 |
       |
|
|
DIAGNOSIS AND MANAGEMENT OF FAECAL FISTULA
Shirish K Bhansali
Consultant Surgeon, Jaslok Hospital and Research Centre, Breach Candy Hospital and Research Centre, Lilavati Hospital and Research Centre.
Development of intestinal fistula, spontaneous (due to disease) or iatrogenic is a serious complication.Small intestinal fistula have worse prognosis than large intestinal fistula right sided colonic fistula behave almost like small intestinal fistula. In this communication, various facets of diagnosis and management of acquired colonic fistula are discussed with reference to literature and to author's experience of 50 acquired colonic fistula (12 of these had associated small intestinal fistula, and 10 had associated pancreatic fistula).
Aetiology
The commonest aetiological factor is an abdominal operation. The few operations following which we encountered colo-cutaneous fistula were laparotomy for infected pancreatic necrosis, left colonic resections (especially low anterior resection), intestinal resection for gangrene, closure of colonic perforation, difficult adhesiolysis, difficult hysterectomy and ovariectomy, difficult nephrectomy, etc. Incidence is higher during second surgery, and is highest with laparostoma.[2] Colonic injuries were also discerned following laparoscopic pelvic surgery. Cautery burns to colon are more likely with the use of monopolar than bipolar cautery. Recently colonic fistula have also been reported following percutaneous drainage of abdominal abscesses.
Colovesical or colovaginal fistula may develop following a Wertheim's hysterectomy, L.A.R. etc. Spontaneous fistulation due to disease occurs usually into an internal organ. Thus colovesical fistula or colovaginal fistula may complicate colonic divertionalisitis, carcinoma, Crohn's disease etc.
Predisposing factors for postoperative fistula
These are local, regional, and general
The local factors are tension on bowel ends, poor vasculature (devitalisation) of margins, specific disease at cut ends, local infection etc.
The regional factors are abscesses, distal obstruction to bowel, acute necrotising pancreatitis (12 of our patients had this disease).
The general co-morbidity which predispose to the fistula are acute catabolic illness (like acute necrotising, pancreatitis), severe malnutrition (with serum-albumin / less than 2 Gm%), marked diabetes mellitus, morbid obesity, prolonged steroid therapy etc.
Classification of fistula
According to output per 24 hours, the fistula are classified by us as :
High output > 1000 ml per day
Medium output 500 - 1000 ml per day
Low output < 500 ml per day
Literature classifies these are low and high output[3]Classification of fistula according to the anatomy
1.Single, simple fistula - One, single, direct tract between bowel and skin or another viscus
2.Branched fistula - Single opening in bowel with branch tract, and two or more openings on skin
3.Multiple fistula - Multiple openings in bowel at different sites (especially seen in 10-20% of operation cases of infected pancreatic necrosis[4]
4.Fistula with distal obstruction in bowel
5.Fistula with complete disruption of bowel - usually due to total or near total anastomotic breakdown e.g. due to (a) tension in LAR, b) recurrence of gangrene
Knowledge regarding output, local anatomy, level, etc., is important for management and prognostication.Problem created by fistula especially colocutaneous ones
1.Bacterial contamination of usually sterile areas like peritoneal cavity, parietes etc. - This is encountered in upwards of 75% of cases.2.Bypass of a variable length of the intestine, leading to "loss" of digestive juices, and impaired digestion and absorption.
3.Fluid and electrolyte imbalance due to loss of intestinal contents.
4.Secondary infection in the wound if the fistula is through the wound.
5.Skin excoriation by digestive enzymes.
The extent of these problems is governed by level, location and magnitude of the fistula. In addition to these, there will always be problems due to underlying disease and co-morbidity. The problems are maximum with small intestinal fistula a minimum with left colonic fistula.Management
Time oriented flexible approach with overlapping phases is necessary. Treatment is both medical and surgical but surgery is not always required. Immediate surgery is rarely required, and early surgery (in the first 3-5 days) is infrequently necessary.
The sequential management considerations (with overlap) are :
1.Fluid and electrolytes imbalance2.Skin protection
3.Nutritional support
4.Pharmacological treatment (antibiotics, octreotide, H2 blockers)
5.Ultrasonogram or CT scan of abdomen and pelvis to look for collections, abscesses.
6.Contrast radiology to localise the fistula and to study local anatomy.
7.Surgery : Immediate? Or early? Or delayed? Or not at all?
1. Resuscitation
This is especially required in right colonic and small intestinal fistula. Initially adequate fluid and electrolyte administration is necessary to correct dehydration, achieve haemodynamic stability and maintain renal function. If serum albumin is less than 2 gm%, human albumin should be administered initially. Haemoglobin should be restored to around 10 gm%.
2. Skin Protection
Skin is protected by the dual use of stoma bag and stoma adhesive / Karaya paste / Zinc peroxide paste. Other advantages of the stoma bag are the measurement of fluid losses and to an extent, of the electrolytes and hence better maintenance of fluid and electrolyte balance and also better evaluation of the response of the quantum of the effluent to the therapy.
However not all fistula can be fitted adequately with a stoma bag. Fistulae through drainage tube sites and which are away from a) main wound, b) scars of previous surgery, c) umbilicus, d) bony prominences, accept the stoma bag snugly. On the other hand, it is impossible to fit a bag on a fistula (or multiple fistulae) through a partially dehisced wound. In situations in between these two extremes, ingenuity of clinician and of the stoma nurse as well as versatility of appliance are necessary.
3. Pharmacological Treatment
a) Octreotide (synthetic analogue of somotostatin) in doses of 100 mg (SC inj.) has been proved to decrease g.i. secretions and thus reduce fistulous effluent.[5] It may even lead to heading of the fistula.
b) H2 Blockers and Other Antacids : Reber et al[6] reported substantial decrease in fistulous output after the use of H2 blocker by reducing the stimultory effect of acid gastric secretion on the bowel. We use it when there is a small bowel fistula following ileo (or jejuno) - colic anastomosis (after resection) but do not advocate its use for pure colonic fistula.
c) Antibiotics : They are essential because more than 75% of the patients have a local infection with or without abscess. Most of the patients who develop fistula postoperatively have a "stormy" pre-fistula phase after the operation and have received considerable antibiotic therapy. Hence both anti anaerobes and higher antibiotics are essential. Third generation cepholoridines are usually required. It may be necessary to counter B-lactamase producing organisms. Guidelines are obtained by culture and sensitivity of swab from wound if this is infected. Patients with proved fungus infection (and often those without such a positive culture but with a protracted febrile and infective illness) are also administered fluconazole or other appropriate antifungals. MRSA should be looked for adequately treated.
4. Sonogram / CT scan
Ultra sonogram or CT scan (with full oral and i.v. contrast) should be carried out to rule out local collection(s) or abscess(es). If found these should be aspirated or therapeutically drained by a pigtail catheter(s) through a "safe window" devoid of bowels and important structures. A complete microbiological examination of the fluid is essential. Technical inability to drain such a collection or an abscess is an indication for an early surgery.
5. Nutritional Support
Many of these patients (especially those with right colonic fistula and those with small intestinal fistula in relation to colon) are malnourished and therefore need nutritional support to bolster their immune system, counter infection, and promote healing, nutritional support is the single most important therapy which in recent times has contributed single most important therapy which in recent times has contributed to a decrease in morbidity / mortality and even lead to spontaneous closure of many low output and some moderate output fistulae.
The need for nutritional support is assessed after nutritional evaluation and a consideration of the previous illness. It is started only after achieving haemodynamic stability and normalisation of renal status. Initially it is given parenterally and then enteral feeding is introduced and increased. Higher the fistula in the intestinal tract lesser is the tolerance and effectiveness of enteral nutrition.
Even if the nutritional support does not heal the fistula it does decrease the fistulous effluent, minimises local infection, and increases tissue integrity. All these ensure a better outcome of the reparative operation.
6. Localisation Studies
These should be completed as early as possible because crisis requiring surgery at an early stage may arise. The possible localisation studies are :
a)Gastrografin per rectum - usually requiredb)Oral gastrografin studies (infrequently required - done if rectal gastrografin study isinconclusive)
c)Gastrografin study through the fistula.
The aims of these studies are to locate the level of the fistula, define its type if possible, assess the completeness or otherwise of the rupture, and detect a distal obstruction if present. X-ray films at 3, 6, and 12 hours after transluminal contrast studies may show up a delayed filling of an abscess cavity.For colovesical fistula, cystoscopy is mandatory. Colonoscopy is essential in spontaneous fistula to assess the colonic pathology but is best avoided in the postoperative fistulae.
7. Surgery
According to the timing of surgery, three categories of patients are recognised (also there is a fourth group which may not require surgery because the fistula may heal with medical treatment in these).
Group A : Those who require immediate surgery (in less than 24 hours). This includes patients with (a) proved or suspected gangrene, b) severe peritonitis, c) life threatening infection or, d) total disruption of intestinal continuity. In these, apart from intestinal resection if required, the two intestinal ends are best brought out as diversion proximal stoma, and mucous distal stoma, (to be joined at a later date, as a second stage). Primary anastomosis in these circumstances invariably breaks down.
Group B : Early surgery (with 3-5 days) is required in those with a) distal obstruction, b) specific disease, c) colovesical fistula and d) an abscess / collection which cannot be adequately drained percutaneously under USG/CT guidance. In this group, majority of patients require a diversion proximal loop stoma (others are treated as in Group A). Formerly loop transverse colostomy was the preferred diversion whenever it was feasible, but during the last 5-10 years (loop-ileostomy which is nearly always possible, is the diversion of choice[7],[8] because of the following reasons :
a) stoma bag fits better over ileostomy
b) odour is less with ileostomy
c) ileostomy interferes less with subsequent colonic resection and anastomosis.
d) closure of ileostomy is simpler.
Groups C and D : All the patients who do not belong to Groups A and B are closely monitored with various medical therapeutic measures. The fistulous effluent, nutrition, and local infection are especially evaluated.
Majority of low level fistula heal with medical treatment outlined earlier. Almost 15-20% of the fistula with medium output (500-1000 ml per day) will also heal, whereas majority of high output fistula will require surgery.
Regardless of the initial quantum of output, surgery is delayed (and medical measures continued) till the output reaches a minimum plateau of quantity, over 3-4 day. But if the output suddenly increases or local infection / collection increases, operation is carried out without delay.
The type of operation is inidividualised to local condition and patients general status. It may entail a) simple freshening of fistulous edges and a transverse closure or b) resection with anastomosis (ideal is side to side stoma) or c) occasionally exteriorisation of the two ends.
As mentioned earlier, this deliberate delaying of surgery with nutritional support results in better tissue integrity, and decreased local infection, and better nutritional status, all leading to better outcome of the operation. And of course some of the patients may not need operation at all.
Postoperative Complications
The important among these are :
1.Recurrence of fistula - especially if original disease process is like acute necrotising pancreatitis, gangrene etc.2.Multiple organ dysfunction syndrome (MODS) leading to multiple organ failure (MOF) (lungs, kidney, and liver are especially vulnerable).
3.Stress GI bleeding
4.Secondary haemorrhage
Prognosis
This depends on several factors.
1.Site of fistula : Small intestinal fistula have worse prognosis (mortality over 15% - whereas the left colonic fistula have the best prognosis with practically no mortality.[1]2.Higher the output, worse is the prognosis
3.Patient with multiple fistula, and various regional unfavourable factors (like acute necrotising pancreatitis) and also co-morbidity factors have poorer prognosis as also patients with complete disruption of the bowel.
4.Patients whose original operation was for infected pancreatic necrosis, gangrene, etc., have adverse outcome.
Recent therapeutic measures which have decreased morbidity and mortality in these patients are :
1.Nutritional support2.Octreotide therapy
3.Higher antibiotics and antifungals
4.Percutaneous drainage of collections / abscess
5.Better organ support, especially in intensive case units.
REFERENCES
1.Sitges-Serra A, Jaurrieta E, Sitges-Creus A. Management of postoperative enterocutaneous fistulas : The roles of parenteral nutrition and surgery. Br J Surg 1982; 69 : 147-50.
2.Bradley EL III. A 15 year experience with open drainage for infected pancreatic necrosis. Surg Gynecol Obstet 1993; 177 : 215-22.
3.Sancho JJ, Hernandez R, Giruent M, et al. Management of enterocutaneous fistulas. Dig Surg 1997; 14 : 483-91.
4.Sarr MG. Gastrointestinal fistulas complicating surgical management of necrotising pancreatitis in "The Pancreas", Eds. HG Beger, AL Warshaw, MW Buchler et al, Blackwell Science 1988 57 : 594-98.
5.Nubiola-Calonge P, Badia JM, Sancho J, et al. Blind evaluation of the effect of octreotide (SMS) 20+995) a somatostatin analogue, on small bowel fistula output. Lancet 1987; ii : 672-9.
6.Reber H, Roberts C, Way L, et al. Management of external gastrointestinal fistulas. Ann Surg 1978; 188 : 460-7.
7.Fasth S, Hulten L. Loop ileostomy : a superior diverting stoma in colorectal surgery. World J Surg 1984; 8 : 401.
8.Williams NS, Nasmyth DG, Jones D, et al. Defunctioning stomas : a prospective controlled trial comparing loop ileostomy with low transverse colostomy. Br J Surg 1986; 73 : 560.
9.Levy E, Frileux P, Cugnenc PH, et al. High output external fistulae of the small bowel : Management with continuous enteral nutrition. Br J Surg 1989; 76 : 676-9.
 |