LOUISVILLE, KY. — Treatment of enterocutaneous fistulas continues to rely primarily on surgery, augmented in some cases by octreotide or vacuum-assisted closure, according to a review of 106 patients.
The management of enterocutaneous fistulas continues to be a problem because of 5%–15% mortality, a spontaneous healing rate of less than 30%, and an overall healing rate hovering around 80%–90%, Dr. John M. Draus said at the annual meeting of the Central Surgical Association.
He and his colleagues in the surgery department at the University of Louisville, Ky., reviewed cases of gastrointestinal-cutaneous fistulas that occurred from 1997 to 2005 at two large teaching hospitals. The investigators excluded patients with inflammatory bowel disease and those with esophageal, pancreatic, or anorectal fistulas.
Among the 106 patients, fistulas most often resulted from a previous operation (81) and originated from the small bowel (67), colon (26), stomach (8), or duodenum (5). These operations were performed to treat cancer, adhesions, small bowel obstruction, gynecologic problems, or ventral hernias. Other fistula etiologies included trauma (15), hernia mesh erosion (6), diverticulitis (2), and radiation (2).
The group of patients comprised 31 with high-output fistulas (leaking more than 200 mL/day), 44 with low-output fistulas (less than 200 mL/day), and 31 whose fistula output was managed with a single gauze dressing. The analysis showed that the rate of healing or need for operation did not differ among patients with low- or high-output fistulas.
In general, the initial treatment plan for each of the patients consisted of the correction of fluid and electrolyte imbalances, nutritional support through total parenteral or enteral feeding, wound care, diagnostic imaging, early recognition of sepsis, drainage of abscesses, and an operation when necessary.
Among 13 patients who received treatment with vacuum-assisted closure (VAC), all had improved wound care and overall healing, but only 1 patient had complete healing that was attributed to VAC. The other 12 required an operation. No septic complications occurred with the use of VAC and fistula output did not increase, Dr. Draus noted.
VAC should be used in the subset of patients whose wounds are free of active infection, have no exposed bowel, and have a healthy layer of surrounding soft tissue, he advised.
The application of fibrin glue resulted in only one completely healed fistula among eight patients, all of whom had small bowel fistulas with high output. The glue transiently healed one patient's fistula for 11 days.
Of 24 patients who received treatment with octreotide, 8 responded with at least a 50% decrease in fistula output within 3 days; 4 of these 8 patients healed without an operation.
“Octreotide responders appear more likely to heal their fistula without operation” than those who receive other nonoperative modalities, Dr. Draus said. He recommended that most patients with an enterocutaneous fistula be given at least a 3-day trial of octreotide. If there isn't a dramatic decrease in fistula output by the end of 3 days, there is probably not much benefit in continuing octreotide, he suggested. Most patients in the study began receiving 100 mcg of octreotide three times per day, but the dose was increased to 500 mcg three times per day in one patient.
Audience member Dr. Bruce A. Harms of the University of Wisconsin, Madison, found this recommendation to be “a little bit of a stretch” in light of the fact that there is no hard efficacy data to back up Dr. Draus' advice. Octreotide should fall into the category of an ancillary treatment, said Dr. Harms.
Of 77 patients who had a planned operation, 69 (90%) healed. The average time from fistula formation to operation was 12 weeks. Some audience members said that 12 weeks was too short of an interval to wait to repair the fistulas, but Dr. Draus noted that the operation occurred after 3–6 months in about half of the patients.
“It is frustrating that in 2006 we still have made so little progress in treating fistulas with nothing but an operation,” said audience member Dr. Merril Dayton of the State University of New York at Buffalo.
Fistulas healed in 82% of patients regardless of which treatment was used. Seven of the patients in the study died as a result of continued sepsis or persistent cancer.