- Email: [email protected]
Revisiting the role of routine retropleural drainage after repair of esophageal atresia with distal tracheoesophageal fistula
Revisiting the Role of Routine Retropleural Drainage After Repair of Esophageal Atresia With Distal Tracheoesophageal Fistula By Saundra Kay and Ken Shaw Montreal, Quebec
Purpose:The ral drainage geal fistula
aim of this study was to review routine retropleuin esophageal atresia with distal tracheoesopha(TEF).
Methods:
The charts of 52 patients diagnosed with TEF between 1987 and 1997 were reviewed. Data collected included gestational age, birth weight, associated congenital anomalies, respiratory status, operative technique and timing, size of gap, tension on the anastomosis, complications related to the drain, and esophageal anastomotic leak (incidence, diagnosis, and treatment).
Results: A total
of 52 cases of TEF were identified. Data on sex, gestation age, birth weights, and malformations are summarized. Forty-seven patients underwent a single-stage extrapleural repair. Only one leak was identified in the single-stage group on routine day 7 esophagogram: a Waterson C patient who was clinically well with minimal drainage.
E
SOPHAGEAL ATRESIA and tracheoesophageal fistula are reported to occur in 1 in 3,520 to 1 in 4,500 live births with a slight male predominance.‘,’ This malformation comes in many forms, but the most common is esophageal atresia with distal tracheoesophageal fistula (TEF).2,3 In 50% to 70% of affected infants, one or more associated anomalies are detected, most commonly cardiac (35%), genitourinary (20%), gastrointestinal (24%), and skeletal (13%).‘~~ Chromosomal anomalies occur in 4.8% to 6.6% of these infants.1,2 In an effort to classify patients into risk categories for prognosis and to aid in choosing the most appropriate surgical option, Waterston in 1962 developed his classification scheme based on birth weight, presence of respiratory disease, and associated congenital anomalies.4 With advances in neonatal care and improved surgical technique, morbidity and mortality rates in esophageal atresia have decreased substantially over the years, and this classification scheme has become less relevant.‘-3,5 A new
From the Department of Pediatric Surgery, Montreal Children’s Hospital, McGill Universit?/, Montreal. Quebec. Canada. Presented at the 30th Annual Meeting of the Canadian Association of Paediatric Surgeons, Toronto, Ontario, Canada, September 25-28, 1998. Address reprint requests to Dr Kenneth S. Shaw, Department of Surgery Montreal Children’s Hospital, 2300 Tupper St, Suite C-1134, Montreal, Quebec, Canada. H3H IP3. Copyright 0 1999 by WB. Saunders Company 0022-3468/99/3407-0010$03.00/0 1082
The only other leakwas in a class C patient who had undergone a staged repair because of an initial long gap. None of the favorable Waterston class patients suffered a leak. Conclusions: A routine retropleural drain placed near the anastomosis may not be necessary in all cases of TEF. Good prognosis patients (Waterston class A and B) who undergo an uncomplicated extrapleural repair without undue tension do not appear to benefit from having a chest drain in place, and there is potential for complications. In complicated cases, however, retropleural drainage remains a reasonable adjunct.
J Pediatr Surg 34:1082-1085. Saunders Company. INDEX
WORDS:
Tracheoesophageal
Copyright
o 1999 by W.B.
fistula.
classification that would more accurately predict outcome and guide treatment was proposed by Poenaru et al5 in 1993 when they discovered that only severe pulmonary dysfunction with preoperative ventilator dependence and severe associated anomalies had any prognostic significance. The surgical approach has gone through many changes since the first successful primary repair of an esophageal atresia was performed by Cameron Haight in 194 1. A left extrapleural approach with fistula ligation and a singlelayer end-to-end esophageal anastomosis was used. Soon after, Haight altered his technique, performing instead a right extrapleural dissection and using a two-layer telescoping anastomosis for the esophagus.’ Favored surgical approach is retropleural; however, a transpleural approach is used by some. 1,3,6,7For some time the end-toside esophageal anastomosis was felt to be superior to the end-to-end anastomosis, but this too has been refuted, and the end-to-end anastomosis is again the preferred method.8 A single layer anastomosis has replaced the double layer anastomosis in most centers.8*9The use of silk for the anastomosis has fallen out of favor, being replaced with absorbable suture material.‘~9J0 Finally, routine gastrostomy has been abandoned.9,*1J2 A drain left adjacent to the anastomosis has been considered standard of care for a long time to identify and treat any postoperative leak. l”,13With the lower complication rates now reported compared with the past, and the JournalofPediatricSurgery,Vol34,
No 7 (July),
1999: pp 1082-1085
ROUTINE
RETROPLEURAL
DRAINAGE
AFTER
EA REPAIR
1083
Table 1. Patient
improved outcome of patients overall, this is becoming less important. 2*9With the extrapleural approach, a leak is not as devastating as it had been with a transpleural approach.13J4 Operative texts that once stressed drain placement as a standard part of the operation, now acknowledge that a drain is not necessary in all cases.‘3.10,11 They do not, however, offer any guidelines as to when it may be preferable to leave a drain and when it may not be necessary. This study was undertaken to determine which groups of patients with esophageal atresia and distal tracheoesophageal fistula are at greatest risk for a leak developing postoperatively and thus may benefit from the placement of a drain close to the anastomosis at the time of the operation.
Data ” 1%)
Sex Male Female
32 (61.5) 20 (38.5)
Gestational >36 wk
age 40 (77)
32-36 wk <32 wk
6 (I 1.5) 6 (I 1.5)
Birth weights >2,500 g 1,800-2,500 <1,800
33 (63.5) 11 (21.1)
g
g
Associated Cardiac
8 (15.4) anomalies 25 (48.1)
Genitourinary Vertebral
6 (11.5) 6 (11.5)
Anal Other
MATERIALS
AND
METHODS
The charts of all patients diagnosed with esophageal atresia and distal tracheoesophageal fistula at the Montreal Children’s Hospital from 1987 to 1997 were reviewed retrospectively. Data collected included gestattonal age: birth weight; the presence of any associated congenital anomalies or chromosomal aberrations; respiratory status; operative technique and timing: size of gap; tension on the anastomosis; complications related to the drain; and the inctdence, method of diagnosis. and treatment of any esophageal anastomotic leak. Patients underwent stratificatton using Waterston’s classification as well as the newer “Montreal classification.“5
RESULTS
A total of 52 cases of esophageal atresia and distal tracheoesophageal fistula were identified, 32 boys (61.5%) and 20 girls (38.5%). The majority of patients (40 of 52, 77%) were born at a gestational age of 236 weeks. Six (11.5%) were born at each of 32 to 36 weeks’ gestation and 132 weeks’ gestation. Birth weights were as follows: 33 (63.5%) were ~2,500 g, 11 (21.1%) weighed between 1,800 and 2,500 g, and eight (15.4%) were 5 1,800 g. A total of 27 infants (5 1.9%) had at least one associated congenital anomaly. Twenty-five patients (48.1%) had a cardiac anomaly, six (11.5%) had genitourinary, five (9.6%) had anal, six (11.5%) had vertebral, and 15 (28.8%) had other types of malformations. Three infants (5.8%) had an abnormal karyotype. Patients were classified by Waterston’s criteria (Table 1) and using the Montreal classification (Table 2). Forty-seven patients (90.4%) underwent a single-stage extrapleural repair with end-to-end single-layer esophagoesophagostomy. All but one had a drain placed close to the anastomosis. Two infants (3.8%) were found at the time of planned single-stage repair to have a gap that was too large to permit a safe anastomosis, so this was deferred. Three infants (5.8%), too unstable for complete repair, were to undergo a staged procedure but died before esophagoesophagostomy could be performed. One had extreme intrauterine growth retardation, one was
5 (9.6) 15 (28.8)
Karyotype
anomalres
3 (5.8)
diagnosed with a gastric perforation soon after birth, and the third was born with hypoplastic lungs. The size of the gap between ends of the esophagus was subjectively determined by the surgeon at the time of procedure and broken down into small (< 1 cm), moderate (1 to 2 cm) and long (>2 cm). Of the 47 patients who underwent single-stage repair, 25 (53.2%) had a small gap, five (10.7%) had a moderate gap, two (4.2%) had a long gap, and 15 (3 1.9%) had a gap of unspecified length. The tension on the anastomosis, similarly determined, was minimal in 32 (68.1%), moderate in six (12.8%), and high in one (2.1%). In eight cases (17%). there was no indication as to the estimated tension on the anastomosis. Only three cases specified the need for extensive proximal and distal dissection of the esophagus, and two of these additionally required esophagomyotomy to increase length. This resulted in successful primary anastomosis in two cases, but in one, despite these maneuvers, anastomosis had to be deferred. In all 49 patients who underwent esophagoesophagostomy, only two leaks were identified. One occurred in a patient classified as Waterston class C and Montreal Table 2. TEF Risk Classification Waterston’s
Classification
A Brrth weight
>2,500
B Brrth weight moderate C Brrth weight
1,800.2,500 g and well -OR- higher birth weight with pneumonia or anomaly
severe
pneumonra
g and well
or severe Montreal
I
anomaly Classrfrcation
All non-class II patients including major anomalres alone and ventilator dependence alone, or assocrated with minor anomalres II Life-threatenrng anomalies -OR- ventilator dependence with major
anomalres
KAY AND
1084
category II who had undergone a staged repair for an initial long gap. Extensive dissection of the esophagus and esophagomyotomy had been performed at the time of the first operation. Via the transpleural route, the anastomosis was performed with minimal tension. On postoperative day 2, saliva and milk (from gastrostomy feedings) were seen in the chest tube drainage. The leak was treated conservatively. Eleven days postesophagoesophagostomy, the baby died presumably from aspiration. The only other leak occurred in a baby classified as Waterston class C and Montreal category I, post-single-stage repair whereby the tension at the anastomosis was felt to be minimal. The baby was doing well postoperatively, chest drainage was minimal and serosanguinous. There was no suspicion of a leak. Routine esophagogram on postoperative day 7 showed a small localized leak. The patient was treated conservatively, and no problems developed. Four days later the baby was brought to the operating room for repair of an aortic coarctation and hypoplastic arch that was uneventful. In all, seven patients (13.5%) died: three before esophagoesophagostomy could be performed, three post-singlestage repair who did well initially but later died of complications related to major cardiac malformations, and one died of presumed aspiration 10 days postdetection of anastomotic leak. No complications directly related to the retropleural drains were detected. DISCUSSION
The profile of our patients was comparable to previous reports with respect to associated congenital and chromosomal anomalies, although the male to female ratio of 1.6 was slightly higher than that reported by others (1.26)’ The incidence of postoperative anastomotic leaks is reported to be in the range of 5% to 20% depending on the type of anastomosis and degree of tension.1.9J1,14An interrupted single-layer end-to-end esophageal anastomosis appears to have the lowest rate of leakage.2 Factors that contribute to anastomotic leaks include poor surgical technique (inadequate approximation of tissues, sutures tied too tightly, trauma to the esophageal ends), ischemia of the esophageal ends and thus at the anastomosis, excess tension on the anastomosis, and the use of silk suture
X9,14.15
In the past, anastomotic leaks were frequent and often fatal6 Now, with improved surgical technique and neonatal care, including nutritional and respiratory support, this
SHAW
complication is seen less frequently and rarely leads to death.2,6.9With the use of an extrapleural approach, the consequences of a leak are diminished even further.14 Major leaks, usually detected within the first 48 hours, require reexploration in most cases, but these account for only a small percentage of all leaks.1,9 Most leaks are clinically insignificant, detected on routine postoperative esophagogram, and can be treated conservatively with nutritional support and antibiotics with spontaneous healing in most cases.1,2,7,9 Our anastomotic leak rate of 4.1% compares favorably with those reported.1*9J1J4Only one was clinically significant. This was in a baby with congenital heart disease who underwent a two-stage repair. At the first stage, all attempts were made for a primary anastomosis including extensive proximal and distal esophageal dissection and esophagomyotomy, but this was not enough. At the second stage 8 weeks later, the two ends came together well, although the blood supply may have been tenuous from the extensive dissection previously. The other leak, in a baby with congenital heart disease, was clinically insignificant and resolved spontaneously. It did not even delay repair of the hypoplastic aortic arch and coarctation. Earlier operative textbooks and reviews all included the placement of a drain close to the anastomosis as a standard part of the operation for repair of esophageal atresia and tracheoesophageal anastomosis.‘0~‘3 Newer texts, however, acknowledge that a drain may not be necessary in all cases and its use should be left at the The risks and benefits of discretion of the surgeon. 1+2,3~10~11 any procedure must be weighed against each other. Although our review did not detect any complications directly related to the drains, there is, nonetheless, the potential for infection, and although we did not look closely at pain or requirement for pain medications in our study, this is another potential concern with a chest drain in a newborn. In good-prognosis patients (Waterston class A and B; Montreal category I) who undergo an uncomplicated extrapleural repair using careful surgical technique with a small to moderate gap and minimal tension on the anastomosis, the rate of leakage is very low and does not support the use of a drain routinely. In more complicated cases, however, retropleural drainage remains a reasonable adjunct.
REFERENCES 1. Harmon CM, Coran AG: Congenital anomalies of the esophagus, in O’Neill JA, Rowe MI, Grosfeld IL, et al (eds): Pediatric Surgery. Toronto, Ont, Mosby, 1998, pp 941-967 2. Beasley SW: Esophageal atresia and tracheoesophageal fistula, in
Oldham KT, Colombani PM, Foglia RP (eds): Surgery of Infants and Children: Scientific Principles and Practice. Philadelphia, PA, LippincottRaven Publishers, 1997, pp 1021-1034 3. Spitz L: Esophageal atresia with and without trachecesophageal
ROUTINE
AETROPLEURAL
DRAINAGE
AFTER
EA REPAIR
fistula, in Spitz L, Coran AG (eds): Rob & Smith’s Operative Surgery, Pediatric Surgery. New York, NY, Chapman & Hall Medical, 1995, pp 111-120 4. Waterston DJ, Bonham-Carter RE, et al: Oesophageal atresia: Tracheoesophageal fistula. A study of survival in 218 infants. Lancet 1:819-822, 1962 5. Poenaru D, Laberge JM, Neilson IR, et al: A new prognostic classification for esophageal atresia. Surgery 113:426-432, 1993 6. Connolly B, Gumey EJ: Trends in tracheoesophageal fistula. Surg Gynecol Obstet 164:308-312, 1987 7. Randolph JG: Esophageal atresia and congenital stenosis, in Welch KJ, Randolph JG (eds): Pediatric Surgery (ed 4). Chicago, IL. Year Book Medical, 1986, pp 682-693 8. Poenaru D, Laberge JM, Neilson IR, et al: A more than 25 year experience wtth end-to-end versus end-to-side repair for esophageal atresia. J Pediatr Surg 26:472-477, 1991 9. Engum SA, Grosfeld JL, West KW, et al: Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades. Arch Surg 130:502-508, 1995
10. Waldhauser JA, Pierce WS: Surgery of the esophagus, m Waldhauser JA, Pierce WS (eds): Johnson’s Surgery of the Chest (ed 5). Chicago, IL, Year Book Medical, 1985, pp 195-202 11. Spitz L, Kiely E, Brereton RJ, et al: Management of esophageal atresia. World J Surg 17:296-300, 1993 12. Shaul DB. Schwartz MZ, Marr CC, et al: Primary repair without routine gastrostomy 1s the treatment of choice for neonates with esophageal atresta and tracheoesophageal fistula. Arch Surg 124: 1188. 1191, 1989 13. Coran AG: Congenital abnormahties of the esophagus. in Zuidema CD, Orringer MB (eds): Shackelford’s Surgery of the Alimentary Tract Vol 1, (ed 3). Montreal, Que. WB Saunders, 1991, pp 127-146 14. McKinnon LJ, Kosloske AM: Prediction and prevention of anastomottc complications of esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 25:778-781, 1990 15. Welch KJ: Esophagus, in Welch KJ (ed): Complications of Pediatric Surgery, Prevention and Management. Philadelphia. PA. WB Saunders, 1982, pp 201-202
Purpose:The ral drainage geal fistula
aim of this study was to review routine retropleuin esophageal atresia with distal tracheoesopha(TEF).
Methods:
The charts of 52 patients diagnosed with TEF between 1987 and 1997 were reviewed. Data collected included gestational age, birth weight, associated congenital anomalies, respiratory status, operative technique and timing, size of gap, tension on the anastomosis, complications related to the drain, and esophageal anastomotic leak (incidence, diagnosis, and treatment).
Results: A total
of 52 cases of TEF were identified. Data on sex, gestation age, birth weights, and malformations are summarized. Forty-seven patients underwent a single-stage extrapleural repair. Only one leak was identified in the single-stage group on routine day 7 esophagogram: a Waterson C patient who was clinically well with minimal drainage.
E
SOPHAGEAL ATRESIA and tracheoesophageal fistula are reported to occur in 1 in 3,520 to 1 in 4,500 live births with a slight male predominance.‘,’ This malformation comes in many forms, but the most common is esophageal atresia with distal tracheoesophageal fistula (TEF).2,3 In 50% to 70% of affected infants, one or more associated anomalies are detected, most commonly cardiac (35%), genitourinary (20%), gastrointestinal (24%), and skeletal (13%).‘~~ Chromosomal anomalies occur in 4.8% to 6.6% of these infants.1,2 In an effort to classify patients into risk categories for prognosis and to aid in choosing the most appropriate surgical option, Waterston in 1962 developed his classification scheme based on birth weight, presence of respiratory disease, and associated congenital anomalies.4 With advances in neonatal care and improved surgical technique, morbidity and mortality rates in esophageal atresia have decreased substantially over the years, and this classification scheme has become less relevant.‘-3,5 A new
From the Department of Pediatric Surgery, Montreal Children’s Hospital, McGill Universit?/, Montreal. Quebec. Canada. Presented at the 30th Annual Meeting of the Canadian Association of Paediatric Surgeons, Toronto, Ontario, Canada, September 25-28, 1998. Address reprint requests to Dr Kenneth S. Shaw, Department of Surgery Montreal Children’s Hospital, 2300 Tupper St, Suite C-1134, Montreal, Quebec, Canada. H3H IP3. Copyright 0 1999 by WB. Saunders Company 0022-3468/99/3407-0010$03.00/0 1082
The only other leakwas in a class C patient who had undergone a staged repair because of an initial long gap. None of the favorable Waterston class patients suffered a leak. Conclusions: A routine retropleural drain placed near the anastomosis may not be necessary in all cases of TEF. Good prognosis patients (Waterston class A and B) who undergo an uncomplicated extrapleural repair without undue tension do not appear to benefit from having a chest drain in place, and there is potential for complications. In complicated cases, however, retropleural drainage remains a reasonable adjunct.
J Pediatr Surg 34:1082-1085. Saunders Company. INDEX
WORDS:
Tracheoesophageal
Copyright
o 1999 by W.B.
fistula.
classification that would more accurately predict outcome and guide treatment was proposed by Poenaru et al5 in 1993 when they discovered that only severe pulmonary dysfunction with preoperative ventilator dependence and severe associated anomalies had any prognostic significance. The surgical approach has gone through many changes since the first successful primary repair of an esophageal atresia was performed by Cameron Haight in 194 1. A left extrapleural approach with fistula ligation and a singlelayer end-to-end esophageal anastomosis was used. Soon after, Haight altered his technique, performing instead a right extrapleural dissection and using a two-layer telescoping anastomosis for the esophagus.’ Favored surgical approach is retropleural; however, a transpleural approach is used by some. 1,3,6,7For some time the end-toside esophageal anastomosis was felt to be superior to the end-to-end anastomosis, but this too has been refuted, and the end-to-end anastomosis is again the preferred method.8 A single layer anastomosis has replaced the double layer anastomosis in most centers.8*9The use of silk for the anastomosis has fallen out of favor, being replaced with absorbable suture material.‘~9J0 Finally, routine gastrostomy has been abandoned.9,*1J2 A drain left adjacent to the anastomosis has been considered standard of care for a long time to identify and treat any postoperative leak. l”,13With the lower complication rates now reported compared with the past, and the JournalofPediatricSurgery,Vol34,
No 7 (July),
1999: pp 1082-1085
ROUTINE
RETROPLEURAL
DRAINAGE
AFTER
EA REPAIR
1083
Table 1. Patient
improved outcome of patients overall, this is becoming less important. 2*9With the extrapleural approach, a leak is not as devastating as it had been with a transpleural approach.13J4 Operative texts that once stressed drain placement as a standard part of the operation, now acknowledge that a drain is not necessary in all cases.‘3.10,11 They do not, however, offer any guidelines as to when it may be preferable to leave a drain and when it may not be necessary. This study was undertaken to determine which groups of patients with esophageal atresia and distal tracheoesophageal fistula are at greatest risk for a leak developing postoperatively and thus may benefit from the placement of a drain close to the anastomosis at the time of the operation.
Data ” 1%)
Sex Male Female
32 (61.5) 20 (38.5)
Gestational >36 wk
age 40 (77)
32-36 wk <32 wk
6 (I 1.5) 6 (I 1.5)
Birth weights >2,500 g 1,800-2,500 <1,800
33 (63.5) 11 (21.1)
g
g
Associated Cardiac
8 (15.4) anomalies 25 (48.1)
Genitourinary Vertebral
6 (11.5) 6 (11.5)
Anal Other
MATERIALS
AND
METHODS
The charts of all patients diagnosed with esophageal atresia and distal tracheoesophageal fistula at the Montreal Children’s Hospital from 1987 to 1997 were reviewed retrospectively. Data collected included gestattonal age: birth weight; the presence of any associated congenital anomalies or chromosomal aberrations; respiratory status; operative technique and timing: size of gap; tension on the anastomosis; complications related to the drain; and the inctdence, method of diagnosis. and treatment of any esophageal anastomotic leak. Patients underwent stratificatton using Waterston’s classification as well as the newer “Montreal classification.“5
RESULTS
A total of 52 cases of esophageal atresia and distal tracheoesophageal fistula were identified, 32 boys (61.5%) and 20 girls (38.5%). The majority of patients (40 of 52, 77%) were born at a gestational age of 236 weeks. Six (11.5%) were born at each of 32 to 36 weeks’ gestation and 132 weeks’ gestation. Birth weights were as follows: 33 (63.5%) were ~2,500 g, 11 (21.1%) weighed between 1,800 and 2,500 g, and eight (15.4%) were 5 1,800 g. A total of 27 infants (5 1.9%) had at least one associated congenital anomaly. Twenty-five patients (48.1%) had a cardiac anomaly, six (11.5%) had genitourinary, five (9.6%) had anal, six (11.5%) had vertebral, and 15 (28.8%) had other types of malformations. Three infants (5.8%) had an abnormal karyotype. Patients were classified by Waterston’s criteria (Table 1) and using the Montreal classification (Table 2). Forty-seven patients (90.4%) underwent a single-stage extrapleural repair with end-to-end single-layer esophagoesophagostomy. All but one had a drain placed close to the anastomosis. Two infants (3.8%) were found at the time of planned single-stage repair to have a gap that was too large to permit a safe anastomosis, so this was deferred. Three infants (5.8%), too unstable for complete repair, were to undergo a staged procedure but died before esophagoesophagostomy could be performed. One had extreme intrauterine growth retardation, one was
5 (9.6) 15 (28.8)
Karyotype
anomalres
3 (5.8)
diagnosed with a gastric perforation soon after birth, and the third was born with hypoplastic lungs. The size of the gap between ends of the esophagus was subjectively determined by the surgeon at the time of procedure and broken down into small (< 1 cm), moderate (1 to 2 cm) and long (>2 cm). Of the 47 patients who underwent single-stage repair, 25 (53.2%) had a small gap, five (10.7%) had a moderate gap, two (4.2%) had a long gap, and 15 (3 1.9%) had a gap of unspecified length. The tension on the anastomosis, similarly determined, was minimal in 32 (68.1%), moderate in six (12.8%), and high in one (2.1%). In eight cases (17%). there was no indication as to the estimated tension on the anastomosis. Only three cases specified the need for extensive proximal and distal dissection of the esophagus, and two of these additionally required esophagomyotomy to increase length. This resulted in successful primary anastomosis in two cases, but in one, despite these maneuvers, anastomosis had to be deferred. In all 49 patients who underwent esophagoesophagostomy, only two leaks were identified. One occurred in a patient classified as Waterston class C and Montreal Table 2. TEF Risk Classification Waterston’s
Classification
A Brrth weight
>2,500
B Brrth weight moderate C Brrth weight
1,800.2,500 g and well -OR- higher birth weight with pneumonia or anomaly
severe
pneumonra
g and well
or severe Montreal
I
anomaly Classrfrcation
All non-class II patients including major anomalres alone and ventilator dependence alone, or assocrated with minor anomalres II Life-threatenrng anomalies -OR- ventilator dependence with major
anomalres
KAY AND
1084
category II who had undergone a staged repair for an initial long gap. Extensive dissection of the esophagus and esophagomyotomy had been performed at the time of the first operation. Via the transpleural route, the anastomosis was performed with minimal tension. On postoperative day 2, saliva and milk (from gastrostomy feedings) were seen in the chest tube drainage. The leak was treated conservatively. Eleven days postesophagoesophagostomy, the baby died presumably from aspiration. The only other leak occurred in a baby classified as Waterston class C and Montreal category I, post-single-stage repair whereby the tension at the anastomosis was felt to be minimal. The baby was doing well postoperatively, chest drainage was minimal and serosanguinous. There was no suspicion of a leak. Routine esophagogram on postoperative day 7 showed a small localized leak. The patient was treated conservatively, and no problems developed. Four days later the baby was brought to the operating room for repair of an aortic coarctation and hypoplastic arch that was uneventful. In all, seven patients (13.5%) died: three before esophagoesophagostomy could be performed, three post-singlestage repair who did well initially but later died of complications related to major cardiac malformations, and one died of presumed aspiration 10 days postdetection of anastomotic leak. No complications directly related to the retropleural drains were detected. DISCUSSION
The profile of our patients was comparable to previous reports with respect to associated congenital and chromosomal anomalies, although the male to female ratio of 1.6 was slightly higher than that reported by others (1.26)’ The incidence of postoperative anastomotic leaks is reported to be in the range of 5% to 20% depending on the type of anastomosis and degree of tension.1.9J1,14An interrupted single-layer end-to-end esophageal anastomosis appears to have the lowest rate of leakage.2 Factors that contribute to anastomotic leaks include poor surgical technique (inadequate approximation of tissues, sutures tied too tightly, trauma to the esophageal ends), ischemia of the esophageal ends and thus at the anastomosis, excess tension on the anastomosis, and the use of silk suture
X9,14.15
In the past, anastomotic leaks were frequent and often fatal6 Now, with improved surgical technique and neonatal care, including nutritional and respiratory support, this
SHAW
complication is seen less frequently and rarely leads to death.2,6.9With the use of an extrapleural approach, the consequences of a leak are diminished even further.14 Major leaks, usually detected within the first 48 hours, require reexploration in most cases, but these account for only a small percentage of all leaks.1,9 Most leaks are clinically insignificant, detected on routine postoperative esophagogram, and can be treated conservatively with nutritional support and antibiotics with spontaneous healing in most cases.1,2,7,9 Our anastomotic leak rate of 4.1% compares favorably with those reported.1*9J1J4Only one was clinically significant. This was in a baby with congenital heart disease who underwent a two-stage repair. At the first stage, all attempts were made for a primary anastomosis including extensive proximal and distal esophageal dissection and esophagomyotomy, but this was not enough. At the second stage 8 weeks later, the two ends came together well, although the blood supply may have been tenuous from the extensive dissection previously. The other leak, in a baby with congenital heart disease, was clinically insignificant and resolved spontaneously. It did not even delay repair of the hypoplastic aortic arch and coarctation. Earlier operative textbooks and reviews all included the placement of a drain close to the anastomosis as a standard part of the operation for repair of esophageal atresia and tracheoesophageal anastomosis.‘0~‘3 Newer texts, however, acknowledge that a drain may not be necessary in all cases and its use should be left at the The risks and benefits of discretion of the surgeon. 1+2,3~10~11 any procedure must be weighed against each other. Although our review did not detect any complications directly related to the drains, there is, nonetheless, the potential for infection, and although we did not look closely at pain or requirement for pain medications in our study, this is another potential concern with a chest drain in a newborn. In good-prognosis patients (Waterston class A and B; Montreal category I) who undergo an uncomplicated extrapleural repair using careful surgical technique with a small to moderate gap and minimal tension on the anastomosis, the rate of leakage is very low and does not support the use of a drain routinely. In more complicated cases, however, retropleural drainage remains a reasonable adjunct.
REFERENCES 1. Harmon CM, Coran AG: Congenital anomalies of the esophagus, in O’Neill JA, Rowe MI, Grosfeld IL, et al (eds): Pediatric Surgery. Toronto, Ont, Mosby, 1998, pp 941-967 2. Beasley SW: Esophageal atresia and tracheoesophageal fistula, in
Oldham KT, Colombani PM, Foglia RP (eds): Surgery of Infants and Children: Scientific Principles and Practice. Philadelphia, PA, LippincottRaven Publishers, 1997, pp 1021-1034 3. Spitz L: Esophageal atresia with and without trachecesophageal
ROUTINE
AETROPLEURAL
DRAINAGE
AFTER
EA REPAIR
fistula, in Spitz L, Coran AG (eds): Rob & Smith’s Operative Surgery, Pediatric Surgery. New York, NY, Chapman & Hall Medical, 1995, pp 111-120 4. Waterston DJ, Bonham-Carter RE, et al: Oesophageal atresia: Tracheoesophageal fistula. A study of survival in 218 infants. Lancet 1:819-822, 1962 5. Poenaru D, Laberge JM, Neilson IR, et al: A new prognostic classification for esophageal atresia. Surgery 113:426-432, 1993 6. Connolly B, Gumey EJ: Trends in tracheoesophageal fistula. Surg Gynecol Obstet 164:308-312, 1987 7. Randolph JG: Esophageal atresia and congenital stenosis, in Welch KJ, Randolph JG (eds): Pediatric Surgery (ed 4). Chicago, IL. Year Book Medical, 1986, pp 682-693 8. Poenaru D, Laberge JM, Neilson IR, et al: A more than 25 year experience wtth end-to-end versus end-to-side repair for esophageal atresia. J Pediatr Surg 26:472-477, 1991 9. Engum SA, Grosfeld JL, West KW, et al: Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades. Arch Surg 130:502-508, 1995
10. Waldhauser JA, Pierce WS: Surgery of the esophagus, m Waldhauser JA, Pierce WS (eds): Johnson’s Surgery of the Chest (ed 5). Chicago, IL, Year Book Medical, 1985, pp 195-202 11. Spitz L, Kiely E, Brereton RJ, et al: Management of esophageal atresia. World J Surg 17:296-300, 1993 12. Shaul DB. Schwartz MZ, Marr CC, et al: Primary repair without routine gastrostomy 1s the treatment of choice for neonates with esophageal atresta and tracheoesophageal fistula. Arch Surg 124: 1188. 1191, 1989 13. Coran AG: Congenital abnormahties of the esophagus. in Zuidema CD, Orringer MB (eds): Shackelford’s Surgery of the Alimentary Tract Vol 1, (ed 3). Montreal, Que. WB Saunders, 1991, pp 127-146 14. McKinnon LJ, Kosloske AM: Prediction and prevention of anastomottc complications of esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 25:778-781, 1990 15. Welch KJ: Esophagus, in Welch KJ (ed): Complications of Pediatric Surgery, Prevention and Management. Philadelphia. PA. WB Saunders, 1982, pp 201-202