 60%

of exposed animals, and is reported to produce other anomalies such as duodenal atresia, small bowel atresia, imperforate anus, agenesis of the urinary bladder, radial hypoplasia, vertebral anomalies, and esophageal duplication (21, 27, 28). Esophageal duplication is noted to occur in 24% of experimental rats in this model (28). This occurs in conjunction with EA-TEF in the vast majority of Adriamycin exposed embryos though this association is clinically rare in humans (28, 29). In a study by Quan Qi and Beasley, the importance of the notochord in the development of foregut defects, especially EA-TEF and duplication is again demonstrated. They conclude that their experimental results concur with earlier observations regarding both normal relationships between the notochord and foregut and those of early models of disease (30, 31). In normal early embryologic development, the notochord is indistinct from the endoderm destined to become the foregut. Over time, the notochord becomes a separate structure posterior to the foregut. Throughout this time of development, the notochord is influencing directional patterning through signaling, as described above, before becoming the nucleus pulposis of the spine. In the Adriamycin rat model, the notochord is seen in many cases to tether to the primitive foregut leading to the formation of cysts or diverticula, thus, according to many authors, accounting for the frequent association between foregut anomalies and spinal pathology. It is interesting to note that tubular esophageal duplication has not been demonstrated in these experimental models (28). The mechanisms behind CES are even less well characterized than those of esophageal duplication. CES due to tracheobronchial remnant (TBR) is widely thought to be secondary to faulty tracheobronchial separation as well (32, 33). The studies supporting this are older studies postulating that splanchnic mesenchyme is sequestered in the esophageal wall during formation of the lateral ridges, leading to the appearance of tracheal cartilage. The majority of clinical studies regarding CES due to TBR do not attempt an explanation for its etiology. Mucosal webs are thought to be secondary to

failure of the esophagus to fully re-canalize during development, though descriptions of this event in contemporary studies are difficult to find (34). The process of luminal obliteration and recanalization would be akin to processes noted in the duodenum (35). However, direct experimental evidence for the formation of mucosal webs via this mechanism is lacking (28). Luminal obstruction and partial recanalization has also been postulated as a mechanism for tubular esophageal duplication, again stemming from classical studies (36). The fibromuscular variant of CES also has not been observed in an experimental model. One theory proposed by Singaram et al postulates that the fibromuscular variant may represent a disease akin to achalasia of the esophagus or Hirschsprung’s disease. They observed a reduction of myenteric nitrinergic neurons in the surgical specimens of two young adults with CES – resulting in the failure of the diseased segment of esophagus to relax. They implicated prominent neutrophilic infiltrates present in these specimens as a potential cause of myenteric neural destruction, postulating that the fibro muscular variant may be due to an autoimmune process (37). While thought provoking, these results have not been confirmed either in animal models or in human studies. Although no current singular theory can explain the spectrum of disease noted clinically, given the well-differentiated nature of foregut duplications and CES, one can suspect the derangements resulting in these anomalies occur early on in fetal development - at about the same time as tracheal budding. It may be that islands of deranged cell signaling/patterning occur in the primitive foregut that are subsequently influenced to differentiate in parallel to the normal processes. In order to become the well-differentiated lesions we see in esophageal duplication and CES due to TBR, these islands would need to be subjected to normal, local signaling pathways. Notochordal signaling and dorsal/ventral signaling would likely still be involved to influence differentiation along the respiratory or gastrointestinal path. Current models cannot account for this occurrence. Additionally, models of

tracheoesophageal morphogenesis that demonstrate tracheal, not esophageal, separation cannot account for communicating esophageal duplication. The process of separation from the foregut of a respiratory lined cyst can partially be explained by the morphogenic models outlined above, as respiratory elements are noted to normally separate from the primitive foregut. Theories whereby the notochord is abnormally tethered to the foregut, and/or the so-called “split notochord” may explain some of the spinal and vertebral abnormalities seen clinically, especially neurenteric cysts, but cannot explain cases of foregut duplication that occur in isolation (38). More sophisticated modeling will be needed in the future before any definitive conclusions can be made (15).

Clinical Characterization of Esophageal Malformations Esophageal Duplication Epidemiology Gastrointestinal duplication is noted to occur in 1/4500 live births with approximately 20% of these malformations representing esophageal duplications (1, 35, 39-41). This is in contrast with EA/TEF, which occurs in 1/3500 live births, making esophageal duplication extremely rare. As pointed out by Holcomb, III and Keckler, outside of case reports, only one hundred and nineteen patients with esophageal duplication are represented in the literature (41). Twenty-two additional cases of thoracoabdominal duplication are also reported. Despite their rarity, foregut duplications are second only to neuronal tumors as a cause of posterior mediastinal mass in children (35). Esophageal duplication has been known to be associated with EA/TEF in a handful of cases (29), associated with spinal or vertebral anomalies (scoliosis, hemivertebrae, or spina bifida) in up to 20% of cases, associated with spinal or vertebral anomalies (scoliosis, hemivertebrae, or spina bifida) in up to 20% of cases (7,

42), and are associated with abdominal duplications in up to 25% of cases (41). Other rare associations have been noted as well, such as congenital diaphragmatic hernia (36, 43). Esophageal duplications seem to occur in equal frequency amongst male and female patients (35). Three morphologies of esophageal duplication are observed: cystic, tubular, or diverticular. All three morphologies typically have the characteristics described earlier: 1) a well-developed coat of smooth muscle, 2) an epithelial lining representing some portion of the alimentary tract - though some respiratory components may be present as well, and 3) an attachment to the esophagus - often via muscular attachment (4). The cystic variants can be communicating or non-communicating to the esophageal lumen and are by far the most common morphology. Tubular variants are rare, usually communicate with the true esophageal lumen and can share a long portion of the wall of the native esophagus (3). Often, the separation of the tubular duplication is via a thickened fibrous septum. Otherwise, the distinction of cystic vs. tubular duplication seems to be purely based on the shape of the malformation. Diverticular variants are rarely observed (44). Esophageal duplication usually occurs in the lower esophagus and is positioned in the posterior mediastinum with projection into the right or left thorax (35, 45). Esophageal duplication has also been described in the abdominal and cervical esophagus. Tubular duplication can originate in the abdomen or thorax (36, 42). Histologically, in addition to respiratory epithelium, which is commonly encountered, ectopic gastric mucosa is noted in up to 50% of esophageal duplications (46). The vast majority of these lesions are benign though malignancy has been described in specimens resected in adults (47-49). It is unknown whether malignant degeneration is due to properties inherent to the lesions or to the chronic inflammation that they are subjected to (2).

Presentation and Diagnosis Symptoms of esophageal duplication usually appear early in life with most patients presenting before two years of age (46). Presentation of foregut duplication depends on four factors: 1) the anatomic level/location of the lesion, 2) mass effect of the lesion, 3) complications secondary to luminal secretions, and 4) cyst infection (47). The majority of esophageal duplications are found incidentally in asymptomatic patients (4, 35). When symptomatic, the most common presentation is that of respiratory symptoms secondary to airway compression/mass effect (39). Due to this, it has been suggested that duplication should be considered in the differential for all cases of respiratory obstruction in the newborn (39). Other less common presenting symptoms are thought to be due to the presence of ectopic gastric tissue and may include pain - secondary to inflammation or rupture, distention, hemoptysis, and peptic ulceration secondary to atopic gastric mucosa - sometimes leading to gastrointestinal hemorrhage (2, 36, 50, 51). Identification of the duplication cyst is usually made via chest x-ray obtained either secondary to respiratory symptoms or incidentally. Duplication cysts usually appear as a mediastinal mass and, therefore, should be worked up accordingly. Posterior-anterior and lateral films have been shown to detect over 90% of lesions (4). Some have suggested that, in a child with vertebral anomalies, a mediastinal mass seen on chest x-ray is more likely to be a foregut duplication (4). Further characterization of the mass should be accomplished with CT scan of the chest (Figure 3). CT can help to identify anatomic relationships of the cyst and characterize vertebral anomalies, if present. Given the 30% concurrence with abdominal duplication, the abdomen should be screened as well (47). It is unknown whether screening with abdominal ultrasound is sufficient in these cases, though some suggest it may be as gastrointestinal duplications have been identified using this imaging modality (4, 52). Characteristically, the masses are spherical, approximately 2-10cm in size and are often seen to

have a hyperechoic inner mucosal layer and a hypoechoic outer muscular layer with possible internal debris or hemorrhage (52, 53). Peristalsis in a cyst associated with the bowel can be a very specific finding for duplication (54). Importantly, children with vertebral anomalies or suspected neurenteric cyst should undergo MRI to determine the extent, if any, of any spinal involvement prior to surgical therapy (5). Tc-99m pertechnetate scintigraphy (Meckel’s scan) can be used to identify ectopic gastric mucosa in these cysts (43, 51, 55). With modern axial imaging this technique has fallen out of favor. In the past, most patients undergoing Meckel’s scan presented with gastrointestinal bleeding. More recently, the main indication for resection is in an asymptomatic patient as the majority of these lesions can easily be excised thoracoscopically without the need for further workup. One potential use would be in aiding with the decision to operate in asymptomatic patients with small lesions, where the presence of gastric mucosa may make one more inclined to operate. An esophagram with water soluble contrast followed by endoscopy can add value toward surgical planning if the lumen of the duplication fills with contrast. Knowledge of the relationship of a communicating duplication to the esophageal hiatus and GE junction could be important to know preoperatively (56, 57). In some cases of tubular duplication, esophagography can also delineate the length of the common wall, and may help determine the feasibility of possible endoscopic therapy (58, 59). Tubular duplications that are separate from the true esophageal lumen by a mucosal bridge may be amenable to endoscopic management, as a handful of case reports describe the use of an endoscopic knife cautery to convert the duplicated lumens into one (59, 60). Endoscopic ultrasound (EUS) is a modality that is becoming more widely accessible for adults and older children. Its role in esophageal duplication is not well defined. It can be performed simultaneously with endoscopy to clarify anatomy in addition to other pre-operative imaging modalities.

The temptation to perform EUS-guided fine needle aspiration (FNA), however, should be avoided as infection rates as high as 14% have been reported (54). FNA may be indicated if the lesion has an indeterminate appearance, or if there is a concern for malignancy (54). Prenatal diagnosis of foregut duplication has been described (47, 61, 62). The ultrasound appearance of duplication is characteristically described as a thick walled cystic structure. The presence of peristalsis is especially distinguishing (62). MRI has been described as a follow on study of ultrasonographically identified gastric duplication, though no reports of using this modality for a thoracic duplication exist (63). Four out of seven patients in the series by Bratu et al with prenatal ultrasound demonstrating thoracic duplication were symptomatic at birth and prenatal identification was helpful in facilitating surgical management (47). Additionally, thoracic duplication has been described as a cause of fetal nonimmune hydrops secondary to compression of the vena cava, impeding venous return. For these patients, thoracoamniotic shunting has been described with success (4, 64). Therapy and Outcomes In general, surgical therapy is recommended for all esophageal duplications. The reasons for this are multifold, but stem from the reported risk of cyst infection, bleeding, erosion, perforation, and the small risk for malignant degeneration (2, 65). These conclusions are based on sparse data given the rarity of the disease (46, 66). No series are reported whereby asymptomatic patients are followed (67). Wiechowska-Kozlowska et al describe four cases of adults with asymptomatic esophageal duplication characterized by EUS (67). Follow up and outcomes for those patients had not been described at the time of publication with follow up endoscopy and EUS planned at two years (67). Thus, the natural history of asymptomatic esophageal duplications is largely unknown. Often cited is the series of patients reported by St-Georges et al who reported increased surgical complication rates, up to 14%, when surgery is performed in adults and adolescents with bronchogenic

cysts (4, 66). This observation may be secondary to chronic inflammation. It is unknown if these outcomes can be extrapolated to include the behavior of esophageal duplication in adults. In general, we recommend minimally invasive resection of these duplications if feasible. Surgical planning is performed based on the characteristics of the individual duplication. The approach is initially dictated by the level of the lesion – cervical vs. thoracic vs. thoracoabdominal. Cervical duplications are less frequently reported than thoracic duplications, and are best excised via cervical incision (35, 38, 46). The second decision is cyst excision vs. marsupialization. Complete excision remains the standard as cyst fenestration and marsupialization lead to high rates of recurrence (2, 46). Third is open excision vs. minimal access surgery. There has been a trend towards minimal access surgery across surgical disciplines. Thoracoscopy has been demonstrated to be feasible and safe in resection of thoracic foregut duplication (47, 68). Cyst aspiration prior to resection may facilitate thoracoscopic visualization and facilitate specimen removal without the need for mini-thoracotomy (47, 68). If dissection leads to a muscular defect in the wall of the esophagus, we recommend repair followed by air insufflation to ensure luminal integrity (65, 68). In the authors’ opinion, thoracoscopic resection should at least be attempted in the majority of cases. Cysts that are infected at presentation should be treated by staged approach with antibiotics and drainage if needed, followed by resection (47). A staged approach, however, should not be performed for neurenteric cysts. Simultaneous cyst resection with spinal decompression may lead to fewer complications (46). The series by Holcomb, III et al remains the most robust for analyzing outcomes after surgical resection. An extremely high success rate is reported in this series in isolated cases of duplication that are able to be completely excised at a single operation (46). Our experience has been previously reported in the literature (Hirose, Clifton, et al) and is supportive of those observations (68). Though a

small series, all three patients who underwent complete resection of esophageal duplication were reported to be doing well with 3-8 years of follow up (68). Azzie et al postulated that, in many of these patients, associated malformations such as spinal and vertebral anomalies are the elements contributing the most to long term morbidity(4). These additional malformations are likely the source of secondary complications in these patients (4). Thoracoabdominal cysts represent a special clinical challenge (4, 43, 46, 47, 69, 70). The connection to the bowel is located in the abdomen in many of these cases, though connection to the esophagus and gastroesophageal junction has been described, making these a subset of esophageal duplications (70). Vertebral defects may be more common with this entity, though spinal involvement still seems rare (46). Anatomic characterization pre-operatively with axial imaging by CT and MRI is particularly important for these lesions to differentiate these from neurenteric cysts (4, 70). Single or multi-staged approach are both reasonable. If a multi-staged approach is performed, care must be taken to ensure that the remaining portion of the cyst is well drained (4). Separate thoracic and abdominal approaches are required, and open surgery remains the standard with no reports of combined laparoscopy and thoracoscopy thus far in the literature, although that approach seems reasonable. Stringer et al report the largest single experience with thoracoabdominal duplication and report a complication in five of their six patients. Of these, two complications lead to deaths. These deaths were due to bleeding and meningitis. The bleeding was thought to be due to retained ectopic mucosa while meningitis was thought to be secondary to a connection to the central nervous system. This latter case, therefore, likely represents a neurenteric cyst. Attempt at complete resection should be accomplished as incomplete resection led to the majority of the complications reported. Complete excision can be difficult given the delicate nature of some of the cystic tracts (43).

Congenital Esophageal Stenosis Epidemiology The true incidence of CES remains unknown. Reports of its incidence range from 1/2500050000 live births by extrapolating and combining rates reported by single institutions (71). At this time, no data exists detailing the incidence of occult, asymptomatic, or well compensated CES. The incidence seems to be equal amongst males and females (34, 72). The simplest and most widely accepted definition of CES is put forth by Nihoul-Fekete et al: “an intrinsic stenosis of the esophagus present although not necessarily symptomatic at birth, which is caused by congenital malformation of the esophageal wall architecture. The malformation may include: (a) the presence of ectopic tracheobronchial tissue, (b) the presence of a membranous diaphragm, or (c) segmental hypertrophy of the muscularis and submucosal layers with diffuse fibrosis (fibromuscular stenosis) (8, 71, 73).” Since this description in 1987, Ramesh et al. have proposed an alternative categorization scheme (Table 1) (74). They intended this classification to group CES lesions according to etiology and associations. The simplicity of the classical categorization continues to make this the preferred nomenclature. A systematic review was recently published by Terui et al that included a detailing of the frequency of each of the three categories of CES (71). Based off of studies by Nihoul-Fekete et al (1987), Takamizawa et al (2002), and Michaud et al (2013) who provided detailed clinical information in their retrospective case series, Terui et al calculated the frequency of fibromuscular hypertrophy (FMS), tracheobronchial remnant (TBR), and mucosal webs (MW) at 53.8%, 29.9%, and 16.2%, respectively (8, 71, 75). MW occur mainly in the upper or middle third of the esophagus, FMS in the middle or lower third, and TBR mainly in the lower third (within 1cm of the gastroesophageal junction) (71, 76). Major associations (up to 33% in some series) include EA/TEF (24.8%), esophageal duplication, congenital heart

disease, trisomy 21, anorectal malformation, duodenal atresia, tracheomalacia, and esophageal hiatal hernia (34, 73, 77). The similarity in the spectrum of associated disease between esophageal duplication, CES, and EA/TEF are, potentially, another sign of their similar developmental origin. Presentation and Diagnosis The most commonly reported symptoms of CES are secondary to esophageal obstruction, namely vomiting and dysphagia. Other symptoms include excess salivation, respiratory distress, regurgitation, recurrent aspiration pneumonia, impacted foreign body, or failure to thrive (73). One common presentation is initiation of symptoms with the introduction of solid foods within the first year of life, usually between 4-10 months of age (73, 77), though many delayed presentations have been reported in adolescents and adults (37, 71, 78). CES associated with EA/TEF is usually diagnosed earlier compared to patients with isolated CES although diagnosis in asymptomatic patients at routine follow up esophagram after EA/TEF repair is common (72, 76). The diagnosis of CES can be difficult to differentiate from other entities. The goal is to rule out acquired forms of stenosis/obstruction. Acquired etiologies of esophageal stenosis include peptic stricture, caustic injury, infection, neoplasia, extrinsic compression, and achalasia as well as other motility disorders (8, 71). Contrast esophagography is the initial diagnostic maneuver along with anterior/posterior and lateral chest x-ray. Esophagram can either demonstrate an abrupt or tapered stenosis (Figure 4) (34). Generally, stenosis owing to TBR are more abrupt than those of FMS or MW (79). pH monitoring to rule out peptic stricture plus endoscopy with biopsies to rule out esophagitis is help narrow the differential diagnosis. As TBRs are not a mucosal-based lesion, endoscopic mucosal biopsy is inadequate to exclude TBR as the cause for stenosis (8, 73). To exclude achalasia, manometry should be performed. Manometry may show a pathologic focal high pressure zone at the location of the stricture in addition

to the normal high pressure zone at the lower esophageal sphincter in 75% of patients (34). This abnormal high pressure zone has been noted to resolve with successful surgery or dilation (34). Once acquired stenosis has been ruled out, endoscopic ultrasound should be performed to look for TBR (71, 75, 76, 80-83). Takamizawa et al used EUS in 5 cases of CES to distinguish TBR from FMS with ultrasound findings corresponding with pathology in all 5 cases (75). In this series, cartilaginous structures were visualized as hypoechoic structures. Usui et al reported two cases where EUS was used to distinguish FMS from TBR, and stated that, while the inside of a homogenous cartilage layer should be hypoechoic, the interface with other tissues may be hyperechoic. Differences in echogenic characteristics between studies may be caused by differences in the thickness of the ectopic cartilaginous structures and its interaction with surrounding tissue (82). Therapy and Outcomes Options for treatment of congenital esophageal stenosis include dilation or surgery. The importance of distinguishing TBR vs. MW or FMH pre-operatively is still of some debate. However, most authors suggest that patients identified as having CES owing to TBR forego dilation in favor of surgical management (8, 34, 75, 79, 82, 84, 85). The theory behind this is that dilation against the fixed cartilaginous structures will not only be ineffective but may also lead to a higher rate of perforation in this subgroup. One of the largest cohorts, however, is at odds with this recommendation (76). Romeo et al attempted dilation in all 47 patients in their cohort and reported a 95.7% success rate regardless of the subtype. Two patients in this series required surgery. TBR was identified in one of these two cases with both preoperative EUS and postoperative histology. This is out of a total of 6 patients identified as having TBR in their study. Therefore, only 1 of 6 patients with identified TBR after EUS was available required surgery. They recommend that all patients undergo attempted dilation prior to surgery. Mean number of dilations in this series was 3 (range 1-9) with 15 days in between sessions (mean follow

up 9.6 years [range 1-29 years]). Others report generally low success rates in patients with the TBR subtype, but do not completely characterize patients who are successfully dilated owing to lack of availability of EUS and assumption of subtype based on esophagram (8, 72, 75). They characterize CES patients who are successfully dilated but who are negative for MW at endoscopy as having FMS (8, 72). Therefore, there may exist a subpopulation of TBR patients who are able to successfully undergo nonoperative management. Esophageal dilation is accomplished with bougienage or by balloon dilation. The two techniques have not been compared prospectively, but proponents of balloon dilation cite the shear caused by bougienage as a source for potential increased perforation using this technique (71). Balloon dilation may be safer and more effective given radial pressures exhibited in a more focused manner (75). Techniques for each are still being refined with maximal balloon sizes and pressures still a matter of debate (71). Romeo et al report a 10.6% post-dilation perforation rate (5/47 patients) in CES patients. It is unknown how many of these had CES owing to TBR since 4 of 5 patients underwent non-operative management of the perforation. The histologic subtype of the patient who required operation was not described. Three perforations occurred after balloon dilation and two after dilation with bougienage in this series. Michaud et al used dilation as first line treatment in 49 of their 58 patients with only two perforations (3.4%) reported (72). One was after balloon dilation and one after bougienage. Both patients were, however, TBR subtype. For the MW type, some are adding endoscopic web ablation with CO2 laser or cautery to dilation (75). It is unknown if this method increases the risk of perforation over that of dilation alone. We feel that pre-operative EUS is important for identifying the TBR subtype and can be easily accomplished at the same time as endoscopy. Given the appeal of non-operative management, and the success reported by Romeo et al, we feel that it is reasonable to offer patients with TBR a trial of

dilation. In the author’s experience dilation of these patients is often unsuccessful, therefore surgeons should not be hesitant to offer operative management in the TBR subtype. An interesting area for future study may be to correlate the burden of cartilaginous disease seen on EUS with the success of dilation in these patients. It is difficult with the available evidence to recommend balloon dilation over bougienage and leave this decision to the operative surgeon based upon available resources and expertise. Segmental esophagectomy with end to end anastomosis has been and continues to be the standard surgical treatment for CES of all types. Fundoplication has been added in some cases and provides theoretical anastomotic protection and can provide reconstruction of the LES if the gastroesophageal junction must be sacrificed, or ends up in the thorax (34). Additionally, gastroesophageal reflux has been cited as a source of postoperative morbidity of segmental esophagectomy, and recommend fundoplication as added protection. The gastroesophageal reflux in these cases may be an intrinsic complication of the disease itself, especially in the case of TBR and FMH (86). In some cases, TBR has been successfully enucleated after palpation of a localized segment of ectopic cartilaginous tissue corresponding with the level of stenosis (75). Circular myomectomy with suture repair of the muscular layers has been performed successfully leaving the mucosal layer intact, and theoretically decreasing the risk of leak (84, 87). As an alternative to resection, the FMH subtype may be effectively treated with longitudinal myotomy (87). MW subtypes refractory to endoscopic management must be treated with resection as well. Though, in some cases, the mucosal web can be excised after longitudinal esophagotomoy without segmental resection (45). The esophagus in the vast majority of cases was approached through the left chest, as most lesions are found in the distal esophagus. This has been traditionally through an open thoracotomy. We have had of success with thoracoscopic resection and anastomosis and feel that this approach aids in

visualization for vagal preservation in cases where segmental esophagectomy is necessary (83). Similarly, lesions of the distal esophagus have been treated with a transabdominal/transhiatal approach. Longitudinal myotomy for the FMH subtype can be accomplished similarly to a laparoscopic Heller myotomy for achalasia. This approach allows fundoplication, gastric feeding access, and hiatal repair if necessary (76). Again, no direct comparisons of open versus minimal access surgery exists for this disease. Given the proven benefit of minimal access surgery in other disease states in experienced hands, it is reasonable to approach CES similarly. Surgical complications reported include but are not limited to anastomotic leakage (treated operatively and nonoperatively), anastomotic stenosis, hiatal hernia, and reflux esophagitis (79). Malnutrition may contribute to anastomotic leakage, as seen by some authors, with 35% of patients in one series being classified as malnourished preoperatively (8, 72). Michaud et al published the largest series and provide the most detailed follow up data available. They demonstrated persistent symptoms in 66% of patients who underwent primary surgical management (5/9) and in 73% of patients who underwent surgery after attempted dilation (11/15). EUS was not performed in this study, limiting the ability to characterize the responders to dilation by subtype, or to allow some pre-procedural selection. They attribute persistent symptoms despite anatomic abnormality to esophageal dysmotility. Dysmotility has been demonstrated in this patient population, but post-operative manometric data is unavailable in the literature at this time, and may ultimately provide to be valuable in determining long term outcomes in CES patients (86). Commentary Esophageal duplication and CES are rare diseases. Given this, impediments to their study are numerous. Consistent use of the definitions outlined here are encouraged. Embryologic modeling of normal/disease states must be able account for CES and duplications. In general, treatment of isolated

duplication is very successful, and with the availability of thoracoscopy, the majority can be easily resected. The question of what to do with small, asymptomatic cysts remains unanswered, but we think that following them with interval imaging is reasonable. Minimal access surgery should be attempted for both esophageal duplication and CES at capable centers. In CES, we recommend the use of EUS to identify patients with TBR. The vast majority of these patients will need surgical resection vs. patients with FMS or MW, where dilation and endoscopic therapy are feasible at centers with these capabilities.

References 1. El-Gohary Y, Gittes GK, Tovar JA. Congenital anomalies of the esophagus. Semin Pediatr Surg. 2010;19(3):186-93. 2. Nobuhara KK, Gorski YC, La Quaglia MP, Shamberger RC. Bronchogenic cysts and esophageal duplications: common origins and treatment. J Pediatr Surg. 1997;32(10):1408-13. 3. Saha AK, Kundu AK. Tubular duplication of the oesophagus presenting with dysphagia. Singapore Medical Journal. 2014;55(6). 4. Azzie G, Beasley S. Diagnosis and treatment of foregut duplications. Semin Pediatr Surg. 2003;12(1):46-54. 5. Fernandes ET, Custer MD, Burton EM, Boulden TF, Wrenn EL, Jr., Whittle AP, et al. Neurenteric cyst: surgery and diagnostic imaging. J Pediatr Surg. 1991;26(1):108-10. 6. Alrabeeah A, Gillis DA, Giacomantonio M, Lau H. Neurenteric cysts—A spectrum. Journal of Pediatric Surgery. 1988;23(8):752-4. 7. Superina RA, Ein SH, Humphreys RP. Cystic duplications of the esophagus and neurenteric cysts. J Pediatr Surg. 1984;19(5):527-30. 8. Nihoul-Fekete C, DeBacker A, Lortat-Jacob S, al. e. Congenital esophageal stenosis: a review of 20 cases. Pediatric Surgery International. 1987;2:86-92. 9. Jacobs J, Ku W, Que J. Genetic and cellular mechanisms regulating anterior foregut and esophageal development. Developmental Biology. 2012;369(1):54-64. 10. Metzger R, Wachowiak R, Kluth D. Embryology of the early foregut. Seminars in Pediatric Surgery. 2011;20:136-44. 11. Moore K, Persaud T. The Developing Human. 8 ed. Philadelphia, PA: Elsevier; 2008. 12. Kluth D, Seidl W. The embryology of foregut malformations. Journal of pediatric surgery. 1987;22(5):389-93. 13. Rosenthal A. Congenital atresia of the esophagus with tracheoesophageal fistula: Report of eight cases. Archives of Pathology. 1931;12:756-22. 14. His W. Zur bildungsgeschischte der lungen beim menschlischen embryo. Archives of anatomy. 1887:89-106. 15. Billmyre KK, Hutson M, Klingensmith J. One shall become two: Separation of the esophagus and trachea from the common foregut tube. Dev Dyn. 2015;244(3):277-88.

16. Que J. The initial establishment and epithelial morphogenesis of the esophagus: a new model of tracheal-esophageal separation and transition of simple columnar into stratified squamous epithelium in the developing esophagus. Wiley Interdiscip Rev Dev Biol. 2015;4(4):419-30. 17. O'Rahilly R, Muller F. Respiratory and alimentary relations in staged human embryos. New embryological data and congenital anomalies. Annals of otology rhinology and laryngology. 1984;93:421-9. 18. Zaw-Tun H. The tracheo-esophageal septum - fact or fantasy? Origin and development of the respiratory primordium and esophagus. Acta Anatomica. 1982;114:1-21. 19. Sasaki T, Kusafuka T, A O. Analysis of the development of normal foregut and tracheoesophageal fustula in an adriamycin rat model using three-dimensional image reconstruction. Surgery Today. 2001;31:133-9. 20. Qi B, Beasley S. Stages of normal tracheo-bronchial development in rat embryos: resolution of a controversy. Developmental Growth and DIfferentiation. 2000;42:145-53. 21. Merei J, Hasthorpe S, Farmer P, al. e. Relationship between esophageal atresia with tracheoesophageal fistula and vertebral anomalies in mammalian embryos. Journal of pediatric surgery. 1998;33:58-63. 22. Mortell A, O’Donnell AM, Giles S, Bannigan J, Puri P. Adriamycin induces notochord hypertrophy with conservation of sonic hedgehog expression in abnormal ectopic notochord in the adriamycin rat model. Journal of Pediatric Surgery. 2004;39(6):859-63. 23. Possoegel A, Diez-Pardo J, Morales C, al. e. Notochord involvement in experimental esophageal atresia. Pediatric Surgery International. 1999;15:201-5. 24. Vleesch Dubois V, Quan Qi B, Beasley S, al. e. Abnormal branching and regression of the notochord and its relationship to foregut abnormalities. European Journal of Pediatric Surgery. 2002;12:83-99. 25. Qi B, Beasley S. Relationship of the notochord to foregut development in the fetal rat model of esophageal atresia. Journal of pediatric surgery. 1999;34:1593-8. 26. Diez-Pardo J, Baoquan Q, Navarro C, al. e. A new rodent experimental model of esophageal atresia and tracheoesophageal fistula: preliminary report. Journal of pediatric surgery. 1996;31:498-502. 27. Merei J, Kotsios C, Hutson J, al. e. Histopathological study of esophageal atresia and tracheoesophageal fistula in an animal model. Journal of pediatric surgery. 1997;32(1):12-4. 28. Quan Qi B, Beasley S, Williams A. Evidence of a common pathogenesis for foregut duplications and esophageal atresia with tracheoesophageal fistula. The Anatomical Record. 2001;264:93-100. 29. Knod JL, Garrison AP, Frischer JS, Dickie B. Foregut duplication cyst associated with esophageal atresia and tracheoesophageal fistula: a case report and literature review. J Pediatr Surg. 2013;48(5):E57. 30. Jurand A. Some aspects of the development of the notochord in mouse embryos. Journal of embryology and experimental morphology. 1974;32(1):1-33. 31. Veeneklaas GM. Pathogenesis of intrathoracic gastrogenic cysts. AMA American journal of diseases of children. 1952;83(4):500-7. 32. Kumar R. A case of congenital oesophageal stricture due to a cartilaginous ring. The British journal of surgery. 1962;49:533-4. 33. Nishina T, Tsuchida Y, Saito S. Congenital esophageal stenosis due to tracheobronchial remnants and its associated anomalies. J Pediatr Surg. 1981;16(2):190-3. 34. Amae S, Nio M, Kamiyama T, Ishii T, Yoshida S, Hayashi Y, et al. Clinical characteristics and management of congenital esophageal stenosis: a report on 14 cases. J Pediatr Surg. 2003;38(4):565-70. 35. Bissler JJ, Klein RL. Alimentary tract duplications in children: case and literature review. Clinical pediatrics. 1988;27(3):152-7. 36. Gross R, Holcomb G, Farber S. Duplications of the alimentary tract. Pediatrics. 1952;9:449-67.

37. Singaram C, Sweet MA, Gaumnitz EA, Cameron AJ, Camilleri M. Peptidergic and nitrinergic denervation in congenital esophageal stenosis. Gastroenterology. 1995;109(1):275-81. 38. Huang Y, Wang D, Liu X, Wang W, Zhang Z. Communicating esophageal tubular duplication in a newborn infant. J Pediatr Surg. 2011;46(8):1655-7. 39. Bower RJ, Sieber WK, Kiesewetter WB. Alimentary tract duplications in children. Annals of surgery. 1978;188(5):669-74. 40. Schalamon J, Schleef J, Hollwarth ME. Experience with gastro-intestinal duplications in childhood. Langenbeck's archives of surgery / Deutsche Gesellschaft fur Chirurgie. 2000;385(6):402-5. 41. Holcomb G, Keckler S. Alimentary Tract Duplications. In: Holcomb G, Murphy J, Ostlie D, editors. Ashcraft's Pediatric Surgery. 6th ed. Philadelphia, PA: Elsevier; 2014. p. 539-47. 42. Carachi R, Azmy A. Foregut duplications. Pediatr Surg Int. 2002;18(5-6):371-4. 43. Stringer MD, Spitz L, Abel R, Kiely E, Drake DP, Agrawal M, et al. Management of alimentary tract duplication in children. The British journal of surgery. 1995;82(1):74-8. 44. Ohbatake M, Muraji T, Yamazato M, Higashimoto Y, Nishijima E, Tsugawa C. Congenital true diverticula of the esophagus: a case report. J Pediatr Surg. 1997;32(11):1592-4. 45. Snyder CL, Bickler SW, Gittes GK, Ramachandran V, Ashcraft KW. Esophageal duplication cyst with esophageal web and tracheoesophageal fistula. J Pediatr Surg. 1996;31(7):968-9. 46. Holcomb GI, Gheissari A, O'Neill J, al. e. Surgical management of alimentary tract duplications. Annals of surgery. 1989;209:167-74. 47. Bratu I, Laberge JM, Flageole H, Bouchard S. Foregut duplications: is there an advantage to thoracoscopic resection? J Pediatr Surg. 2005;40(1):138-41. 48. Lee MY, Jensen E, Kwak S, Larson RA. Metastatic adenocarcinoma arising in a congenital foregut cyst of the esophagus: a case report with review of the literature. American journal of clinical oncology. 1998;21(1):64-6. 49. Olsen JB, Clemmensen O, Andersen K. Adenocarcinoma arising in a foregut cyst of the mediastinum. The Annals of thoracic surgery. 1991;51(3):497-9. 50. Neo EL, Watson DI, Bessell JR. Acute ruptured esophageal duplication cyst. Diseases of the esophagus : official journal of the International Society for Diseases of the Esophagus / ISDE. 2004;17(1):109-11. 51. Macpherson RI. Gastrointestinal tract duplications: clinical, pathologic, etiologic, and radiologic considerations. Radiographics : a review publication of the Radiological Society of North America, Inc. 1993;13(5):1063-80. 52. Barr LL, Hayden CK, Jr., Stansberry SD, Swischuk LE. Enteric duplication cysts in children: are their ultrasonographic wall characteristics diagnostic? Pediatric radiology. 1990;20(5):326-8. 53. Wootton-Gorges SL, Thomas KB, Harned RK, Wu SR, Stein-Wexler R, Strain JD. Giant cystic abdominal masses in children. Pediatric radiology. 2005;35(12):1277-88. 54. Liu R, Adler DG. Duplication cysts: Diagnosis, management, and the role of endoscopic ultrasound. Endoscopic ultrasound. 2014;3(3):152-60. 55. Whitaker JA, Deffenbaugh LD, Cooke AR. Esophageal duplication cyst. Case report. The American journal of gastroenterology. 1980;73(4):329-32. 56. Kim JH, Kwon CI, Rho JY, Han SW, Kim JS, Shin SP, et al. Communicating Tubular Esophageal Duplication Combined with Bronchoesophageal Fistula. Clinical endoscopy. 2016;49(1):81-5. 57. Barabino A, Nardi F, Arrigo S, Gandullia P, Vignola S, Muraca M, et al. Tubular esophageal duplication: further evidence of a possible endoscopic treatment. J Pediatr Gastroenterol Nutr. 2014;58(6):e53. 58. Garge S, Samujh R. Isolated complete tubular esophageal duplication in a neonate. Diseases of the esophagus : official journal of the International Society for Diseases of the Esophagus / ISDE. 2013;26(3):342.

59. Coumaros D, Schneider A, Tsesmeli N, Geiss S, Becmeur F. Endoscopic management of a tubular esophageal duplication diagnosed in adolescence (with videos). Gastrointestinal endoscopy. 2010;71(4):827-30. 60. Tahri N, Mnif L, Chtourou L, Boudabbous M, Yaich K, Fourati H, et al. Complete endoscopic management of tubular esophageal duplication in a young woman. Endoscopy. 2012;44 Suppl 2 UCTN:E261-2. 61. Jacquemard F, Palaric JC, Cousin C, d'Herve D, Conan H, Giraud JR. [Prenatal echographic study of esophageal duplication. A case report]. Journal de gynecologie, obstetrique et biologie de la reproduction. 1991;20(7):963-7. 62. Richards DS, Langham MR, Anderson CD. The prenatal sonographic appearance of enteric duplication cysts. Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology. 1996;7(1):17-20. 63. Granata C, Dell'Acqua A, Lituania M, Oddone M, Rossi U, Toma P. Gastric duplication cyst: appearance on prenatal US and MRI. Pediatric radiology. 2003;33(2):148-9. 64. Martinez Ferro M, Milner R, Voto L, Zapaterio J, Cannizzaro C, Rodriguez S, et al. Intrathoracic alimentary tract duplication cysts treated in utero by thoracoamniotic shunting. Fetal diagnosis and therapy. 1998;13(6):343-7. 65. Benedict LA, Bairdain S, Paulus JK, Jackson C-C, Chen C, Kelleher C. Esophageal duplication cysts and closure of the muscle layer. Journal of Surgical Research. 2016;206(1):231-4. 66. St-Georges R, Deslauriers J, Duranceau A, Vaillancourt R, Deschamps C, Beauchamp G, et al. Clinical spectrum of bronchogenic cysts of the mediastinum and lung in the adult. The Annals of thoracic surgery. 1991;52(1):6-13. 67. Wiechowska-Kozlowska A, Wunsch E, Majewski M, Milkiewicz P. Esophageal duplication cysts: endosonographic findings in asymptomatic patients. World J Gastroenterol. 2012;18(11):1270-2. 68. Hirose S, Clifton MS, Bratton B, Harrison MR, Farmer DL, Nobuhara KK, et al. Thoracoscopic resection of foregut duplication cysts. Journal of laparoendoscopic & advanced surgical techniques Part A. 2006;16(5):526-9. 69. Pokorny WJ, Goldstein IR. Enteric thoracoabdominal duplications in children. The Journal of thoracic and cardiovascular surgery. 1984;87(6):821-4. 70. Mazziotti MV, Ternberg JL. Continuous communicating esophageal and gastric duplication. J Pediatr Surg. 1997;32(5):775-8. 71. Terui K, Saito T, Mitsunaga T, Nakata M, Yoshida H. Endoscopic management for congenital esophageal stenosis: A systematic review. World journal of gastrointestinal endoscopy. 2015;7(3):18391. 72. Michaud L, Coutenier F, Podevin G, Bonnard A, Becmeur F, Khen-Dunlop N, et al. Characteristics and management of congenital esophageal stenosis: findings from a multicenter study. Orphanet journal of rare diseases. 2013;8:186. 73. Murphy S, Yazbeck S, Russo P. Isolated congenital esophageal stenosis. Journal of pediatric surgery. 1995;30(8):1238-41. 74. Ramesh JC, Ramanujam TM, Jayaram G. Congenital esophageal stenosis: report of three cases, literature review, and a proposed classification. Pediatr Surg Int. 2001;17(2-3):188-92. 75. Takamizawa S, Tsugawa C, Mouri N, Satoh S, Kanegawa K, Nishijima E, et al. Congenital esophageal stenosis: Therapeutic strategy based on etiology. J Pediatr Surg. 2002;37(2):197-201. 76. Romeo E, Foschia F, de Angelis P, Caldaro T, Federici di Abriola G, Gambitta R, et al. Endoscopic management of congenital esophageal stenosis. J Pediatr Surg. 2011;46(5):838-41. 77. Achildi O, Grewal H. Congenital anomalies of the esophagus. Otolaryngol Clin North Am. 2007;40(1):219-44, viii.

78. McNally PR, Collier EH, 3rd, Lopiano MC, Brewer TG, Wong RK. Congenital esophageal stenosis. A rare cause of food impaction in the adult. Digestive diseases and sciences. 1990;35(2):263-6. 79. Zhao LL, Hsieh WS, Hsu WM. Congenital esophageal stenosis owing to ectopic tracheobronchial remnants. J Pediatr Surg. 2004;39(8):1183-7. 80. Kouchi K, Yoshida H, Matsunaga T, Ohtsuka Y, Nagatake E, Satoh Y, et al. Endosonographic evaluation in two children with esophageal stenosis. J Pediatr Surg. 2002;37(6):934-6. 81. Bocus P, Realdon S, Eloubeidi MA, Diamantis G, Betalli P, Gamba P, et al. High-frequency miniprobes and 3-dimensional EUS for preoperative evaluation of the etiology of congenital esophageal stenosis in children (with video). Gastrointestinal endoscopy. 2011;74(1):204-7. 82. Usui N, Kamata S, Kawahara H, Sawai T, Nakajima K, Soh H, et al. Usefulness of endoscopic ultrasonography in the diagnosis of congenital esophageal stenosis. J Pediatr Surg. 2002;37(12):1744-6. 83. Quiros JA, Hirose S, Patino M, Lee H. Esophageal tracheobronchial remnant, endoscopic ultrasound diagnosis, and surgical management. J Pediatr Gastroenterol Nutr. 2013;56(3):e14. 84. Saito T, Ise K, Kawahara Y, Yamashita M, Shimizu H, Suzuki H, et al. Congenital esophageal stenosis because of tracheobronchial remnant and treated by circular myectomy: a case report. J Pediatr Surg. 2008;43(3):583-5. 85. Vasudevan SA, Kerendi F, Lee H, Ricketts RR. Management of congenital esophageal stenosis. J Pediatr Surg. 2002;37(7):1024-6. 86. Kawahara H, Oue T, Okuyama H, Kubota A, Okada A. Esophageal motor function in congenital esophageal stenosis. J Pediatr Surg. 2003;38(12):1716-9. 87. Maeda K, Hisamatsu C, Hasegawa T, Tanaka H, Okita Y. Circular myectomy for the treatment of congenital esophageal stenosis owing to tracheobronchial remnant. J Pediatr Surg. 2004;39(12):1765-8.

Table 1. Ramesh et. al Classification – Congenital Esophageal Stenosis (CES) Type I: Isolated CES – Segmental Type

a. Tracheobronchial Remnants (TBR) b. Fibromuscular Stenosis (FMS)

Type II: Isolated CES – Diaphragm Type

a. Membranous web (MW)

Type III: Combined Lesions

a. Segmental stenosis occurring distal to EA/TEF b. Segmental stenosis occurring distal to a MW

Figure 1. Models of tracheal-esophageal separation: (1) The outgrowth model – Trachea extends from the common foregut tube as the lung buds grow, while common foregut tube becomes the esophagus. (2) The watershed model – both the trachea and esophagus elongate while separated by a mesenchymal septum that serves as a wedge to prevent the extension of the lateral wall at the dorsal-ventral midline. The empty arrowhead represents the hypothetical mesenchymal condensation which has yet to be identified. (3) The septation model – epithelial cells at the dorsal-ventral midline make contact across the lumen and fuse to form a septum. (4) Splitting and extension – proposes that the separation of the trachea and esophagus initiates at the level where the lung grows out and moves rostrally. A saddle-like structure (red arc) splits the anterior foregut. Abbreviations: Ve, ventral; Do, dorsal; Pr, proximal (rostral); Di, distal (caudal); CF, common foregut; Lu, lung; St, stomach; Es, esophagus; Tr, trachea. With permission from Que J. Wiley Interdiscip Rev Dev Biol. 2015;4(4):419-30.

Figure 2. Dorsal-ventral pattern of signaling during tracheal-esophageal separation. With permission from Que J. Wiley Interdiscip Rev Dev Biol. 2015;4(4):419-30.

Figure 3. Axial CT of the chest. The esophageal duplication cyst is labeled in the posterior mediastinum. Note the proximity to the vertebral body. RL, right lung; LL, left lung. With permission from Hirose S. 2006;16(5):526-9.

Figure 4. Upper gastrointestinal contrast study demonstrates a dilated esophagus proximal to a distal esophageal stenosis. After ruling out other pathologies, congenital esophageal stenosis owing to tracheobronchial remnant was confirmed on endoscopic ultrasound. With permission from Quiros JA. J Pediatr Gastroenterol Nutr. 2013;56(3):e14.

Figure 5. Endoscopic ultrasound demonstrating hypoechoic structure in the esophageal wall. This correlates with the cartilaginous structure demonstrated in the final pathology below. With permission from Quiros JA. J Pediatr Gastroenterol Nutr. 2013;56(3):e14.