Primary Repair of Esophageal Atresia Without Staging or Preliminary Gastrostomy Kenneth R. T. Tyson, M.D. ABSTRACT Twenty-two neonates with esophageal
atresia and a distal tracheoesophageal fistula were treated by primary repair. Only if an anastomotic leak was demonstrated was a tube gastrostomy for decompression with transgastric placement of a duodenal tube for feeding performed. Twelve newborns were classified as high-risk by the presence of prematurity with low birth weight (less than 2,250 gm), significant pneumonia, or obvious significant associated anomalies. There was 1 operative death and 1 late death. Three babies requireda tube gastrostomy and duodenal tube for anastomotic leaks. All 20 survivors are symptom free and thriving. Progress in perioperative management of neonates with esophageal atresia may obviate the need for routine gastrostomy and for staging of high-risk newborns.
In 1943Haight and Towsley [3] reported primary correction of esophageal atresia with a distal tracheoesophageal (T-E) fistula in a newborn. Tube gastrostomy was utilized as an integral part of the operative and postoperative management. The initial enthusiasm was broad and well received for application of primary repair with tube gastrostomy to all newborns with the anomaly of esophageal atresia and distal T-E fistula. By the middle 1950s it was apparent that most robust, full-term babies with this anomaly but without significant associated anomalies or pneumonia would survive and thrive following primary repair and gastrostomy. On the other hand, those babies with esophageal atresia and distal T-E fistula who were premature and of low birth weight, or who had significant pneumonia or associated congenital anomalies, did not often survive gastrostomy and primary repair of the defect. In 1962 Holder and associates [61 advocated a From the Department of Surgery, University of Texas Medical Branch, Galveston, TX 77550. Presented at the Twenty-second Annual Meeting of the Southern Thoracic Surgical Association, Nov 64,1975, New Orleans, LA.
378
staged approach for those newborns who at that time appeared to be high-risk candidates for gastrostomy and primary repair of esophageal atresia and distal T-E fistula. That group included babies of low birth weight, those with significant associated congenital anomalies that were apparent at or shortly after birth, and those with significant aspiration pneumonia. The staging consisted of (1)Stamm gastrostomy, (2) retropleural division of the T-E fistula, and (3) sump suction of the proximal esophageal pouch. Anastomosis of the esophagus was then carried out at a later date when the baby was heavier, when other anomalies were treated, and when pulmonary complications had cleared. A definite improval in survival was obtained by staging the repair for these sick infants [7]. As with primary repair, the concept of staging the repair of esophageal atresia with a distal T-E fistula in a neonate who is premature or whose condition is otherwise compromised was well received and widely applied. By 1972, however, several pediatric surgical centers were questioning the application of staging in light of advances in perinatal care for critically ill neonates [l,4, 91. This report deals with a recent series of babies born with esophageal atresia and a distal T-E fistula in whom primary repair without routine use of a tube gastrostomy was applied irrespective of low birth weight, associated anomalies, or significant pneumonia.
Materials and Methods In the six-year period from 1968 to 1974,22 newborns with esophageal atresia and a distal T-E fistula were treated by primary closure of the fistula and esophagoesophagostomy. All babies with the anomaly seen during that interval had operative management; a staged repair was not utilized. There were 12 boys and 10 girls in the series. Birth weights ranged from 1,100 to 3,650 gm; 10 of the babies weighed less than 2,250 gm. One
379 Tyson: Repair of Esophageal Atresia
baby (birth weight 2,500 gm) was admitted at 4 days of age with severe bilateral aspiration pneumonia. One other baby (birth weight 2,450 gm) had an obvious imperforate anus and a bilateral cleft lip and palate (Table 1).The re-
maining 10 babies were of normal birth weight and had no obvious significant associated anomalies (Table 2). Sixteen of the 22 babies had roentgenographic evidence of right upper lobe pulmonary infiltrate or atelectasis. The interval
Table 1 . High-Risk Group Sex & Birth PneuWeight (gm) monia
F, 1,100 M , 1,600 F, 1,750 M, 1,850 F, 1,950 F, 2,000 M, 2,050 M, 2,100 F, 2,150
+ 0 0 0
+ + 0
+
M, 2,200
+ +
M, 2,450
0
M, 2,500
+++
Obvious Associated Anomalies
Occult Anomalies
Complications
Outcome Well Well Well Well Well Well after pulmonary artery banding; awaits repair Well Well Late death 6 mo infectious diarrhea Well after pullthrough Died 3 days (13-15 trisomy)
0 0 0 0 0 0
0 0 0 0 0 Ventricular septa1 defect
Leak at anastomosis 0 0 0 Esophageal stricture Leak at anastomosis
0 0 0
0 0 0
0 0 0
Imperforate anus Imperforate anus, cleft lip & palate 0
0
0
Aortic atresia
0
0
Esophageal stricture
Well
Table 2. Low-Risk Group Sex & Birth Weight (gm)
Pneumonia
Obvious Associated Anomalies
Occult Anomalies
Complications
Outcome
M, 2,250
+
0
Double aortic arch
Well
F, 2,700 F, 2,800
+ +
0 0
0 Patent ductus arteriosus
Airway obstruction, tracheostomy, leak at anastomosis, esophageal stricture 0 0
F, 2,900 M, 2,950 F, 3,000 M, 3,150
+ + + +
0 0 0 0
0 0 0 Crossed renal ectopia; double-outlet right ventricle transposition 0 0 0
F, 3,300 M, 3,450 M, 3,650
+ + 0
0 0 0
0 0 0 0
0 0 0
Well Well after ductus division Well Well Well Well after Rastelli reconstruction Well Well Well
380 The Annals of Thoracic Surgery Vol 21
No 5 May 1976
from birth to primary repair ranged from 12 hours to 5 days (average, 2% days). Each baby was initially treated by placement of a Replogle [8] sump tube through the nose into the proximal esophageal pouch for continuous suction. A head-up position was maintained throughout the procedure. An intravenous infusion of 10% dextrose solution was begun, and each baby was placed on a parenteral regimen of ampicillin and gentamicin. If necessary, rehydration was carried out. The interval between the beginning of initial treatment and operation varied from 1 to 24 hours with an average of 4 hours. Each child then had primary repair of the esophageal atresia and distal T-E fistula. General endotracheal anesthesia was utilized with the baby in the head-up position. The operative approach was a retropleural dissection through a right posterolateral thoracotomy except in 1 child with an obvious right aortic arch in whom a left posterolateral thoracotomy approach was utilized. Repair was accomplished by suture ligation and division of the T-E fistula and primary one-layered end-to-end anastomosis of the proximal esophagus to the distal esophagus. A small extrapleural chest tube was placed a short distance from the anastomosis, and the thoracotomy incision was closed. The 2 babies (both male) with imperforate anus then had divided right transverse colostomies performed. Postoperatively, each baby was maintained on intravenous infusions of hypertonic glucose and parenteral antibiotics. The head-up position was continued and intensive general neonatal care was maintained. If there was evidence of an esophageal anastomotic leak through the extrapleural chest tube in the first 5 days postoperatively, the leak was confirmed roentgenographically and a Stamm gastrostomy was performed with operative placement of a duodenal feeding tube. The baby was maintained in the head-up position, the tube gastrostomy was connected to gravity drainage, and formula feeding through the duodenal tube was begun. This was continued until the esophageal leak had closed, at which time oral feedings were begun. If there was no external evidence of an esophageal leak by the fifth or sixth postopera-
tive day, an esophagogram was obtained to confirm esophageal integrity, and oral feedings were begun on the sixth or seventh postoperative day.
Results There was 1 operative death. The baby with bilateral cleft lip and palate and imperforate anus died suddenly on the third postoperative day. Clinical suspicion of trisomy 13-15 was confirmed at postmortem examination. The immediate cause of death was aortic atresia with closure of the ductus arteriosus. There was 1late death (birth weight, 2,150 gm) at age 6 months from infectious diarrhea. This child had had an uneventful postoperative course, and the death was thought to be unrelated to the esophageal repair. The 20 remaining children are alive and well one to seven years following primary repair of esophageal atresia with a distal T-E fistula. A circumtracheal double aortic arch was unexpectedly encountered in 1 newborn (birth weight 2,250 gm). The smaller anterior arch was divided and primary repair of the esophagus carried out. Postoperatively the baby developed airway obstruction that eventually necessitated a tracheostomy. An esophageal anastomotic leak developed on the third postoperative day that required tube gastrostomy and placement of a duodenal tube. The tracheostomy was successfully removed. The esophageal leak healed spontaneously within three weeks, and oral feedings were begun. An anastomotic stricture subsequently developed, and the infant required esophageal dilation on three occasions. The stricture has not recurred in three years. Two other patients (birth weights 1,100 and 2,000 gm) developed esophageal anastomotic leaks that required tube gastrostomy and placement of a duodenal feeding tube. The leaks closed spontaneously in 8 and 22 days, respectively. Duodenal feedings were uneventful; formula did not reflux into the stomach. Two other children (birth weights 1,950 and 2,500 gm) developed esophageal anastomotic strictures that required two and one esophageal dilations, respectively. Three other babies (birth weights 2,000,2,800, and 3,150 gm) were found to have significant congenital heart disease. The first 2 had conges-
381 Tyson: Repair of Esophageal Atresia
tive heart failure from a ventricular septal defect and a patent ductus arteriosus, respectively. The other child developed profound cyanosis at age 9 months and was found to have a double-outlet right ventricle transposition with subvalvular pulmonic stenosis. He was treated first with a Blalock shunt and ultimately had repair with a Rastelli reconstruction. The baby with the large patent ductus arteriosus was treated successfully by ductus division, and the baby with the large ventricular septal defect was treated by pulmonary artery banding and now awaits complete repair.
Comment Fifteen or 20 years ago, babies born weighing 1,250 gm had an expected mortality of up to SOo/0 even in the absence of esophageal atresia [2]. The least thing that one could do operatively to tide a baby over this tenuous period of life was certainly justified and desirable. In the past decade, however, giant strides in the technological support and scientific understanding of the neonate have been made. In perinatal intensive care units in pediatric surgical centers today, the baby with a birth weight of 1,250 gm has about an SO% chance for survival.* In this milieu, it seems prudent to go ahead with early complete correction of esophageal atresia with a distal T-E fistula even in low birth weight infants. This small series tends to support that concept. Severe associated congenital anomalies no longer need be an indication for a staged repair of esophageal atresia with a T-E fistula. Most other congenital anomalies are also more successfully managed. For instance, in 1964 only 1 child in 5 born with esophageal atresia and congenital heart disease survived [5]. This series *Richardson, CJ: Unpublished data, 1974
suggests that the results are much better today. In general, most babies with associated anomalies so severe that they will not survive primary correction of esophageal atresia and T-E fistula will not survive a series of staged procedures [ll. Routine gastrostomy was not used in this group of patients. Only when there was evidence of an esophageal anastomotic leak was a gastrostomy performed. At the time of gastrostomy, a duodenal feeding tube was easily placed through the gastrostomy. The duodenal feeding tube greatly simplifies alimentation in the presence of an esophageal anastomotic leak.
References 1. Abrahamson J, Shandling B: Esophageal atresia in the underweight baby: a challenge. J Pediatr Surg 7:608, 1972 2. Battaglia FC, Lubchenco LO: A practical classification of newborn infants by weight and gestational age. J Pediatr 71:159, 1967 3. Haight C, Towsley HA: Congenital atresia of the esophagus with tracheoesophageal fistula. Surg Gynecol Obstet 76:672, 1943 4. Holder T: Discussion of Abrahamson and Shandling [l]. 5. Holder TM, Cloud DT, Lewis JE Jr, et al: Esophageal atresia and tracheoesophageal fistula: a survey of its members by the surgical sections of the American Academy of Pediatrics. Pediatrics 34:542, 1964 6. Holder TM, MacDonald VG Jr, Wooley MM: The premature or critically ill infant with esophageal atresia: increased success with a staged approach. J Thorac Cardiovasc Surg 44:344, 1962 7. Koop CE, Hamilton JP: Atresia of the esophagus: increased survival with staged procedures in the poor-risk infant. Ann Surg 162:389, 1965 8. Replogle RL: Esophageal atresia: plastic sump catheter for drainage of the proximal pouch. Surgery 54:296, 1963 9. Wooley MM: Discussion of Abrahamson and Shandling [l].
atresia and a distal tracheoesophageal fistula were treated by primary repair. Only if an anastomotic leak was demonstrated was a tube gastrostomy for decompression with transgastric placement of a duodenal tube for feeding performed. Twelve newborns were classified as high-risk by the presence of prematurity with low birth weight (less than 2,250 gm), significant pneumonia, or obvious significant associated anomalies. There was 1 operative death and 1 late death. Three babies requireda tube gastrostomy and duodenal tube for anastomotic leaks. All 20 survivors are symptom free and thriving. Progress in perioperative management of neonates with esophageal atresia may obviate the need for routine gastrostomy and for staging of high-risk newborns.
In 1943Haight and Towsley [3] reported primary correction of esophageal atresia with a distal tracheoesophageal (T-E) fistula in a newborn. Tube gastrostomy was utilized as an integral part of the operative and postoperative management. The initial enthusiasm was broad and well received for application of primary repair with tube gastrostomy to all newborns with the anomaly of esophageal atresia and distal T-E fistula. By the middle 1950s it was apparent that most robust, full-term babies with this anomaly but without significant associated anomalies or pneumonia would survive and thrive following primary repair and gastrostomy. On the other hand, those babies with esophageal atresia and distal T-E fistula who were premature and of low birth weight, or who had significant pneumonia or associated congenital anomalies, did not often survive gastrostomy and primary repair of the defect. In 1962 Holder and associates [61 advocated a From the Department of Surgery, University of Texas Medical Branch, Galveston, TX 77550. Presented at the Twenty-second Annual Meeting of the Southern Thoracic Surgical Association, Nov 64,1975, New Orleans, LA.
378
staged approach for those newborns who at that time appeared to be high-risk candidates for gastrostomy and primary repair of esophageal atresia and distal T-E fistula. That group included babies of low birth weight, those with significant associated congenital anomalies that were apparent at or shortly after birth, and those with significant aspiration pneumonia. The staging consisted of (1)Stamm gastrostomy, (2) retropleural division of the T-E fistula, and (3) sump suction of the proximal esophageal pouch. Anastomosis of the esophagus was then carried out at a later date when the baby was heavier, when other anomalies were treated, and when pulmonary complications had cleared. A definite improval in survival was obtained by staging the repair for these sick infants [7]. As with primary repair, the concept of staging the repair of esophageal atresia with a distal T-E fistula in a neonate who is premature or whose condition is otherwise compromised was well received and widely applied. By 1972, however, several pediatric surgical centers were questioning the application of staging in light of advances in perinatal care for critically ill neonates [l,4, 91. This report deals with a recent series of babies born with esophageal atresia and a distal T-E fistula in whom primary repair without routine use of a tube gastrostomy was applied irrespective of low birth weight, associated anomalies, or significant pneumonia.
Materials and Methods In the six-year period from 1968 to 1974,22 newborns with esophageal atresia and a distal T-E fistula were treated by primary closure of the fistula and esophagoesophagostomy. All babies with the anomaly seen during that interval had operative management; a staged repair was not utilized. There were 12 boys and 10 girls in the series. Birth weights ranged from 1,100 to 3,650 gm; 10 of the babies weighed less than 2,250 gm. One
379 Tyson: Repair of Esophageal Atresia
baby (birth weight 2,500 gm) was admitted at 4 days of age with severe bilateral aspiration pneumonia. One other baby (birth weight 2,450 gm) had an obvious imperforate anus and a bilateral cleft lip and palate (Table 1).The re-
maining 10 babies were of normal birth weight and had no obvious significant associated anomalies (Table 2). Sixteen of the 22 babies had roentgenographic evidence of right upper lobe pulmonary infiltrate or atelectasis. The interval
Table 1 . High-Risk Group Sex & Birth PneuWeight (gm) monia
F, 1,100 M , 1,600 F, 1,750 M, 1,850 F, 1,950 F, 2,000 M, 2,050 M, 2,100 F, 2,150
+ 0 0 0
+ + 0
+
M, 2,200
+ +
M, 2,450
0
M, 2,500
+++
Obvious Associated Anomalies
Occult Anomalies
Complications
Outcome Well Well Well Well Well Well after pulmonary artery banding; awaits repair Well Well Late death 6 mo infectious diarrhea Well after pullthrough Died 3 days (13-15 trisomy)
0 0 0 0 0 0
0 0 0 0 0 Ventricular septa1 defect
Leak at anastomosis 0 0 0 Esophageal stricture Leak at anastomosis
0 0 0
0 0 0
0 0 0
Imperforate anus Imperforate anus, cleft lip & palate 0
0
0
Aortic atresia
0
0
Esophageal stricture
Well
Table 2. Low-Risk Group Sex & Birth Weight (gm)
Pneumonia
Obvious Associated Anomalies
Occult Anomalies
Complications
Outcome
M, 2,250
+
0
Double aortic arch
Well
F, 2,700 F, 2,800
+ +
0 0
0 Patent ductus arteriosus
Airway obstruction, tracheostomy, leak at anastomosis, esophageal stricture 0 0
F, 2,900 M, 2,950 F, 3,000 M, 3,150
+ + + +
0 0 0 0
0 0 0 Crossed renal ectopia; double-outlet right ventricle transposition 0 0 0
F, 3,300 M, 3,450 M, 3,650
+ + 0
0 0 0
0 0 0 0
0 0 0
Well Well after ductus division Well Well Well Well after Rastelli reconstruction Well Well Well
380 The Annals of Thoracic Surgery Vol 21
No 5 May 1976
from birth to primary repair ranged from 12 hours to 5 days (average, 2% days). Each baby was initially treated by placement of a Replogle [8] sump tube through the nose into the proximal esophageal pouch for continuous suction. A head-up position was maintained throughout the procedure. An intravenous infusion of 10% dextrose solution was begun, and each baby was placed on a parenteral regimen of ampicillin and gentamicin. If necessary, rehydration was carried out. The interval between the beginning of initial treatment and operation varied from 1 to 24 hours with an average of 4 hours. Each child then had primary repair of the esophageal atresia and distal T-E fistula. General endotracheal anesthesia was utilized with the baby in the head-up position. The operative approach was a retropleural dissection through a right posterolateral thoracotomy except in 1 child with an obvious right aortic arch in whom a left posterolateral thoracotomy approach was utilized. Repair was accomplished by suture ligation and division of the T-E fistula and primary one-layered end-to-end anastomosis of the proximal esophagus to the distal esophagus. A small extrapleural chest tube was placed a short distance from the anastomosis, and the thoracotomy incision was closed. The 2 babies (both male) with imperforate anus then had divided right transverse colostomies performed. Postoperatively, each baby was maintained on intravenous infusions of hypertonic glucose and parenteral antibiotics. The head-up position was continued and intensive general neonatal care was maintained. If there was evidence of an esophageal anastomotic leak through the extrapleural chest tube in the first 5 days postoperatively, the leak was confirmed roentgenographically and a Stamm gastrostomy was performed with operative placement of a duodenal feeding tube. The baby was maintained in the head-up position, the tube gastrostomy was connected to gravity drainage, and formula feeding through the duodenal tube was begun. This was continued until the esophageal leak had closed, at which time oral feedings were begun. If there was no external evidence of an esophageal leak by the fifth or sixth postopera-
tive day, an esophagogram was obtained to confirm esophageal integrity, and oral feedings were begun on the sixth or seventh postoperative day.
Results There was 1 operative death. The baby with bilateral cleft lip and palate and imperforate anus died suddenly on the third postoperative day. Clinical suspicion of trisomy 13-15 was confirmed at postmortem examination. The immediate cause of death was aortic atresia with closure of the ductus arteriosus. There was 1late death (birth weight, 2,150 gm) at age 6 months from infectious diarrhea. This child had had an uneventful postoperative course, and the death was thought to be unrelated to the esophageal repair. The 20 remaining children are alive and well one to seven years following primary repair of esophageal atresia with a distal T-E fistula. A circumtracheal double aortic arch was unexpectedly encountered in 1 newborn (birth weight 2,250 gm). The smaller anterior arch was divided and primary repair of the esophagus carried out. Postoperatively the baby developed airway obstruction that eventually necessitated a tracheostomy. An esophageal anastomotic leak developed on the third postoperative day that required tube gastrostomy and placement of a duodenal tube. The tracheostomy was successfully removed. The esophageal leak healed spontaneously within three weeks, and oral feedings were begun. An anastomotic stricture subsequently developed, and the infant required esophageal dilation on three occasions. The stricture has not recurred in three years. Two other patients (birth weights 1,100 and 2,000 gm) developed esophageal anastomotic leaks that required tube gastrostomy and placement of a duodenal feeding tube. The leaks closed spontaneously in 8 and 22 days, respectively. Duodenal feedings were uneventful; formula did not reflux into the stomach. Two other children (birth weights 1,950 and 2,500 gm) developed esophageal anastomotic strictures that required two and one esophageal dilations, respectively. Three other babies (birth weights 2,000,2,800, and 3,150 gm) were found to have significant congenital heart disease. The first 2 had conges-
381 Tyson: Repair of Esophageal Atresia
tive heart failure from a ventricular septal defect and a patent ductus arteriosus, respectively. The other child developed profound cyanosis at age 9 months and was found to have a double-outlet right ventricle transposition with subvalvular pulmonic stenosis. He was treated first with a Blalock shunt and ultimately had repair with a Rastelli reconstruction. The baby with the large patent ductus arteriosus was treated successfully by ductus division, and the baby with the large ventricular septal defect was treated by pulmonary artery banding and now awaits complete repair.
Comment Fifteen or 20 years ago, babies born weighing 1,250 gm had an expected mortality of up to SOo/0 even in the absence of esophageal atresia [2]. The least thing that one could do operatively to tide a baby over this tenuous period of life was certainly justified and desirable. In the past decade, however, giant strides in the technological support and scientific understanding of the neonate have been made. In perinatal intensive care units in pediatric surgical centers today, the baby with a birth weight of 1,250 gm has about an SO% chance for survival.* In this milieu, it seems prudent to go ahead with early complete correction of esophageal atresia with a distal T-E fistula even in low birth weight infants. This small series tends to support that concept. Severe associated congenital anomalies no longer need be an indication for a staged repair of esophageal atresia with a T-E fistula. Most other congenital anomalies are also more successfully managed. For instance, in 1964 only 1 child in 5 born with esophageal atresia and congenital heart disease survived [5]. This series *Richardson, CJ: Unpublished data, 1974
suggests that the results are much better today. In general, most babies with associated anomalies so severe that they will not survive primary correction of esophageal atresia and T-E fistula will not survive a series of staged procedures [ll. Routine gastrostomy was not used in this group of patients. Only when there was evidence of an esophageal anastomotic leak was a gastrostomy performed. At the time of gastrostomy, a duodenal feeding tube was easily placed through the gastrostomy. The duodenal feeding tube greatly simplifies alimentation in the presence of an esophageal anastomotic leak.
References 1. Abrahamson J, Shandling B: Esophageal atresia in the underweight baby: a challenge. J Pediatr Surg 7:608, 1972 2. Battaglia FC, Lubchenco LO: A practical classification of newborn infants by weight and gestational age. J Pediatr 71:159, 1967 3. Haight C, Towsley HA: Congenital atresia of the esophagus with tracheoesophageal fistula. Surg Gynecol Obstet 76:672, 1943 4. Holder T: Discussion of Abrahamson and Shandling [l]. 5. Holder TM, Cloud DT, Lewis JE Jr, et al: Esophageal atresia and tracheoesophageal fistula: a survey of its members by the surgical sections of the American Academy of Pediatrics. Pediatrics 34:542, 1964 6. Holder TM, MacDonald VG Jr, Wooley MM: The premature or critically ill infant with esophageal atresia: increased success with a staged approach. J Thorac Cardiovasc Surg 44:344, 1962 7. Koop CE, Hamilton JP: Atresia of the esophagus: increased survival with staged procedures in the poor-risk infant. Ann Surg 162:389, 1965 8. Replogle RL: Esophageal atresia: plastic sump catheter for drainage of the proximal pouch. Surgery 54:296, 1963 9. Wooley MM: Discussion of Abrahamson and Shandling [l].
