Diseases of the

COLON & R E C T U M

Vol. 32 October 1989 No. 10

Original Contributions

Rectovaginal Fistula in Crohn's Disease JEFFREY L. COHEN, M.D.,* JAMES W. STRICKER, M.D. ,~ DAVID J. SCHOETZ, JR., M.D. ,

JOHN A. COLLER, M.D., MALCOLM C. VEIDENHEIMER, M.D.

Cohen JL, Stricker JW, Schoetz DJ Jr, Coller JA, Veidenheimer MC. Rectovaginal fistula in Crohn's disease. Dis Colon Rectum 1989;32:825- 828.

Rectovaginal fistulas in the setting of Crohn's disease present a difficult management dilemma. Some patients with this problem require proctocolectomy, yet other patients with minimal symptoms never require an operation for treatment of the rectovaginal fistula. For a small percentage of patients, local surgical repair of the fistula may be warranted. Since 1980, this study has attempted local repair in seven patients with symptomatic rectovaginal fistulas from Crohn's disease. Five patients underwent staged repair of the fistula. Closure of the colostomy was eventually possible in three of these patients. Two of the three patients have had no evidence of recurrence at follow- up in excess of two years. The third patient required an ileostomy for intestinal disease and had no recurrence of the fistula. Two patients underwent primary repair of the rectovaginal fistula without fecal diversion; in one of these patients, the fistula recurred ten days after operation, necessitating a diverting ileostomy. The other patient remains cured 26 months after repair. The results of this review indicate that in the setting of quiescent rectal disease, an attempt to repair the fistula can be expected to have a reasonable chance of success. The presence of a rectovaginal fistula in a patient with Crohn's disease does not mandate removal of the rectum. [Key words: Rectovaginal fistula; Crohn's disease]

Poster presentation at the meeting of the American Society of Colon and Rectal Surgeons, Anaheim, California, June 12 to 17, 1988.

Address reprint requests to Dr. Schoetz: Department of Colon and Rectal Surgery, Lahey Clinic Medical Center, 41 Mall Road, Bur- lington, Massachusetts 01805.

*Present address: 85 Seymour Street, Suite 425, Hartford, Connecticut 06106.

tPresent address: San Francisco-Kaiser, 2200 O'Farrell Street, San Francisco, California 94115.

From the Department of Colon and Rectal Surgery, Lahe~ Clinic Medical Center

Burlington, Massachusetts

PERIANAL INVOLVEMENT in pa t i en t s w i th C r o h n ' s disease is common . I , z T h e d e v e l o p m e n t of p e r i a n a l disease can be a devas ta t ing p r o b l e m for the pa t ien t , especial ly when c o m p l i c a t e d by the d e v e l o p m e n t of a rec tovagina l fistula. U n c o n t r o l l e d fecal d r a inage f rom the vag ina is bo th unc omfo r t a b l e a n d embar r a s s ing to the p a t i e n t . T h i s c o n d i t i o n p r e s e n t s c h a l l e n g i n g m a n a g e m e n t p r o b l e m s for the c l in ic ian .

As wi th o ther man i fes ta t ions of anorec ta l C roh n ' s disease, s y m p t o m a t i c t rea tment of rec tovag ina l f is tulas is of ten the mos t a p p r o p r i a t e therapy. T h e l i te ra ture is replete wi th studies r e p o r t i n g p o o r results f rom surgical t r ea tment of rec tovag ina l f is tulas in pa t ien ts w i th Crohn ' s disease. ~-5 G o l i g h e r a n d coau thors 6 devoted on ly one p a r a g r a p h to this topic a n d conc luded tha t for mos t pa t ients , exc is ion of the r ec tum shou ld be performed.

Recen t ly , there has been a t r e n d t o w a r d m o r e aggressive local t rea tment of, selected pa t ien t s w i th a rec tovagina l f i s tu la w h e n the s u r r o u n d i n g anorec ta l disease is quiescent.7, 8 T h e favorable results achieved wi th this a p p r o a c h have s t imu la t ed a r eappra i sa l of the t r ad i t iona l a t t i tude toward the t rea tment of this p rob l em. In this study, we r epor t the results of ou r t rea tment of pa t ien ts w i th rec tovagina l f i s tu la in the se t t ing of C roh n ' s disease.

Dis. Col. g: Reck 826 COHEN, ET AL. October 1989

TABLE 1. Operative Procedures and Patient Characteristics

Total Local Repair Proctocolectomy

(N = 7) (Y = 7)

Median age (years) 30 49 Median duration of Crohn's

disease (years) 4 10 Number of patients with active

anorectal disease 3 7 Number of patients taking steroids

at time of operation 3 6

Materials and Methods

All patients admitted to the Lahey Clinic Medical Center between 1980 and 1986 with the diagnosis of Crohn's disease and a rectovaginal fistula were studied. A retrospective chart review was conducted to determine patient characteristics, operative treatment, and follow- up data. The diagnosis of Crohn's disease was based on the clinical impression of the gastroenterologist or surgeon, with either radiologic or pathologic confir- mation. All patients were seen in follow-up at the Lahey Clinic. Telephone inquiries were sometimes required to achieve a m in im um one-year follow-up in all patients who underwent local repair.

During the period of review, 14 women with the diagnosis of Crohn's disease and a rectovaginal fistula underwent operative treatment. A total proctocolectomy was performed in seven patients without prior attempt at repair of the rectovaginal fistula. The other seven patients underwent attempted repair of the fistula.

All seven patients who underwent total proctocolec- tomy had severe anorectal Crohn's disease at the time of operation. In six of these patients, anorectal disease was the main indication for operation. In one patient, both a cecal carcinoma and active perianal disease were the indications for total proctocolectomy. Although three of the seven patients had symptoms from the rectovaginal fistula at the time of operation, this was not the primary indication for operation.

All seven patients who underwent attempted local repair had a symptomatic rectovaginal fistula t h a t was the major indication for operation. Compared with the proctocolectomy group, patients in this group were younger, had a shorter duration of Crohn's disease, and were less likely to be taking steroids at the time of operation (Table 1). Only three patients in the group undergoing local repair of the rectovaginal fistula had associated active anorectal disease.

Local repair of the rectovaginal fistula was accomp- lished by one of two operative approaches. The more frequently employed technique involved division of the

TABLE 2. Local Repair o] Rectovaginal Fistula Associated with Crohn's Disease

Follow-up Number (Number of

of Months After Type of Repair Patients Recurrence/Comment Repair)

Diversion 2 1 patient failed--abdomi- - - noperineal resection

1 patient asymptomatic recurrence at 2 years

+ layered repair 3 No recurrences--1 patient 33 still with ileostomy

No Diversion Layered repair 1 No recurrence 26 Advancement

flap 1 Recurrence on tenth post- - - operative day

rectovaginal septum up to and sometimes above the level of the fistula. After curettage of the fistulous tracts, a layered closure was performed with repair of the external sphincter and perineal body. The alternative technique employed was an endorectal advancement flap described by Rothenberger et al. 9

Results

The patients in whom an attempt at local repair of the rectovaginal fistula was initiated are listed in Table 2 and are separated by whether diversion of the fecal stream was a component of treatment. Of the five patients who underwent diversion, two had no further treatment of the fistula. In one of the patients the rectovaginal fistula closed spontaneously. Closure of the colostomy in this patient was followed by an asymp- tomatic recurrence of the fistula two years later. The other patient treated with fecal diversion alone had progressive anorectal disease and underwent proctoco- lectomy six months later.

Three patients were treated by diversion of the fecal stream with local repair of the rectovaginal fistula. Two of these patients have had their colostomies closed four and six months, respectively, after repair, with no recurrence over a 12- and 48-month follow-up. Th e third patient also has had no evidence of recurrence two years after local repair; she has, however, required an ileostomy for management of persistent intestinal disease.

T w o patients were treated by local repair wi thout diversion of the fecal stream. Th e patient treated by layered repair has had no evidence of recurrence over a 26-month follow-up. The patient treated by endorectal advancement flap had a recurrence in the immediate postoperative period.

Volume 32 Number 10 R E C T O V A G I N A L FISTULAS IN C R O H N ' S DISEASE 827

The three patients who underwent a successful repair and who presently have intestinal continuity are fully continent. One of these patients required almost two years to achieve full continence.

The only perioperative complication in this group of patients occurred in the patient who required abdominoperineal resection for progression of anorectal disease, A perineal hematoma developed that required reexploration. This was followed by a persistent perineal sinus that closed spontaneously eight months later.

Of the seven patients treated initially by proctocolec- tomy, perineal wound complications requiring readmis- sion developed in three patients. Two of these patients required additional operations for nonhealing perineal wounds.

Discussion

The development of a rectovaginal fistula is a relatively uncommon manifestation of Crohn's disease with a reported incidence ranging from 1.7 percent to 23 percent.l~ n Most of these fistulas represent a component of more extensive anorectal disease. Although the incidence of rectovaginal fistulas is low, perianal disease is a well-recognized component of the Crohn's disease spectrum. Depending on the criteria employed in diagnosis and the degree to which evidence of perianal disease is sought, there is a wide variation in its reported incidence. Fielding 1 used broad criteria, including the presence of skin tags and skin maceration, and found perianal lesions in 80 percent of 167 patients with intestinal Crohn's disease. Likewise, Lockhart- Mummery ~ found a 61 percent incidence in 304 patients. Not surprisingly, Williams et a112 reported a somewhat lower incidence when a more strict definition of perianal disease was used, consisting only of a fistula, fissure, or abscess. In all of these studies, the more distal the involvement of intestinal Crohn's disease, the more likely was the presence of perianal disease.

In our experience, associated perianal disease was re- sponsible for many of the symptoms in patients with a rectovaginal fistula. In the group of seven patients who underwent proctectomy as the initial treatment for anorectal disease and who had a rectovaginal fistula, only three patients complained of symptoms directly attributable to the presence of the rectovaginal fistula. Furthermore, in none of these patients was the presence of the fistula the primary indication for proctectomy. This finding is in agreement with that of Radcliffe et

al. 8 who studied 886 women with Crohn's disease treated at St. Mark's Hospital in London. They reported that 67 percent of the 90 patients with rectovaginal fistulas thought the symptoms of the intestinal disease over- shadowed those of the fistula. In only five of 34 patients

who underwent proctectomy as initial treatment was the rectovaginal fistula a prominent indication for operation.

In patients with Crohn's disease the rectovaginal fistula occasionally represents the only significant manifestation of disease. In the past, local repair of such fistulas has been discouraged because of the panenteric nature of the disease, the lack of satisfactory normal tissue to incorporate in the repair, and the tendency for recurrent perianal disease to develop, s-6 Faulconer and Muldoon 4 treated 11 of 15 patients with Crohn's disease and a rectovaginal fistula by proctectomy and concluded that local reparative anorectal surgery would likely fail in patients with colitis. They further suggested that therapy should be directed toward medical management of the colitis. Tuxen and Castro 3 believed that the development of a rectovaginal fistula in Crohn's disease was an unfavorable prognostic sign that usually indicated the necessity for proximal diversion or possible resection of the rectum. They proposed closure of the fistula as an alternative, however, and treated two of the six patients in their group with this approach, achieving good results. 3 They stressed that the rectum must appear healthy for the local repair to be successful.

After this report of success, Bandy and colleagues 7 attempted local repair of rectovaginal fistulas in ten patients with underlying Crohn's disease. Only patients with quiescent intestinal Crohn's disease and no active rectal disease were considered for repair. Their technique consisted of episioproctotomy with meticulous layered closure. Only one patient underwent prior fecal diversion. Nine of ten patients had a successful repair of the fistula. There were two late recurrences in a follow-up period ranging from 13 to 53 months. Employing a variety of techniques for local repair, Radcliffe et al. 8 achieved similar results. Through careful selection of patients to be treated by closure of the rectovaginal fistula, eight of 12 patients were ultimately asymptomatic. Two of these patients required multiple attempts at repair to achieve success.

Our experience also supports the use of local repair in selected patients. In seven patients with rectovaginal fistulas, we initiated a course of local treatment with the goal of preserving sphincter function. Of the five patients who eventually underwent an attempt at local repair, four had successful results. The only failure was in a patient who underwent an endorectal flap advancement. This approach failed in the immediate postoperative period, and subsequent fecal diversion was performed.

Radcliffe et al. 8 proposed that the endorectal flap advancement technique is the preferred method for local repair of these rectovaginal fistulas. The advantages of the endorectal advancement flap technique as outlined

Dis. Col . gc Rect 828 COHEN, ET AL. October 1989

by Radcl i f fe et al. s a n d R o t h e n b e r g e r e t al. 9 are a lso a~c~omplished by a layered c losure a p p r o a c h . We have actf ieved a h i g h degree of success w i th a m e t i c u l o u s layered c losure after d i v i d i n g the r ec tovag ina l s e p t u m u p to a n d above the level of the f istula. We bel ieve tha t the mos t i m p o r t a n t d e t e r m i n a n t s of successful local r epa i r are tha t the r epa i r be u n d e r t a k e n f rom the rectal (h igh-pressure ) side of the f i s tu la a n d tha t hea l thy t issue is used for a tens ion-free closure.

T h e need for c o n c o m i t a n t fecal d ivers ion to e n h a n c e the success rate of local r epa i r is cont rovers ia l . Fecal d ive r s ion as the sole t r e a tmen t for p e r i a n a l C r o h n ' s disease in genera l , a n d r ec tovag ina l f i s tu la in pa r t i cu l a r , has a l i m i t e d role w i t h a h i g h l i k e l i h o o d of p r o g r e s s i o n of disease. ~3,~ Bandy et al. 7 p e r f o r m e d fecal d ive r s ion in on ly one p a t i e n t before local r epa i r a n d be l ieved tha t fecal d ive r s ion was unnecessa ry as l o n g as pa t i en t s were selected a p p r o p r i a t e l y . Radcl i f fe e t al. s agreed w i th this c o n c l u s i o n b u t e m p h a s i z e d the need for c o m p l e t e m e c h a n i c a l bowe l p r e p a r a t i o n as a n in t eg ra l pa r t of the repai r . A l t h o u g h the m a j o r i t y of ou r pa t i en t s u n d e r w e n t fecal d ive r s ion at some stage of the a t t e m p t e d repai r , success was a lso ach ieved w i t h o u t fecal d ivers ion.

Based on ou r exper ience , we bel ieve tha t the t rend t o w a r d loca l r epa i r of r ec tovag ina l f is tulas in selected pa t i en t s w i th C r o h n ' s disease is w o r t h w h i l e . As l o n g as the anorec ta l disease is m i n i m a l a n d the in t e s t ina l C r o h n ' s disease is qu iescent , success rates wi l l a p p r o a c h those for r epa i r of r ec tovag ina l f is tulas a r i s i ng f rom o the r causes. S h o u l d the f i s tu la recur, the pa t i en t wi l l st i l l have benef i ted f rom prese rva t ion of the r ec tum u n t i l

tha t t ime a n d m a y be cons ide red for a no the r a t t e m p t at local r epa i r if the anorec ta l disease r e ma ins quiescent .

References

1. Fielding JF. Perianal lesions in Crohn's disease. J R Coll Surg Edinb 1972;17:32-7.

2. Lockhart-Mummery HE. Crohn's disease: anal lesions. Dis Colon Rectum 1975;18:200-2.

3. Tuxen PA, Castro AF. Rectovaginal fistula in Crohn's disease. Dis Colon Rectum 1979;22:58-62.

4. Faulconer HT, Muldoon JP. Rectovaginal fistula in patients with colitis: review and report of a case. Dis Colon Rectum 1975;18:413-15,

5. Holland RM, Greiss FC Jr. Perineal Crohn's disease. Obstet Gynecol 1983;62:527-9.

6. Goligher JC, Duthie H, Nixon H~ Surgery ol the anus, rectum, gc colon. 5th ed. London: Baillie~e-Tindall, 1984.

7. Bandy LC, Addison A, Parker RT. Surgical management of rectovaginal fistulas in Crohn's disease. Am J Obstet Gynecol 1983; 147:359-63.

8. Radcliffe AG, Ritchie JK, Hawley PR, Lennard-Jones JE, Northover JM. Anovaginal and rectovaginal fistulas in Crohn's disease. Dis Colon Rectum 1988;31:94-99.

9. Rothenberger DA, Christenson CE, Balcos EG, Schottler JL, Nemer FD, Nivatvongs S, Goldberg SM. Endorectal advance- ment flap for treatment of simple rectovaginal fistula. Dis Colon Rectum 1982;25:297-300.

10. Van Patter WN, Bargen JA, Dockerty MB, Feldman WH, Mayo CW, Waugh JM. Regional enteritis. Gastroenterology 1954;26:347-450.

11. Ritchie JK, Lennard-Jones JE. Crohn's disease of the distal large bowel. Scand J Gastroenterol 1976;11:433-6.

12. Williams DR, Coller JA, Corman ML, Nugent FW, Veidenheimer MC. Anal complications in Crohn's disease. Dis Colon Rectum 1981 ;24:22-24.

13. Harper PH, Kettlewell MG, Lee EC. The effect of split ileostomy on perianal Crohn's disease. Br J Surg 1982;69:608-10.

14. Grant DR, Cohen Z, McLeod RS. Loop ileostomy for anorectal Crohn's disease. Can J Surg 1986;29:32-35.