Oesophageal atresia without fistula — anastomosis or replacement

Pediatr Surg lnt (1987) 2:216-222 Pediatric Surgery

International © Springer-Vertag 1987

Oesophageal atresia without fistula- anastomosis or replacement?

N. A. Myers, s . w . Beasley, A. W. Auldist, M. Kent, V. Wright*, and P. Chetcuti

Department of Surgery, Royal Children's Hospital, Flemington Road, Parkville, Victoria 3052, Australia

Abstract. At the Royal Children's Hospital, Mel- bourne, 553 babies with oesophageal atresia and / or tracheo-oesophageal fistula have been admit- ted during the past 39 years; 36 (6.5%) of these had oesophageal atresia without a tracheo-oeso- phageal fistula. Definitive surgery was performed in 27 patients: the primary definitive procedure was oesophageal anastomosis in 15 and oesopha- geal replacement in 12. Aspects of diagnosis and selection of the most appropriate treatment mo- dality are discussed, with the results of treatment presented. Our current policy is to perform an oesophageal anastomosis whenever possible, and this has been successful in 7 of the last t 1 patients.

Key words: Oesophageat atresia - Gasless abdo- men - Oesophageal replacement - Long gap

Introduction

Oesophageal atresia without an accompanying tracheo-oesophageal fistula is a relatively rare anomaly and varies between 5% and 10% in most reported series of patient~ with oesophageal atre- sia and tracheo-oesophageal fistula [1, 2, 5, 6, 10, l l , 14, 15, 17, 19, 20, 24, 26, 33, 35, 38]. Many centres therefore have a limited experience of the anomaly; at the Royal Children's Hospital in Melbourne there have been 36 patients who have had atresia without fistula. Thus, we have aver- aged tess than one patient a year during a period which has seen many advances in neonatal care.

*Present address: Queen Elizabeth Hospital for Children, Hackney Road, London E2 8PS, UK. Offprint requests to: N.A. Myers

Despite these advances and increasing experience controversy still exists as to the most appropriate management of the condition [3, 4, 7, 9, 12, 13, 25, 27, 28, 32]. This paper reports our experience and suggests a protocol which may assist in deter- mining whether primary anastomosis or replace- ment is appropriate.

Materials and methods

During the period 1948 - 1986, a total of 553 babies with oeso- phageal atresia and/or tracheo-oesophageal fistula were ad- mitted to the Royal Children's Hospital, Melbourne. Oe- sophageal atresia without fistula was seen in 36 (6.5%) of these (Table I).

Much of the data was collected prospectively by one of the authors (NAM) and supplemented by careful review of each patient's hospital records.

Table 1. 553 patients with oesophageal atresia and/or tracheo- oesophageal fistula (1948- 1986)

Oesophageal atresia with distal tracheo-oesophageal fistula

Oesophageal atresia with proximal tracheo-oesophageal fistula

Oesophageal atresia with proximal and distal tracheo-oesophageal fistula

Oesophageal atresia without fistula

Tracheo-oesophageal fistula ("H')

Total

475

11

1

36

30

553

Results

Polyhydramnios was present in 34 patients. Birth weight ranged from 1.6 to 3.6 kg (mean = 2.3 kg). One or more associated congenital anomalies were re¢ognised in 21 patients (Table 2). Of 3 t pa- tients in whom the gestation was recorded, t7 (55%) were tess than 37 weeks-

Table 2. Associated congenital anomalies in 21 patients

Chromosomal aberrations 6 Trisomy 21 . . . . . . . . . . . 5 Trisomy 18 . . . . . . . . . . . 1

Gastrointestinal 15 Anorectal . . . . . . . . . . . . 7 Other . . . . . . . . . . . . . . 8

Cardiac (excluding PDA) 8

Genitourinary 6

Skeletal 12 Vertebral . . . . . . . . . . . . 4 Digital . . . . . . . . . . . . . . 6 Other . . . . . . . . . . . . . . 2

Miscel laneous:hydrocephalus, hypospadias, cleft palate, biliary tract abnormali ty

Table 3. Patients not receiving definitive surgery

No. Year Procedure Comment

1 1948 Thoracotomy Gastrostomy Oesophagostomy

2 1957 Gastrostomy Oesophagostomy

1958 Nil

1959 Gastrostomy Oesophagostomy

1961 Nil

1968 Nil

1968 Nil

1970 Gastrostomy

1980 Nil

Died following rectoplasty

Hydrocephalus

Transposition of great vessels

Moribund at presentation

Down's syndrome

Trisomy E

Down°s syndrome

Down's syndrome; cardiac anomaly

0ESOPHAGEAL ATRESIA WITHOUT FISTULA 38 I

I ALIMENTARY CONTINUITY ESTABLISHED

27

OESOPIHAGEAL * OESOPHIAGEAL ANASTOMOSIS REPLACEMENT

15 12

~3~ 1 I NO DEFINITIVE OESOPHAGO- OESOPHAGO ** PROSTHETIC SURGERY (TABLE 3} COLOPLAST¥ GASTROPLASTY OESOPHAGUS

9 5 6 1 (9) (4) (1 ) (1)

Fig. 1. Management of the 36 patients (9: no definitive sur- gery; 27: alimentary continuity established). Number of deaths indicated in parenthesis. *2 cases required oesophageal re- placement; **one case required later coloplasty

217

A definitive attempt to provide alimentary continuity was made in 27 patients: the remaining 9 either had no surgery or a gastrostomy and /o r oesophagostomy alone (Table 3).

The management of patients in whom a defin- itive attempt was made to establish continuity is summarised in Fig. 1. One colonic replacement followed failed gastric replacement, and one greater curvature gastric tube followed failure of a Rehbein procedure. The prosthetic oesophagus using polythene was inserted in 1953. Two pa- tients who had a primary oesophageal anastomo- sis subsequently required oesophageal replace- ment. In 1 patient excision of a complete oeso- phageal web was the only procedure required.

Oesophageal anastomosis was achieved in 2 patients by mobilisation of the lower oesophagus through an abdominal approach at the same time as thoracotomy. The lower oesophagus was freed

Table 4. Oesophageal anastomosis

No. Year Age at Comment anastomosis

1 1963 16 weeks

2 1964 14 weeks

3 1964 9 weeks

4 1968 5 weeks

5 1968 4 weeks

6 1969 7 weeks

7 1971 8 weeks

8 1972 6 weeks

9 1974 12 weeks

10 1974 8 weeks

11 1975 14 weeks

12 1976 11 weeks

13 1977 11 weeks

14 1980 12 weeks

15 1985 In 1st week

Upper pouch bougienage Disruption Retrosternal coloplasty

Successful Death from other anomalies

Successful Upper pouch bougienage

Successful

Successful Anti-reflux operation

Successful Anti-reflux operation

Disruption Death

Successful Later resection of stricture

Oesophago-pulmonary fistula Death

Successful

Successful

Disruption Subsequent greater curvature tube

Successful (including abdominal mobilization) Anti-reflux operation

Successful (including abdominal mobilisation) Anti-reflux operation

Excision of complete web

218

Table 5, Survivors - anastomotic group

No. Comments Current status Follow-up (years)

1 Anastomotic leak - coloplasty No dysphagia 25 Early dysphagia

2 Early problems with gastro-oesophageal reflux and 22 oesophageal stricture. Stricture dilation

3 Early problems with reflux and stricture 18

4 18

5 17

6 14

7 12

8 11

9 10

10 9

11 6

12 1

Required resection of stricture and anti-reflux operation. Recent oesophagoscopy

Required anti-reflux operation

Haematemesis Early stricture resection

Stricture which responded to dilation

Stricture which responded to dilation

Anastomotic leak - greater curvature tube. Early dysphagia. Pycloroplasty for vomiting

Problems with reflux and infection of food early. Anti-reflux operation

Problem with reflux and dysphagia. Anti-reflux operation

Oesophageal web

Problem free No dysphagia

Occasional dysphagia (3 per year). No reflux problems

Moderate dysphagia with food "sticking" once a month. Heartburn frequently

Asthma

No dysphagia or recent haematemesis

Occasional dysphagia (once a month), mild "heartburn", mild reflux occasionally

Occasional dysphagia (2 - 3 per year). No problem with reflux

No problem with swallowing but daily episodes of reflux

Now foods stick only once per year. No reflux

Occasionally food sticks. Tends to avoid meat and poultry

Food sticks every 2 weeks

Table 6. Survivors - replacement group

No. Comments Current status Follow-up (years)

Gross mental retardation

Ulceration and perforation of greater curvature tube

Rehbein procedure - greater curvature tube

Kasai operation for biliary tract anomaly. Early respiratory tract infections

Greater curvature tube - coloplasty, multiple complications

Early - several oesophagoscopies

Chronic lung disease 26

Multiple problems. Dysphagia 18 Respiratory tract infections Vomiting/reflux

Recent diagnosis of oesophagitis. 15 Recent rectoplasty

Occasional regurgitation and heartburn. 14 Food sticks occasionally. Drinks with meals

No recent information 10

Regurgitation a problem and food sticks "about once a week"

within the hiatus and allowed to ride up into the chest, facilitating direct anastomosis of the oe- sophageal ends through the thoracotomy incision.

There were 6 deaths in the 12 patients with "primary" oesophageal replacement. The further 2 patients in whom primary anastomosis was un- successful, necessitating subsequent oesophageal replacement, are alive, bringing to 8 the total number of survivors with oesophageal replace- ment.

There were 3 deaths in the 15 patients having an oesophageal anastomosis: 2 died of complica- tions of the procedure and 1 of congenital sub- glottic stenosis. Ten are alive with a functioning oesophagus and 2 required replacement. The age at anastomosis varied from 4 weeks to 16 weeks (Table 4). The baby with a complete oesophageal web was repaired during the 1st week of life

Follow-up has ranged from 1 to 27 years: de- tails are summarised in Tables 5 and 6.

D i s c u s s i o n

Until the 1960s, oesophageal anastomosis in oe- sophageal atresia without fistula was considered impossible and the standard management in- volved establishment of a salivary fistula in the neck as a cervical oesophagostomy. In 1965 Koop and Hamilton [16, 22] introduced the concept of "staged" management for oesophageal atresia with distal fistula, advocating early gastrostomy

219

followed by a period of waiting before oesopha- geal anastomosis. This concept was extended to the baby with atresia without fistula. Preservation of the upper pouch produced new problems: the baby had to remain in hospital; there was risk of aspiration from accumulation of saliva in the upper pouch; and the child could not "sham feed". Aspiration was reduced by instituting an intensive programme of nursing care, attention to the posture of the baby, and continuous or fre-

Fig. 2. Contrast study of upper oesophageal pouch in baby with airless abdomen, Upper pouch fistula excluded

Fig. 3. Metal bougie in upper pouch - baby had airless abdomen

Fig. 4. X-ray taken at time of gastrostomy; sounds in both oesophageal segments

Fig. 5. Sounds introduced into both oesophageal segments; antero-posterior view in baby aged 3 weeks (same patient as Fig. 4)

220

quent suction of the upper pouch. If further eval- uation showed that oesophageal anastomosis was not feasible, a cervical oesophagostomy was per- formed. In 1963 upper pouch bougienage was in- troduced. In retrospect, we believe that the main advantage of bougienage was to encourage upper pouch preservation in babies who would other- wise have had a cervical oesophagostomy.

The use of upper pouch bougienage as first de- scribed by Howard and Myers [21] led to the de- velopment of other techniques such as Rehbein's Olive technique [29, 30], electro-magnetic bougie- nage [18], and circular myotomy [23, 31, 34, 36, 37, 39], each of which had its proponents. We no longer perform bougienage, but do encourage large bolus gastrostomy feeds with the object of promoting gastro-oesophageal reflux in the belief that spontaneous growth of the lower oesophageal segment may be increased by reflux.

Once a child with oesophageal atresia has been shown to have a gasless abdomen two ques- tions need answering: (1) is there a proximal tracheo-oesophageal fistula present? This is de- monstrated by a careful upper pouch contrast study and in our experience about one in four will have an upper-pouch fistula [8]; and (2) what is the length of the gap between the oesophageal ends and is it such that a primary anastomosis can be performed?

The gap between the oesphageal segments in isolated oesophageal atresia tends to be much greater than when a tracheo-oesophageal fistula is present. However, we know from our own experi- ence that direct end-to-end oesophageal anasto- mosis is possible in many of these patients and has been successfully performed in 7 of our last 11 patients.

To establish the suitability for anastomosis it is necessary to evaluate the lengths of both oesopha- geal segments. Initially, the length of the upper segment is shown by contrast study (Fig. 2) and subsequently by the use of a metal bougie (Fig. 3). The length of the lower segment is evaluated at the time of gastrostomy, by the introduction of a bougie or sound, and this can be combined with the simultaneous introduction of a bougie into the upper segment by the anaesthetist (Fig. 4). Further evaluation is performed at monthly intervals by simultaneous insertion of metal sounds into both oesophageal segments until a decision is made re- garding the definitive operative management (Figs. 5-7). After 3 months further decrease in the gap is unlikely. The vertebral column is used as the reference point for estimating the gap between the segments.

Fig. 6. Sounds introduced into both oesophageal segments; lateral view in baby aged 4 weeks (same patient as Fig. 4)

Fig. 7. Lateral view with sounds in both segments in baby aged 4 weeks. Ruler to facilitate assessment of progress

221

I GAP LESS THAN TWO VERTEBRAE

PROCEED TO ANASTOMOSIS

OESOPHAGEAL ATRESIA GASLESSARDOMEN

DYE STUDY OF UPPER POUCH TO EXCLUDE PROXIMAL FISTULA

GASTROSTOM¥

I EVALUATE LENGTH OF OESOPHAGEAL

SEGMENTS

t GAP BETWEEN TWO AND FOUR VERTEBRAE

RE-EVALUATE AT MONTHLY INTERVALS UP TO THREE MONTHS

1 PROCEED TO ANASTOMOSIS

I GAP GREATER THAN FOUR VERTEBRAE

I RE-EVALUATE AT MONTHLY INTERVALS UP TO THREE MONTHS

I PROCEED TO ANASTOMOSIS +/-- INTRA ABDOMINAL MOBILISATION OF LOWER OESOPHAGUS

I IF FAILS

CERVICAL OESOPHAGOSTOMY AND SUBSEQUENT OESOPHAGEAL REPLACEMENT

Fig. 8. Oesophageal atresia with airless abdomen: scheme of management

In our experience a gap of less than 2 verte- bral bodies justifies oesophageal anastomosis (Fig. 8). A gap between 2 and 4 vertebrae merits monthly re-evaluation in the expectation that anastomosis will be possible at a later stage even if it necessitates full mobilisation of the lower seg- ment. We have performed abdominal mobilisa- tion of the lower end of the oesophagus to allow successful intrathoracic oesophageal anastomosis on two occasions. Both patients have subsequent- ly required an abdominal Nissen fundoplication for severe gastro-oesophageal reflux but are now well. Circular myotomy was not used in this se- ries, although other authors [23, 31] have suggest- ed it may facilitate anastomosis when the gap is large.

When the radiological or operative evidence indicates that anastomosis is unlikely to succeed, or in the event of early and major anastomotic disruption, cervical oesophagostomy should be performed with oesophageal replacement when the baby reaches 10 kg. Our current preference is for replacement by means of a greater curvature tube of stomach.

C o n c l u s i o n s

The diagnosis of oesophageal atresia without fis- tula should only be accepted after a careful upper

pouch study to exclude a fistula. Initial manage- ment should involve a gastrostomy with evalua- tion of both oesophageal segments at the time. If the gap between the ends is small (less than two vertebral bodies) anastomosis can proceed. If the gap is larger' re-evaluation at 4-weekly intervals is indicated, as the gap tends to decrease up till about 12 weeks of age. Successful anastomosis is precluded in most cases where the gap is more than 4 vertebral bodies in length.

Where an anastomosis is performed it is to be anticipatedthat gastro-oesophageal reflux will oc- cur and may necessitate anti-reflux surgery.

The long-term follow-up suggests that oesoph- ageal anastomosis is preferable to replacement, and in our current practice is possible in most in- stances.

Acknowledgements. The authors are grateful to Miss Vicki Holt who typed the manuscript, Miss Jocelyn Brady, Research As- sistant, and to the Department of Radiology and Education Resource Centre, Royal Children's Hospital. The study was supported by a grant from the Royal Children's Hospital Re- search Foundation and Oesophageal Atresia Foundation.

References 1. Aberdeen E, Myers NA (1979) The esophagus. In:

Ravitch MM (eds) Pediatric Surgery, 3rd edn. Year Book Medical Publishers, Chicago, 450

2. Ahmed S (1972-1974) Congenital tracheo-oesophageal fistula without atresia (H-fistula). South Australian Clinics 6:388-391

3. Anderson KD (1984) Esophageal substitution. Proceed- ings of International Congress of Paediatric Surgery, Perth

4. Anderson KD (1984) Oesophageal substition. Aust NZ J Surg 54:447-449

5. Atwell JD, Beard RC (1974) Congenital anomalies of the upper urinary tract associated with oesophageal atresia and tracheo-oesophageal fistula. J Pediatr Surg 9: 825-831

6. Bishop PJ (1985) Transpleural repair of oesophageal atre- sia without a primary gastrostomy: 240 patients treated between 1951 and 1983. J Pediatr Surg 6:823-828

7. Booss D, Hollwarth M, Sauer H (1982) Endoscopic esophageal anastomosis. J Pediatr Surg 17: 138-143

8. Cass D, Auldist AW (1987) Oesophageal atresia with proximal tracheo-oesophageal fistula. Pediatr Surg Int 2: 212-215

9. Cohen D (1970) Oesophageal reconstruction using a gas- tric tube: a preliminary report. Aust Paediatr J 6:22-24

10. Cohen SJ (1965) Unusual types of oesophageal atresia and tracheo-oesophageal fistula. Diagnostic aids and proce- dures. Clin Pediatr 4:271-275 Cumming WA (1975) Esophageal atresia and tracheo-oe- sophageal fistula. Radiol Clin North Am 13:277-295 Foker JE, Ring WS, Varco RL (1982) Technique of je- junal interposition for esophageal replacement. J Thorac Cardiovasc Surg 83:928-933 Freeman NV, Cass DT (1982) Colon interposition: a mod- ification of the Waterston technique using the normal oes- ophageal route. J Pediatr Surg 17:17-21

11.

12.

13.

222

14. Haight C (1957) Some observations on esophageal atresia and tracheoesophageal fistulas of congenital origin. J Thorac Surg 34: 141-172

15. Haight C (1969) Congenital oesophageal atresia and tracheo-oesophageal fistula. In: Mustard WT, Pediatric surgery, 2nd Edn., Year Book Medical Publishers, Chica- go, pp 357-358

16. Hamilton JP (1966) Esophageal atresia. Technical points in the staged procedures leading to esophageal anastomo- sis. J Pediatr Surg 1 : 253-255

17. Hays DM, Woolley MM and Snyder WH (1966) Esopha- geal atresia and tracheo-esophageal fistula: management of the uncommon types. J Pediatr Surg 1 : 240-252

18. Hendren WH, Hale JR (1975) Electromagnetic bougie- nage to lengthen esophageal segments in congenital esophageal atresia. N Engl J Med 293:428-432

19. Holden MP, Woller GN (1970) Tracheo-oesophageal fis- tula and oesophageal atresia: results of 30 years experi- ence. Thorax 25:406-412

20. Holder TM (1964) Esophageal atresia and tracheo- esophageal fistula. A survery of its members by the Surgi- cal Section of the American Academy of Pediatrics. Pedi- atrics 34:542-549

21. Howard R, Myers NA (1965) Oesophageal atresia: a tech- nique for elongating the upper pouch. Surgery 58: 725-727 Koop CE, Hamilton JP (1965) Atresia of the esophagus. Increased survival with staged procedures in the poor-risk infant. Ann Surg 162:389-401 Livaditis A (1973) Esophageal atresia; a method of over- bridging large segmental gaps. Z Kinderchir 13:298-306 Louhimo I, Lindahl H (1983) Esophageal atresia. Primary results of 500 consecutively treated patients. J Pediatr Surg 18:217-299 Myers NA (1974) Oesophageal atresia: the epitome of modern surgery. Ann R Coll Surg Engl 54:277-287 Myers NA (1984) The long gap in oesophageal atresia. Why not the oesophagus? Proceedings of International Congress of Paediatric Surgery, Perth

22.

23.

24.

25.

26.

27. Puri P (1981) Delayed primary anastomosis following spontaneous growth of esophageal segments in esopha- geal atresia. J Pediatr Surg 16: 180-183

28. Randolph J (1986) In: Welch (eds) Pediatric surgery, p 682

29. Rehbein F, Schweder N (1971) Reconstruction of the esophagus without colon transplantation in cases of atre- sia. J Pediatr Surg 6:746-752

30. Rehbein F, Schweder N (1972) Neue Wege in der Rekon- struktion der kindlichen Speiser6hre. Dtsch Med Woch- enschr 97 : 757-770

31. Ricketts RR, Luck SR, Raffensperger JG (1981)Circular esophagomyotomy for primary repair of long-gap esopha- geal atresia. J Pediatr Surg 16:365-369

32. Ring WS (1982) Esophageal replacement with jejunum in children: an 18- to 33-year follow up. J Thorac Cardiovasc Surg 83:918-927

33. Roberts KD (1958) Congenital oesophageal atresia and tracheo-oesophageal fistula. Thorax 13 : 116-129

34. Rosselo PJ, Lebrow H, France AAR (1980) The technique of myotomy in esophageal reconstruction. Anexperimen- tal study. J Pediatr Surg 15:430-432

35. Swenson O (1958) Pediatric Surgery, 2nd edn., Appleton- Century-Crofts, New York, p 155

36. Takada Y, Kent G, Filler RM (1981) Circular myotomy and esophageal length and safe esophageal anastomosis. An experimental study. J Pediatr Surg 16:343-348

37. Vizas D, Ein SH, Simpson JS (1978) The value of circular myotomy for esophageal atresia. J Pediatr Surg 13: 357-359

38. Waterson DJ, Bonham-Carter RE, Aberdeen E (1963) Congenital tracheo-oesophageal fistula in association with oesophageal atresia. Lancet 2:55-57

39. Zigiotti GL, Domini R (1974) La miotomia cincolare del moncone superiore secondo livaditis nell' anastomosis primara per atresia esofager. Riv Chit Pediatr 16:49

Received March 4, 1987