Ann Thorac Surg 2003;76:304-306
© 2003 The Society of Thoracic Surgeons
How to do it
Repair of a recurrent benign Tracheoesophageal fistula with a Gore-Tex membrane
Romeo Bardini, MDa*,
Valentina Radicchi, MDa,
Paolo Parimbelli, MDa,
Sara Maria Tosato, MDa,
Surendra Narne, MDb
a Clinica Chirurgica I, Dipartimento di Scienze Chirurgiche e Gastroenterologiche, Padova, Italy
b Chirurgia Endoscopica delle vie Aeree, Università degli Studi di Padova, Padova, Italy
Accepted for publication November 25, 2002.
* Address reprint requests to Dr Bardini, Dipartimento di Scienze Chirurgiche e Gastroenterologiche, Clinica Chirurgica I, via Giustiniani 2, 35100 Padova, Italy
e-mail: romeo.bardini{at}unipd.it
 |
Abstract
|
|---|
A case is reported of a recurrent postintubation tracheoesophageal fistula treated with the interposition of a Gore-Tex patch between the trachea and the esophagus.
|
Introduction
|
|---|
In spite of low-pressure cuff tracheal tubes, prolonged mechanical ventilation still accounts for the majority of acquired tracheoesophageal nonneoplastic fistulas [1, 2]. The fistula is usually caused by ischemia of the tracheal and esophageal walls, due to the overinflation of the cuff and favored by the presence of a rigid nasogastric tube [2].
The treatment is usually the direct closure of both tracheal and esophageal defects, with the interposing of a muscle flap or, more recently, tracheal resection and anastomosis with primary esophageal repair [2–4]. The early recurrence of the fistula after tracheal resection is rare, but it is always a life-threatening condition [2, 3].
|
Technique
|
|---|
The patient was a 70-year-old woman who underwent a coronary bypass using the mammary artery in December 2000 and a new bypass with the autologous saphenous vein in January 2001. The postoperative course was complicated by a poststernotomy mediastinitis and respiratory distress that required 33 days of tracheal intubation. Immediately after extubation, the patient complained of dysphagia and coughing, and signs of sepsis were evident. An esophagoscopy and a tracheoscopy demonstrated the presence of a large tracheoesophageal fistula. The patient was intubated once more; a feeding jejunostomy and a gastrostomy were performed. No nasogastric tube was used. The cuff of the orotracheal tube was positioned below the fistula under visual control. Antibiotics, enteral and parenteral feeding, and mechanical ventilation helped to clear up the sepsis and respiratory distress.
In April 2001, the patient was transferred to the Otolaryngology Department of our University Hospital. An esophagoscopy located the fistula immediately below the upper esophageal sphincter, and a tracheoscopy revealed the absence of the posterior wall of the trachea from the first to the fifth tracheal ring. Through a U-shaped cervicotomy, tracheal resection, and anastomosis, esophageal repair and interposing of the left sternocleiodomastoid muscle were performed. The endotracheal tube was removed 24 hours after surgery. On the 5th postoperative day, coughing and dyspnoea occurred, and a tracheoscopy demonstrated an edema and granulating tissue on the posterior aspect of the tracheal suture. A recurrence of the tracheoesophageal fistula was clinically evident 1 week later. This was confirmed by tracheoscopy, esophagoscopy, and esophagography. The patient needed to be reintubated because of serious respiratory insufficiency, and was then transferred to the intensive care unit.
Our opinion was required, and we discussed the patient’s situation with the anesthesiologists, who pointed out the absolute need to eliminate saliva inhalation as soon as possible, as the inflated cuff of the tracheal tube was not sufficient to hinder the passage of saliva into the trachea and bronchi. Due to the patient’s serious general condition, the anesthesiologists consented to minor surgery at the cervical level only.
We suggested a laryngectomy, definitive tracheostomy, and esophageal repair as life-saving procedure. The patient refused a definitive tracheostomy; therefore, we made the decision to perform a new conservative treatment.
Through a left lateral cervicotomy, the fistula was difficult to isolate due to marked inflammation of tissues and the interposed sternocleidomastoid muscle. The left recurrent laryngeal nerve could not be identified. The esophageal suture was dehiscent, as were the posterior and lateral aspects of the tracheal suture, and the membranaceous part of the trachea was lacking at least for 1 cm below and above the suture line (Fig 1A).
Two absorbable stitches were used to approximate the left lateral wall of the trachea, and the membranaceous part was left open due to the distance of the margins. At the level of the esophageal leak, only the mucosa could be repaired with a single layer of running absorbable suture after removing of the necrotic margins. Esophageal muscle was unavailable to cover the esopahgeal suture line (Fig 1B). Due to the high risk of a new recurrence of the fistula as a consequence of the tension on the fragile esophageal mucosa, a patch of Gore-Tex membrane (W.L. Gore & Associates, Flagstaff, AZ) was interposed between the esophageal suture and the posterior tracheal defect (Fig 2A).
The patch was fixed using absorbable stitches to the right side of the esophagus and to the upper esophageal sphincter to keep it in place. The left lateral extremity of the patch was sutured to the skin, allowing eventual removal. A small soft drain was positioned on the left side of the esophagus, close to the esophageal suture. Therefore, a complete separation between the esophageal suture and the tracheal defect was obtained by a mechanical barrier (Fig 2B). The endotracheal tube was removed on the 4th postoperative day. As expected, an esophageal leakage was already clinically evident because saliva was present along the cervical drain. At tracheoscopy, the Gore-Tex membrane was clearly visible through the hole in the membranaceous part of the trachea, but no saliva could be found in the airways. On the 13th postoperative day, a mild tracheal stenosis, due to the presence of granulating tissue at the level of the suture, required the positioning of a Montgomery T-tube.
View larger version (62K):
[in this window]
[in a new window]
|
Fig 1. (A) Surgical view of the recurrent tracheoesophageal fistula: dehiscence of the esophageal suture and wide opening of the posterior and lateral wall of the trachea. (B) Suture of the cartilaginous aspect of the trachea and closure of the esophagus.
|
|
View larger version (38K):
[in this window]
[in a new window]
|
Fig 2. (A) The Gore-Tex patch is sutured to the right side of the esophagus and to the upper esophageal sphincter. A separation between the tracheal defect and the esophageal suture is obtained. The arrow indicates the posterior part of the trachea that was left open to close spontaneously. (B) Final view: a soft drain is positioned next to the esophageal suture.
|
|
The salivary output, about 30 mL per day, spontaneously stopped 3 weeks after the operation. After an esophagography, which confirmed the complete healing of the esophageal leak, the patient was able to start drinking and eating soft food. A tracheoscopy surprisingly demonstrated also the complete repair of the posterior tracheal wall.
On the 33rd postoperative day, the cervical drain was removed, as was the Gore-Tex patch by applying a gentle traction on the margin sutured to the skin. The patient was discharged a few days later. Further tracheoscopies during the follow-up demonstrated no tracheal stenosis but a hypomobility of the left vocal cord.
|
Comment
|
|---|
Early recurrence of a tracheoesophageal fistula after a long tracheal resection and anastomosis along with esophageal repair is rare, but it is a life-threatening complication, as reported by Macchiarini and associates [2] and Mathisen and associates [4].
The poor general condition of our patient allowed only for a cervical approach to eliminate saliva inhalation. Alternative surgery included an esophageal diversion or laryngectomy with definitive tracheostomy. An esophageal diversion was nevertheless technically impossible and the patient refused a permanent tracheostomy. Then, as more conventional techniques could not be employed, we interposed a patch of Gore-Tex membrane between the esophageal suture and the tracheal wall, which perfectly acted as a mechanical barrier, thus hindering the passage of saliva into the trachea through the esophageal leak. This technical artifice allowed the spontaneous healing of both esophageal and tracheal leaks, preventing further pulmonary complications.
In our opinion, this technical detail is to be warranted for the treatment of tracheoesophageal fistulas when other more conventional techniques can not be used.
|
References
|
|---|
- Thomas A.N. Management of tracheo-esophageal fistula caused by cuffed tracheal tubes. Am J Surg 1972;124:181-189.[Medline]
- Macchiarini P., Verhoye J.P., Chapelier A., Fadel E., Dartevelle P. Evaluation and outcome of different surgical techniques for postintubation tracheoesophageal fistulas. J Thorac Cardiovasc Surg 2000;119:268-276.[Abstract/Free Full Text]
- Semlacher RA, Bharadwaj J, Nixon A. Management of a post-traumatic tracheo-esophageal fistula following primary repair. J Cardiovasc Surg 1994;35:83–7
- Mathisen D.J., Grillo H.C., Wain J.C., Hilgenberg A.D. Management of acquired nonmalignant tracheoesophageal fistula. Ann Thorac Surg 1991;52:759-765.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
Y. Han, K. Liu, X. Li, X. Wang, Y. Zhou, Z. Gu, Q. Ma, T. Jiang, L. Huang, T. Zhang, et al.
Repair of massive stent-induced tracheoesophageal fistula
J. Thorac. Cardiovasc. Surg.,
April 1, 2009;
137(4):
813 - 817.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
