Distal Congenital
Esophageal Esophageal
Stenosis Atresia
Associated
With
By Ian R. Neilson, Dan P. Croitoru, Frank M. Guttman, Sami Youssef, and Jean-Martin Montreal, l A series of six patients with congenital esophageal stenosis associated with esophageal atresia (EA) and distal tracheoesophageal fistula is presented. Three patients required only repeated dilatations, and have had good results. Two patients required limited resections of the distal esophagus, with excellent results. One patient died following a Heller myotomy. Tracheobronchial rests were present in the distal esophagus in the latter three patients. Diagnosis of congenital distal esophageal stenosis following repair of EA requires a high index of suspicion and a careful review of previous esophagrams. It is important to exclude anastomotic stricture and stenosis associated with gastroesophageal reflux. This requires barium esophagram, esophagoscopy with biopsy, and esophageal pH monitoring. Once a congenital basis for distal esophageal stenosis is suspected, management consists of dilatation by bouginage followed by balloon dilatation. Resection is reserved for persistent stenoses from tracheobronchial rests, which usually do not respond to dilatations. Copyright o 1991 by W.B. Saunders Company INDEX WORDS: stenosis.
Esophageal
atresia; congenital
esophageal
E
SOPHAGEAL atresia (EA) may be complicated by postoperative stenosis, most frequently at the site of anastomosis or in the distal esophagus secondary to gastroesophageal reflux (GER). However, stenosis can also occur in the distal esophagus from a congenital esophageal stenosis (CES). The first case of CES associated with an EA was reported from the Montreal Children’s Hospital by Dunbar in 1958.’ Forty other cases of CES associated with EA and/or tracheoesophageal fistula (TEF) have previously been reported.lm8 We report a further 6 cases of CES following repair of an EA with distal TEF. The investigation and management of CES in this setting is discussed. MATERIALS AND METHODS Patients with EA presenting from January 1962 to December 1989 were included in this review. Records of patients with
From The Montreal Children’s Hospital, McGill University, Montreal, Quebec. Presented at the 21si Annual Meeiing of the American Pediahic Surgical Association, Vancouver, British Columbia, May 19-22, 1990. Address reprint requests to F.M. Guttman, MD, Director, Pediatric General Surgery, Montreal Children’s Hospital, 2300 Tupper, Montreal, Quebec H3H lP3, Canada. Copyright o 1991 by U?B. Saunders Company 0022-3468/9l/2604-0021$03.00l0
478
Laberge
Quebec esophageal stenosis were evaluated to determine if the stenosis was at the anastomotic site, associated with GER, or whether it could be a distal CES. The management of the latter cases was then reviewed in detail. Case 1 An infant boy underwent an end-to-end repair of EA with division of a distal TEF. A contrast esophagram on the 7th day of life showed reflux to the upper third of the esophagus and in retrospect, a narrowing of the distal third of the esophagus. The esophagram was repeated 1 week later following a suspected right lower lobe aspiration pneumonia. There was persistent reflux and a stricture was noted 2 cm distal to the anastomosis (Fig 1A). He was treated with an antireflux regimen, but was lost to follow-up. He presented again at 15 months of age with a foreign body at the level of the 7th to 8th thoracic vertebra. After removal of a nickel, an esophagram showed a stricture in the lower third of the esophagus. This was dilated and he was next seen at 19 months of age for suspected recurrent pulmonary aspiration. The esophagram remained unchanged (Fig 1B). Esophagoscopy showed a stenosis 6 cm distal to the anastomosis. A biopsy proximal to the stenosis was consistent with esophagitis. Dilatation was performed with a size 34F dilator. He required a second dilatation at 20 months of age using a 15-mm hydrostatic balloon inflated to 1 atmosphere pressure for 1 minute (Fig 1C). Balloon dilatation was repeated two further times. An esophageal pH study did not demonstrate GER. Because of failure to respond to dilatations, a left thoracotomy and a limited 2-cm resection of the distal esophageal stenosis was undertaken at 22 months of age. The stricture was difficult to palpate intraoperatively and was identified with the assistance of a flexible endoscope. Histological examination of the resected specimen showed submucosal mucus glands of bronchial origin, mixed salivary glands, a ruptured cyst lined with respiratory epithelium, and several pieces of cartilage (Fig 1D). Postoperatively there was a contained leakage of contrast, which was managed nonoperatively with resolution. After 9 months he is asymptomatic and eating a normal diet. Case
2
On the first day of life, a boy underwent an uncomplicated end-to-end repair of an EA and division of a distal TEF. The initial postoperative esophagram was normal apart from a slight narrowing in the distal esophagus and minimal GER. At 7 weeks, he developed cough and apneic episodes after feedings. An esophagram showed a distal esophageal stenosis. This was dilated, but he was lost to general surgical follow-up and over the first 3 years of life he required endoscopic removal of 6 foreign bodies and 12 further dilatations. The maximum size of dilatation achieved was 36F at 2 years of age, and then there was a progressive decrease in size to 28F. At 3 years of age, he underwent a distal 2-cm resection of the stenosis with the finding of cartilage and respiratory epithelium in a disrupted sinus tract within the esophageal wall (Fig 2). The patient is now 7 years old and is asymptomic with a normal esophagram.
JournalofPediatric
Surgery, Vol26,
No 4 (April),
1991: pp 478-482
DISTAL CONGENITAL
ESOPHAGEAL
STENOSIS
Fig 1. (A) Esophagram (case 1) at 2 weeks of age showing distal esophageal narrowing, dilated proximal esophagus, and patent anastomosis. (8) Esophagram (case 1) at 19 months of age after bouginage showing progression of stenosis. (C) Hydrostatic balloon dilatation (case 1) at 20 months of age using 15-mm baHoon at 1 atmosphere under fluoroscopic control. Note “waist” on the balloon outlining the stenosis. (D) Transverse section of esophageal wall (case 1) showing tracheobronchial rest with nodule of cartilage (arrow) surrounded by glands (hematoxylin phloxin, safrin; original magnification x30).
Case 3 An infant girl underwent primary end-to-end repair of EA and division of a distal TEF. An anastomotic leak was noted on the first esophagram done at 10 days of age. This was managed conservatively. A follow-up esophagram at 4 weeks of age showed a midesophageal stenosis. She did well until solids were introduced, and at 8 months an esophagram showed a stricture at the junction
of the middle and distal thirds of the esophagus. Dilatations were carried out to a size 30F dilator. Her esophagram was normal at 2 and 9 years of age. She is now asymptomatic at 11 years of age.
Case 4 A boy underwent repair of EA and division the first day of life. The initial esophagram
of a distal TEF on was normal and
NEILSON ET AL
Fig 2. Transverse section of esophageal wall (case 2) showing tracheobronchial rest with nodule of cartilage (arrow) surrounded by glands and a sinus tract (asterisk) lined with ciliated epithelium (hematoxylin, phloxin, safrin; original magnification x25).
symptoms did not begin until 6 months of age, when an esophagram confirmed a distal stenosis. This has been dilated 19 times over 5 years and he has required esophagoscopy for removal of two foreign bodies in the past 2 years. He is now 6 years old and tolerates a normal diet.
Case 5 A 36-week gestation girl underwent repair of an EA with distal TEF at birth. One week later, a leak was demonstrated by an esophagram and managed nonoperatively. At 2 months of age a distal esophageal stenosis was demonstrated by esophagram. This has been dilated 6 times. The patient is now 6 years old with minimal symptoms except for occasional cough after eating.
Case 6 In 1963, this full-term baby underwent primary anastomosis of an EA with division of a distal TEF on the first day of life. A transanastomotic 5F gastric feeding tube was passed to the stomach with some difficulty. At 5 days of age he was reexplored for an enlarging air-fluid mass in the posterior mediastinum. The distal esophagus was found distended from 2 cm above the diaphragm to the level of the anastomosis. The esophagus was opened and dilators up to 8F were passed to the stomach and a gastrostomy was performed. Two weeks later an esophagram showed two small anastomotic leaks and a distal esophageal stenosis. From 2 to 5 months of age he underwent retrograde dilatations from 12F to 20F. Due to failure to respond to dilatations and severe failure to thrive, a Heller myotomy was performed at 6 months of age for presumed achalasia. The patient died 10 days later of respiratory failure. At autopsy, nodules of cartilage were demonstrated within the esophagus at the level of the myotomy. DISCUSSION
Spitz, 2 in 1973, was the first to demonstrate a congenital basis for distal esophageal stenosis associated with EA by showing the presence of tracheobronchial rests in an esophagectomy specimen. A recent
review of 20 cases of CES by Nihoul-FCketC et al3 includes only 2 patients with associated EA or TEF. These authors define CES as an intrinsic stenosis present, although not necessarily symptomatic, at birth, which is caused by a congenital malformation of the esophageal wall. They classify the etiology as: (1) tracheobronchial rest (cartilage, respiratory mucous glands, and/or ciliated epithelium); (2) a membranous diaphragm; or (3) segmental hypertrophy of the muscularis and diffuse fibrosis of submucosa (fibromuscular stenosis). They stress that it is essential to exclude postnatally acquired stenoses (peptic, caustic, infectious, neoplastic), extrinsic compression, and achalasia. The association of CES with EA and the more distal location of CES caused by tracheoesophageal rests can be explained on an embryological basis. At the time of the separation of the respiratory tract from the foregut around the 25th day of gestation, tracheobronchial tissue is sequestered in the esophageal wall and then carried caudally by the normal growth of the esophagus.4 In series where the type of esophageal anomaly has been stated,‘.’ CES has been most frequently associated with EA with distal TEF (64%), followed by isolated TEF (20%) and isolated EA (16%). All the cases in the present series had EA with distal TEF, corresponding to the expected incidence of various esophageal anomalies. Histological descriptions of CES associated with EAD’EF were available in only 10 of the 40 previously reported patients. Of these, the majority (7/10) had tracheobronchial rests2,4,6and 3 had fibromuscular stenoses.3,7 Three cases in the present series had tracheobronchial rests in the distal esophageal wall, all with cartilage, two with ciliated epithelial-lined cyst or sinus, and one with bronchial glands. Tracheobronchial rests may be overrepresented in the literature because most require resection, whereas fibromuscular stenoses and membranous diaphragms usually respond to dilatation alone and, hence, are not examined histologically. The diagnosis of CES associated with EA begins with a high index of suspicion. It is important to verify patency of the distal esophagus at the time of primary anastomosis by passage of a tube to the stomach intraoperatively. Symptoms may present in the early postoperative period with leakage at the anastomosis exacerbated by distal obstruction as seen in 4 of the cases in this series (cases 3, 4, 5, and 6). Early symptoms also include aspiration manifest by recurrent pneumonia (cases 1 and 5) or apnea (case 2). Fluoroscopy of the initial postoperative esophagram must be carefully reviewed and repeated looking
DISTAL CONGENITAL
ESOPHAGEAL
481
STENOSIS
specifically for areas of persistent narrowing at the junction of the middle and distal thirds of the esophagus. The initial postoperative esophagram showed a CES in three of the patients in this series (cases 1, 2, and 6), but the correct diagnosis was not made. The majority of cases reported in the literature have presented at 4 to 6 months of age following introduction of solid foods, which results in dysphagia, regurgitation, and cough. Presentation with distal esophageal foreign bodies is also common, as occurred in 3 of the present patients (cases 1,2, and 4). It must be remembered that there are other causes of esophageal stenosis in a patient who has had a repair of EA and/or TEF. The site of the stenosis often suggests the etiology. Stenoses can be located either at the anastomosis or in the distal esophagus. Distal stenoses can be associated either with GER or CES. Therefore, investigation should include a feeding history, a review of the initial and a repeat contrast tine-esophagram, esophagoscopy with biopsy, and esophageal pH monitoring. Even with these investigations it may be difficult to distinguish CES from strictures due to GER. After these investigations, we recommend a single bouginage that can be combined with initial esophagoscopy. Of the two
common types of CES, those caused by fibromuscular hyperplasia often respond to single or intermittent dilation. If the esophagram shows persistence of the stenosis, a series of hydrostatic balloon dilatations to 1 atmosphere pressure should be undertaken. Hydrostatic dilatation is safer than bouginage because disrupting forces are transmitted axially rather than longitudinally and the progress or extent of the dilatation can be followed under fluoroscopic control.’ If there is failure of improvement after 2 to 3 serial hydrostatic dilatations, and if symptoms are present, then operative treatment is recommended. It is possible that some of the presumed fibromuscular cases are actually due to tracheobronchial rests, which may respond to nonoperative therapy. However, we would assume that failure of dilatation implies the presence of cartilage in the CES. The presence of cartilage is also suggested by the narrow “waist” seen on the hydrostatic balloon when it is inflated with dilute contrast solution (Fig 1C). Because the tracheobronchial rests are limited to a short l- to 2-cm length, a limited resection rather than a more extensive resection with esophageal substitution is all that is usually required.
REFERENCES 1. Dunbar JS: Congenital oesophageal North Am 5:433-455,1958
stenosis. Pediatr Clin
2. Spitz L: Congenital esophageal stenosis distal to associated esophageal atresia. J Pediatr Surg 8:973-974, 1973 3. Nihoul-FCkCtC C, De Backer A, Lortat-Jacob S, et al: Congenital esophageal stenosis: A review of 20 cases. Pediatr Surg Int 2:86-92,1987 4. Deiraniya AK: Congenital oesophageal stenosis due tracheobronchial remnants. Thorax 29:720-725,1974 5. Thomason MA, Gay BB: Esophageal stenosis with esophageal atresia. Pediatr Radio1 17:197-201,1987
6. Nishina R, Tsuchida Y, Saitos S: Congenital esophageal stenosis due to tracheobronchial remnants and its associated anomalies. J Pediatr Surg 16:190-193, 1981 7. Tuqan NA: Annular stricture of the esophagus distal to congenital tracheoesophageal fistula. Surgery 52:394-395, 1962 8. Mahour GH, Johnston PW, Gwinn JL: Congenital esophageal stenosis distal to esophageal atresia. Surgery 69:936-939, 1971 9. Hoffer FA, Winter HS, Fellows KE, et al: The treatment of postoperative and peptic esophageal strictures after esophageal atresia repair: A program including dilatation with balloon catheters. Pediatr Radio1 17:454-458,1987
Discussion R Sonnino (Cleveland, OH): I think it is a problem that we are not really looking for as much as we should and, as you have shown, it can cause a whole set of problems when it goes unrecognized. I do have a couple of questions for you though, relating both to the manuscript and the presentation. In particular, you are really including all 6 of the babies in the CES group. But as to the 3 who didn’t undergo resection, or as far as I can tell biopsy, I guess I am questioning how sure you are that these indeed had congenital stenosis rather than reflux or some other problem. Another question, although you mentioned it briefly, can you speculate in any way what the actual inci-
dence of congenital stenosis is both in patients with an atresia and in normal babies? In other words, how many babies might there be running around that actually do have this type of stenosis? It seems to me from the embryology you presented that there is no reason why it wouldn’t occur fairly frequently in babies without an atresia as well. Another question that I would like to ask you is this. Obviously, all these children were diagnosed after their atresia was repaired. Can you think of some way that we can look for it while we are fixing the atresia and not discover later on when the baby leaks or gets a distended esophagus or any of the complications you have
482
discussed? And finally, do you think that esophagoscopy with just a punch biopsy, as you would probably do as a first stage in diagnosis, is going to be adequate, or are your chances of missing these tracheal bronchial risks going to be too high? Is that going to be a good way to monitor for these? M.E. Matlak (Salt Lake City, UT): To answer Dr Soninno’s question, this can be diagnosed in the operating room if you would cannulate the lower esophageal segment with an 8F or 10F red rubber catheter before you repair the esophagus; you can feel the stenosis and then have a high index of suspicion of checking it postoperatively with your esophagram. We have picked up several of these simply having that be part of our surgical armamentarium. The other thing I would emphasize is that you should be very conservative in doing your dilatation with these youngsters, because I think a high incidence of the risk problem or the fibromuscular hyperplasia and leads toward a resection rather than the dilatation because these youngsters are prone to having their esophaguses perforate. Why did your one patient die with a Heller myotomy? Was it a perforation or mediastinitis? I.R. Neilson (response): Dr Sonnino, it is difficult to tell if these are reflux strictures or not when they
NEILSON ET AL
appear late. The early patients did not have pH studies; their clinical history suggested that it wasn’t a reflux stricture. In reviewing the initial esophagram at 1 week, you can often see distal narrowing and not much reflux. Even if reflux is present, it is difficult to imagine a stricture at 1 week. Also, the stenoses were clearly due to anatomical malformations. There are roughly 400 case reports of the isolated congenital esophageal stenoses. The largest series was by NihoulFCkCtCwith 20 patients. In that series there were 2 or 3 with associated esophageal atresia so the isolated stenosis is much more common than the association. I agree that these should probably be looked for in the operating room. We routinely pass a 10F catheter into the stomach after the posterior wall has been joined. However, this could not rule out a stenosis. I wouldn’t treat them at the time of surgery. If there is a high-grade stenosis, this may cause a postoperative leak. As to why the patient died after the Heller myotomy, this was in 1963, before TPN and advanced neonatal care. This patient weighed 3.8 kg at 6 months of age and he was a full-term baby so this really reflects a very severe degree of failure to thrive and he died of purely respiratory complications, ie, bronchopneumonia.
Esophageal Esophageal
Stenosis Atresia
Associated
With
By Ian R. Neilson, Dan P. Croitoru, Frank M. Guttman, Sami Youssef, and Jean-Martin Montreal, l A series of six patients with congenital esophageal stenosis associated with esophageal atresia (EA) and distal tracheoesophageal fistula is presented. Three patients required only repeated dilatations, and have had good results. Two patients required limited resections of the distal esophagus, with excellent results. One patient died following a Heller myotomy. Tracheobronchial rests were present in the distal esophagus in the latter three patients. Diagnosis of congenital distal esophageal stenosis following repair of EA requires a high index of suspicion and a careful review of previous esophagrams. It is important to exclude anastomotic stricture and stenosis associated with gastroesophageal reflux. This requires barium esophagram, esophagoscopy with biopsy, and esophageal pH monitoring. Once a congenital basis for distal esophageal stenosis is suspected, management consists of dilatation by bouginage followed by balloon dilatation. Resection is reserved for persistent stenoses from tracheobronchial rests, which usually do not respond to dilatations. Copyright o 1991 by W.B. Saunders Company INDEX WORDS: stenosis.
Esophageal
atresia; congenital
esophageal
E
SOPHAGEAL atresia (EA) may be complicated by postoperative stenosis, most frequently at the site of anastomosis or in the distal esophagus secondary to gastroesophageal reflux (GER). However, stenosis can also occur in the distal esophagus from a congenital esophageal stenosis (CES). The first case of CES associated with an EA was reported from the Montreal Children’s Hospital by Dunbar in 1958.’ Forty other cases of CES associated with EA and/or tracheoesophageal fistula (TEF) have previously been reported.lm8 We report a further 6 cases of CES following repair of an EA with distal TEF. The investigation and management of CES in this setting is discussed. MATERIALS AND METHODS Patients with EA presenting from January 1962 to December 1989 were included in this review. Records of patients with
From The Montreal Children’s Hospital, McGill University, Montreal, Quebec. Presented at the 21si Annual Meeiing of the American Pediahic Surgical Association, Vancouver, British Columbia, May 19-22, 1990. Address reprint requests to F.M. Guttman, MD, Director, Pediatric General Surgery, Montreal Children’s Hospital, 2300 Tupper, Montreal, Quebec H3H lP3, Canada. Copyright o 1991 by U?B. Saunders Company 0022-3468/9l/2604-0021$03.00l0
478
Laberge
Quebec esophageal stenosis were evaluated to determine if the stenosis was at the anastomotic site, associated with GER, or whether it could be a distal CES. The management of the latter cases was then reviewed in detail. Case 1 An infant boy underwent an end-to-end repair of EA with division of a distal TEF. A contrast esophagram on the 7th day of life showed reflux to the upper third of the esophagus and in retrospect, a narrowing of the distal third of the esophagus. The esophagram was repeated 1 week later following a suspected right lower lobe aspiration pneumonia. There was persistent reflux and a stricture was noted 2 cm distal to the anastomosis (Fig 1A). He was treated with an antireflux regimen, but was lost to follow-up. He presented again at 15 months of age with a foreign body at the level of the 7th to 8th thoracic vertebra. After removal of a nickel, an esophagram showed a stricture in the lower third of the esophagus. This was dilated and he was next seen at 19 months of age for suspected recurrent pulmonary aspiration. The esophagram remained unchanged (Fig 1B). Esophagoscopy showed a stenosis 6 cm distal to the anastomosis. A biopsy proximal to the stenosis was consistent with esophagitis. Dilatation was performed with a size 34F dilator. He required a second dilatation at 20 months of age using a 15-mm hydrostatic balloon inflated to 1 atmosphere pressure for 1 minute (Fig 1C). Balloon dilatation was repeated two further times. An esophageal pH study did not demonstrate GER. Because of failure to respond to dilatations, a left thoracotomy and a limited 2-cm resection of the distal esophageal stenosis was undertaken at 22 months of age. The stricture was difficult to palpate intraoperatively and was identified with the assistance of a flexible endoscope. Histological examination of the resected specimen showed submucosal mucus glands of bronchial origin, mixed salivary glands, a ruptured cyst lined with respiratory epithelium, and several pieces of cartilage (Fig 1D). Postoperatively there was a contained leakage of contrast, which was managed nonoperatively with resolution. After 9 months he is asymptomatic and eating a normal diet. Case
2
On the first day of life, a boy underwent an uncomplicated end-to-end repair of an EA and division of a distal TEF. The initial postoperative esophagram was normal apart from a slight narrowing in the distal esophagus and minimal GER. At 7 weeks, he developed cough and apneic episodes after feedings. An esophagram showed a distal esophageal stenosis. This was dilated, but he was lost to general surgical follow-up and over the first 3 years of life he required endoscopic removal of 6 foreign bodies and 12 further dilatations. The maximum size of dilatation achieved was 36F at 2 years of age, and then there was a progressive decrease in size to 28F. At 3 years of age, he underwent a distal 2-cm resection of the stenosis with the finding of cartilage and respiratory epithelium in a disrupted sinus tract within the esophageal wall (Fig 2). The patient is now 7 years old and is asymptomic with a normal esophagram.
JournalofPediatric
Surgery, Vol26,
No 4 (April),
1991: pp 478-482
DISTAL CONGENITAL
ESOPHAGEAL
STENOSIS
Fig 1. (A) Esophagram (case 1) at 2 weeks of age showing distal esophageal narrowing, dilated proximal esophagus, and patent anastomosis. (8) Esophagram (case 1) at 19 months of age after bouginage showing progression of stenosis. (C) Hydrostatic balloon dilatation (case 1) at 20 months of age using 15-mm baHoon at 1 atmosphere under fluoroscopic control. Note “waist” on the balloon outlining the stenosis. (D) Transverse section of esophageal wall (case 1) showing tracheobronchial rest with nodule of cartilage (arrow) surrounded by glands (hematoxylin phloxin, safrin; original magnification x30).
Case 3 An infant girl underwent primary end-to-end repair of EA and division of a distal TEF. An anastomotic leak was noted on the first esophagram done at 10 days of age. This was managed conservatively. A follow-up esophagram at 4 weeks of age showed a midesophageal stenosis. She did well until solids were introduced, and at 8 months an esophagram showed a stricture at the junction
of the middle and distal thirds of the esophagus. Dilatations were carried out to a size 30F dilator. Her esophagram was normal at 2 and 9 years of age. She is now asymptomatic at 11 years of age.
Case 4 A boy underwent repair of EA and division the first day of life. The initial esophagram
of a distal TEF on was normal and
NEILSON ET AL
Fig 2. Transverse section of esophageal wall (case 2) showing tracheobronchial rest with nodule of cartilage (arrow) surrounded by glands and a sinus tract (asterisk) lined with ciliated epithelium (hematoxylin, phloxin, safrin; original magnification x25).
symptoms did not begin until 6 months of age, when an esophagram confirmed a distal stenosis. This has been dilated 19 times over 5 years and he has required esophagoscopy for removal of two foreign bodies in the past 2 years. He is now 6 years old and tolerates a normal diet.
Case 5 A 36-week gestation girl underwent repair of an EA with distal TEF at birth. One week later, a leak was demonstrated by an esophagram and managed nonoperatively. At 2 months of age a distal esophageal stenosis was demonstrated by esophagram. This has been dilated 6 times. The patient is now 6 years old with minimal symptoms except for occasional cough after eating.
Case 6 In 1963, this full-term baby underwent primary anastomosis of an EA with division of a distal TEF on the first day of life. A transanastomotic 5F gastric feeding tube was passed to the stomach with some difficulty. At 5 days of age he was reexplored for an enlarging air-fluid mass in the posterior mediastinum. The distal esophagus was found distended from 2 cm above the diaphragm to the level of the anastomosis. The esophagus was opened and dilators up to 8F were passed to the stomach and a gastrostomy was performed. Two weeks later an esophagram showed two small anastomotic leaks and a distal esophageal stenosis. From 2 to 5 months of age he underwent retrograde dilatations from 12F to 20F. Due to failure to respond to dilatations and severe failure to thrive, a Heller myotomy was performed at 6 months of age for presumed achalasia. The patient died 10 days later of respiratory failure. At autopsy, nodules of cartilage were demonstrated within the esophagus at the level of the myotomy. DISCUSSION
Spitz, 2 in 1973, was the first to demonstrate a congenital basis for distal esophageal stenosis associated with EA by showing the presence of tracheobronchial rests in an esophagectomy specimen. A recent
review of 20 cases of CES by Nihoul-FCketC et al3 includes only 2 patients with associated EA or TEF. These authors define CES as an intrinsic stenosis present, although not necessarily symptomatic, at birth, which is caused by a congenital malformation of the esophageal wall. They classify the etiology as: (1) tracheobronchial rest (cartilage, respiratory mucous glands, and/or ciliated epithelium); (2) a membranous diaphragm; or (3) segmental hypertrophy of the muscularis and diffuse fibrosis of submucosa (fibromuscular stenosis). They stress that it is essential to exclude postnatally acquired stenoses (peptic, caustic, infectious, neoplastic), extrinsic compression, and achalasia. The association of CES with EA and the more distal location of CES caused by tracheoesophageal rests can be explained on an embryological basis. At the time of the separation of the respiratory tract from the foregut around the 25th day of gestation, tracheobronchial tissue is sequestered in the esophageal wall and then carried caudally by the normal growth of the esophagus.4 In series where the type of esophageal anomaly has been stated,‘.’ CES has been most frequently associated with EA with distal TEF (64%), followed by isolated TEF (20%) and isolated EA (16%). All the cases in the present series had EA with distal TEF, corresponding to the expected incidence of various esophageal anomalies. Histological descriptions of CES associated with EAD’EF were available in only 10 of the 40 previously reported patients. Of these, the majority (7/10) had tracheobronchial rests2,4,6and 3 had fibromuscular stenoses.3,7 Three cases in the present series had tracheobronchial rests in the distal esophageal wall, all with cartilage, two with ciliated epithelial-lined cyst or sinus, and one with bronchial glands. Tracheobronchial rests may be overrepresented in the literature because most require resection, whereas fibromuscular stenoses and membranous diaphragms usually respond to dilatation alone and, hence, are not examined histologically. The diagnosis of CES associated with EA begins with a high index of suspicion. It is important to verify patency of the distal esophagus at the time of primary anastomosis by passage of a tube to the stomach intraoperatively. Symptoms may present in the early postoperative period with leakage at the anastomosis exacerbated by distal obstruction as seen in 4 of the cases in this series (cases 3, 4, 5, and 6). Early symptoms also include aspiration manifest by recurrent pneumonia (cases 1 and 5) or apnea (case 2). Fluoroscopy of the initial postoperative esophagram must be carefully reviewed and repeated looking
DISTAL CONGENITAL
ESOPHAGEAL
481
STENOSIS
specifically for areas of persistent narrowing at the junction of the middle and distal thirds of the esophagus. The initial postoperative esophagram showed a CES in three of the patients in this series (cases 1, 2, and 6), but the correct diagnosis was not made. The majority of cases reported in the literature have presented at 4 to 6 months of age following introduction of solid foods, which results in dysphagia, regurgitation, and cough. Presentation with distal esophageal foreign bodies is also common, as occurred in 3 of the present patients (cases 1,2, and 4). It must be remembered that there are other causes of esophageal stenosis in a patient who has had a repair of EA and/or TEF. The site of the stenosis often suggests the etiology. Stenoses can be located either at the anastomosis or in the distal esophagus. Distal stenoses can be associated either with GER or CES. Therefore, investigation should include a feeding history, a review of the initial and a repeat contrast tine-esophagram, esophagoscopy with biopsy, and esophageal pH monitoring. Even with these investigations it may be difficult to distinguish CES from strictures due to GER. After these investigations, we recommend a single bouginage that can be combined with initial esophagoscopy. Of the two
common types of CES, those caused by fibromuscular hyperplasia often respond to single or intermittent dilation. If the esophagram shows persistence of the stenosis, a series of hydrostatic balloon dilatations to 1 atmosphere pressure should be undertaken. Hydrostatic dilatation is safer than bouginage because disrupting forces are transmitted axially rather than longitudinally and the progress or extent of the dilatation can be followed under fluoroscopic control.’ If there is failure of improvement after 2 to 3 serial hydrostatic dilatations, and if symptoms are present, then operative treatment is recommended. It is possible that some of the presumed fibromuscular cases are actually due to tracheobronchial rests, which may respond to nonoperative therapy. However, we would assume that failure of dilatation implies the presence of cartilage in the CES. The presence of cartilage is also suggested by the narrow “waist” seen on the hydrostatic balloon when it is inflated with dilute contrast solution (Fig 1C). Because the tracheobronchial rests are limited to a short l- to 2-cm length, a limited resection rather than a more extensive resection with esophageal substitution is all that is usually required.
REFERENCES 1. Dunbar JS: Congenital oesophageal North Am 5:433-455,1958
stenosis. Pediatr Clin
2. Spitz L: Congenital esophageal stenosis distal to associated esophageal atresia. J Pediatr Surg 8:973-974, 1973 3. Nihoul-FCkCtC C, De Backer A, Lortat-Jacob S, et al: Congenital esophageal stenosis: A review of 20 cases. Pediatr Surg Int 2:86-92,1987 4. Deiraniya AK: Congenital oesophageal stenosis due tracheobronchial remnants. Thorax 29:720-725,1974 5. Thomason MA, Gay BB: Esophageal stenosis with esophageal atresia. Pediatr Radio1 17:197-201,1987
6. Nishina R, Tsuchida Y, Saitos S: Congenital esophageal stenosis due to tracheobronchial remnants and its associated anomalies. J Pediatr Surg 16:190-193, 1981 7. Tuqan NA: Annular stricture of the esophagus distal to congenital tracheoesophageal fistula. Surgery 52:394-395, 1962 8. Mahour GH, Johnston PW, Gwinn JL: Congenital esophageal stenosis distal to esophageal atresia. Surgery 69:936-939, 1971 9. Hoffer FA, Winter HS, Fellows KE, et al: The treatment of postoperative and peptic esophageal strictures after esophageal atresia repair: A program including dilatation with balloon catheters. Pediatr Radio1 17:454-458,1987
Discussion R Sonnino (Cleveland, OH): I think it is a problem that we are not really looking for as much as we should and, as you have shown, it can cause a whole set of problems when it goes unrecognized. I do have a couple of questions for you though, relating both to the manuscript and the presentation. In particular, you are really including all 6 of the babies in the CES group. But as to the 3 who didn’t undergo resection, or as far as I can tell biopsy, I guess I am questioning how sure you are that these indeed had congenital stenosis rather than reflux or some other problem. Another question, although you mentioned it briefly, can you speculate in any way what the actual inci-
dence of congenital stenosis is both in patients with an atresia and in normal babies? In other words, how many babies might there be running around that actually do have this type of stenosis? It seems to me from the embryology you presented that there is no reason why it wouldn’t occur fairly frequently in babies without an atresia as well. Another question that I would like to ask you is this. Obviously, all these children were diagnosed after their atresia was repaired. Can you think of some way that we can look for it while we are fixing the atresia and not discover later on when the baby leaks or gets a distended esophagus or any of the complications you have
482
discussed? And finally, do you think that esophagoscopy with just a punch biopsy, as you would probably do as a first stage in diagnosis, is going to be adequate, or are your chances of missing these tracheal bronchial risks going to be too high? Is that going to be a good way to monitor for these? M.E. Matlak (Salt Lake City, UT): To answer Dr Soninno’s question, this can be diagnosed in the operating room if you would cannulate the lower esophageal segment with an 8F or 10F red rubber catheter before you repair the esophagus; you can feel the stenosis and then have a high index of suspicion of checking it postoperatively with your esophagram. We have picked up several of these simply having that be part of our surgical armamentarium. The other thing I would emphasize is that you should be very conservative in doing your dilatation with these youngsters, because I think a high incidence of the risk problem or the fibromuscular hyperplasia and leads toward a resection rather than the dilatation because these youngsters are prone to having their esophaguses perforate. Why did your one patient die with a Heller myotomy? Was it a perforation or mediastinitis? I.R. Neilson (response): Dr Sonnino, it is difficult to tell if these are reflux strictures or not when they
NEILSON ET AL
appear late. The early patients did not have pH studies; their clinical history suggested that it wasn’t a reflux stricture. In reviewing the initial esophagram at 1 week, you can often see distal narrowing and not much reflux. Even if reflux is present, it is difficult to imagine a stricture at 1 week. Also, the stenoses were clearly due to anatomical malformations. There are roughly 400 case reports of the isolated congenital esophageal stenoses. The largest series was by NihoulFCkCtCwith 20 patients. In that series there were 2 or 3 with associated esophageal atresia so the isolated stenosis is much more common than the association. I agree that these should probably be looked for in the operating room. We routinely pass a 10F catheter into the stomach after the posterior wall has been joined. However, this could not rule out a stenosis. I wouldn’t treat them at the time of surgery. If there is a high-grade stenosis, this may cause a postoperative leak. As to why the patient died after the Heller myotomy, this was in 1963, before TPN and advanced neonatal care. This patient weighed 3.8 kg at 6 months of age and he was a full-term baby so this really reflects a very severe degree of failure to thrive and he died of purely respiratory complications, ie, bronchopneumonia.
