Esophageal Atresia and Tracheo-Esophageal Fistula

Abstract

Esophageal atresia may be suspected on prenatal sonography by the absence of a gastric bubble, polyhydramnios, and distension of the upper esophagus during swallowing attempts by the fetus. After birth, a child with EA presents with excessive salivation, mucus coming out of the mouth or nose, and noisy breathing with episodes of choking or cyanosis. These symptoms worsen if oral feedings are attempted. The diagnosis is confirmed when a 10F Replogle tube passed through the mouth or nose cannot be passed beyond about 10 cm. Smaller or more flexible catheters should be avoided because they can coil in the upper esophagus and give a false impression of esophageal patency. The tube is placed on suction to clear the excess secretions. An AP and lateral radiograph that includes the neck, chest and abdomen (“babygram”) is then obtained while gentle pressure is maintained on the Replogle and 10 mL of air is injected through it. This delineates very well the location of the upper pouch in relation to the vertebral bodies. Routine contrast studies are not indicated and might lead to aspiration.

Appendices

Summary Points

Diagnosis usually straightforward, may be suspected prenatally

Associated anomalies and prematurity common and may affect prognosis

Rigid bronchoscopy useful to eliminate a proximal fistula and verify location of distal fistula

For the usual EATEF, repair usually done under some tension; if unable to approximate despite usual maneuvers, Foker technique may be considered but classic method is to close distal esophagus and tack it to prevertebral fascia, feed by gastrostomy and return later (definitely safest method in premature babies or those with major associated anomalies)

Postoperative complications common

Most leaks can be treated conservatively and seal spontaneously

Strictures common, treated with balloon dilatation

All patients should be treated medically for GER initially

Be aware of symptoms of severe tracheomalacia (“dying spells”)

Despite these complications and the fact that all patients have esophageal dysmotility to varying degrees, long-term outcome is excellent in most.

Final word: an esophageal atresia is not a “TEF”

Editor’s Comments

Esophageal atresia with tracheoesophageal fistula is often considered the quintessential pediatric general surgical operation. It is usually straightforward, elegant, and gratifying, but only if the two ends come together easily. When they do not, it can be one of the most challenging and disheartening experiences for young and seasoned pediatric surgeons alike. The key in these situations is to always have a back-up plan and to avoid irrevocable errors: (1) excessive mobilization of the distal esophagus, which can compromise the blood supply; (2) multiple attempts to approximate the ends under tension, which results in loss of length when the sutures tear through, and (3) creation of a cervical esophagostomy, which often commits the patient to esophageal replacement and is associated with multiple complications.

I consider repair of the typical patient with EA to be a semi-elective operation, unless a patient with a distal fistula is intubated and mechanically ventilated, in which case I consider it an emergency. This is because regardless of the position of the tip of the endotracheal tube, positive pressure inevitably results in massive abdominal distension and ventilatory compromise. The thoracoscopic approach holds some promise of a better way and it is important that there are pioneers who are advancing the field, but, given the initial results, it is hard not to conclude that the leak and stricture rates are high, perhaps suggesting that the technique needs to be significantly more refined before it can become the standard approach.

For most patients, a small, posterolateral, muscle-sparing thoracotomy and extrapleural approach to the esophagus are preferable. Occasionally, one encounters a variation of the major vascular anatomy, such as double aortic arch or aberrant subclavian artery. Generally, it is possible to work around these structures but it is important to be vigilant regarding the vagus nerves and thoracic duct, whose course in the chest might be altered, exposing them to injury. When a right aortic arch is encountered, it is occasionally best to close the chest and reposition the patient for a thoracotomy on the opposite side. When approximating the two ends of the esophagus while the back-row sutures are being tied, ring forceps work well and are less traumatic than DeBakey forceps. The choice of suture material for the repair is individual, but using absorbable suture is probably better than using nonabsorbable suture, which has been known to cause foreign body reaction, granulation tissue, and, in rare cases, fistulas. The trachea can be oversewn with absorbable or nonabsorbable suture but it should probably always be a monofilament suture.

All patients with EA have reflux, but only those with failure to thrive, recurrent stricture, or complications will need a fundoplication. I still think a partial wrap, preferably done laparoscopically, better prevents dysphagia, but some surgeons insist that a loose Nissen works just as well. In patients who are difficult to extubate or who have other signs of severe tracheomalacia, one should have a low threshold for performing an aortopexy. It is a safe operation with excellent results and can be performed thoracoscopically. Finally, balloon dilatation under fluoroscopic guidance is clearly the best way to manage anastomotic strictures at any age. It is safer, more effective, and lasts longer than traditional bougienage.

Differential Diagnosis

Prenatally

• Absent/small stomach seen with neuromuscular disorders with inadequate swallowing

After birth

• Esophageal atresia with/without fistula, possibility of proximal fistula

• Esophageal perforation (especially prematures)

H-type

• LTEC, GE reflux

Diagnostic Studies

Attempt to pass a 10 Fr Replogle (8 Fr Replogle may be adequate in premature babies)

AP plain “babygram” with 10 mL of air injected in the Replogle

Contrast only if Replogle blocks at unusual level or blood-tinged aspirate, especially if premature

Bronchoscopy before positioning for thoracotomy/thoracoscopy

H-type: contrast injected by feeding tube in esophagus under fluoroscopy in prone or lateral decubitus; rigid bronchoscopy

Parental Preparation

Major surgery, multiple possible complications, but likelihood of a good long-term outcome in term babies without major associated anomalies.

Pamphlet explaining frequent complications and giving addresses of useful web sites

Preoperative Preparation

Replogle to suction

Cross-match

Prophylactic antibiotics

Rigid bronchoscopy preop +/− flexible intraop

Technical Points

Limited posterolateral thoracotomy

Use flexible bronchoscopy intraop PRN if unexplained desaturations

Use continuous traction on esophageal ends intraop PRN if tension excessive

Pure EA: scope to R/O proximal fistula, gastrostomy and wait

H-type fistula:

• Bronchoscopy to identify location of fistula and catheterize it.

• Cervical approach in most cases.