Ann Thorac Surg 1998;66:927
© 1998 The Society of Thoracic Surgeons
Original articles: general thoracic
Invited commentary
Helmut W. Unruh, MDa
a Section of Thoracic Surgery, University of Manitoba, Winnipeg, GH604-820 Sherbrook St, Health Sciences Center, Winnipeg, Manitoba, Canada, R3A 1R9
 |
Invited commentary
|
|---|
 Top
 Invited commentary
|
|---|
Isolated case reports of postoperative bronchopleural fistula closure with endoscopic techniques began to appear in the medical literature in the 1980s. Most reports, including this series by Hollaus and associates, employed fibrin glue. This report however, is important in that it is one of the largest reported series of patients treated in this manner and we can now begin to predict an expected outcome and ultimately define a role for this type of therapy in the management of patients with postoperative fistulas.
The first important observation to make from this article is that 35% of the patients in this series were not treated endoscopically. Hollaus and associates, referencing their results to the entire group of patients, report an overall closure rate of 35% with endoscopic gluing. Seven of these patients were, however, left with a chronic empyema cavity, so that only 20% were actually free of any drainage tubes or stomas in their chest walls. Another way to look at this is that 31% of patients in whom endoscopic gluing was attempted were cured and a total of 51% had the fistula closed, although some still required drainage. Size appears to be an important factor in predicting outcome. No fistulas greater than 8 mm were treated endoscopically. The average size in the cured group was 1 mm, whereas those that closed but still remained with an empyema cavity were 2.8 mm. The average size of fistulas associated with death was even larger, 4.2 mm. There was, however, considerable overlap, with failure of closure as well as death occurring in some patients with smaller fistulas. This type of treatment can also be protracted with up to five applications used by Hollaus and associates over the course of several months. Hollaus and associates also comment that submucosal injection is better that simple local instillation.
What, then, is the role of endoscopic closure in a patient who has pulmonary resection complicated by bronchopleural fistula formation? The initial therapy remains controlled drainage and management of sepsis. This will suffice for many patients, and chronic open window drainage is very acceptable therapy, especially for the elderly, medically complicated, or poorer prognosis patients. Although they do not provide the evidence to substantiate the fact, Hollaus and associates comment that smaller fistulas, those less than 3 mm, in their experience are best suited for this approach. Larger fistulas require more complex therapy such as placement of a bony matrix to retain the glue and multiple applications, often without success. More complex approaches such as thoracoplasty, muscle transposition, or transpericardial fistula closure should be reserved for the larger fistulas in patients who cannot or do not want to be managed by open drainage, for even they are associated with a significant complication and failure rate.
