- Email: [email protected]
Anal fistula plug and fibrin glue versus conventional treatment in repair of complex anal fistulas
The American Journal of Surgery (2009) 197, 604 – 608
The North Pacific Surgical Association
Anal fistula plug and fibrin glue versus conventional treatment in repair of complex anal fistulas Wiley Chung, B.Sc., Pooya Kazemi, David Ko, B.Sc., Clare Sun, B.Sc., Carl J. Brown, M.D., M.Sc., Manoj Raval, M.D., M.Sc., Terry Phang, M.D., M.Sc.* Division of General Surgery, St. Paul’s Hospital, University of British Columbia, 1081 Burrard Street, Vancouver, BC, Canada V6Z 1Y6 KEYWORDS: Anal fistula; Anal fistula plug repair; Advancement flap closure
Abstract INTRODUCTION: High transsphincteric fistulas are difficult to treat because fistulotomy of involved sphincter muscle results in incontinence. We compare our outcomes for anal fistula plug, fibrin glue, advancement flap closure, and seton drain insertion. METHODS: This is a retrospective study of patients treated for high transsphincteric anal fistulas. The primary outcome was full healing at 12 weeks postoperatively. RESULTS: Between 1997 and 2008, 232 patients with anal fistula were identified in the St. Paul’s Hospital Anal Fistula Database. Postoperative healing rates at the 12-week follow-up for the fistula plug, fibrin glue, flap advancement, and seton drain groups were 59.3%, 39.1%, 60.4%, and 32.6%, respectively (P ⬍ .0001). CONCLUSIONS: Closure of the primary fistula opening using a biological anal fistula plug and anal flap advancement result in similar fistula healing rates in patients with high transsphincteric fistulae. These 2 strategies are superior to seton placement and fibrin glue. Given the low morbidity and relative simplicity of the procedure, the anal fistula plug is a viable alternative treatment for patients with high transsphincteric anal fistulas. © 2009 Elsevier Inc. All rights reserved.
Perianal fistulas are common in the population. The incidence in men and women is 12.3 per 100,000 and 5.6 per 100,000, respectively.1 The disease presents predominantly in the 3rd and 4th decades of life.2 According to the cryptoglandular hypothesis, intersphincteric gland infection is the initiating event in the formation of perianal fistulas.3 Parks et al4 suggested a classification of intersphincteric, transsphincteric, suprasphincteric, and extrasphincteric fisDisclosure Presentation expenses for David Ko were supported from an unrestricted education grant provided by Cook. * Corresponding author: Tel.: ⫹1-604-806-8711; fax: ⫹1-604-8069604. E-mail address: [email protected] Manuscript received October 29, 2008; revised manuscript December 13, 2008
0002-9610/$ - see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2008.12.013
tulas. This is the most widely used classification.5 However, a more practical approach is based on whether the internal opening is high or low within the anal canal. Low transsphincteric fistulas involve the lower 3rd of the external anal sphincter mechanism and are generally treated by fistulotomy with a high success rate for cure. High transsphincteric fistulas involving the upper two thirds of the external sphincter remain a surgical challenge because incontinence will result from the division of muscle involving more than one third of the sphincter. Surgical procedures for treating high transsphincteric fistulas include advancement flap closure, loose or cutting setons, and injection of fibrin glue. These procedures have variable success rates. Fistula recurrence after advancement flap closure is reported to range between 0% and 63%.6 – 8
W. Chung et al.
Anal fistula procedure outcomes
Recurrence rates from 22% to 39% have been reported after seton drain(s) are inserted.9,10 Fibrin glue has long-term closure rates that range from 14% to 60%.11–13 A potential cause for the failure of fibrin glue in the management of fistulas is early extrusion of the fibrin clot secondary to increased pressure from coughing or straining.14 In 2006, Johnson et al15 reported their use of a new biological anal fistula plug (Surgisis; Cook Surgical, Inc, Bloomington, IN) for high transsphincteric perianal fistulas. The plug is a bioabsorbable xenograft composed of lyophilized porcine intestinal submucosa. It has inherent resistance to infection, produces no foreign body or giant cell reaction, and becomes repopulated with host cell tissue during a period of 3 months. The material was fashioned into a conical plug and secured into the primary opening of the fistula tract. Johnson et al15 achieved promising results in a prospective study of 15 patients with high anorectal fistulas treated with the anal fistula plug with closure rates of 87% at a median follow-up of 13.8 ⫾ 3.1 weeks. After this initial study, several authors have reported further experience using the anal fistula plug with success ranging from 41% to 89%.16 –20 In this study, we reviewed our outcomes for the new anal fistula plug in comparison to advancement flap closure, seton drain placement, and fibrin glue injection.
Methods All patients treated for anal fistulas by a single colorectal surgeon at St. Paul’s Hospital, University of British Columbia, Vancouver, BC, Canada, from January 1997 to December 2008 were identified from a database. Simple fistulas were defined as low fistulas including subcutaneous, intersphincteric, and low transsphincteric fistulas. Low transsphincteric fistulas involved the lower 3rd of the external anal sphincter mechanism. Conversely, complex anal fistulas were defined as Crohn fistulas, low compromised fistulas, and high transsphincteric fistulas. Low compromised fistulas occurred in the lower 3rd of the external anal sphincter in patients who had preexisting tendency for incontinence. High transsphincteric fistulas involved the upper two thirds of the external sphincter and were further subclassified as direct transsphincteric, horseshoe, and supralevator fistulas. Inclusion criteria for this study were age above 18 years and high transsphincteric fistulas of cryptoglandular origin. Exclusion criteria were simple fistulas, rectovaginal fistulas, and fistulas of Crohn etiology.
Surgical procedures All procedures were performed under general anesthesia. Fistula tracts were identified, and the primary opening was located using a fistula probe, hydrogen peroxide installation,
605 or fistuloscopy. All tracts were irrigated with hydrogen peroxide. Fibrin glue. Five milliliters of fibrin glue (Tisseal; Baxter, Inc, Mississauga, Ontario) was reconstituted according to the manufacturer’s directions. The internal opening was closed with a figure-of-8 Vicryl suture (Ethicon, Inc, Cincinnati, OH). Fibrin glue was instilled into the fistula tract by way of the external opening. A nonadherent dressing was placed, and the patient was instructed to avoid lifting and vigorous physical activity for the next 7 days. Patients were also instructed not to take a sitz bath. Bioprosthetic plug. After flushing the fistula tract with hydrogen peroxide, a conical biological plug (Surgisis ES) was inserted through the tract with the wide end of the plug wedged firmly in the internal opening of the fistula tract. Excess plug material was trimmed flush with the mucosa at the internal fistula opening and at the external fistula opening at skin level. An absorbable suture was used to secure the plug to the internal sphincter muscle and to cover the mucosal opening of the fistula and the internal end of the plug. The external end of the plug was secured to 1 side of the external fistula opening with a single absorbable suture. Seton drain placement. The fistula tracts were probed and were opened external to the anal sphincter. Transsphincteric Ethibond sutures (Ethicon, Inc) tied loosely were used as seton drains. Advancement flap closure. The fistula tracts were probed and then opened external to the anal sphincter. A mucosal flap was raised proximally from the internal fistula opening by undermining rectal mucosa for at least 4 cm to prevent tension on the suture line. The distal end of the flap with the internal fistula opening was excised. The transsphincteric fistula opening in the internal sphincter was closed using Vicryl sutures, and the mucosal flap was advanced and sutured in place with interrupted Vicryl sutures. The fistulotomy external to the sphincter was left open for drainage.
Data collection and follow-up Demographics, previous treatments, details of surgery, and follow-up results were obtained from a review of patient charts. All patients were evaluated in the outpatient clinic. The primary endpoint was full healing defined as closure of the external fistula opening with no drainage or infection at 12 weeks postoperatively for the fistula plug, fibrin glue, and flap advancement groups. Healing for the seton drain group was defined as a persistent fistula opening at the seton site but absence of drainage or infection. Telephone interviews were conducted subsequently to confirm the durability of healing or elicit symptoms and signs of recurrence. Group means were assessed by using analysis of variance and proportional data using a chi-square test. Statisti-
606 Table 1
The American Journal of Surgery, Vol 197, No 5, May 2009 Patient demographics and outcomes
Number of patients (n) Median age Range Sex Men Women Classification of fistula Direct transsphincteric Horseshoe Supralevator Outcomes Fistulas healed at 12 wk (%)
Fistula plug
Fibrin glue
Seton drain
Flap advancement
27 46 23–68
23 49 22–68
86 46 21–82
96 46 28–75
18 9
22 1
70 16
71 25
13 14 0
6 15 2
14 67 5
33 58 5
16/27 (59.3)
9/23 (39.1)
28/86 (32.6)
58/96 (60.4)
cally significant differences were defined by a P value less than .05. Multiple logistic regression was used to evaluate factors considered to affect healing. Statistical analysis was performed by using SPSS, version 9.0 (SPSS Inc, Chicago, IL). The study received approval from the University of British Columbia/Providence Health Care Research Ethics Board.
including azathioprine, 6-mercaptopurine, methotrexate, and prednisone). These were not found to be associated with healing at 12 weeks using multivariate analysis. However, with only 27 patients, this model was underpowered to show a significant difference.
Comments Results Two hundred thirty-two patients who had high transsphincteric fistulas of cryptoglandular origin were included. Table 1 shows the demographic information. The median age was 47 (range 21– 82). Twenty-seven patients were treated with the anal fistula plug, 23 with fibrin glue injection, 86 with seton drain placement, and 96 with an advancement flap. There was no significant difference in age between treatment groups (P ⫽ .645). There was a significant difference in the sex ratio between groups, with the fibrin glue group having the highest proportion of men (P ⫽ .04). Fistulas were classified as direct transsphincteric, horseshoe, and supralevator (Table 1). The healing rates at week 12 are reported in Table 1 and range from 39% for fibrin glue to 60% for flap advancement. Healing rates were significantly different between treatment groups (P ⬍ .05). Further follow-up of patients treated using the plug showed continued improvement with healing in 3 out of 27, bringing the healing rate at 24 weeks to 19 out of 27 (70.4%). The other groups were not assessed at 24 weeks. Of the 11 fistula plug patients with persistent fistulae at 12 weeks, 3 had infection and 5 had plug extrusion. Four out of 5 plug extrusions occurred within the first week, and the 5th extrusion occurred by 4 weeks. Of the remaining 3 patients, 2 healed by 24 weeks postoperatively, and one had a persistent fistula opening at 24 weeks. Risk factors assessed for possible association with plug failure were age, fistula classification, and comorbidites (diabetes, HIV, ulcerative colitis, and immunosuppression
We retrospectively evaluated healing of high transsphincteric anal fistulas of cryptoglandular etiology using the newer treatment techniques (fistula plug and fibrin glue) and conventional treatments (flap advancement and seton drain placement). Postoperative healing rates at 12 weeks for the plug repair and flap advancement were similar at 59.3% and 60.4%, respectively. The fibrin glue and seton drain treatments had significantly lower healing rates at 39.1% and 32.6%, respectively. Our healing rate for the fistula plug is similar to previously reported small series that have shown success rates between 41% and 89% (Table 2). Both fibrin glue and fistula plug treatments have a minimal risk of incontinence. Failure of these newer treatments does not compromise subsequent surgical options. Lower healing rates seen with fibrin glue treatment may be secondary to the liquid consistency of the glue in a short fistula tract causing early extrusion. The higher healing rate seen with the fistula plug is likely caused by a more secure
Table 2 Healing rates in published studies of fistula treatment using the plug repair Authors
Year
Healing rate Median follow-up (%) time (mo)
Thekkinkattil et al16 Christoforidis et al17 Ky et al18 Van Koperen et al19 Ellis20 Johnson et al15
2008 2008 2008 2007 2007 2006
19/43 20/47 28/45 7/17 16/18 13/15
(44) (43) (62) (41) (89) (87)
11 6.5 6.5 7 10 3
W. Chung et al.
Anal fistula procedure outcomes
fixation than glue in the fistula tract, resulting in less early extrusion. The conical shape of the plug inserted in the high-pressure area of the primary opening and suture fixation to the internal sphincter also adds to mechanical stability.15 However, plug extrusion remains a significant problem and accounted for 45% of plug failures in the current study. A potential solution for early extrusion is the addition of a button sewn to the plug at the internal opening. Active infection was observed in 11.1% of the patients who had failed plug repairs and is recognized to be a significant cause of plug failure by others. We support the strategy of using plug repair only in the absence of significant infection. In the presence of significant acute infection, we treat the infection using a loose seton drain and antibiotics. The plug repair can be used once the acute infection has resolved. Our study result is limited by a relatively shorter follow-up time of 12 weeks compared with other studies with a follow-up duration of 6.5 months to 11 months.16 –20 Longer-term follow-up is currently in progress. The retrospective methodology used in our study has potential significant inaccuracy and bias that would be improved by using prospective data collection. A larger randomized, multicenter prospective trial of plug repair and advancement flap repair is indicated to determine the more effective treatment for outcomes of healing, morbidity, and cost. Presently, there is a trial based in The Netherlands that is designed to achieve this goal (http://www.controlled-trials. com/ISRCTN97376902/anal⫹fistula). In summary, we found similar healing rates for the fistula plug and conventional advancement flap closure for treating complex anal fistulas in our preliminary study. Given the low morbidity and relative simplicity of the procedure, the anal fistula plug should be considered an alternative effective treatment for patients with complex anal fistulas.
References 1. Sainio P. Fistula-in-ano in a defined population. Incidence and epidemiological aspects. Ann Chir Gynaecol 1984;73:219 –24. 2. Marks CG, Ritchie JK. Anal fistulas at St Mark’s Hospital. Br J Surg 1977;64:84 –91. 3. Parks AG. Pathogenesis and treatment of fistula-in-ano. Br Med J 1961;1:463–9. 4. Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-inano. Br J Surg 1976;63:1–12. 5. Rickard MJ. Anal abscesses and fistulas. ANZ J Surg 2005;75:64 –72. 6. Ortiz H, Marzo J. Endorectal flap advancement repair and fistulectomy for high trans-sphincteric and suprasphincteric fistulas. Br J Surg 2000;87:1680 –3. 7. Hagen SJ, Baeten CG, Soeters PB, et al. Long-term outcome following mucosal advancement flap for high perianal fistulas and fistulotomy for low perianal fistulas: recurrent perianal fistulas: failure of treatment or recurrent patient disease? Int J Colorectal Dis 2006;21:784 –90. 8. Buchanan GN, Owen HA, Torkington J, et al. Long-term outcome following loose-Seton technique for external sphincter preservation in complex anal fistula. Br J Surg 2004;91:476 – 80. 9. Theerapol A, So BYJ, Ngoi SS. Routine use of setons for the treatment of anal fistulae. Singapore Med J 2002;43:305–7.
607 10. Williams JG, MacLeod CA, Rothenberger DA, et al. Seton treatment of high anal fistulae. Br J Surg 1991;78:1159 – 61. 11. Buchanan GN, Bartram CI, Phillips RK, et al. Efficacy of fibrin sealant in the management of complex anal fistula: a prospective trial. Dis Colon Rectum Sentovich SM. Fibrin glue for anal fistulas: long-term results. Dis Colon Rectum 2003;46:498 –502. 12. Sentovich SM. Fibrin glue for anal fistulas: long-term results. Dis Colon Rectum 2003;46:498 –502. 13. Gisbertz SS, Sosef MN, Festen S, et al. Treatment of fistulas in ano with fibrin glue. Digest Surg 2005;22:91– 4. 14. Hammond TM, Grahn MF, Lunniss PJ. Fibrin glue in the management of anal fistulae. Colorectal Dis 2004;6:308 –19. 15. Johnson EK, Gaw JU, Armstrong DN. Efficacy of anal fistula plug vs. fibrin glue in closure of anorectal fistulas. Dis Colon Rectum 2006; 49:371– 6. 16. Thekkinkattil D, Botterill I, Ambrose S, et al. Efficacy of the anal fistula plug in complex anorectal fistulae. Colorectal Dis 2008 [Epub ahead of print]. 17. Christoforidis D, Etzioni DA, Goldberg SM, et al. Treatment of complex anal fistulas with the collagen fistula plug. Dis Colon Rectum 2008;51:1482–7. 18. Ky AJ, Sylla P, Steinhagen R, et al. Collagen fistula plug for the treatment of anal fistulas. Dis Colon Rectum 2008;51:838 – 43. 19. Van Koperen PJ, D’hoore A, Wolthuis AM, et al. Anal fistula plug for closure of difficult anorectal fistula: a prospective study. Dis Colon Rectum 2007;50:2168 –72. 20. Ellis CN. Bioprosthetic plugs for complex anal fistulas: an early experience. J Surg Educ 2007;64:36 – 40.
Discussion Kim C. Lu, M.D. (Portland, OR): High transsphincteric anal fistulas have vexed surgeons for a long time. Historically, a “lay-open” fistulotomy would definitively treat the underlying infection but invariably lead to incontinence. The cure seemed worse than the disease. Hence, every wave of innovation sought to minimize the cutting of the anal sphincter. For example, rather than cutting any muscle, a loose seton could be placed into a high fistula to control sepsis. Once the sepsis was resolved, the internal fistula opening could be covered by advancing rectal mucosa. This would prevent stool from entering the fistula and perpetuating chronic infection of the perianal tissue. Unfortunately, even this technique has been associated with compromised fecal continence.1,2 Surgeons have tried treating anal fistula with newer techniques including fibrin glue or a porcine intestinal submucosal “plug.” Although initial reports had spectacular results with no theoretic effect on continence,3,4 subsequent reports results from other institutions have tempered enthusiasm.5,6 In the article by Dr. Chung et al, the author described the healing rates of high transsphincteric cryptoglandular fistulas from a retrospective cohort study. They treated 232 patients. They had treated 86 patients with setons only, 23 patients with fibrin glue, 27 patients with a fistula plug, and 96 patients with rectal mucosal flap advancement. They followed them for 12 weeks and defined healing as closure of the external opening (with nonseton techniques), no drainage, and no signs of infection.
608
The American Journal of Surgery, Vol 197, No 5, May 2009
The healing rates were 32.6% (set on), 39.1% (glue), 59.3% (plug), and 60.4% (flap). At first glance, these results were not surprising. These results were within the range of published healing rates for each technique. These results confirmed the surgeon is facile with all 4 techniques. Curiously, at the 24-week follow-up, the plug healing rate improved to 70.4%. Another group found that the healing rate with fistula plug deteriorated with time.5 With early failures attributed to the fistula plug falling out of the fistula, in more recent publications, surgeons have omitted suturing the plug to the external opening.6 Given that the authors used the technique originally described in 2006, their extrusion rate (5/27 patients) was not surprising. The new finding of this study was that a lot of perioperative recommendations may be unnecessary to effectively use the fistula plug. Other authors have selectively placed setons for 3 months to mature the tract.6 These authors did not use preoperative setons, unless there were active infection. Some authors have prescribed postoperative topical metronidazole 3 times a day, limited the postoperative diet to clear liquids for 48 hours, and asked the patients to avoid strenuous activity.3,4 The authors did not describe the postoperative care. Still, they had reasonable results after the use of a single plug in each fistula. There were a few limitations. First, there was no comment on why a patient would receive a seton, glue, plug, or flap. Their successes may be because of not only technical skill but also the choice of procedure. Second, the statistical analysis only showed that the healing rates of the 4 techniques were different. A more interesting comparison would have been between flap and plug healing rates. However, because these rates were so similar (59.3% for the plug versus 60.4% for the flap), this pilot study was underpowered to detect a difference. Third, no plugs were performed for suprasphincteric fistulas. If the healing rate for the flap were also calculated after excluding those fistulas, a more
meaningful comparison between the plug and flap could be made. Fourth, fecal incontinence rates were not reported for the flap. Even if the healing rates for the plug and flap were equivalent, if the fecal incontinence rates for the flap were substantial (up to 40%),2 the plug would be preferable. Finally, as the authors admitted, the follow-up was somewhat short. In a recent study, the healing rates with the plug deteriorated over time (eg, decreased to 54.6% at 6.5 months). In summary, this article confirmed reasonable healing rates of high transsphincteric cryptoglandular fistulas when using the anal fistula plug. They proposed further study of the anal fistula plug versus the rectal advancement flap. Recently, a prospective randomized blinded multicenter trial has been started in The Netherlands.7
References 1. Uribe N, Millan M, Minguez M, et al. Clinical and manometric results of endorectal advancement flaps for complex anal fistula. Int J Colorectal Dis 2007;22:259 – 64. 2. van Koperen PJ, Wind J, Bemelman WA, et al. Long-Term functional outcome and risk factors for recurrence After surgical treatment for low and high perianal fistulas of cryptoglandular origin. Dis col rectum 2008;51:1475– 81. 3. Johnson EK, Gaw JU, Armstrong DN. Efficacy of anal fistula plug vs. fibrin glue in closure of anorectal fistulas. Dis col rectum 2006;49: 371– 6. 4. Champagne BJ, O’Connor LM, Ferguson M, et al. Efficacy of anal fistula plug in closure of cryptoglandular fistulas: long-term follow-up. Dis col rectum 2006;49:1817–21. 5. Ky AJ, Sylla P, Steinhagen R, et al. Collagen fistula plug for the treatment of anal fistulas. Dis Col Rectum 2008;51:838 – 43. 6. Christoforidis D, Etzioni DA, Goldberg SM, et al. Treatment of complex anal fistulas with the collagen fistula plug. Dis Col Rectum 2008; 51:1482– 87. 7. van Koperen PJ, Bemelman WA, Bossuyt PMM, et al. The anal fistula plug versus the mucosal advancement flap for the treatment of anorectal fistula (PLUG trial). Study protocol. BMC Surg 2008;8:11.
The North Pacific Surgical Association
Anal fistula plug and fibrin glue versus conventional treatment in repair of complex anal fistulas Wiley Chung, B.Sc., Pooya Kazemi, David Ko, B.Sc., Clare Sun, B.Sc., Carl J. Brown, M.D., M.Sc., Manoj Raval, M.D., M.Sc., Terry Phang, M.D., M.Sc.* Division of General Surgery, St. Paul’s Hospital, University of British Columbia, 1081 Burrard Street, Vancouver, BC, Canada V6Z 1Y6 KEYWORDS: Anal fistula; Anal fistula plug repair; Advancement flap closure
Abstract INTRODUCTION: High transsphincteric fistulas are difficult to treat because fistulotomy of involved sphincter muscle results in incontinence. We compare our outcomes for anal fistula plug, fibrin glue, advancement flap closure, and seton drain insertion. METHODS: This is a retrospective study of patients treated for high transsphincteric anal fistulas. The primary outcome was full healing at 12 weeks postoperatively. RESULTS: Between 1997 and 2008, 232 patients with anal fistula were identified in the St. Paul’s Hospital Anal Fistula Database. Postoperative healing rates at the 12-week follow-up for the fistula plug, fibrin glue, flap advancement, and seton drain groups were 59.3%, 39.1%, 60.4%, and 32.6%, respectively (P ⬍ .0001). CONCLUSIONS: Closure of the primary fistula opening using a biological anal fistula plug and anal flap advancement result in similar fistula healing rates in patients with high transsphincteric fistulae. These 2 strategies are superior to seton placement and fibrin glue. Given the low morbidity and relative simplicity of the procedure, the anal fistula plug is a viable alternative treatment for patients with high transsphincteric anal fistulas. © 2009 Elsevier Inc. All rights reserved.
Perianal fistulas are common in the population. The incidence in men and women is 12.3 per 100,000 and 5.6 per 100,000, respectively.1 The disease presents predominantly in the 3rd and 4th decades of life.2 According to the cryptoglandular hypothesis, intersphincteric gland infection is the initiating event in the formation of perianal fistulas.3 Parks et al4 suggested a classification of intersphincteric, transsphincteric, suprasphincteric, and extrasphincteric fisDisclosure Presentation expenses for David Ko were supported from an unrestricted education grant provided by Cook. * Corresponding author: Tel.: ⫹1-604-806-8711; fax: ⫹1-604-8069604. E-mail address: [email protected] Manuscript received October 29, 2008; revised manuscript December 13, 2008
0002-9610/$ - see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2008.12.013
tulas. This is the most widely used classification.5 However, a more practical approach is based on whether the internal opening is high or low within the anal canal. Low transsphincteric fistulas involve the lower 3rd of the external anal sphincter mechanism and are generally treated by fistulotomy with a high success rate for cure. High transsphincteric fistulas involving the upper two thirds of the external sphincter remain a surgical challenge because incontinence will result from the division of muscle involving more than one third of the sphincter. Surgical procedures for treating high transsphincteric fistulas include advancement flap closure, loose or cutting setons, and injection of fibrin glue. These procedures have variable success rates. Fistula recurrence after advancement flap closure is reported to range between 0% and 63%.6 – 8
W. Chung et al.
Anal fistula procedure outcomes
Recurrence rates from 22% to 39% have been reported after seton drain(s) are inserted.9,10 Fibrin glue has long-term closure rates that range from 14% to 60%.11–13 A potential cause for the failure of fibrin glue in the management of fistulas is early extrusion of the fibrin clot secondary to increased pressure from coughing or straining.14 In 2006, Johnson et al15 reported their use of a new biological anal fistula plug (Surgisis; Cook Surgical, Inc, Bloomington, IN) for high transsphincteric perianal fistulas. The plug is a bioabsorbable xenograft composed of lyophilized porcine intestinal submucosa. It has inherent resistance to infection, produces no foreign body or giant cell reaction, and becomes repopulated with host cell tissue during a period of 3 months. The material was fashioned into a conical plug and secured into the primary opening of the fistula tract. Johnson et al15 achieved promising results in a prospective study of 15 patients with high anorectal fistulas treated with the anal fistula plug with closure rates of 87% at a median follow-up of 13.8 ⫾ 3.1 weeks. After this initial study, several authors have reported further experience using the anal fistula plug with success ranging from 41% to 89%.16 –20 In this study, we reviewed our outcomes for the new anal fistula plug in comparison to advancement flap closure, seton drain placement, and fibrin glue injection.
Methods All patients treated for anal fistulas by a single colorectal surgeon at St. Paul’s Hospital, University of British Columbia, Vancouver, BC, Canada, from January 1997 to December 2008 were identified from a database. Simple fistulas were defined as low fistulas including subcutaneous, intersphincteric, and low transsphincteric fistulas. Low transsphincteric fistulas involved the lower 3rd of the external anal sphincter mechanism. Conversely, complex anal fistulas were defined as Crohn fistulas, low compromised fistulas, and high transsphincteric fistulas. Low compromised fistulas occurred in the lower 3rd of the external anal sphincter in patients who had preexisting tendency for incontinence. High transsphincteric fistulas involved the upper two thirds of the external sphincter and were further subclassified as direct transsphincteric, horseshoe, and supralevator fistulas. Inclusion criteria for this study were age above 18 years and high transsphincteric fistulas of cryptoglandular origin. Exclusion criteria were simple fistulas, rectovaginal fistulas, and fistulas of Crohn etiology.
Surgical procedures All procedures were performed under general anesthesia. Fistula tracts were identified, and the primary opening was located using a fistula probe, hydrogen peroxide installation,
605 or fistuloscopy. All tracts were irrigated with hydrogen peroxide. Fibrin glue. Five milliliters of fibrin glue (Tisseal; Baxter, Inc, Mississauga, Ontario) was reconstituted according to the manufacturer’s directions. The internal opening was closed with a figure-of-8 Vicryl suture (Ethicon, Inc, Cincinnati, OH). Fibrin glue was instilled into the fistula tract by way of the external opening. A nonadherent dressing was placed, and the patient was instructed to avoid lifting and vigorous physical activity for the next 7 days. Patients were also instructed not to take a sitz bath. Bioprosthetic plug. After flushing the fistula tract with hydrogen peroxide, a conical biological plug (Surgisis ES) was inserted through the tract with the wide end of the plug wedged firmly in the internal opening of the fistula tract. Excess plug material was trimmed flush with the mucosa at the internal fistula opening and at the external fistula opening at skin level. An absorbable suture was used to secure the plug to the internal sphincter muscle and to cover the mucosal opening of the fistula and the internal end of the plug. The external end of the plug was secured to 1 side of the external fistula opening with a single absorbable suture. Seton drain placement. The fistula tracts were probed and were opened external to the anal sphincter. Transsphincteric Ethibond sutures (Ethicon, Inc) tied loosely were used as seton drains. Advancement flap closure. The fistula tracts were probed and then opened external to the anal sphincter. A mucosal flap was raised proximally from the internal fistula opening by undermining rectal mucosa for at least 4 cm to prevent tension on the suture line. The distal end of the flap with the internal fistula opening was excised. The transsphincteric fistula opening in the internal sphincter was closed using Vicryl sutures, and the mucosal flap was advanced and sutured in place with interrupted Vicryl sutures. The fistulotomy external to the sphincter was left open for drainage.
Data collection and follow-up Demographics, previous treatments, details of surgery, and follow-up results were obtained from a review of patient charts. All patients were evaluated in the outpatient clinic. The primary endpoint was full healing defined as closure of the external fistula opening with no drainage or infection at 12 weeks postoperatively for the fistula plug, fibrin glue, and flap advancement groups. Healing for the seton drain group was defined as a persistent fistula opening at the seton site but absence of drainage or infection. Telephone interviews were conducted subsequently to confirm the durability of healing or elicit symptoms and signs of recurrence. Group means were assessed by using analysis of variance and proportional data using a chi-square test. Statisti-
606 Table 1
The American Journal of Surgery, Vol 197, No 5, May 2009 Patient demographics and outcomes
Number of patients (n) Median age Range Sex Men Women Classification of fistula Direct transsphincteric Horseshoe Supralevator Outcomes Fistulas healed at 12 wk (%)
Fistula plug
Fibrin glue
Seton drain
Flap advancement
27 46 23–68
23 49 22–68
86 46 21–82
96 46 28–75
18 9
22 1
70 16
71 25
13 14 0
6 15 2
14 67 5
33 58 5
16/27 (59.3)
9/23 (39.1)
28/86 (32.6)
58/96 (60.4)
cally significant differences were defined by a P value less than .05. Multiple logistic regression was used to evaluate factors considered to affect healing. Statistical analysis was performed by using SPSS, version 9.0 (SPSS Inc, Chicago, IL). The study received approval from the University of British Columbia/Providence Health Care Research Ethics Board.
including azathioprine, 6-mercaptopurine, methotrexate, and prednisone). These were not found to be associated with healing at 12 weeks using multivariate analysis. However, with only 27 patients, this model was underpowered to show a significant difference.
Comments Results Two hundred thirty-two patients who had high transsphincteric fistulas of cryptoglandular origin were included. Table 1 shows the demographic information. The median age was 47 (range 21– 82). Twenty-seven patients were treated with the anal fistula plug, 23 with fibrin glue injection, 86 with seton drain placement, and 96 with an advancement flap. There was no significant difference in age between treatment groups (P ⫽ .645). There was a significant difference in the sex ratio between groups, with the fibrin glue group having the highest proportion of men (P ⫽ .04). Fistulas were classified as direct transsphincteric, horseshoe, and supralevator (Table 1). The healing rates at week 12 are reported in Table 1 and range from 39% for fibrin glue to 60% for flap advancement. Healing rates were significantly different between treatment groups (P ⬍ .05). Further follow-up of patients treated using the plug showed continued improvement with healing in 3 out of 27, bringing the healing rate at 24 weeks to 19 out of 27 (70.4%). The other groups were not assessed at 24 weeks. Of the 11 fistula plug patients with persistent fistulae at 12 weeks, 3 had infection and 5 had plug extrusion. Four out of 5 plug extrusions occurred within the first week, and the 5th extrusion occurred by 4 weeks. Of the remaining 3 patients, 2 healed by 24 weeks postoperatively, and one had a persistent fistula opening at 24 weeks. Risk factors assessed for possible association with plug failure were age, fistula classification, and comorbidites (diabetes, HIV, ulcerative colitis, and immunosuppression
We retrospectively evaluated healing of high transsphincteric anal fistulas of cryptoglandular etiology using the newer treatment techniques (fistula plug and fibrin glue) and conventional treatments (flap advancement and seton drain placement). Postoperative healing rates at 12 weeks for the plug repair and flap advancement were similar at 59.3% and 60.4%, respectively. The fibrin glue and seton drain treatments had significantly lower healing rates at 39.1% and 32.6%, respectively. Our healing rate for the fistula plug is similar to previously reported small series that have shown success rates between 41% and 89% (Table 2). Both fibrin glue and fistula plug treatments have a minimal risk of incontinence. Failure of these newer treatments does not compromise subsequent surgical options. Lower healing rates seen with fibrin glue treatment may be secondary to the liquid consistency of the glue in a short fistula tract causing early extrusion. The higher healing rate seen with the fistula plug is likely caused by a more secure
Table 2 Healing rates in published studies of fistula treatment using the plug repair Authors
Year
Healing rate Median follow-up (%) time (mo)
Thekkinkattil et al16 Christoforidis et al17 Ky et al18 Van Koperen et al19 Ellis20 Johnson et al15
2008 2008 2008 2007 2007 2006
19/43 20/47 28/45 7/17 16/18 13/15
(44) (43) (62) (41) (89) (87)
11 6.5 6.5 7 10 3
W. Chung et al.
Anal fistula procedure outcomes
fixation than glue in the fistula tract, resulting in less early extrusion. The conical shape of the plug inserted in the high-pressure area of the primary opening and suture fixation to the internal sphincter also adds to mechanical stability.15 However, plug extrusion remains a significant problem and accounted for 45% of plug failures in the current study. A potential solution for early extrusion is the addition of a button sewn to the plug at the internal opening. Active infection was observed in 11.1% of the patients who had failed plug repairs and is recognized to be a significant cause of plug failure by others. We support the strategy of using plug repair only in the absence of significant infection. In the presence of significant acute infection, we treat the infection using a loose seton drain and antibiotics. The plug repair can be used once the acute infection has resolved. Our study result is limited by a relatively shorter follow-up time of 12 weeks compared with other studies with a follow-up duration of 6.5 months to 11 months.16 –20 Longer-term follow-up is currently in progress. The retrospective methodology used in our study has potential significant inaccuracy and bias that would be improved by using prospective data collection. A larger randomized, multicenter prospective trial of plug repair and advancement flap repair is indicated to determine the more effective treatment for outcomes of healing, morbidity, and cost. Presently, there is a trial based in The Netherlands that is designed to achieve this goal (http://www.controlled-trials. com/ISRCTN97376902/anal⫹fistula). In summary, we found similar healing rates for the fistula plug and conventional advancement flap closure for treating complex anal fistulas in our preliminary study. Given the low morbidity and relative simplicity of the procedure, the anal fistula plug should be considered an alternative effective treatment for patients with complex anal fistulas.
References 1. Sainio P. Fistula-in-ano in a defined population. Incidence and epidemiological aspects. Ann Chir Gynaecol 1984;73:219 –24. 2. Marks CG, Ritchie JK. Anal fistulas at St Mark’s Hospital. Br J Surg 1977;64:84 –91. 3. Parks AG. Pathogenesis and treatment of fistula-in-ano. Br Med J 1961;1:463–9. 4. Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-inano. Br J Surg 1976;63:1–12. 5. Rickard MJ. Anal abscesses and fistulas. ANZ J Surg 2005;75:64 –72. 6. Ortiz H, Marzo J. Endorectal flap advancement repair and fistulectomy for high trans-sphincteric and suprasphincteric fistulas. Br J Surg 2000;87:1680 –3. 7. Hagen SJ, Baeten CG, Soeters PB, et al. Long-term outcome following mucosal advancement flap for high perianal fistulas and fistulotomy for low perianal fistulas: recurrent perianal fistulas: failure of treatment or recurrent patient disease? Int J Colorectal Dis 2006;21:784 –90. 8. Buchanan GN, Owen HA, Torkington J, et al. Long-term outcome following loose-Seton technique for external sphincter preservation in complex anal fistula. Br J Surg 2004;91:476 – 80. 9. Theerapol A, So BYJ, Ngoi SS. Routine use of setons for the treatment of anal fistulae. Singapore Med J 2002;43:305–7.
607 10. Williams JG, MacLeod CA, Rothenberger DA, et al. Seton treatment of high anal fistulae. Br J Surg 1991;78:1159 – 61. 11. Buchanan GN, Bartram CI, Phillips RK, et al. Efficacy of fibrin sealant in the management of complex anal fistula: a prospective trial. Dis Colon Rectum Sentovich SM. Fibrin glue for anal fistulas: long-term results. Dis Colon Rectum 2003;46:498 –502. 12. Sentovich SM. Fibrin glue for anal fistulas: long-term results. Dis Colon Rectum 2003;46:498 –502. 13. Gisbertz SS, Sosef MN, Festen S, et al. Treatment of fistulas in ano with fibrin glue. Digest Surg 2005;22:91– 4. 14. Hammond TM, Grahn MF, Lunniss PJ. Fibrin glue in the management of anal fistulae. Colorectal Dis 2004;6:308 –19. 15. Johnson EK, Gaw JU, Armstrong DN. Efficacy of anal fistula plug vs. fibrin glue in closure of anorectal fistulas. Dis Colon Rectum 2006; 49:371– 6. 16. Thekkinkattil D, Botterill I, Ambrose S, et al. Efficacy of the anal fistula plug in complex anorectal fistulae. Colorectal Dis 2008 [Epub ahead of print]. 17. Christoforidis D, Etzioni DA, Goldberg SM, et al. Treatment of complex anal fistulas with the collagen fistula plug. Dis Colon Rectum 2008;51:1482–7. 18. Ky AJ, Sylla P, Steinhagen R, et al. Collagen fistula plug for the treatment of anal fistulas. Dis Colon Rectum 2008;51:838 – 43. 19. Van Koperen PJ, D’hoore A, Wolthuis AM, et al. Anal fistula plug for closure of difficult anorectal fistula: a prospective study. Dis Colon Rectum 2007;50:2168 –72. 20. Ellis CN. Bioprosthetic plugs for complex anal fistulas: an early experience. J Surg Educ 2007;64:36 – 40.
Discussion Kim C. Lu, M.D. (Portland, OR): High transsphincteric anal fistulas have vexed surgeons for a long time. Historically, a “lay-open” fistulotomy would definitively treat the underlying infection but invariably lead to incontinence. The cure seemed worse than the disease. Hence, every wave of innovation sought to minimize the cutting of the anal sphincter. For example, rather than cutting any muscle, a loose seton could be placed into a high fistula to control sepsis. Once the sepsis was resolved, the internal fistula opening could be covered by advancing rectal mucosa. This would prevent stool from entering the fistula and perpetuating chronic infection of the perianal tissue. Unfortunately, even this technique has been associated with compromised fecal continence.1,2 Surgeons have tried treating anal fistula with newer techniques including fibrin glue or a porcine intestinal submucosal “plug.” Although initial reports had spectacular results with no theoretic effect on continence,3,4 subsequent reports results from other institutions have tempered enthusiasm.5,6 In the article by Dr. Chung et al, the author described the healing rates of high transsphincteric cryptoglandular fistulas from a retrospective cohort study. They treated 232 patients. They had treated 86 patients with setons only, 23 patients with fibrin glue, 27 patients with a fistula plug, and 96 patients with rectal mucosal flap advancement. They followed them for 12 weeks and defined healing as closure of the external opening (with nonseton techniques), no drainage, and no signs of infection.
608
The American Journal of Surgery, Vol 197, No 5, May 2009
The healing rates were 32.6% (set on), 39.1% (glue), 59.3% (plug), and 60.4% (flap). At first glance, these results were not surprising. These results were within the range of published healing rates for each technique. These results confirmed the surgeon is facile with all 4 techniques. Curiously, at the 24-week follow-up, the plug healing rate improved to 70.4%. Another group found that the healing rate with fistula plug deteriorated with time.5 With early failures attributed to the fistula plug falling out of the fistula, in more recent publications, surgeons have omitted suturing the plug to the external opening.6 Given that the authors used the technique originally described in 2006, their extrusion rate (5/27 patients) was not surprising. The new finding of this study was that a lot of perioperative recommendations may be unnecessary to effectively use the fistula plug. Other authors have selectively placed setons for 3 months to mature the tract.6 These authors did not use preoperative setons, unless there were active infection. Some authors have prescribed postoperative topical metronidazole 3 times a day, limited the postoperative diet to clear liquids for 48 hours, and asked the patients to avoid strenuous activity.3,4 The authors did not describe the postoperative care. Still, they had reasonable results after the use of a single plug in each fistula. There were a few limitations. First, there was no comment on why a patient would receive a seton, glue, plug, or flap. Their successes may be because of not only technical skill but also the choice of procedure. Second, the statistical analysis only showed that the healing rates of the 4 techniques were different. A more interesting comparison would have been between flap and plug healing rates. However, because these rates were so similar (59.3% for the plug versus 60.4% for the flap), this pilot study was underpowered to detect a difference. Third, no plugs were performed for suprasphincteric fistulas. If the healing rate for the flap were also calculated after excluding those fistulas, a more
meaningful comparison between the plug and flap could be made. Fourth, fecal incontinence rates were not reported for the flap. Even if the healing rates for the plug and flap were equivalent, if the fecal incontinence rates for the flap were substantial (up to 40%),2 the plug would be preferable. Finally, as the authors admitted, the follow-up was somewhat short. In a recent study, the healing rates with the plug deteriorated over time (eg, decreased to 54.6% at 6.5 months). In summary, this article confirmed reasonable healing rates of high transsphincteric cryptoglandular fistulas when using the anal fistula plug. They proposed further study of the anal fistula plug versus the rectal advancement flap. Recently, a prospective randomized blinded multicenter trial has been started in The Netherlands.7
References 1. Uribe N, Millan M, Minguez M, et al. Clinical and manometric results of endorectal advancement flaps for complex anal fistula. Int J Colorectal Dis 2007;22:259 – 64. 2. van Koperen PJ, Wind J, Bemelman WA, et al. Long-Term functional outcome and risk factors for recurrence After surgical treatment for low and high perianal fistulas of cryptoglandular origin. Dis col rectum 2008;51:1475– 81. 3. Johnson EK, Gaw JU, Armstrong DN. Efficacy of anal fistula plug vs. fibrin glue in closure of anorectal fistulas. Dis col rectum 2006;49: 371– 6. 4. Champagne BJ, O’Connor LM, Ferguson M, et al. Efficacy of anal fistula plug in closure of cryptoglandular fistulas: long-term follow-up. Dis col rectum 2006;49:1817–21. 5. Ky AJ, Sylla P, Steinhagen R, et al. Collagen fistula plug for the treatment of anal fistulas. Dis Col Rectum 2008;51:838 – 43. 6. Christoforidis D, Etzioni DA, Goldberg SM, et al. Treatment of complex anal fistulas with the collagen fistula plug. Dis Col Rectum 2008; 51:1482– 87. 7. van Koperen PJ, Bemelman WA, Bossuyt PMM, et al. The anal fistula plug versus the mucosal advancement flap for the treatment of anorectal fistula (PLUG trial). Study protocol. BMC Surg 2008;8:11.