- Email: [email protected]
Endorectal mucosal advancement flap: the preferred method for complex cryptoglandular fistula-in-ano
Endorectal Mucosal Advancement Flap: The Preferred Method for Complex Cryptoglandular Fistula-in-Ano Richard W. Golub, M.D., William E. Wise, Jr., M.D., Bruce A. Kerner, M.D., Karamjit S. Kbanduja, M.D., Pedro S. Aguilar, M.D.
Cryptoglandular fistula-in-ano is a common affliction that usually responds well to conventional surgical procedures such as fistulectomy, fistulotomy, and seton placement. These procedures, however, can be associated with varying degrees of fecal incontinence. Endorectal mucosal advancement flap has been advocated as an alternative procedure that avoids this problem. This study was undertaken to determine the risks and benefits associated with endorectal mucosal advancement flap in the treatment of complex fistula-in-ano. One hundred sixty-four patients underwent 167 endorectal mucosal advancement flap procedures for complex cryptoglandular fistula-in-ano between January 1982 and December 1990. There were 126 men and 38 women whose mean age was 42.1 years (range 20 to 79 years). The majority of the patients (70%) had complex fistulas (transsphincteric, suprasphincteric, or extrasphincteric). Fifteen patients (9%) had an intersphincteric fistula. All patients were available for short-term follow-up (6 weeks). Postoperative morbidity was minimal and included urinary retention in 13 patients (7.8%) and bleeding in one patient. Healing time averaged 6 weeks. Long-term follow-up, ranging from 19 to 135 months, was carried out in 61 patients. There were two recurrences (3.28%). Nine patients (15%) complained of varying degrees of fecal incontinence. Six patients complained of incontinence to flatus and three patients complained of incontinence to liquid stool. No patient was incontinent of solid stool. Sixty patients (98%) rated their functional result as excellent or good. Endorectal mucosal advancement flap is a safe and effective technique for the treatment of complex cryptoglandular fistula-in-ano. It can be performed with minimal morbidity, no mortality, an acceptable recurrence rate, and little alteration in anorectal continence. (J GASTROINTESTSURf_;1997; 1:487-491 .)
Cryptoglandular fistula-in-ano is a c o m m o n afflicdon that usually responds well to conventional operative procedures such as fistulectomy, fistulotomy, and seton placement. T h e s e procedures, however, can be associated with varying degrees of fecal incontinence ranging from 6% to 4 5 % ) -3 T h e risk of fecal incontinence is particularly high with complex fistulas (transsphincteric, suprasphincteric, and extrasphincteric), and therefore these fistulas require careful preoperative evaluation and often alternative surgical techniques. One such technique is the endorectal mucosal advancement flap. First proposed by N o b l e 4 in 1902 for repair of a rectovaginal fistula, and later modified by Elting 5 and Laird, 6 this technique employs a broad-based flap of rectal wall to obliterate the
origin of the fistula. In 1985 Aguilar et al. 7 proposed the use of an endorectal mucosal advancement flap to treat complex cryptoglandular fistula-in-ano. This report is a continuation of that series. SURGICAL
TECHNIQUE
Most of the patients undergoing an endorectal mucosal advancement flap procedure for cryptoglandular fistula-in-ano received a standard anorectal bowel preparation consisting of a clear liquid diet and a Fleet enema (C.B. Fleet Company Inc., Lynchburg, Va.) on the evening before and the morning of surgery. Intravenous antibiotics were employed only for endocarditis or prosthetic joint prophylaxis when indicated.
From the Division of Colon and Rectal Surgery, Grant Medical Center, Columbus, Ohio. Presented in part at the annual meeting of the American Society of Colon and Rectal Surgeons, Chicago, Ill., May 2-7, 1993. Reprint requests: Richard W. Gx)lub, M.D_, EA.C.S., State University of New York Health Science Center at BrooHyn, 450 Clarkson Ave., Box 40, Brooklyn NY 11203. 487
488
Jour,lal of Gastrointestinal Surgery
(-;olub et al.
T h e technique of endorectal mucosal advancement flap has been described extensively elsewhere and is essentially a modification of the techniques originally developed by Noble 4 and Elting. 5 General, regional, or local anesthesia with intravenous sedation can be used. Intraoperative fluids are limited. Most patients are placed in the prone jackknife position with the buttocks taped apart, although the lithotomy position can be advantageous for posterior fistulas. Illumination with a headlamp can be helpful. T h e perineum and anal canal are prepared and irrigated with povidone-iodine solution. Anesthesia is supplemented with a perianal field block consisting of bupivacaine 0.25%, lidocaine 0.5% with 1/200,000 epinephrine, and 150 units of hyaluronidase (Wydase, Wyeth-Ayerst Laboratories, Philadelphia, Pa.). A Hill-Ferguson retractor is placed in the anal canal and a thorough anorectal evaluation is undertaken. Associated pathologic conditions are noted. Internal and external openings are located, and the course of the fistulous tract is carefully assessed by palpation and gentle probing. T h e fistulous tract is removed subcutaneously from the external opening to the level of the external sphincter muscle. T h e residual tract and any granulation tissue is removed by curettage. Secondary tracts are noted and treated in a similar fashion. T h e perianal skin is widely saucerized. Using sharp scissors dissection, a flap consisting of anoderm, mucosa, and submucosa is raised from the intersphincteric groove to a point approximately 3 to 4 cm proximal to the internal opening. T h e configuration of the flap is trapezoidal with the base wider than the apex to ensure an adequate distal blood supply. Meticulous electrocautery hemostasis is maintained t h r o u g h o u t the course of the dissection. T h e opening in the internal anal sphincter is closed with interrupted 3-0 chromic sutures. T h e flap is advanced distally, without tension, to cover the defect in the internal sphincter and the excess mucosa (including the internal opening) is amputated. T h e distal edge is sutured to the internal anal sphincter at the level of the intersphincteric groove, and the lateral aspects are reapproximated to adjacent rectal mucosa using 2-0 or 3-0 chromic catgut sutures. Sterile dressings are applied to the wound. Packing, topical antibiotics, and suction catheters are not routinely used. Fecal diversion is avoided. Repair of the external voluntary sphincter was not undertaken in this series of patients. However, in patients presenting with anteriorly placed fistulas, perineal body reconstruction and/or simple reefing of the anterior sphincter musculature can be added to the procedure if preoperative history suggests inconti-
nence or physical examination demonstrates lax musculature or an attenuated perineal body. Postoperatively the patient is started on a regular diet supplemented with bulk fiber agents and stool softeners. Oral analgesics, diazepam, and warm sitz baths are employed to promote local hygiene and to control pain and muscle spasms. Patients are discharged home when they are able to void without difficulty. Follow-up is conducted every 2 to 3 weeks until all wounds are healed. RESULTS One hundred sixty-four patients underwent 167 endorectal mucosal advancement flap procedures for cryptoglandular fistula-in-ano from January 1982 through December 1990. Fistulas secondary to inflammatory bowel disease, actinomycosis, foreign bodies, and malignancy were excluded from this study, as were rectovaginal and rectourethral fistulas. There were 126 men and 38 women whose mean age was 42.1 years (range 20 to 79 years). Sixty-two percent of patients had undergone prior anorectal surgery, including 18 patients (11%) who had previous fistula surgery. Perirectal pain and drainage were the most common presenting sTmptoms with a preoperative duration of 1 to 3 months. Five patients (3 %) presented with an acute fistulous abscess. All patients underwent an endorectal mucosal advancement flap as their definitive surgical procedure. One hundred fifteen patients (70%) were classified as having complex fistula-in-ano (transsphincteric, suprasphincteric, or extrasphincteric), whereas 15 patients (9%) had intersphincteric fistula. T h e majority of these 15 patients had anteriorly placed fistulas and were deemed at risk for incontinence with conventional surgical technique. T h e type of fistula was not recorded in 34 patients (Fig. 1). An internal opening was located in the posterior midline in 87 patients (53%). All 164 patients were available for short-term follow-up. Complications and healing rates are based on this group of patients. Healing time averaged 6 weeks for all patients. Ten patients (5.98%) experienced prolonged healing. All of these patients developed either small subcutaneous (7 patients) or superficial fistulas in the early postoperative period. Six patients were treated with incision and drainage, three patients had an in-office fistulotomy, and one patient required a repeat mucosal advancement flap. Postoperative morbidity also included urinary retention (13 patients) and bleeding (1 patient) (Table I). T h e r e were no operative deaths. Long-term follow-up was carried out by office visit or telephone questionnaire in 61 patients (37%) and
Vol. 1, No. 5 1997
Mucosal Advancement Flap
Suprasphincteric (1%) Unclassified ~ / Intersphincteric (20%) \ ~ (9%)
489
Table I. Summary of early complications (N = 167) Complication
No. of patients
Urinary retention
13
7.78
1 7 3
0.6 4.2 1.8
Bleeding Abscess Fistula
%
Table II. Summary of late complications (N = 61) Complication
Transsphincteric (70%) Fig. 1. Summary of fistula classifications.
ranged from 19 to 135 months (mean 71.2 months). L o n g - t e r m results regarding incontinence, recurrence, and functional outcome are based on this group of 61 patients. Ninety-eight percent of patients questioned (60 of 61) rated their outcome as either good or excellent. Two patients (3.28%) developed a fistula-in-ano at a site remote from their index fistula, requiring a second mucosal advancement flap for definitive treatment. These recurrences were at 33 and 47 months, respectively. One patient developed anal stenosis and one developed a mucosal ectropion. Nine patients (15 %) complained of varying degrees of incontinence following repair. Six patients complained of incontinence to flatus, whereas three patients complained of incontinence to liquid stool. N o patient was disabled by his or her incontinence and no patient complained of incontinence to solid stool ('Ihble II). All complications developed in patients with either complex or unclassified fistulas. DISCUSSION Cryptoglandular fistulas-in-ano originate from an anal crypt at the dentate line and traverse varying amounts of the anal sphincter mechanism. Traditionally, simple fistulas (superficial, intersphincteric, and low transsphincteric) have been treated by fistulotomy with a low recurrence rate and minimal clinical alteration in continence. There is, however, a diminution in anal canal resting pressure, s and varying degrees of fecal incontinence have been reported.l-3 More corn-
No. of patients
%
Fistula
2
3.28
Incontinence Flatus Liquid Ectropion Anal stenosis
6 3 i 1
9.8 4.9 0.6 0.6
plex fistulas (high transsphincteric, suprasphincteric, and extrasphincteric) have been treated with seton placement and staged fistulotomy with similar alteration in continence. 9,1~ Recently there has been renewed interest in the endorectal mucosai advancement flap because of continuing concerns regarding sphincter damage and incontinence following "conventional" fistula surgery. Presumed advantages of this procedure include less alteration in anatomic integrity, preservation of the sphincter mechanism, rapid healing, reduced pain, and avoidance of a diverting stoma. Additionally, the technique has the advantage that recurrence does not preclude a repeat advancement flap or a "traditional" surgical approach. Evaluation of the surgical treatment of fistula-inano includes assessment of healing time, recurrence, and fecal incontinence. Healing time averaged 6 weeks in our series with 10 patients (5.98%) experiencing delayed healing of their surgical wounds. T h e majority of these patients had external wounds that either closed prematurely or developed subcutaneous fistula as a result of bridging at the site of the external wound. These patients all required reopening of the external wound or superficial fistulotomy. One patient required a repeat mucosal advancement flap procedure. T h e delayed healing in our series compares favorably with the 8.75% of patients whose wounds took longer than 26 weeks to heal in the report by Vasilevsky and Gordon." Recurrence rates range from 3.7% 11 to 40.6% 1_,in the literature. Complex fistula, previous history of
490
Journal of Gastrointestinal Surgery
Golub et al.
anorectal abscess, and failure to identify all of the tracts at the index operation have all been implicated in the development of a recurrent abscess or fistula. The initial series from this institution reported a recurrence rate of 1.5%. 7 The recurrence rate of 3.28% in this series compares very favorably with all published reports. T h e two patients with recurrent abscess and fistula presented at 33 and 47 months postoperatively, and both had abscesses and fistulas remote from the index abscess. These may actually represent new fistulous tracts rather than recurrent disease. Fecal incontinence is reported in almost any series published on the surgical treatment of fistula-in-ano. Comparison of various studies is difficult because of a lack of standard reporting techniques for patients with altered continence. In this series, six patients available for long-term follow-up complained of incontinence to flatus and three complained of incontinence to liquid stool. The mean follow-up in this group of 61 patients was 71.2 months. N o patient complained of gross incontinence, and no patient was disabled by his or her incontinence. Reports in the literature demonstrate rates of incontinence ranging from 0.00% 13 to 30%. 14These reports all rely on "conventional" surgical therapy with either fistulotomy or seton placement. Our incontinence rate in 61 patients followed for a mean of six years compares favorably with the rates quoted in the literature. T h e endorectal advancement flap was originally used in the treatment of rectovaginal fistula, but there are now several reports documenting the success of this technique in the repair of complex cryptoglandular fistula-in-ano and rectourethral fistulas. 15,16In the original series from this institution, Aguilar et al. 7 reported on 189 patients who underwent advancement flap for fistula-in-ano. An 80% follow-up, ranging from 8 months to 7 years, revealed a 90% asymptomatic group and a 10% group who had minor symptoms. Eight percent of the symptomatic patients had minor soiling; 7% were incontinent of gas and 6% were incontinent of loose stool. N o patient was incontinent of solid feces. Three patients (1.5%) developed recurrent fistula-in-ano. Two of them had a second mucosal advancement flap performed with good results and the other patient declined further therapy. Similarly, in a study of 15 patients with high recurrent anal fistula, Oh 17 described satisfactory healing in 87% and Wedell et al. 18reported complete healing in 29 of 30 patients. Although it is generally accepted that anorectal surgery should be avoided in patients with Crohn's disease, some authors have recently begun using endorectal mucosal advancement flaps in selected Crohn's patients with encouraging results. In a retro-
spective study from the Cleveland Clinic, Ozuner et al. 19 reported on 101 patients undergoing advancement flap procedures including 47 with Crohn's disease. Overall recurrence was seen in 29 patients (29%); however, the etiology of the fistula did not statistically affect outcome. Failure was only influenced by previous number of repairs. In a smaller study of eight patients, six of whom had Crohn's disease, healing was achieved in six. Early recurrence occurred in one patient with acute Crohn's proctitis and in one patient with an idiopathic horseshoe track. 2~ In a larger series Kodner et al. 21 reported on 107 patients undergoing endorectal advancement flap over a 10-year period. The causes were obstetric injury in 48, cryptoglandular abscess fistula in 31, Crohn's disease in 24, and trauma in four. Complete primary healing was achieved in 90 patients (84%). However, of the 17 patients with persistent fistula, nine underwent successful secondary surgical correction (superficial fistulotomy or repeat advancement flap) for a final success rate of 94%. As with the previous reports, this technique worked well in patients with Crohn's disease. Seventeen (71%) of 24 patients healed after initial repair and 22 (92%) of 24 patients healed after more than one procedure. Recently anorectal manometry was introduced in the evaluation of these patients in an attempt to determine the physiologic consequences of this procedure. Manometric evaluation in 10 patients both preand postoperatively found patients with cryptoglandular fistula-in-ano to have normal physiologic parameters and, most important, endorectal mucosal advancement did not statistically alter anorectal motor or sensory function in the postoperative period. 2" Other studies have reported similar results. In a series of 11 patients undergoing endorectal advancement flap for idiopathic, infralevator, transsphincteric fistulas, Finan 23 recorded successful healing in 10. One patient developed an abscess under the flap and the fistula was laid open in the standard manner. All patients remained continent. Additionally, pre- and postoperative manometric testing revealed no change in anal sensation, median maximal resting pressure, or the resting pressure profile of the anal canal. CONCLUSION Endorectal mucosal advancement flap is a safe, effective, and easily learned technique for the treatment of complex cryptoglandular fistula-in-ano. The advantages of this procedure include preservation of normal anorectal anatomy, avoidance of large perineal wounds, preservation of the anal sphincter mechanism, and avoidance of a diverting stoma. It can be performed with minimal morbidity, no deaths, and a
Vol. 1, No. 5 1997
l o w r e c u r r e n c e rate. W e b e l i e v e t h a t e n d o r e c t a l m u cosal advancement flap should be considered the t r e a t m e n t o f c h o i c e f o r c o m p l e x c r y p t o g l a n d u l a r fistula-in-ano. REFERENCES 1. Sainio P, Husa A. Fistula-in-ano: Clinical features and longterm results of surgery in 199 adults. Acta Chir Scand 1985; 151:169-176. 2. Vasilevsky CA, Gordon PH. Results of treatment of fistulain-ano. Dis Colon Rectum 1985;28:225-231. 3. Bennett RC. A review of the results of orthodox treatment for anal fistula. Proc R Soc Med 1962;55:756-757. 4. Noble GH. New operation for complete laceration of the perincum designed for the purpose of eliminating infection from the rectum. Trans Am Gynecol Soc 1902;27:357-363. 5. Elting AW. The treatment of fistula-in-ano. Ann Surg
1912;56:744-752. 6. Laird DR. Procedures used in the treatment of complicated fistulas. AmJ Surg 1948;76:701-708. 7. Aguilar PS, Plasencia G, Hardy TGJr, Hartmann Rb, Steward WRC. Mucosal advancement in the treatment of anal fistula. Dis Colon Rectum 1985;28:496-498. 8. Sainio P, Husa A. A prospective manometric study of the effect of anal fistula surgery on anorectal function. Acta Chir Scand
1985;151:279-288. 9. WilliamsJG, Macleod CA, Rothenberger DA, Goldberg SM. Seton treatment of high anal fistula. BrJ Surg 1991;78:11591161. 10. Parks AG, Stitz RW. The treatment of high fistula-in-ano. Dis Colon Rectum 1976;19:487-499. 11. Ramanujam PS, Prasad MI,, Abcarian H, Tan AB. Perianal abscesses and fistulas. Dis Colon Rectum 1984;27:593-597.
Mucosal Advancement Flap
491
12. Schouten WR, Van Vroonhaven TJMV. Treaunent of anorectal abscess with or without primary fistulotomy: Results of a prospective trial. Dis Colon Rectum 1991;34:60-63. 13. Mazier WP. The treatment and care of anal fistulas: A study of 1,000 patients. Dis Colon Rectum 1971;14:134-144. 14. Parks AG, Gordon PH, HardcastleJI). A classification of fistula-in-ano. BrJ Surg 1976;63:1-12. 15. Fazio VW, Jones IT, Jagelman DG, Weakley FL. Rectourethral fistulas in Crohn's disease. Surg Gynecol Obstet 1987;169:148-150. 16. Parks AG, Motson RW. Perianal repair of rectuprostatic fistula. BrJ Surg 1983;70:725-726. 17. Oh C. Management of high recurrent anal fistula. Surgery 1983;93:330-332. 18. Wedell J, Meier zu Eissen P, Banzhaf G, Kleine L. Sliding flap advancement for the treatment of high level fistulae. BrJ Surg 1987;74:390-391. 19. Ozuner G, Hull TL, Cartmill J, Fazio VW. Long-term analysis of the use of transanal rectal advancement flaps for complicated anorectal/vaginal fistulas. Dis Colon Rectum 1996; 39:10-14. 20. Bartolo DCC, Lewis P. "I]'eatment oftrans-sphincteric fistulae by full thickness anorectal advancement flaps. Br J Surg 1990;77:1187-1189. 21. Kodner IJ, Mazor A, Shemesh El, Fry RD, Fleshmen JW, Birnbaum EH. Endorectal advancement flap repair of rectovaginal and other complicated anorectal fistulas. Surgery 1993;114:682-690. 22. Wise WE, Kerner BA, Golub RW, Khandjua KS, Aguilar PS. Endorectal mucosal advancement flap: A physiologic and clinical evaluation [Abst]. Dis Colon Rectum 1993;36:PI 3. 23. Finan PJ. Management by advancement flap technique. In Phillips R, Lunnjiss PJ, eds. Anal Fistula. London: Chapman & Hall, 1996.
Cryptoglandular fistula-in-ano is a common affliction that usually responds well to conventional surgical procedures such as fistulectomy, fistulotomy, and seton placement. These procedures, however, can be associated with varying degrees of fecal incontinence. Endorectal mucosal advancement flap has been advocated as an alternative procedure that avoids this problem. This study was undertaken to determine the risks and benefits associated with endorectal mucosal advancement flap in the treatment of complex fistula-in-ano. One hundred sixty-four patients underwent 167 endorectal mucosal advancement flap procedures for complex cryptoglandular fistula-in-ano between January 1982 and December 1990. There were 126 men and 38 women whose mean age was 42.1 years (range 20 to 79 years). The majority of the patients (70%) had complex fistulas (transsphincteric, suprasphincteric, or extrasphincteric). Fifteen patients (9%) had an intersphincteric fistula. All patients were available for short-term follow-up (6 weeks). Postoperative morbidity was minimal and included urinary retention in 13 patients (7.8%) and bleeding in one patient. Healing time averaged 6 weeks. Long-term follow-up, ranging from 19 to 135 months, was carried out in 61 patients. There were two recurrences (3.28%). Nine patients (15%) complained of varying degrees of fecal incontinence. Six patients complained of incontinence to flatus and three patients complained of incontinence to liquid stool. No patient was incontinent of solid stool. Sixty patients (98%) rated their functional result as excellent or good. Endorectal mucosal advancement flap is a safe and effective technique for the treatment of complex cryptoglandular fistula-in-ano. It can be performed with minimal morbidity, no mortality, an acceptable recurrence rate, and little alteration in anorectal continence. (J GASTROINTESTSURf_;1997; 1:487-491 .)
Cryptoglandular fistula-in-ano is a c o m m o n afflicdon that usually responds well to conventional operative procedures such as fistulectomy, fistulotomy, and seton placement. T h e s e procedures, however, can be associated with varying degrees of fecal incontinence ranging from 6% to 4 5 % ) -3 T h e risk of fecal incontinence is particularly high with complex fistulas (transsphincteric, suprasphincteric, and extrasphincteric), and therefore these fistulas require careful preoperative evaluation and often alternative surgical techniques. One such technique is the endorectal mucosal advancement flap. First proposed by N o b l e 4 in 1902 for repair of a rectovaginal fistula, and later modified by Elting 5 and Laird, 6 this technique employs a broad-based flap of rectal wall to obliterate the
origin of the fistula. In 1985 Aguilar et al. 7 proposed the use of an endorectal mucosal advancement flap to treat complex cryptoglandular fistula-in-ano. This report is a continuation of that series. SURGICAL
TECHNIQUE
Most of the patients undergoing an endorectal mucosal advancement flap procedure for cryptoglandular fistula-in-ano received a standard anorectal bowel preparation consisting of a clear liquid diet and a Fleet enema (C.B. Fleet Company Inc., Lynchburg, Va.) on the evening before and the morning of surgery. Intravenous antibiotics were employed only for endocarditis or prosthetic joint prophylaxis when indicated.
From the Division of Colon and Rectal Surgery, Grant Medical Center, Columbus, Ohio. Presented in part at the annual meeting of the American Society of Colon and Rectal Surgeons, Chicago, Ill., May 2-7, 1993. Reprint requests: Richard W. Gx)lub, M.D_, EA.C.S., State University of New York Health Science Center at BrooHyn, 450 Clarkson Ave., Box 40, Brooklyn NY 11203. 487
488
Jour,lal of Gastrointestinal Surgery
(-;olub et al.
T h e technique of endorectal mucosal advancement flap has been described extensively elsewhere and is essentially a modification of the techniques originally developed by Noble 4 and Elting. 5 General, regional, or local anesthesia with intravenous sedation can be used. Intraoperative fluids are limited. Most patients are placed in the prone jackknife position with the buttocks taped apart, although the lithotomy position can be advantageous for posterior fistulas. Illumination with a headlamp can be helpful. T h e perineum and anal canal are prepared and irrigated with povidone-iodine solution. Anesthesia is supplemented with a perianal field block consisting of bupivacaine 0.25%, lidocaine 0.5% with 1/200,000 epinephrine, and 150 units of hyaluronidase (Wydase, Wyeth-Ayerst Laboratories, Philadelphia, Pa.). A Hill-Ferguson retractor is placed in the anal canal and a thorough anorectal evaluation is undertaken. Associated pathologic conditions are noted. Internal and external openings are located, and the course of the fistulous tract is carefully assessed by palpation and gentle probing. T h e fistulous tract is removed subcutaneously from the external opening to the level of the external sphincter muscle. T h e residual tract and any granulation tissue is removed by curettage. Secondary tracts are noted and treated in a similar fashion. T h e perianal skin is widely saucerized. Using sharp scissors dissection, a flap consisting of anoderm, mucosa, and submucosa is raised from the intersphincteric groove to a point approximately 3 to 4 cm proximal to the internal opening. T h e configuration of the flap is trapezoidal with the base wider than the apex to ensure an adequate distal blood supply. Meticulous electrocautery hemostasis is maintained t h r o u g h o u t the course of the dissection. T h e opening in the internal anal sphincter is closed with interrupted 3-0 chromic sutures. T h e flap is advanced distally, without tension, to cover the defect in the internal sphincter and the excess mucosa (including the internal opening) is amputated. T h e distal edge is sutured to the internal anal sphincter at the level of the intersphincteric groove, and the lateral aspects are reapproximated to adjacent rectal mucosa using 2-0 or 3-0 chromic catgut sutures. Sterile dressings are applied to the wound. Packing, topical antibiotics, and suction catheters are not routinely used. Fecal diversion is avoided. Repair of the external voluntary sphincter was not undertaken in this series of patients. However, in patients presenting with anteriorly placed fistulas, perineal body reconstruction and/or simple reefing of the anterior sphincter musculature can be added to the procedure if preoperative history suggests inconti-
nence or physical examination demonstrates lax musculature or an attenuated perineal body. Postoperatively the patient is started on a regular diet supplemented with bulk fiber agents and stool softeners. Oral analgesics, diazepam, and warm sitz baths are employed to promote local hygiene and to control pain and muscle spasms. Patients are discharged home when they are able to void without difficulty. Follow-up is conducted every 2 to 3 weeks until all wounds are healed. RESULTS One hundred sixty-four patients underwent 167 endorectal mucosal advancement flap procedures for cryptoglandular fistula-in-ano from January 1982 through December 1990. Fistulas secondary to inflammatory bowel disease, actinomycosis, foreign bodies, and malignancy were excluded from this study, as were rectovaginal and rectourethral fistulas. There were 126 men and 38 women whose mean age was 42.1 years (range 20 to 79 years). Sixty-two percent of patients had undergone prior anorectal surgery, including 18 patients (11%) who had previous fistula surgery. Perirectal pain and drainage were the most common presenting sTmptoms with a preoperative duration of 1 to 3 months. Five patients (3 %) presented with an acute fistulous abscess. All patients underwent an endorectal mucosal advancement flap as their definitive surgical procedure. One hundred fifteen patients (70%) were classified as having complex fistula-in-ano (transsphincteric, suprasphincteric, or extrasphincteric), whereas 15 patients (9%) had intersphincteric fistula. T h e majority of these 15 patients had anteriorly placed fistulas and were deemed at risk for incontinence with conventional surgical technique. T h e type of fistula was not recorded in 34 patients (Fig. 1). An internal opening was located in the posterior midline in 87 patients (53%). All 164 patients were available for short-term follow-up. Complications and healing rates are based on this group of patients. Healing time averaged 6 weeks for all patients. Ten patients (5.98%) experienced prolonged healing. All of these patients developed either small subcutaneous (7 patients) or superficial fistulas in the early postoperative period. Six patients were treated with incision and drainage, three patients had an in-office fistulotomy, and one patient required a repeat mucosal advancement flap. Postoperative morbidity also included urinary retention (13 patients) and bleeding (1 patient) (Table I). T h e r e were no operative deaths. Long-term follow-up was carried out by office visit or telephone questionnaire in 61 patients (37%) and
Vol. 1, No. 5 1997
Mucosal Advancement Flap
Suprasphincteric (1%) Unclassified ~ / Intersphincteric (20%) \ ~ (9%)
489
Table I. Summary of early complications (N = 167) Complication
No. of patients
Urinary retention
13
7.78
1 7 3
0.6 4.2 1.8
Bleeding Abscess Fistula
%
Table II. Summary of late complications (N = 61) Complication
Transsphincteric (70%) Fig. 1. Summary of fistula classifications.
ranged from 19 to 135 months (mean 71.2 months). L o n g - t e r m results regarding incontinence, recurrence, and functional outcome are based on this group of 61 patients. Ninety-eight percent of patients questioned (60 of 61) rated their outcome as either good or excellent. Two patients (3.28%) developed a fistula-in-ano at a site remote from their index fistula, requiring a second mucosal advancement flap for definitive treatment. These recurrences were at 33 and 47 months, respectively. One patient developed anal stenosis and one developed a mucosal ectropion. Nine patients (15 %) complained of varying degrees of incontinence following repair. Six patients complained of incontinence to flatus, whereas three patients complained of incontinence to liquid stool. N o patient was disabled by his or her incontinence and no patient complained of incontinence to solid stool ('Ihble II). All complications developed in patients with either complex or unclassified fistulas. DISCUSSION Cryptoglandular fistulas-in-ano originate from an anal crypt at the dentate line and traverse varying amounts of the anal sphincter mechanism. Traditionally, simple fistulas (superficial, intersphincteric, and low transsphincteric) have been treated by fistulotomy with a low recurrence rate and minimal clinical alteration in continence. There is, however, a diminution in anal canal resting pressure, s and varying degrees of fecal incontinence have been reported.l-3 More corn-
No. of patients
%
Fistula
2
3.28
Incontinence Flatus Liquid Ectropion Anal stenosis
6 3 i 1
9.8 4.9 0.6 0.6
plex fistulas (high transsphincteric, suprasphincteric, and extrasphincteric) have been treated with seton placement and staged fistulotomy with similar alteration in continence. 9,1~ Recently there has been renewed interest in the endorectal mucosai advancement flap because of continuing concerns regarding sphincter damage and incontinence following "conventional" fistula surgery. Presumed advantages of this procedure include less alteration in anatomic integrity, preservation of the sphincter mechanism, rapid healing, reduced pain, and avoidance of a diverting stoma. Additionally, the technique has the advantage that recurrence does not preclude a repeat advancement flap or a "traditional" surgical approach. Evaluation of the surgical treatment of fistula-inano includes assessment of healing time, recurrence, and fecal incontinence. Healing time averaged 6 weeks in our series with 10 patients (5.98%) experiencing delayed healing of their surgical wounds. T h e majority of these patients had external wounds that either closed prematurely or developed subcutaneous fistula as a result of bridging at the site of the external wound. These patients all required reopening of the external wound or superficial fistulotomy. One patient required a repeat mucosal advancement flap procedure. T h e delayed healing in our series compares favorably with the 8.75% of patients whose wounds took longer than 26 weeks to heal in the report by Vasilevsky and Gordon." Recurrence rates range from 3.7% 11 to 40.6% 1_,in the literature. Complex fistula, previous history of
490
Journal of Gastrointestinal Surgery
Golub et al.
anorectal abscess, and failure to identify all of the tracts at the index operation have all been implicated in the development of a recurrent abscess or fistula. The initial series from this institution reported a recurrence rate of 1.5%. 7 The recurrence rate of 3.28% in this series compares very favorably with all published reports. T h e two patients with recurrent abscess and fistula presented at 33 and 47 months postoperatively, and both had abscesses and fistulas remote from the index abscess. These may actually represent new fistulous tracts rather than recurrent disease. Fecal incontinence is reported in almost any series published on the surgical treatment of fistula-in-ano. Comparison of various studies is difficult because of a lack of standard reporting techniques for patients with altered continence. In this series, six patients available for long-term follow-up complained of incontinence to flatus and three complained of incontinence to liquid stool. The mean follow-up in this group of 61 patients was 71.2 months. N o patient complained of gross incontinence, and no patient was disabled by his or her incontinence. Reports in the literature demonstrate rates of incontinence ranging from 0.00% 13 to 30%. 14These reports all rely on "conventional" surgical therapy with either fistulotomy or seton placement. Our incontinence rate in 61 patients followed for a mean of six years compares favorably with the rates quoted in the literature. T h e endorectal advancement flap was originally used in the treatment of rectovaginal fistula, but there are now several reports documenting the success of this technique in the repair of complex cryptoglandular fistula-in-ano and rectourethral fistulas. 15,16In the original series from this institution, Aguilar et al. 7 reported on 189 patients who underwent advancement flap for fistula-in-ano. An 80% follow-up, ranging from 8 months to 7 years, revealed a 90% asymptomatic group and a 10% group who had minor symptoms. Eight percent of the symptomatic patients had minor soiling; 7% were incontinent of gas and 6% were incontinent of loose stool. N o patient was incontinent of solid feces. Three patients (1.5%) developed recurrent fistula-in-ano. Two of them had a second mucosal advancement flap performed with good results and the other patient declined further therapy. Similarly, in a study of 15 patients with high recurrent anal fistula, Oh 17 described satisfactory healing in 87% and Wedell et al. 18reported complete healing in 29 of 30 patients. Although it is generally accepted that anorectal surgery should be avoided in patients with Crohn's disease, some authors have recently begun using endorectal mucosal advancement flaps in selected Crohn's patients with encouraging results. In a retro-
spective study from the Cleveland Clinic, Ozuner et al. 19 reported on 101 patients undergoing advancement flap procedures including 47 with Crohn's disease. Overall recurrence was seen in 29 patients (29%); however, the etiology of the fistula did not statistically affect outcome. Failure was only influenced by previous number of repairs. In a smaller study of eight patients, six of whom had Crohn's disease, healing was achieved in six. Early recurrence occurred in one patient with acute Crohn's proctitis and in one patient with an idiopathic horseshoe track. 2~ In a larger series Kodner et al. 21 reported on 107 patients undergoing endorectal advancement flap over a 10-year period. The causes were obstetric injury in 48, cryptoglandular abscess fistula in 31, Crohn's disease in 24, and trauma in four. Complete primary healing was achieved in 90 patients (84%). However, of the 17 patients with persistent fistula, nine underwent successful secondary surgical correction (superficial fistulotomy or repeat advancement flap) for a final success rate of 94%. As with the previous reports, this technique worked well in patients with Crohn's disease. Seventeen (71%) of 24 patients healed after initial repair and 22 (92%) of 24 patients healed after more than one procedure. Recently anorectal manometry was introduced in the evaluation of these patients in an attempt to determine the physiologic consequences of this procedure. Manometric evaluation in 10 patients both preand postoperatively found patients with cryptoglandular fistula-in-ano to have normal physiologic parameters and, most important, endorectal mucosal advancement did not statistically alter anorectal motor or sensory function in the postoperative period. 2" Other studies have reported similar results. In a series of 11 patients undergoing endorectal advancement flap for idiopathic, infralevator, transsphincteric fistulas, Finan 23 recorded successful healing in 10. One patient developed an abscess under the flap and the fistula was laid open in the standard manner. All patients remained continent. Additionally, pre- and postoperative manometric testing revealed no change in anal sensation, median maximal resting pressure, or the resting pressure profile of the anal canal. CONCLUSION Endorectal mucosal advancement flap is a safe, effective, and easily learned technique for the treatment of complex cryptoglandular fistula-in-ano. The advantages of this procedure include preservation of normal anorectal anatomy, avoidance of large perineal wounds, preservation of the anal sphincter mechanism, and avoidance of a diverting stoma. It can be performed with minimal morbidity, no deaths, and a
Vol. 1, No. 5 1997
l o w r e c u r r e n c e rate. W e b e l i e v e t h a t e n d o r e c t a l m u cosal advancement flap should be considered the t r e a t m e n t o f c h o i c e f o r c o m p l e x c r y p t o g l a n d u l a r fistula-in-ano. REFERENCES 1. Sainio P, Husa A. Fistula-in-ano: Clinical features and longterm results of surgery in 199 adults. Acta Chir Scand 1985; 151:169-176. 2. Vasilevsky CA, Gordon PH. Results of treatment of fistulain-ano. Dis Colon Rectum 1985;28:225-231. 3. Bennett RC. A review of the results of orthodox treatment for anal fistula. Proc R Soc Med 1962;55:756-757. 4. Noble GH. New operation for complete laceration of the perincum designed for the purpose of eliminating infection from the rectum. Trans Am Gynecol Soc 1902;27:357-363. 5. Elting AW. The treatment of fistula-in-ano. Ann Surg
1912;56:744-752. 6. Laird DR. Procedures used in the treatment of complicated fistulas. AmJ Surg 1948;76:701-708. 7. Aguilar PS, Plasencia G, Hardy TGJr, Hartmann Rb, Steward WRC. Mucosal advancement in the treatment of anal fistula. Dis Colon Rectum 1985;28:496-498. 8. Sainio P, Husa A. A prospective manometric study of the effect of anal fistula surgery on anorectal function. Acta Chir Scand
1985;151:279-288. 9. WilliamsJG, Macleod CA, Rothenberger DA, Goldberg SM. Seton treatment of high anal fistula. BrJ Surg 1991;78:11591161. 10. Parks AG, Stitz RW. The treatment of high fistula-in-ano. Dis Colon Rectum 1976;19:487-499. 11. Ramanujam PS, Prasad MI,, Abcarian H, Tan AB. Perianal abscesses and fistulas. Dis Colon Rectum 1984;27:593-597.
Mucosal Advancement Flap
491
12. Schouten WR, Van Vroonhaven TJMV. Treaunent of anorectal abscess with or without primary fistulotomy: Results of a prospective trial. Dis Colon Rectum 1991;34:60-63. 13. Mazier WP. The treatment and care of anal fistulas: A study of 1,000 patients. Dis Colon Rectum 1971;14:134-144. 14. Parks AG, Gordon PH, HardcastleJI). A classification of fistula-in-ano. BrJ Surg 1976;63:1-12. 15. Fazio VW, Jones IT, Jagelman DG, Weakley FL. Rectourethral fistulas in Crohn's disease. Surg Gynecol Obstet 1987;169:148-150. 16. Parks AG, Motson RW. Perianal repair of rectuprostatic fistula. BrJ Surg 1983;70:725-726. 17. Oh C. Management of high recurrent anal fistula. Surgery 1983;93:330-332. 18. Wedell J, Meier zu Eissen P, Banzhaf G, Kleine L. Sliding flap advancement for the treatment of high level fistulae. BrJ Surg 1987;74:390-391. 19. Ozuner G, Hull TL, Cartmill J, Fazio VW. Long-term analysis of the use of transanal rectal advancement flaps for complicated anorectal/vaginal fistulas. Dis Colon Rectum 1996; 39:10-14. 20. Bartolo DCC, Lewis P. "I]'eatment oftrans-sphincteric fistulae by full thickness anorectal advancement flaps. Br J Surg 1990;77:1187-1189. 21. Kodner IJ, Mazor A, Shemesh El, Fry RD, Fleshmen JW, Birnbaum EH. Endorectal advancement flap repair of rectovaginal and other complicated anorectal fistulas. Surgery 1993;114:682-690. 22. Wise WE, Kerner BA, Golub RW, Khandjua KS, Aguilar PS. Endorectal mucosal advancement flap: A physiologic and clinical evaluation [Abst]. Dis Colon Rectum 1993;36:PI 3. 23. Finan PJ. Management by advancement flap technique. In Phillips R, Lunnjiss PJ, eds. Anal Fistula. London: Chapman & Hall, 1996.