Tech Coloproctol (2009) 13:135140 DOI 10.

1007/s10151-009-0470-x

O R I G I NA L A RT I C L E

Innovative technique for the closure of rectovaginal fistula using Surgisis mesh
O. Schwandner A. Fuerst K. Kunstreich R. Scherer

Received: 15 January 2009 / Accepted: 6 March 2009 / Published online: 30 May 2009 Springer-Verlag 2009

Abstract Background The aim of this prospective study was to analyse the efficacy of Surgisis mesh for closure of rectovaginal fistulas. Prospective data were collected from two centres. Methods All patients with a rectovaginal fistula who underwent definitive surgery using Surgisis mesh were prospectively enrolled in this study. Inclusion criteria included a rectovaginal fistula in the lower two-thirds of the rectovaginal septum. Surgery was performed with a standardized technique including combined transrectal and transvaginal excision of the rectovaginal fistula with transvaginal placement of the mesh. Success was defined as closure of both internal and external (perianal and vaginal) openings, absence of drainage without further intervention, and no abscess formation. Results Over a period of 16 months, a total of 21 mesh procedures were performed in two centres. The mean age of the patients was 47 years (1859 years). Of the 21 patients, 18 (86%) had recurrent rectovaginal fistula, and the mean number of prior attempts was 2.3 (08). The majority of patients (nine) had Crohns disease-associat-

ed fistula, followed by six with iatrogenic fistula, two with radiation-induced fistula, two with obstetric injuryinduced fistula, and two with idiopathic fistula. The mesh procedure was performed under faecal diversion in eight patients (38%). The mean operative time was 38 min; no intraoperative morbidity occurred. Patients were discharged from hospital on day 4. After a mean follow-up of 12 months (range, 318 months), the overall success rate after primary mesh procedure was 71% (15/21; 6 patients had failure or recurrence). All patients with failure or recurrence were reoperated upon. Out of these six patients who were reoperated upon, four had definite healing (75%). Among the eight patients who had faecal diversion, four (50%) had reversal of their stoma. Conclusion The preliminary success rate for this innovative technique using Surgisis mesh for the closure of rectovaginal fistulas is promising. Further studies are needed to assess the definite role of this novel technique in comparison to traditional surgical procedures Key words Rectovaginal fistula Bioprosthesis

Introduction Failure, reoperations and recurrences of rectovaginal fistulas have a major impact on anal continence and quality of life. The technical feasibility of closing rectovaginal fistulas using new bioprosthetics such as Surgisis mesh (Cook Medical, Bloomington, IN, USA) has already been demonstrated [1, 2]. As it is generally accepted that both recurrence and reoperation rates are much higher in rectovaginal fistulas associated with Crohns disease, and the risk of failure is influenced by the number of prior attempts to close the fistula, it was the aim of this

O. Schwandner () A. Fuerst Department of Surgery and Pelvic Floor Center, Caritas-Krankenhaus St. Josef, Landshuter Str. 65, Regensburg, 93053, Germany E-mail: oschwandner@caritasstjosef.de K. Kunstreich R. Scherer Department of Coloproctology and Pelvic Floor Surgery, Krankenhaus Waldfriede, Berlin, Germany

136

Tech Coloproctol (2009) 13:135140

prospective two-centre study to analyse the short-term efficacy of Surgisis mesh for the closure of rectovaginal fistulas using a standardized surgical technique.

Methods

Study design All patients suffering from a rectovaginal fistula who underwent surgery using Surgisis mesh were prospectively enrolled in this study. Inclusion criteria included a rectovaginal fistula in the lower two-thirds of the rectovaginal space. Related to the origin of the fistulas, all patients with a recurrent, iatrogenic, postobstetric, radiation-induced and Crohns disease-associated rectovaginal fistula were candidates for the procedure. In those with inflammatory bowel disease, only fistulas with concurrent stable disease and without any evidence of severe inflammatory activity or perianal sepsis were operated upon using the mesh. The data comprised pooled prospective data from two coloproctology centres in Germany (Berlin, Regensburg). Data were collected prospectively using a PC database and included the following variables: patient demographics, origin of the fistula, prior surgical treatment, comorbidity, faecal diversion, time of surgery, morbidity, hospital stay, and follow-up information (success, reoperation). Primary end-points of the study included success or failure. Success was defined as closure of both openings (vaginal and rectal), absence of drainage without further intervention, and no abscess formation. In all patients, informed consent focusing on the unclear long-term outcome of the mesh procedure and the necessity for regular follow-up visits was obtained. Surgical technique Pre- and postoperative management as well as the surgery were performed according to a standardized technique in both centres. Three surgeons (O.S., A.F., Regensburg; R.S., Berlin) performed the mesh procedure. All patients were examined preoperatively in the proctological office. Patients who had seton drainage due to previous drainage of a concomitant abscess were examined by endoanal ultrasonography to exclude residual anorectal or rectovaginal abscess. In recurrent fistulas, faecal diversion (loop ileostomy or loop colostomy) was advised. On the day of surgery, bowel preparation was performed with an enema. Single-dose antibiotic prophylaxis (cefotaxime and metronidazole) was mandatory. Both procedures were performed under either general or spinal anaesthesia and in the lithotomy position. The placement of a urinary catheter was mandatory in all patients. After removal of

the seton drainage, a combined transvaginal and transrectal approach was performed. Primarily, the posterior vaginal wall was opened and the fistula tract was identified with a conventional fistula probe. The vagina was dissected from the rectum within the rectovaginal space approximately 1 cm proximal and distal to the fistula tract. Consequently, the fistula tract was completely excised from a combined transvaginal and transrectal approach including a complete coring-out of the fistula tissue. The rectal component of the operation consisted of an advancement flap repair using PDS 2/0 (muscle layer) and Vicryl 3/0 (rectal mucosa and submucosa) sutures (Ethicon Endo-Surgery, Norderstedt, Germany). After completion of the transrectal advancement flap repair, the rectovaginal space was irrigated with antiseptic solution. Then, the Surgisis mesh was cut to size (approximately 22 cm), placed in the rectovaginal space and fixed with Vicryl 3/0 at each corner. Finally, the posterior vaginal mucosa was reclosed over the mesh with interrupted Vicryl 3/0 sutures. The major steps of the operation are illustrated in Figures 13. Postoperatively, immediate oral feeding was administered if faecal diversion was present. In patients without faecal diversion clear liquids were administered for 2 days. No further antibiotics were given routinely. The urinary catheter was removed on the second or third postoperative day. Patients were discharged on the fourth postoperative day.

Follow-up All patients were followed up at 2 weeks after surgery to monitor wound healing. Specific follow-up information was derived from clinical examination 3, 6, 12 and 18 months postoperatively. This follow-up was per-

Fig. 1 Opening of the vaginal wall

Tech Coloproctol (2009) 13:135140

137

Fig. 2 Transvaginal mesh placement

Fig. 3 Transrectal mobilization of the advancement flap

formed in an office setting with regular clinical examination (including digital examination, proctoscopy, vaginal examination and evaluation of the fistula status using a fistula probe) and assessment of success or failure. No patient was lost to follow-up.

one after transperineal surgery). Finally, two patients each had a radiation-induced fistula, a postobstetric fistula, and an idiopathic fistula. Regarding comorbidity, three patients had prior surgical cancer resection (rectal cancer, anal cancer, vulval cancer), seven were tobacco smokers, and two had diabetes. The mesh procedure was performed under faecal diversion in eight patients (38%). The mean operative time was 38 min (range, 1465 min); no intraoperative morbidity occurred. Postoperatively, four patients had minor complications (urinary tract infection, two; vaginal sore, one; partial dehiscence of the rectal advancement flap requiring proctoscopic lavage and oral antibiotics, one) which were treated conservatively. Patients were discharged from hospital on day 4 (range, 29). After a mean follow-up of 12 months (range, 318 months), overall success rate after primary mesh procedure was 71% (15/21; 6 patients had failure or recurrence of their rectovaginal fistula after the primary mesh procedure). All patients with failure or recurrence were reoperated upon (Table 1); the mean number of reoperations was two (range one to five). Of the six patients who were reoperated upon, four had definite healing (75%). The success rates related to the procedure, the origins of the fistulas and the number of previous operations are shown in Table 2. Despite reoperation, at the time of this report there were still two patients who were suffering from rectovaginal fistula (seton drainage, faecal diversion), and in whom gracilis transposition was planned. Among the eight patients who had faecal diversion, four (50%) had reversal of their stoma ; one reason for the stoma not being reversed was fistula persistence, and in these two patients gracilis transposition was planned. One patient had active Crohns colitis in the further course, and the other patient wanted to extend the followup time prior to stoma reversal.

Discussion Results A total of 21 mesh procedures were performed between July 2007 and October 2008 (10 at Regensburg and 11 at Berlin). The mean age of the female patients was 47 years (range, 1859 years). The majority of patients (18/21, 86%) had recurrent rectovaginal fistula, and the mean number of prior attempts to close the fistula was 2.3 (08). Related to the origin of the rectovaginal fistula, the majority of patients had Crohns disease-associated fistula (nine). Six patients had iatrogenic or surgically induced fistula (three after colorectal resection, two intraoperative injury during gynaecological surgery, and Surgical repair for rectovaginal fistula remains challenging [3]. Particularly in complex rectovaginal fistula associated with Crohns disease or after irradiation of the small pelvis, a variety of surgical options such as fistulotomy [4, 5], long-term seton [6, 7], transanal [816] or transvaginal [17] advancement flap, transperineal procedures, rectal sleeve advancement flap [1821], bilateral gluteus muscle patching [22], bulbocavernous muscle flap [23] and, finally, proctectomy [24, 25], have been applied to close the fistula. As primary healing rates after traditional surgery using flap techniques are somewhat disappointing [26, 27], the aim of this study was to deter-

138 Table 1 Reasons for failure and retreatment Patient age (years) 52 18 Origin of fistula Crohns disease Crohns disease Reason for failure Persistence of fistula, no active Crohns disease Abscess with mesh dislodgement, active Crohns disease Abscess with recurrence Redo surgery Advancement flap repair Excision of abscess, seton drainage, faecal diversion (loop ileostomy) Excision of abscess, seton drainage, proctectomy with end-colostomy redo mesh procedure

Tech Coloproctol (2009) 13:135140

No. of redo procedures 1 2

Healing after redo surgery Yes No

59

Radiation-induced

Yes

31 57

Post-obstetric Idiopathic

Persistence of fistula Persistence of fistula

1 5

Yes No

Closure with anal fistula plug, seton drainage, fistulectomy and closure with flap, preanal repair Excision of abscess, seton drainage, redo mesh procedure

54

Idiopathic

Abscess with recurrence

Yes

Table 2 Success rates in relation to the presence of Crohns disease and the number of previous attempts Success rate Overall (after primary mesh procedure) Crohns disease (after primary mesh procedure) Non-Crohns disease (after primary mesh procedure) Recurrent fistula with only one prior attempt Recurrent fistula with at least two prior attempts 71% (15/21) 78% (7/9) 83% (10/12) 88% (7/8) 62% (8/13)

mine short-term results on the feasibility and efficacy of an innovative management approach. Different centres have reported on the use of acellular cadaveric dermal grafts [28], porcine dermal grafts [29] and plugs [30] for the treatment of rectovaginal fistulas with promising results. Excellent results have been obtained with the use of absorbable synthetic mesh [31]. In the current series reporting preliminary experience from two centres, a commercially available bioprosthetic interposition graft (Surgisis) was used. Surgisis is a biocompatible mesh generated from lyophilized porcine small intestinal submucosa, which allows host cells to replace and repair damage or defective tissue. Once extracted from the porcine small intestine, the mesh is processed to enable all cells to be removed. This leaves only the natural matrix molecules, including collagen, glycosaminoglycans, and glycoproteins, that signal the host cells to repopulate the mesh with host tissue. The mesh is gradually replaced as the host cells rebuild and remodel weakened tissue. Its sterile, acellular properties

enable it to be used without the complication of rejection. Biocompatibility testing has been carried out by an accredited independent testing laboratory using testing methods based on international standards and FDA guidance. There is a significant reduction in fibrosis during the healing process, because the mesh supports the patients own connective tissue and smooth muscle growth [1]. The primary success rate of 71% for this innovative technique using this type of biomaterial in a nonselected patient population including complex rectovaginal fistula (e.g. associated with Crohns disease) is promising. However, we acknowledge that the current experience is based on only 21 procedures and success rates from only short-term follow-up are provided. On this background, we realize that definitive healing rates particularly related to the dynamic course of Crohns disease can only be evaluated with long-term follow-up. However, these short-term results indicate that this technique could potentially be an effective alternative or enrichment to traditional procedures such as advancement flap repair as the main procedure, or closure with muscle transposition (e.g. puborectalis sling or gracilis muscle transposition). However, it remains clear that rectovaginal fistulas associated with Crohns disease, as in the majority (43%) of our patients, have a poorer outcome, even though surgical techniques and medical therapy regimens have been significantly improved [27, 32]. There are two previous reports of the placement of a bioprosthetic mesh for the closure of rectovaginal fistu-

Tech Coloproctol (2009) 13:135140

139

las. Pye et al. reported successful closure of a recurrent rectovaginal fistula in a female patient in whom traditional advancement flap repair had failed [1]. In a recent series, Ellis reported a clearly lower recurrence rate after placement of a bioprosthetic mesh (19%) in comparison to the traditional advancement flap repair (34%) [2]. However, only 2 of 27 patients with a rectovaginal fistulas (7%) who underwent the mesh procedure suffered from Crohns disease [2]. To our knowledge, this is the first series showing a relatively high number of patients who underwent the mesh procedure in a standardized technique with prospective data collection and 12-month follow-up information. Obviously, the low rate of faecal diversion (38%) in our patients is contradictory to generally accepted practice. Analysing our experience with the mesh procedure, our policy relating to faecal diversion has generally changed and potential inconsistencies need to be explained. Prior to the introduction of the mesh in our institutions, faecal diversion for recurrent rectovaginal fistula was mandatory in all patients. However, following the relatively high success rates of our first procedures, we decided that faecal diversion was not a dogmatic prerequisite for the mesh procedure. Consequently, faecal diversion was only advised, and the patients were informed specifically as to the role of faecal diversion in relation to success or failure. Therefore, the issue of faecal diversion was discussed individually, and depended on several variables (e.g. the number of prior surgical procedures, the patients general health status, the extent of the rectovaginal defect). Only the long-term success rates with the mesh procedure will influence our final position regarding the role of faecal diversion in our practice. Critically reflecting our preliminary experience with the use of the mesh, we cannot yet provide a general conclusion as to the adequate and best treatment for rectovaginal fistula. In detail, a variety of issues have to be addressed in the near future including technical aspects (e.g. optimal size of the mesh), postoperative management (e.g. restriction in oral feeding) and ideal indication for or the role of faecal diversion. Finally, the question as to how to proceed in patients with failure remains unclear (e.g. re-mesh, repeat advancement flap or muscle transposition). However, looking at the different procedures of redo surgery in our series, the importance of considering all types of surgical options in these patients including advancement flap and gracilis transposition is a prerequisite to performing surgery for rectovaginal fistula. In particular, if flap repair and mesh procedure have failed, gracilis transposition seems to be the most effective surgical option [29, 33]. In conclusion, the preliminary success rate for this innovative technique using the Surgisis mesh for the clo-

sure of rectovaginal fistulas is promising. Further analysis is needed to assess the definitive role of this innovative technique in comparison to traditional surgical techniques.
Conflict of interest statement The Authors declare that they have no conflict of interest related to the publication of this article.

References
1. Pye PK, Dada T, Duthie G et al (2004) Surgisis mesh: a novel approach to repair of a recurrent rectovaginal fistula. Dis Colon Rectum 47:15541556 2. Ellis CN (2008) Outcomes after repair of rectovaginal fistulas using bioprosthetics. Dis Colon Rectum 51:10841088 3. Saclarides TJ (2002) Rectovaginal fistula. Surg Clin North Am 82:12611272 4. Scott NA, Nair A, Hughes LE (1992) Anovaginal and rectovaginal fistula in patients with Crohns disease. Br J Surg 79:13791380 5. Levy C, Tremaine WJ (2002) Management of internal fistulas in Crohns disease. Inflamm Bowel Dis 8:106111 6. Fry RD, Shemesh EI, Kodner IJ et al (1989) Techniques and results in the management of anal and perianal Crohns disease. Surg Gynecol Obstet 168:4248 7. Lichtenstein GR (2000) Treatment of fistulizing Crohns disease. Gastroenterology 119:11321147 8. Greenwald JC, Hoexter B (1978) Repair of rectovaginal fistulas. Surg Gynecol Obstet 146:443445 9. Cohen JL, Stricker JW, Schoetz DJ et al (1989) Rectovaginal fistula in Crohns disease. Dis Colon Rectum 32:825828 10. Crim RW, Fazio VW, Laveri IC (1990) Rectal advancement flap repair in Crohns disease. Factors predictive of failure. Dis Colon Rectum 33:P3 11. Makowiec F, Jehle EC, Becker HD et al (1999) Clinical course after transanal advancement flap repair of perianal fistula in patients with Crohns disease. Br J Surg 82:603606 12. Hull TL, Fazio VW (1997) Surgical approaches to low anovaginal fistula in Crohns disease. Am J Surg 173:9598 13. Kodner IJ, Mazor A, Shemesh EI et al (1993) Endorectal advancement flap repair of rectovaginal and other complicated anorectal fistulas. Surgery 114:682690 14. Sonoda T, Hull T, Piedmonte MR et al (2002) Outcomes of primary repair of anorectal and rectovaginal fistulas using the endorectal advancement flap. Dis Colon Rectum 45:16221628 15. Zimmerman DD, Gosselink MP, Briel JW et al (2002) The outcome of transanal advancement flap repair of rectovaginal fistulas is not improved by an additional labial fat transposition. Tech Coloproctol 6:3742 16. Mizrahi N, Wexner SD, Zmora O et al (2002) Endorectal advancement flap: are there predictors of failure? Dis Colon Rectum 45:16161621 17. Sher ME, Bauer JJ, Gelernt I (1991) Surgical repair of rectovaginal fistulas in patients with Crohns disease: transvaginal approach. Dis Colon Rectum 34:641648 18. Berman IR (1991) Sleeve advancement anorectoplasty for complicated anorectal/vaginal fistula. Dis Colon Rectum 34:10321037 19. Marchesa P, Hull TL, Fazio VW (1998) Advancement sleeve flaps for treatment of severe perianal Crohns disease. Br J Surg 85:16951698 20. Simmang CL, Lacey SW, Huber PJ (1998) Rectal sleeve advancement: repair of rectovaginal fistula associated with anorectal stricture in Crohns disease. Dis Colon Rectum 41:787789

140 21. Joo JS, Weiss EG, Nogueras JJ et al (1998) Endorectal advancement flap in perianal Crohns disease. Am Surg 64:147150 22. Onodera H, Nagayama S, Kohmoto I et al (2003) Novel surgical repair with bilateral gluteus muscle patching for intractable rectovaginal fistula. Tech Coloproctol 7:198202 23. Chitrathara K, Namratha D, Francis V et al (2001) Spontaneous rectovaginal fistula and repair using bulbocavernosus muscle flap. Tech Coloproctol 5:4749 24. Heyen F, Winslet MC, Andrews H et al (1989) Vaginal fistulas in Crohns disease. Dis Colon Rectum 32:379383 25. Devesa JM, Devesa M, Velasco GR et al (2007) Benign rectovaginal fistulas: management and results of a personal series. Tech Coloproctol 11:128134 26. Andreani SM, Dang HH, Grondona P et al (2007) Rectovaginal fistula in Crohns disease. Dis Colon Rectum 50:22152222 27. Hannaway CD, Hull TL (2008) Current considerations in the management of rectovaginal fistula from Crohns disease. Colorectal Dis 10:747756

Tech Coloproctol (2009) 13:135140

28. Shelton AA, Welton ML (2006) Transperineal repair of persistent rectovaginal fistulas using an acellular cadaveric dermal graft (AlloDerm). Dis Colon Rectum 49:14541457 29. Moore RD, Miklos JR, Kohli N (2004) Rectovaginal fistula repair using a porcine dermal graft. Obstet Gynecol 104:11651167 30. van Koperen PJ, Bemelman WA, Bossuyt PM et al (2008) The anal fistula plug versus the mucosal advancement flap for the treatment of anorectal fistula (PLUG trial). BMC Surg 8:11 31. Walfisch A, Zilberstein T, Walfisch S (2004) Rectovaginal septal repair: case presentations and introduction of a modified reconstruction technique. Tech Coloproctol 8:192194 32. Frst A, Schmidbauer C, Swol-Ben J et al (2008) Gracilis transposition for repair of recurrent anovaginal and rectovaginal fistulas in Crohns disease. Int J Colorectal Dis 23:349353 33. Ruiz D, Bashankaev B, Speranaza J et al (2008) Graciloplasty for rectourethral, rectovaginal and rectovesical fistulas: technique overview, pitfalls and complications. Tech Coloproctol 12:277281