Ann Thorac Surg 2009;88:1698-1700. doi:10.1016/j.athoracsur.2009.04.080
© 2009 The Society of Thoracic Surgeons
Case Reports
Repair of a Postesophagectomy Bronchogastric Tube Fistula With Polyglactin Mesh Supported With a Muscle Flap
Giuseppe Marulli, MDa,
Romeo Bardini, MDb,
Luigi Bortolotti, MDa,
Abdel-Mohsen Hamad, MDa,
Federico Rea, MDa,*
a Department of Cardiothoracic and Vascular Sciences, University of Padova, Padova, Italy
b Department of Gastroenteric Surgical Sciences, University of Padova, Padova, Italy
Accepted for publication April 15, 2009.
* Address correspondence to Dr Rea, Department of Cardiothoracic and Vascular Sciences, Policlinico di Padova, Via Giustiniani, 2, Padova, 35100, Italy (Email: federico.rea{at}unipd.it).
 |
Abstract
|
|---|
A bronchogastric fistula is a very rare complication of transthoracic esophagectomy. We report a case of bronchogastric fistula after transthoracic esophagectomy caused by dehiscence of the staple line in the gastric tube, with subsequent erosion into the right main bronchus. The patient was managed successfully in two surgical stages. First, the bronchial defect was repaired using a polyglactin mesh covered by a serratus anterior muscle flap. Two months later, the esophagogastric continuity was restored with colon interposition.
 |
Introduction
|
|---|
A bronchogastric fistula after esophageal surgery is a very rare, often fatal complication [1–4]. Regardless of the surgical approach for the esophagectomy, a stomach tube is commonly fashioned to re-establish the continuity of the digestive tract. Surgical staplers are usually used for this purpose; stapling has the advantage of saving time without increasing the complication rate. However, the gastric staple line may leak, leading in rare cases to erosion of the adjacent bronchial wall.
A 63-year-old man underwent an esophagectomy with an intrathoracic gastroesophageal anastomosis to treat cancer of the esophagus in another hospital. His early postoperative course was normal. The chest drainage became bilious beginning on postoperative day 4. On postoperative day 8, the patient had a cough develop with intermittent hemoptysis and dyspnea. A barium swallow study showed contrast leakage into the right bronchial tree. Fiberoptic bronchoscopy revealed a 2 x 2-cm communication between the membranous part of the right main bronchus and the gastric tube with bilious fluid flowing into the bronchial tree. The general condition of the patient deteriorated rapidly with the development of respiratory distress and progressive abdominal distension. The patient was transferred to our center for further management; he was stabilized medically and operated on. Through a right re-thoracotomy, the esophagogastric anastomosis was found to have dehisced completely, and a fistula between the right main bronchus and gastric tube at the site of the staple line was identified (Fig 1A). Owing to the contaminated field, definitive repair was deferred; the stomach was debrided, closed, and reintroduced into the abdomen with the creation of a feeding jejunostomy, and a cervical esophagostomy was performed.

View larger version (173K):
[in this window]
[in a new window]
|
Fig 1. (A) Intraoperative view showing the esophagogastric anastomosis completely dehiscent with evidence of the esophageal stump (thick arrow) with a nasogastric tube inside (asterisk); the staple line of the gastric tube had opened completely (star), and there was a fistula in the right main bronchus (thin arrow). (B) Polyglactin mesh was sutured over the bronchial defect (arrow), and (C) a serratus anterior muscle flap was used to cover the mesh (arrow). (D) Follow-up bronchoscopy showing healing of the defect, with complete mucosal coverage (arrow).
|
|
The large size of the bronchial defect precluded direct closure, necessitating some sort of muscle flap. Therefore, a pedicled flap of the serratus anterior was prepared and passed through a window created by removing a small segment of the fourth rib. To avoid bulging of the transposed muscle into the bronchial lumen, we initially closed the fistula with absorbable polyglactin (Vicryl) mesh (Ethicon Inc, Somerville, NJ) sutured with multiple single stitches of 4.0 polyglactin to the borders of the defect. Then, the muscle flap was used to cover the mesh (Figs 1B and 1C).
During the operation, single-lung ventilation was not sufficient to maintain oxygenation, and high-frequency jet ventilation on the right side was required. After the operation, the patient was extubated in the operating room and transferred to the intensive care unit. In the following days, frequent bronchoscopic toilette was performed. Apart from a small hematoma of the chest wall, no complications were observed postoperatively, and the patient was discharged home on postoperative day 17.
Bronchoscopy performed 1 month after the operation showed coverage of the mesh with normal bronchial mucosa (Fig 1D). Three months later, the patient underwent restoration of esophagogastric continuity with colon interposition from which he recovered uneventfully.
 |
Comment
|
|---|
In the early postoperative period, dyspnea and coughing after transthoracic esophagectomy are often signs of pulmonary embolism, atelectasis, or aspiration pneumonitis. Late in the postoperative course, however, an acute onset of these symptoms together with hemoptysis is uncommon and deserves bronchoscopic assessment with a high index of suspicion for the development of a fistula. The rapidity of diagnosis of a fistula between the esophagus or stomach and the airways is of paramount importance. Late diagnosis with prolonged aspiration leads to serious, extensive damage on the bronchial wall resulting from the erosive action of the digestive fluids with subsequent respiratory failure and sepsis.
There is no standard management of esophago-respiratory or gastro-respiratory fistulas. In the presence of mediastinal contamination, as in our case, a technique similar to esophageal exclusion (end-cervical esophagostomy and placing the stomach back into the abdomen with a feeding jejunostomy) is preferable. For airway management, two options are endobronchial stenting and surgical closure of the fistula. Surgical repair, if possible, is the best choice, and in most of the reported cases, the bronchial fistula was repaired primarily with direct sutures and subsequent viable tissue interposition [5]. In our case, the extent of the bronchial wall loss precluded repair with direct sutures or sleeve resection of the right main bronchus. Therefore, we decided to cover the defect with a serratus anterior muscle flap. As the defect was located entirely within the membranous portion, we first covered it with a polyglactin mesh to temporarily create a stout support to avoid bulging of the muscle into the airway lumen. One month after discharge, bronchoscopy revealed complete mucosal coverage of the mesh, with no healing problems or excessive scar.
In conclusion, a gastrobronchial fistula is an extremely rare, life-threatening, complication of transthoracic esophagectomy. Respiratory distress with cough and hemoptysis are the presenting symptoms and should be investigated promptly. In our case, polyglactin mesh provided good support for muscle flap repair of the bronchial defect and provided a surface for mucosal coverage.
 |
References
|
|---|
- Pramesh CS, Sharma S, Saklani AP, Sanghvi BV. Broncho-gastric fistula complicating transthoracic esophagectomy Dis Esophagus 2001;14:271-273.[Medline]
- Baisi A, Fumagalli U, Rosati R, Marinoni M, Bonavina L. Successful primary treatment of bronchial fistula complicating esophagogastrectomy Dis Esophagus 1994;7:209-211.
- Stringer DA, Pablot SM. Broncho-gastric tube fistula as a complication of esophageal replacement J Can Assoc Radiol 1985;36:61-62.[Medline]
- Kron IL, Johnson AM, Morgan RF. Gastrotracheal fistula: a late complication after transhiatal esophagectomy Ann Thorac Surg 1989;47:767-768.[Abstract/Free Full Text]
- Mangi AA, Gaissert HA, Wright CD, et al. Benign bronchoesophageal fistula in the adult Ann Thorac Surg 2002;73:911-915.[Abstract/Free Full Text]