The American Journal of Surgery (2011) 202, 203–206
Clinical Science
Repair of esophageal atresia with tracheoesophageal fistula via thoracotomy: a contemporary series Jeffrey M. Burford, M.D.a, Melvin S. Dassinger, M.D.b, Daniel R. Copeland, M.D.a, Jennifer E. Keller, M.D.b, Samuel D. Smith, M.D.b,* a
Department of Surgery, University of Arkansas for Medical Sciences, Little Rock, AR; bDepartment of Pediatric Surgery, Arkansas Children’s Hospital, Little Rock, AR, USA KEYWORDS: Tracheoesophageal fistula; Esophageal atresia; Thoracotomy; Thoracoscopic surgery
Abstract BACKGROUND: A recent series detailing thoracoscopic repair of esophageal atresia with tracheoesophageal fistula (EA/TEF) reported lower complication rates compared with historic controls. This study provides a contemporary cohort of patients repaired via thoracotomy for comparison with the recent large multi-institutional thoracoscopic series. METHODS: Records of patients with EA/TEF between 1993 and 2008 were reviewed. Attention was focused on demographics and complications including anastomotic leak, recurrent fistulae, stricture formation, and need for fundoplication. RESULTS: Seventy-two patients underwent repair of EA/TEF via thoracotomy. Complication rates in the current series compared with the thoracoscopic series were anastomotic leak, 2.7% versus 7.6%; recurrent fistulae, 2.7% versus 1.9%; stricture, 5.5% versus 3.8%; and need for fundoplication, 12% versus 24%. Differences in complication rates did not reach statistical significance. Two children in this cohort developed mild scoliosis attributed to congenital vertebral anomalies, neither of whom required intervention. CONCLUSIONS: Thoracoscopic repair of EA/TEF yielded complication rates similar to this contemporary series; however, trends toward increased anastomotic leaks and greater need for fundoplication were noted. No musculoskeletal sequelae were directly attributable to thoracotomy. © 2011 Elsevier Inc. All rights reserved.
Haight performed the first successful primary repair of esophageal atresia with tracheoesophageal fistula (EA/TEF) via extrapleural thoracotomy in 1941.1 The practice of surgery has evolved since that time, and minimally invasive techniques have been developed and applied to many conditions, including congenital anomalies.2 This natural evolution led to thoracoscopic repair of EA/TEF, which was first performed in 1999 at the International Pediatric Surgi* Corresponding author. Tel.: 501-364-2942; fax: 501-364-1516. E-mail address:
[email protected] Manuscript received June 3, 2010; revised manuscript September 8, 2010
0002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2010.09.035
cal Endoscopy meeting.3 Since that time, there have been several small series evaluating the thoracoscopic approach, concluding the technique is safe and effective.4 More recently, a large multi-institutional review detailing the results of 104 newborns treated for EA/TEF thoracoscopically was published.5 Thoracoscopic outcomes were compared with those of historic controls, concluding that the minimally invasive approach was comparable with standard open repair. The authors inferred that the minimally invasive approach may have fewer long-term musculoskeletal sequelae. These inferences were made on the basis of results from thoracotomy series performed ⬎20 years ago. The purpose of this study was to describe a contemporary series
204 of EA/TEF repaired via a thoracotomy for adequate comparison with this most recent series of thoracoscopic repair.
The American Journal of Surgery, Vol 202, No 2, August 2011 Table 1 Associated anomalies of 72 patients undergoing thoracotomy repair of EA/TEF Anomaly
Methods A retrospective review including all newborns undergoing EA/TEF repair between 1993 and 2008 was performed following study approval by an institutional review board. Cases were identified via International Classification of Diseases, Ninth Revision, diagnostic code. Neonates with EA without TEF or those with TEF without EA were excluded. Short-gap atresia was defined as a gap ⬍3 vertebral bodies in length. Demographic data points included age, weight, and associated congenital anomalies including cardiac defects. Chart reviews were focused on surgical outcomes and complications such as anastomotic leaks and strictures, recurrent fistulas, need for fundoplication, musculoskeletal sequelae, and mortality. Anastomotic leak was defined as extravasation of contrast on esophagram and stricture was defined as a narrowing noted on esophagram requiring ⱖ2 dilations. Data from this contemporary thoracotomy series were compared with results from a recent multi-institutional thoracoscopic series. Data were then compared with results from several historic thoracotomy series (historic controls) previously used for comparison in the thoracoscopic series.1,6 – 8 Statistical analysis was performed using Fisher’s exact test, with results considered statistically significant at P ⱕ .05.
Results Seventy-two patients underwent repair of EA with distal TEF via thoracotomy. The mean gestational age was 37.3 weeks (range, 28 – 42 weeks), and the average birth weight was 2.68 kg (range, .98 – 4.62 kg). Surgery was performed on day of life 3.7 (range, 1– 64 days), with a mean operative time of 140 minutes (range, 89 –213 minutes). Surgical repair was approached via right thoracotomy in most patients (n ⫽ 71), and left thoracotomy was used in 1 patient with a right aortic arch. Muscle-sparing thoracotomy was used infrequently (n ⫽ 16). Of the 72 patients, 44 (61%) had significant associated anomalies, as displayed in Table 1. Fifteen (21%) had congenital heart defects such as tetralogy of Fallot, dextrocardia, atrial septal defect, ventricular septal defect, and right aortic arch. Two were found to have concomitant duodenal atresia. Seventeen had VACTERL syndrome, of whom 8 patients had associated imperforate anus. One was found to have Down syndrome, and 2 patients had trisomy 18 anomalies. Comparing this contemporary series with the thoracoscopic series reported by Holcomb et al,5 infants were similar in size, averaging 2.6 kg in both studies. Infants
Cardiac ASD/VSD Right aortic arch TOF Dextrocardia PDA (ligation) Gastrointestinal Imperforate anus Duodenal atresia Cloaca Chromosomal abnormalities/syndromes VACTERL syndrome Down syndrome Trisomy 18 Renal Horseshoe kidney Unilateral agenesis Other Vertebral anomalies Radial aplasia Tethered cord
n 13 2 1 1 7 8 2 1 17 1 2 3 1 3 2 1
ASD ⫽ atrial septal defect; PDA ⫽ patent ductus arteriosus; TOF ⫽ tetralogy of Fallot; VSD ⫽ ventricular septal defect.
treated via thoracotomy were taken to surgery on average 2 days later (3.7 vs 1.2 days), and operative times were slightly longer than those in the thoracoscopic series (140 vs 129 minutes). Anastomotic leak occurred in 2 of our patients (2.77%), neither requiring immediate operative intervention. However, recurrent fistulas developed in both of these patients. Four patients developed strictures (5.55%), only 1 of whom required ⬎4 dilations (1.38%). One patient developed a chylothorax, and 1 patient suffered a right phrenic nerve injury. Nine patients eventually underwent fundoplication for reflux (12.5%). Two patients developed mild scoliosis, neither requiring surgery. Of those 2 patients, 1 had undergone muscle-sparing thoracotomy during repair of EA/TEF. Two other patients were found to have a high right shoulder deformity, without functional limitation. Non-muscle-sparing thoracotomy was used on both patients with shoulder deformities. One patient in this contemporary series underwent aortopexy. The mortality rate for the series was 0%. The rates of leak and stricture were not significantly different between our cohort and those who underwent thoracoscopic repair (Table 2). In the contemporary group, 2.7% had leaks compared with 7.6% of those with thoracoscopic repairs (P ⫽ .2021). The anastomotic stricture rate was 5.5% for the contemporary series and 3.8% for the thoracoscopy series (P ⫽ .7177). Forty percent of patients in the contemporary series required ⱖ1 dilation, as opposed to 31.7% in the thoracoscopic group (P ⫽ .2642). Rates of recurrent fistula were also similar between the 2 contemporary studies, 2.7% in the open series and 1.9% in the thoracoscopic series (P ⫽ .6456). In the contemporary series,
J.M. Burford et al. Table 2
Contemporary Series Open Repair Esophageal Atresia
205
Comparison of current thoracotomy series with thoracoscopic series and historic thoracotomy controls for repair of EA/TEF
Variable
Current Series
Holcomb et al5
Historic Controls1,6–8
Number of patients Mean length of hospitalization (d) Anastomotic leak Anastomotic stricture Patients requiring ⱖ1 dilation Anastomotic revision Fundoplication Aortopexy Mortality Recurrent fistula
72 29.1 (5–159) 2.70% 5.50% 40% 2.70% 12.50% 1.30% 0% 2.70%
104 18.1 7.6% 3.8%* 31.7% 1.9% 24% 6.7% .9% 1.9%
340 NR 17.9% 16.7% NR NR 21% NR 73% 7.9%
(6–120) (P ⫽ .20) (P ⫽ .70) (P ⫽ .20) (P ⫽ 1.00) (P ⫽ .07) (P ⫽ .10) (P ⫽ 1.00) (P ⫽ 1.00)
(P ⫽ .0005) (P ⫽ .016) (P ⫽ .10) (P ⫽ .012) (P ⫽ .20)
NR ⫽ not reported for all studies included. *Stricture is defined as requiring ⬎4 dilations.
12.5% of patients required fundoplication as opposed to 24% of thoracoscopic patients (P ⫽ .0796). Conversely, when comparing the contemporary series with the historical controls, rates of leak and stricture were significantly lower, at 2.7% versus 17.9% for leak (P ⫽ .0006) and 5.3% versus 16.7% for stricture (P ⫽ .0098). There were also fewer recurrent fistulas in the contemporary series (2.7% vs 7.9%), although this did not reach statistical significance (P ⫽ .1883). Furthermore, there was also a decreased fundoplication rate in the contemporary series (12% vs 21%, P ⫽ .0772).1,6 – 8
Comments EA with distal TEF has traditionally been repaired via right thoracotomy. Advances in minimally invasive techniques have led to the development of a thoracoscopic approach suitable for repair of EA/TEF.3–5 A recent multiinstitutional review by Holcomb et al5 was published comparing thoracoscopic outcomes with those of several older series with repairs performed via thoracotomy. The study concluded that the thoracoscopic approach for repair of EA/TEF was safe and effective.5,9 A criticism of that report was that the thoracoscopic series was compared with historic rather than contemporary data.1,6 – 8,10 This review was designed to provide a contemporary series of patients undergoing thoracotomy for repair of EA/TEF for comparison with thoracoscopic repair. This study, performed from 1993 to 2008, provides a contemporary comparison with the thoracoscopic review, which used for comparison thoracotomy series conducted from 1971 to 1993, 1980 to 1984, 1982 to 1988, 1977 to 1985, and 1989 to 1999. This series reviewed 72 patients with short-gap EA and distal TEF who underwent repair via thoracotomy. This contemporary series of EA/TEF repair via thoracotomy was performed by on-call pediatric surgeons and yielded similar complication rates for anastomotic leak, stricture, recurrent fistula, and need for fundoplication compared with the thoracoscopic series performed by
experienced endoscopic surgeons, suggesting that the average pediatric surgeon can perform the standard open repair effectively without higher complication rates. Although the results were not significantly different between the 2 modern series, there were trends toward lower rates of leak and fundoplication in the contemporary thoracotomy series. Both thoracoscopic and contemporary series showed significant improvements in leak rate and stricture rate compared with the historic group. Although strictures were not found to be significantly different between the contemporary group and the thoracoscopic group, differences in the definition of stricture leave room for debate. There is considerable variability between studies, both historic and modern, as to the definition of stricture. Historic series used for comparison by Holcomb et al5 defined stricture in multiple ways, including varying numbers of dilations, any dilation after a certain period, or narrowing on radiographic imaging.1,5– 8,10 Our contemporary series defined stricture as requiring ⬎2 dilations. This definition gave a stricture rate of 5.5%. If the contemporary series redefined stricture as in the thoracoscopic series, requiring ⬎4 dilations, the stricture rate decreased to 1.3%. This new rate, although lower, remained not significantly different. It is likely that comparing stricture rates will remain difficult, as various studies consider different degrees of stricture significant. Children with EA are found to have a significant incidence of gastroesophageal reflux after repair. It is postulated that this incidence of reflux is related to esophageal dismotility, delayed gastric emptying, and a superior displacement of the gastroesophageal junction.11 As many as one third of patients with EA/TEF in some series require fundoplication for treatment of associated reflux.12 The rate of surgical treatment for reflux in this series was 12.5%, which is lower than both the thoracoscopic series and various historic series. These lower rates may be explained by variability in surgeon threshold for surgical treatment and better medical management of reflux disease in more recent times.13
206 One of the presumed advantages of thoracoscopy over thoracotomy is thought to be the potential reduction in musculoskeletal sequelae that may occur after thoracotomy.14,15 In our series, we found 2 patients with scoliosis and 2 patients with high right shoulder deformities. Both patients with scoliosis had mild disease, and neither required surgical correction. Neither patient with a high shoulder deformity had a functional limitation. The 4 patients with scoliosis or shoulder deformity were followed for an average of 11 years (range, 6 –15 years). Previous studies have reported a 17.9% rate of severe scoliosis and a 23.6% rate of “winged” scapula, which was not found in the contemporary series.14 We conclude from our contemporary series that the musculoskeletal complications historically seen with thoracotomy may no longer be an advantage of thoracoscopy.
Conclusions This review provides a modern comparison with the recent large multi-institutional series reviewing thoracoscopic repair of EA/TEF. Further analysis of this topic is warranted, and a randomized trial would be beneficial in revealing the true differences between the thoracoscopic and thoracotomy approaches for repair of short-gap EA with fistula. The question of whether the presumed benefits of thoracoscopy, such as reduced musculoskeletal complications and improved visualization, outweigh the trend toward higher rates of anastomotic leak, and an increased need for fundoplication remains. Finally, the fact that musculoskeletal sequelae, thought to be one of the main advantages of thoracoscopy, were not seen in this study suggests that thoracotomy in the modern era remains a comparable and effective technique for treatment of EA/TEF.
The American Journal of Surgery, Vol 202, No 2, August 2011
References 1. Manning PB, Morgan RA, Coran AG, et al. Fifty years’ experience with esophageal atresia and tracheoesophageal fistula. Ann Surg 1986; 204:446 –51. 2. Darzi SA, Munz Y. The impact of minimally invasive surgical techniques. Annu Rev Med 2004;55:223–37. 3. Lobe TE, Rothenburg SS, Waldschmitt J, et al. Thoracoscopic repair of esophageal atresia in an infant: a surgical first. Ped Endosurg Innov Tech 1999;3:141– 8. 4. Das K, Rothberg M. Thoracoscopic surgery: historical perspectives. Neurosurg Focus 2000;9:1–3. 5. Holcomb GW, Rothenberg SS, Bax K, et al. Thoracoscopic repair of esophageal atresia and tracheoesophageal fistula. Ann Surg 2005;242: 422– 8. 6. Spitz L, Kiely E, Brereton RJ. Esophageal atresia: five year experience with 148 cases. J Pediatr Surg 1987;22:103– 8. 7. Yanchar NL, Gordon R, Cooper M, et al. Significance of the clinical course and early upper gastrointestinal studies in predicting complications associated with repair of esophageal atresia. J Pediatr Surg 2001;36:815–22. 8. Randolph JG, Newman KD, Anderson KD. Current results in repair of esophageal atresia with tracheoesophageal fistula using physiologic status as a guide to therapy. Ann Surg. 1989;209:526 –30. 9. Nguyen T, Zainabad K, Bui T, et al. Thoracoscopic repair of esophageal atresia and tracheoesophageal fistula: lessons learned. J Laparoendosc Adv Surg Tech 2006;16:174 – 8. 10. Engum SA, Grosfeld JL, West KA, et al. Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades. Arch Surg 1995;130:502–9. 11. Foker JE, Linden BC, Boyle EM, et al. Development of a true primary repair for the full spectrum of esophageal atresia. Ann Surg 1997;226: 533– 43. 12. Bergmeijer JH, Tibboel D, Hazebroek FW. Nissen fundoplication in the management of gastroesophageal reflux occurring after repair of esophageal atresia. J Pediatr Surg 2000;35:573– 6. 13. Konkin DE, O’Hali WA, Webber EM, et al. Outcomes in esophageal atresia and tracheoesophageal fistula. J Pediatr Surg 2003;38:1726 –9. 14. Jaureguizar E, Vazquez J, Murcia J, et al. Morbid musculoskeletal sequelae of thoracotomy for tracheoesophageal fistula. J Pediatr Surg 1985;20:511– 4. 15. Soucy P, Bass J, Evans M. The muscle-sparing thoracotomy in infants and children. J Pediatr Surg 1991;26:1232–5.
