| Abstract|| |
Based on the modality of treatment, historically the management of enterocutaneous fistula has been grouped into three periods. The era of antibiotics (1945-1960), the era of intensive nurse care (1960-1970) and the era of intravenous hyperalimentation (1970-1975). Schein's modification of Sitges-Serra classification of enterocutaneous fistula is now preferred to the old classification of high-output type and low output type. A major cause of enterocutaneous fistula is technical failure. Serum level of short-turn over proteins such as albumin retinal-binding protein, thyroxin binding pre-albumin as serum transferring are predictors of mortality and spontaneous fistula closure. Immediate surgical correction of the fistula is not a priority of treatment. The control of sepsis and adequate nutrition are the two most important aspects of management. The use of somatostatin and octreotide has been shown to shorten the period of spontaneous closure of the fistula.
Keywords: entercutaeous fistula, short-turnover protein, somatostatin and octreotide.
|How to cite this article:|
Ajao OG, Shehri MY. Enterocutaneous fistula. Saudi J Gastroenterol 2001;7:51-4
| Introduction|| |
The most distressing situation to the patient, his relatives and the surgeon, is the development of enterocutaneous or fecal fistula post-operatively. This is to be distinguished from the internal fistulas, which often occurs spontaneously as a result of some intra-abdominal pathology  . Unfortunately, no recent publication has been found in a research of English literature to refresh our knowledge of the management of this important and dreadful complication of abdominal surgery. Most large series dealing with this complication cover a period of many years back. Unfortunately, most are retrospective studies since enterocutaneous fistula is not a complication that every surgeon look forward to. This paper tries to highlight and review various aspects of this condition as documented by various centers that have succeeded in reducing the morbidity and mortality of this surgical complication.
| Historical Aspects of the Management|| |
The historical aspect of the management of this complication has been grouped into three era corresponding to when major advances were made  . The first era was between 1945-1960 when the use of antibiotics was introduced  . During this period, Soeters et al  reported that mortality rate from enterocutaneous fistula was about 45%. The second era covered a period from 1960-1970, when emphasis was laid on respiratory support, antibiotics, nutrition and intensive care  . During this period, the reported mortality rate  dropped down to about 15%. The third era was from 1970-1975  when the central intravenous nutrition was introduced. Even though the mortality rate did not improve significantly from that of the previous era, spontaneous fistula closure rate rose from about 10% in the 1960/70 era to about 25% in the 1970/75 era  . The authors attributed this relative good result to the use of total parental nutrition , .
| Classification of Fistulas|| |
The old system of classifying enterocutaneous fistula into either high output type or low output type , is no more favored by many workers because this classification bears little or no relevance to the prognosis and to the modality of treatment except, probably, in the two extremes of high output and low output fistulas. Siteges-Sera et al  proposed a classification based on the area of location and this was modified by Schein et al  as follows: type I refers to abdominal esophagus and gastroduodenal fistulas. Type 2 refers to small bowl fistula. Type 3 refers to a large bowl fistula and type 4 refers to fistula at any site but associated with a large abdominal wall defect. This was correlated with mortality rates and it was found that type I fistula carries a mortality rate of about 17%, type 2 about 33%, type 2 about 20% and type 4 about 60%  .
| Causes of Eenterocutaneous Fistula|| |
Causes of enterocutaneous fistula include inadequate surgical technique, diverticulitis, biliary tract obstruction and malignant disease  . In a review of 79 patients with 116 fistulas Kuvsshinoff et al  found 82% to be postoperative, 6.3% due to Crohn's disease, 3.8% due to neoplasm and infection each, 2.5% iatrogenic and 1.3% due to radiation. The postoperative fistulas were due to anastigmatic leak, inadvertent enterotomy, local sepsis, distal obstruction preventing the healing of anastomosis, presence of foreign body (e. g Merlex mesh) dehiscence and complex wound problems , . Reports from some areas in some young developing countries, where surgical experience is limited include the following as causes of enterocutaneous fistulas: after surgical closure of typhoid intestinal perforation , , lysis of adhesions in intestinal obstruction, illegal septic abortions, gynecological operations and operations in or around the duodenum , . Other causes are inguinal hernia mistakenly incised as a groin abscess by an unqualified practitioners, obstructed paraumbilical and groin hernias of the Richter's type and inadvertently catching a loop of bowel with stitch while closing the abdomen  . This can usually be avoided by spreading the omculum over the loops of bowel before closing the abdomen and this maneuver also prevents inadvertent entry into a loop of bowel stuck to the undersurface of the previous abdominal incision scar in a re-operation.
| Prognostic Factors in Enterocutaneous Fistulas|| |
Kuvshinoff et al  reported that serum levels of short turnover proteins such as albumin, retinol binding protein and serum transferrin are predictors of spontaneous enterocutaneous fistulas closure and mortality. Of these, it was found that transferrin level was predictive of spontaneous closure, whereas serum transferrin, retinol-binding protein and thyroxin-binding prealbumin are predictors of mortality in patients with enterocutaneous fistula. Serum albumin measurements were found to be nondiagnostic. Although limited experience at Asir General Hospital, we found low serum albumin level in most elderly patients that developed enterocutaneous fistula after bowel resection and anastomosis. The explanation of this, is that the patients who were not able to respond to stress such as sepsis, neoplasm etc. by increasing hepatic synthesis of acute phase proteins (transferrin, retinol-binding protein and thyroxin-binding prealbumin) were more likely to die  , whereas those who were able were more likely to survive.
Some causes of failed spontaneous closure that have been identified are irradiated field, foreign body in the fistula tract, undrained intra-abdominal abcess, distal obstruction and abdominal wall defect  . Other are  prolapsed mucosa in the fistula tract, tuberculosis fistula and malignancy at and around the fistula. Majority of the deaths of enterocutaneous fistula are due to uncontrolled infection and associated malnutrition.
| Management|| |
Many workers in this field agree that immediate surgical correction of the fistula is usually not a treatment priority ,,,, . The initial surgical treatment if indicated, should be restricted to treatment of intra-abdominal abscess and hemorrhage when present. There is general agreement that nutrition and control of sepsis are two most important aspects of management. When these are instituted, most of the fistulas will close on conservative treatment.
Total parenteral nutrition (TPN), hyperalimentation and enteral feeding have been effective  . The safety of feeding jejunostomy or enteral feeding in comparison with TPN is a distinct advantage  . The enteral feeding may be through a tube  elemental enteral feeding beyond the fistula , or oral intake of high protein, high calorie diet with multivitamins added  . Enteral feeding does not seem to prevent closure of a fistula, especially if the location is lower down in the gastrointestinal tract. The enteral diet routinely used by Levy et al  consists of glucose polymers, protein, hydrolysates, medium chain triglycerides and essential fatty acids. A viscous additive was often added to slow down intestinal transit time and allow better absorption. The osmolality of such diet was usually between 300-350 mosm/L with a caloric concentration of 2.09 KJ/ml (0.5 Kcal/ml) to 4.18 KJ/ml (1.0 Kcal/ml). The amount of enteral nutrient given depends on the severity of the sepsis. The aim is to achieve weight gain and a positive nitrogen balance of 4 to 6 grams per day. In a report from University College Hospital, Ibadan, Nigeria. fourteen patients with types 2 and 3 enterocutaneous fistula were treated by oral feeding using high protein, high calorie diet and vitamins  . Relatively simple local care was established for the fistula. Nine patients survived and five died including two whom have had early surgical intervention for intestinal obstruction. This gives a mortality rate of bout 36%. In those that survived fistulas closed between 7 and 150 days.
The advantage of total parenteral nutrition is that; it allows the gastrointestinal tract to rest, while providing nutrition. An effective treatment consists of total bowel rest and TPN providing 2000-5000 kcal, 70-200 grams amino acids and 500-1000 mls of 10% fat emulsion daily  . Normal intravenous hyperalimentation formula that have been used successfully is: 500 mls 50% dextrose, 500 mls 8.5% amino acids, 6.4 grams nitrogen, electrolytes and vitamins. The caloric and nutrient requirements should be the same whether the mode of nutrition is eteral or parenteral  . In their series, Kuvshinoff et al (9) found that with TPN only, the duration of closure was 50+ 14 days.
| The Use of Somatostatin and Octreotide|| |
In a multi-center trial, Torres et al  found out that the use of somtostantin helps fistula to heal faster, reduces complications, reduces the period of hospitalization and morbidity. Somatostatin is a naturally occurring 14-amino acid peptide that inhibits gastrointestinal hormone secretion  . This is usually administered in the treatment of fistulas as a continuous intravenous infusion.
Octreotide, a synthetic analogue of somatostatin has also been tried in the treatment of fistula. The dose used was 100 micrograms every eight hours subcutaneously, octreotide also shortens the time required for spontaneous closure of fistula  when other adequate treatment is established.
| The Role of Surgery|| |
Even though it is generally agreed that immediate surgical intervention is not a priority, Reber et al  recommended surgical closure for most fistulas that persist beyond 30 days. But many authors ,,, including us believe that when there are no complications one should wait for at least 6 weeks before thinking of surgical intervention. Also, surgery is indicated when the fistula is complicated by intestinal obstruction bleeding or abscess formation  . Most workers in this field, including us do not have any evidence to suggest the use of proximal diverting enterotomy and/or the use of serosal patch are viable options in the management of enterocutaneous fistula.
Prophylactic antibiotics are usually not recommended and should be reserved for the treatment of septicemia and in the immediate preoperative period  . The preferred surgical procedure is complete resection of the bowel segment containing the fistula and an end-to-end anastomosis , . However, bypass of the fistula may be indicated if resection is hazardous such as in lateral duodenal fistulas and bowel fistulas deep within the pelvis. Direct suture closure of the fistula is not recommended because the fistula often breaks down.
Local treatment of the fistula consists of irrigation of the fistula tract and aspiration of the fistula contents  . Levy et al  used a solution of 0.45% lactic acid for irrigation using about 2500 mls 24hr. But even with a simple device such as a colostomy bag and frequent changing of dressings over the fistula, most will close  . Substances like karaya paste may be necessary to prevent esoritation of skin around the fistula  .
Reported mortality rate in enterocutaneous fistula ranges between 0 and 33% ,, .
| Conclusion|| |
Most uncomplicated enterocutaneous fistula will close spontaneously when properly managed. Surgery is usually not an immediate priority except to deal with complications. However, when surgical intervention is required to deal with the fistula resection and anastomosis or bypass procedures are the preferred surgical procedures. Simple suture of the fistula is not recommended.
| References|| |
|1.||Al-Shehri MY. Gastropleural associated with hepatic hydatid cyst. Report of a case and review of literature. Emirates Med. J. 1994; 12: 154-6. |
|2.||McClelland R.M. (Editor). Small bowel fistulas. Selected readings in general surgery 1996;23: 1-17. |
|3.||Soeters PB, Ebeid AM, Fischer JE. Review of 404 patients with gastrointestinal fistulas impact: of parenteral nutrition. Ann Surg. 1979;90: 182-202. |
|4.||Monod Broca O. treatment of intestinal fistulas. Br. J Surg 1977;64 685-9. |
|5.||Ajao OG. Ladipo. JK, Adebamowo CA. Enterocutaneous fistula: Ibadan experience. Intl. J Trop Surg 1991; 2: 48-52. |
|6.||Sitges-Sera A, jaurrieta E, Stiges-Creus A. management of postoperative enterocutaneous fistula: the role of parenteral nutrition and surgery. Br. J. Surg. 1982;69: 147-50. |
|7.||Schein M, Decker GAG. Postoperative extrernal alimentary tract fistulas. The Amer J Surg. 1991; 116:435-8. |
|8.||Schein M and Decker GAG. Gastrointestinal fistulas associated with large abdominal wall defects: experience with 43 patients. Br. J Surg 1990; 77: 97-100. |
|9.||Kuvshinoff BW, Brodish RJ, McFadden DW, Fischer JE. Serum transferrin as prognostic indicator of spontaneous closure and mortality in gastrointestinal cutaneous fistulas Ann Surg 1993; 217: 615-23. |
|10.||Ajao OG. Typhoid perforation: factors affecting mortality and morbidity. Interl Surg 1982; 67: 317-9. |
|11.||II. Adesunkanmi AR, Ajao OG. The prognostic factors in typhoid ileal perforation: a prospective study of 50 patients. JR Coll Surg Edin 1997; 42: 395-9. |
|12.||Levy E, Frileux P, Cognenc PH, Honiger J, Olliveier JM, Parc R. -High-output external fistulae of the small bowel: management with continuous external nutrition. Br J Surg, 1989; 76: 676-9. |
|13.||MacFayden BV, Dudrick SJ, Ruberg RJ. Management of gastroe\intestianl fistulas with parenteral hyperalimentation. Surgery 1973; 74: 100-5. |
|14.||Al-Shehri M, Makarewicz. P, Freeman JB. Feeding jejunostomy: a safety adjunct to laparotomy. Candian J. Surg. 1990,33:181-4. |
|15.||Reber HA, Roberts C, Way LW, Dunphy JE. Management of external gastrointestinal fistulas. Ann Surg 1978.188: 460-6 |
|16.||Torres AJ, Landa JI, Moreno-Azoita M, Arguello JM, et al. Somatostanin in the management of gastrointestinal fistulas: A multicenter trial. Arch Surg 1992; 127: 97-100. |
|17.||Sancho JJ, di Costanzo J, Nubiola P, et al. Randomized double-blind placecbo-controlled trial of early octreotide in patients with postoperative enterocutaneous fistula. Br J Surg. 1995; 82:638-41 |
|18.||Tremolieres J, Bonfils S, Gros J, L'irrigation paracide lactique treatment des fistules cutanees digestive avec secretion trypsique et auto gigestion parietale. Arch Mal appar Dig 1961;50:636-45 |
|19.||Gros E, Irving M. Protection of the skin around intestinal fistulas. Br J Surg 1977; 64: 258-63. |
Oluwole Gholagunte Ajao
College of Medicine P.O. Box 641, Abha
Source of Support: None, Conflict of Interest: None