Ann Thorac Surg 1996;62:286-288
© 1996 The Society of Thoracic Surgeons
Case Report
Trachea–Innominate Artery Fistula and Concomitant Critical Cerebrovascular Disease
Michael D. Black, MD,
Farid M. Shamji, MD,
Thomas R. J. Todd, MD
Division of Thoracic Surgery, Ottawa Civic Hospital, University of Ottawa, Ottawa, Ontario, Canada
Accepted for publication February 13, 1996.
 |
Abstract
|
|---|
To avoid the laborious task of investigating the cerebrovascular circulation in the midst of a trachea–innominate artery fistula, we strongly recommend preoperative cerebrovascular investigations in all patients about to undergo mediastinal tracheostomy. Paramount to this dictum remains the possibility of asymptomatic cerebrovascular disease. Inadequate preoperative cerebrovascular assessment may result in, as described in this report, the possibility of significant postoperative neurologic morbidity or mortality. Angiography should assist the surgeon in deciding which method of cerebral arterial reconstruction is best suited to the individual circumstance. We recommend the avoidance of innominate artery reconstruction even with the interposition of autologous tissues, as the operative field remains grossly infected.
|
Introduction
|
|---|
Trachea–innominate artery fistula has been seen after tracheostomy or after complex segmental tracheal resections. Treatment, always urgent, carries a serious risk of stroke. Our case and its unusual management strategy specifically addresses this problem. If the patient can be temporarily stabilized by securing an airway, demonstration of the intracranial circulation is desirable, so that the fistula can be repaired, minimizing the risk of serious neurologic deficit.
A 70-year-old man, who previously made a complete neurologic recovery after a cerebrovascular accident, had persistent complex pharyngocutaneous and tracheoesophageal fistulas after laryngectomy carried out in 1989 for squamous cell carcinoma (stage T4 N0 M0). He had received a full course of preoperative radiation therapy. Four attempts at closure of these nonmalignant fistulas using multiple flaps and autologous tissue failed. To maintain nutrition and to avoid persistent soilage of the tracheobronchial tree, a feeding jejunostomy was performed in concert with placement of a Montgomery T-tube stent in the esophagus through the tracheoesophageal fistula and catheter intubation of the pharyngocutaneous fistula.
The patient underwent a cervical exenteration including manubriectomy, total esophagectomy and pharyngectomy, pharyngogastric anastomosis, and pyloromyotomy. Mediastinal tracheostomy was fashioned by delivering the trachea inferior and to the right of the innominate artery. The nonmalignant fistulous defect in the membranous trachea was repaired primarily with interrupted 3-0 Vicryl (Ethicon, Somerville, NJ) sutures rather than tracheal resection due to the short length of available distal trachea. Omentum was mobilized during preparation of the gastric conduit and delivered to the neck to provide mechanical separation of the trachea and innominate artery while providing a rich vascular network to the previously irradiated trachea.
Seventeen days later, a 3-mm pharyngogastic anastomotic leak developed, which healed spontaneously. There was also sloughing of the tracheostomy stoma over a 5-mm distance. He had several small bleeds during the preceding 3 days and was initially resuscitated by the ear, nose, and throat specialist. The source of bleeding was thought to be granulation tissue. Twelve weeks after the operation the patient required urgent treatment for significant airway hemorrhage (300 mL), which required immediate resuscitation by the thoracic surgeon including the use of tracheal intubation with a cuffed tracheostomy tube, which temporarily controlled the bleeding. A trachea–innominate arterial fistula was suspected, and a cerebral angiogram was obtained to assess the adequacy of the intracranial circulation because of the previous cerebrovascular accident (Fig 1
). The angiogram depicted complete occlusion of the left internal carotid artery at its origin and stenosis of the proximal right internal carotid artery and vertebral arteries. The operative procedure to control the fistula and maintain adequate blood flow to the brain consisted of first performing a right axillary–femoral arterial bypass graft, followed by resection of the 3-mm fistula and oversewing the divided ends of the innominate artery. The defect in the trachea was approximated with 3-0 Vicryl suture, and autologous tissue was interposed. The patient recovered without neurologic deficit. The right axillo–femoral graft was studied 1 month after the operation and demonstrated blood flow from the femoral to axillary artery.
View larger version (59K):
[in this window]
[in a new window]
|
Fig 1. . Cerebroangiogram obtained to ascertain the adequacy of the intracranial circulation. Note in (A) stenosis in the proximal right internal carotid artery (large arrow) and occluded left internal carotid artery (small arrow). In (B), the small arrow points to stenosis in the right vertebral artery at its origin.
|
|
|
Comment
|
|---|
Trachea–innominate artery fistula is a serious and life-threatening complication of tracheostomy and tracheal reconstructive operations. Most patients with a massive bleed do not survive [1]. The factors that predispose to the formation of this dreaded complication include a low tracheostomy stoma, over-inflation of the tracheostomy cuff, improper positioning of the tracheal stoma, localized sepsis, and an abnormally horizontal or high innominate artery. The fistula is usually small, being 3 to 5 mm in greatest dimension. Significant airway hemorrhage usually occurs after previous sentinel bleeds. The immediate maneuvers required to control bleeding have been previously well described by Cooper [2]. The seriousness of the fistula is such that Grillo and Mathisen [3] recommend prophylactic division of the innominate artery to avoid this complication after cervical exenteration. In addition, Grillo and Mathisen also recommend preoperative cerebral angiograms and electroencephalographic monitoring in every patient who undergoes mediastinal tracheostomy because of the possibility of the complications described in this report.
Division of the innominate artery carries a significant risk of neurologic sequelae (approximately 10%) [4, 5]. It therefore becomes important to perform cerebral angiography before cervical exenteration when possible, and particularly if there is a history of cerebrovascular disease. This could alter the initial intraoperative management, ie, prophylactic division of the innominate artery, with or without an additional operation to protect circulation to the brain.
The surgical management of this fistula requires division of the innominate artery and separation of the oversewn arterial ends from the trachea. Although our patient did not have preoperative angiography initially, cerebrovascular anatomy was eventually delineated by angiography to properly manage the subsequent fistula. Several caveats are important with respect to angiography: It should be readily available, especially to those with a previous history of cerebrovascular accidents. It should only be performed on stable patients with no evidence of continuing life-threatening bleeding. The airway should never be compromised. The patient should be accompanied at all times by a surgeon or senior resident/fellow experienced enough to terminate the procedure and resuscitate the patient if necessary. The best and most viable vascular option to protect the brain before dividing the innominate artery in our patient was the axillo–femoral graft described. Other authors have described axillo–axillary bypass, carotid–carotid bypass, and aorta–right common carotid artery bypass [2, 4, 5].
In conclusion, the physician and surgeon should maintain a high index of suspicion of trachea–innominate artery fistula in any patient with a recent tracheal operation and subsequent hemoptysis. The assumption that the bleed is from a less significant cause such as granulation tissue can be lethal. Management of the fistula concomitant with known cerebrovascular disease requires careful preoperative planning including cerebral angiography. Of greater concern are patients with asymptomatic cerebrovascular disease [1]. If the patient can be temporarily stabilized, anatomic demonstration of cerebral circulation before division of the innominate artery is recommended [6]. The angiogram will assist the surgeon in deciding which method of arterial reconstruction is best suited to the individual circumstance. We do not advocate the reconstruction of the innominate artery even with the interposition of autologous tissues, because the operative field is grossly infected [7, 8].
|
Acknowledgments
|
|---|
We thank Dr Graeme G. Barber (Department of Vascular Surgery, Ottawa Civic Hospital) and Dr Joseph G. Marsan (Department of Otolaryngology, Ottawa Civic Hospital) for assistance with the vascular reconstruction and for providing this fascinating patient, respectively.
|
Footnotes
|
|---|
Address reprint requests to Dr Black, Division of Pediatric Cardiovascular Surgery, The Hospital for Sick Children, 555 University Ave, Toronto, Ont, Canada M5G 1X8.
|
References
|
|---|
- Jones JW, Reynolds M, Hewitt RL, Drapanas T. Tracheoinnominate artery erosion. Ann Surg 1976;184:194–204.[Medline]
- Cooper JD. Trachea–innominate artery fistula. Ann Thorac Surg 1977;24:439–47.[Abstract]
- Grillo H, Mathisen DJ. Cervical exenteration. Ann Thorac Surg 1990;49:401–8.[Abstract]
- AbuRahma AF, Robinson PA, Khan MZ, Khan JH, Boland JP. Brachiocephalic revascularization: a comparison between carotid–subclavian artery bypass and axillo–axillary artery bypass. Surgery 1992;112:84–91.[Medline]
- Melliere D, Becquenmin JP, Benyahia NE, Ecollan P, Fitoussi M. Atherosclerotic disease of the innominate artery: current management and results. J Cardiovasc Surg 1992;33:319–23.[Medline]
- Lane EE, Temes GD, Anderson WH. Tracheal–innominate artery fistula due to tracheostomy. Chest 1975;68:678–83.
- Schuch D, Wolff L. Repair of mycotic aneurysm of the innominate artery with homograft tissue. Ann Thorac Surg 1991;52:863–4.[Abstract]
- Nunn DB, Sanches-Salazar AA, McCullagh JM. Trachea innominate artery fistula following tracheostomy. Ann Chir Gynaecol Fenn 1979;68:9–17.
This article has been cited by other articles:
|
|
|
|
 
M. G. Gasparri, A. C. Nicolosi, and G. H. Almassi
A novel approach to the management of tracheoinnominate artery fistula
Ann. Thorac. Surg.,
April 1, 2004;
77(4):
1424 - 1426.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. Kapural, J. Sprung, I. Gluncic, M. Kapural, S. Andelinovic, D. Primorac, and P. K. Schoenwald
Tracheo-Innominate Artery Fistula After Tracheostomy
Anesth. Analg.,
April 1, 1999;
88(4):
777 - 777.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
