Fistulotomy, although effective for repair of simple and superficial anal fistulas, poses a substantial risk of incontinence in the management of deep transsphincteric or suprasphincteric fistulas.1,2 In addition, fistulas involving more than 30% of the sphincter muscles, anterior fistulas, and fistulas in patients who have preexisting incontinence are not suitable for fistulotomy.3,4 Even the use of a cutting seton to allow for a staged fistulotomy in such cases carries a significant risk of incontinence.5
Popular sphincter-sparing approaches to complex fistulas (which do not divide the sphincter muscle) include advancement flap closure,6 fibrin glue injection or anal fistula plug insertion,7 and most recently, ligation of the intersphincteric fistula tract (LIFT).8 In patients with cryptoglandular fistulas, success rates of 24.1% to 98.5% have been reported for advancement flaps, but the procedure can adversely affect continence in up to 35% of patients.9 Similar to fibrin glue,10 a fistula plug is a simple, minimally invasive option with low to no incontinence risk; however, success rates range from 35% to 87% in prospective studies of complex fistula-in-ano, with more recent studies reporting success rates toward the lower end of that range.11
Total fistulectomy with simple closure of the internal opening is another option for the management of complex fistulas. It is based on the premise that removal of the chronic, epithelialized tract will allow healing by secondary intention of healthier tissue.4 However, this is a more extensive procedure, and there is a lack of support for it in the literature. The purpose of this study was to report the long-term outcomes achieved by our group with this approach in patients with complex cryptoglandular disease.
PATIENTS AND METHODS
The study group consisted of 262 consecutive patients with complex cryptoglandular fistulas who had undergone total fistulectomy with simple closure of the internal opening between January 1997 and May 2007 at our institution. Total fistulectomy with simple closure of the internal opening was the only definitive sphincter-sparing procedure used. Patients were considered eligible if they had an intersphincteric, transsphincteric, or suprasphincteric fistula that involved more than 30% of the sphincter muscle. Sphincter involvement was determined in all patients by means of operative palpation by an expert proctologist. Preoperative endoanal ultrasound examinations were performed routinely in all patients, and 107 patients (42%) also underwent adjunctive magnetic resonance imaging (MRI).
During the same period, 743 patients with simple (subcutaneous, low intersphincteric, or low transsphincteric) fistulas underwent fistulotomy and were excluded from the study, as were patients with Crohn’s disease or rectovaginal fistulas.
The study was approved by the institutional review board of the Dresden University of Technology.
At the initial operation, the anatomy of the fistula was identified by gentle probing of the tract and by injection of indigo carmine blue solution, if needed. The fistulas were classified as anterior, lateral, or posterior depending on the location of the internal opening, and their relationship to the anal sphincter muscles was categorized as intersphincteric, transsphincteric, or suprasphincteric according to a modified Parks classification.12 Setons were placed at the surgeon’s discretion if the fistula tract involved too much sphincter to allow a safe fistulotomy. The setons were left in place for approximately 3 months to provide drainage and maturation of the fistula tract.
All patients were operated on in the lithotomy position under general or spinal anesthesia without formal preoperative bowel preparation. Prophylactic parenteral antibiotics were given just before the procedure. The external opening was circumcised, and the tract was dissected from the surrounding tissue deep into the sphincter. All of the tracts were traced to their origins, ensuring that no deep transsphincteric or suprasphincteric tracts were missed. At this point, the procedure was continued via a transanal approach. The internal opening was excised by using a Lone Star Retractor System (Lone Star Medical Products, Inc., Houston, TX) to efface the anal canal, and a window was created in the full-thickness wall next to the internal opening. The dissected part of the fistula tract was then pulled into the anal canal through that window. This step was essential to completely remove all of the fistula tissue internal to the external sphincter without any sphincter division. Care was taken to drain all of the associated abscesses. Finally, the internal opening was closed with a simple layer of interrupted full-thickness sutures of 3-0 polyglycolic acid. If needed, the full thickness of the proximal edge of the internal opening was slightly mobilized to ensure closure without tension. The external opening was left open to provide drainage.
All procedures were performed on an inpatient basis, and digital examination of the wound was performed daily for approximately 3 to 5 days after surgery to break down granulation tissue bridging the wound, thereby preventing premature closure of the external opening. After discharge from the hospital, the patients were instructed to shower 2 to 3 times a day and after bowel movements until the external wound had closed. There were no dietary restrictions. Follow-up office visits were scheduled at 3 months and as needed until the wound had healed.
At the time of treatment, data were entered in a prospectively managed database. We retrospectively reviewed records from this database for the current study. Follow-up data were obtained at the scheduled office visits and were supplemented by comprehensive chart review, mailed questionnaires, and telephone interviews. The patients were asked about the signs and symptoms of persistence or recurrence of the fistula. They were also asked to complete the Cleveland Clinic Florida Fecal Incontinence (CCF-FI) Scale (Wexner score).13 The data used for this study were from the last follow-up visit or contact with the patient.
Follow-up examination for determination of the success rate took place at least 12 months after the procedure. The treatment was considered successful if the external opening was closed, and if the patient had no drainage at the last follow-up visit.
Statistical analyses were performed using SPSS 16.0 (SPSS, Chicago, IL). Factors associated with success or failure were identified in univariate analyses using the Pearson χ2 or Fisher exact test for proportions and the Mann-Whitney U test for continuous variables. A multivariate logistic regression analysis was performed to identify independent predictors. For all tests, differences were considered statistically significant if the p values were < 0.05.
Of the 262 patients who underwent fistulectomy with closure of the internal opening during the study period, 252 (96%) were available for review. These included 201 (80%) men and 51 (20%) women. The overall mean age was 48.7 (range, 21–81) years. Initial seton drainage was performed in 125 patients (50%), and 33 patients (13%) had at least 1 previous failed attempt at repair, including advancement flaps, fistulotomy, or cutting setons. The location of the internal opening of the fistula and the Parks category are shown in Table 1. During operative palpation, all 252 fistulas (including 50 intersphincteric and 202 transsphincteric or suprasphincteric fistulas) were observed to involve more than 30% of the sphincter muscle.
The median follow-up time was 70 (range, 14–141) months. Success was achieved in 187 patients(74%), failure occurred within the first year after surgery in 58 patients (23%), and a recurrent fistula developed 23 to 87 months after surgery in 7 patients (3%). Of the 58 patients with failure during the first year, 12 patients required early reoperation because of suture line dehiscence and 46 patients experienced persistent drainage from a residual external opening without recognized suture line dehiscence. In the 7 patients with late recurrence, the recurrent abscess or fistula developed at the same site as that of the previously closed fistula tract.
As shown in Table 2, patients for whom the procedure failed were significantly younger than those for whom it was a success (p = 0.010). Success was not associated with the sex of the patient, previous failed repair, seton use, performance of a preoperative MRI, Parks category, or location of the internal fistula opening. Patients with posterior transsphincteric or suprasphincteric fistulas had a significantly higher success rate than those with other types of fistula (p = 0.014). Multivariate analysis showed that older age and the presence of a posterior transsphincteric or suprasphincteric fistula decreased the likelihood of failure and thus were independent predictors of success (Table 3).
Of the 65 patients in whom treatment failed, 18 patients underwent a repeat fistulectomy with closure of the internal opening, and 14 (78%) of these patients achieved secondary success. Of the patients with treatment failure, 32 patients with a more superficial anatomy of the recurrent fistula went on to receive a fistulotomy; 29 (91%) of these fistulas closed. The remaining 15 patients declined repeat attempts at definitive repair and elected to leave a permanent seton in place (11 patients) or subsequently attended another hospital (4 patients).
The mean CCF-FI score for the 187 patients with success was 2.2 (range, 0–18); when those with a history of prior repair were excluded, the mean score was 2.1 (range, 0–18; n = 160). Of the 160 patients with success and without previous surgery, 89 (56%) had no incontinence (CCF-FI score = 0), and 58 (36%) reported some form of incontinence (for gas, liquid, and/or solid stools) “rarely to sometimes.” Thirteen patients (8%) reported having at least 1 form of incontinence “usually to always,” and 10 (77%) of these patients had a posterior transsphincteric or suprasphincteric fistula. Of the 160 total patients with success, 23 (14%) reported lifestyle alterations “rarely to sometimes,” and 11 (7%) reported lifestyle alterations “usually to always.” The patient with a CCF-FI score of 18 was a woman with a posterior transsphincteric fistula.
Our data show the effectiveness of total fistulectomy with simple closure of the internal opening in the management of complex fistulas. We have been using this approach for more than a decade for all but the most superficial fistulas. The overall success rate of 74% is comparable to that reported for other currently accepted sphincter-sparing procedures, such as advancement flaps and the fistula plug.9,11 Of note, total fistulectomy with simple closure of the internal opening was most successful in patients with posterior transsphincteric or suprasphincteric fistulas, with a success rate of 82%. We have also used this technique successfully to manage fistulas in Crohn’s disease and rectovaginal fistulas, but the current study was limited to a more homogenous group of patients with complex cryptoglandular fistulas.
In most reports, a full or partial flap of rectum is advanced for a tension-free, 2-layer closure of the internal opening, whereas the fistula tract is often only curetted.6,9 The creation of a flap can be technically difficult, especially in posterior fistulas.14 In contrast, even though we sometimes mobilized the full thickness of the proximal edge of the internal opening, we performed a simple appositional closure without an advancement flap. Instead, our technique involved a total fistulectomy with removal of the primary and secondary tracts and complete excision of the internal opening. Closure of the internal opening is also the objective of the fistula plug and the more recent LIFT procedure.7,8 However, both of these techniques are new, and neither has long-term data behind it. Few previous studies have used a closure technique similar to ours, and most are small studies with inhomogeneous populations. Our results were better than those reported by 2 smaller studies15,16 and were similar to those in a series of 90 patients with transsphincteric fistulas reported by Athanasiadis et al.17
The strengths of our study are the large number of patients involved and the long follow-up time. The median follow-up time for the patients with success was 70 (range, 14–141) months, and all of these patients had at least 1 year of follow-up. Although this long-term follow-up was based on self-report by mailed or telephone questionnaire in the majority of patients, we consider this subjective determination to be accurate because a patient’s perception of success is mainly based on symptom resolution.
Nevertheless, this was a retrospective study on prospectively collected data and, as such, had some limitations. Although preoperative ultrasound was routinely performed as an adjunct to operative palpation, the decision to perform the procedure was a judgment made by the individual surgeon during the operation, which could have introduced selection bias. The fact that 80% of the patients in this study were men might be evidence of selection bias, because this proportion is higher than the usually reported preponderance of men in populations with fistula disease.18 It is important to note that 79% of the 743 patients who underwent fistulotomy at our institution during the same period were men, although we cannot explain the overall high proportion of men in our population. Additionally, in most studies, intersphincteric fistulas were treated with fistulotomy rather than sphincter-sparing techniques because the risk of incontinence is very low after fistulotomy for such fistulas. Therefore, the fact that 20% of the fistulas in our study were intersphincteric might suggest a selection bias. However, partial division of the internal sphincter for other benign anorectal diseases, such as anal fissure, involves cutting a portion (usually to a maximum of 30%) of the internal sphincter only. From this it might be inferred that if an intersphincteric fistula involves more than 30% of the internal sphincter, then thought should be given to a sphincter-sparing approach.4
In our study, the majority of failures resulted from persistent drainage from a residual external opening, which was likely a consequence of missed and untreated fistula tracts. We have increasingly used MRI-guided surgery to detect all of the associated abscesses. However, only 42% of the patients in this study had an MRI, and there was no statistically significant overall difference between patients with and without MRI.
Our finding that a posterior transsphincteric or suprasphincteric fistula location was an independent predictor of success is encouraging because these are the most difficult types of fistulas to treat, and no other management option has proven to be superior. In our observation, the typical course of a posterior deep fistula is angled like a boomerang, running from the internal opening in a crypt backwards and upwards before passing through the puborectalis or levator ani, and then out again to the skin surface. Because there is no established classification other than the Parks system, we often classified these “boomerang” fistulas as suprasphincteric at surgery.
Age at surgery was also an independent predictor of outcome; patients with failure were significantly younger than those with success. We cannot explain this finding. It may be that the younger patients were more likely than the older patients to be considered for early reintervention, and potential slow healers were thus missed more often in the younger patients. In addition, younger patients may have had more aggressive forms of the disease.
No statistically significant difference between patients with and those without setons was observed in this study. However, total fistulectomy with simple closure of the internal opening was technically easier when a seton was in place. The presence of a seton matures the fistula tract (making the wall more fibrotic), reduces side tracts, and helps to control sepsis.19 Once the fistula tract has matured, the anatomy becomes clearer, making the tract easier to identify and remove completely. We support the idea that a draining seton should be used for all patients with “sphincteric” anal fistulas, regardless of which definitive sphincter-sparing treatment is planned.20
Because no sphincter was divided, the results of our postoperative continence evaluation were quite surprising; 44% of patients reported some degree of incontinence and 21% reported lifestyle-altering incontinence. These findings were even more surprising because 80% of the patients in this study were men and could not have had preexisting sphincter damage due to obstetric trauma. Additionally, men are generally more tolerant of less-than-perfect continence, so the fact that 21% reported lifestyle alteration is of concern. However, it is important to remember that we measured incontinence only postoperatively. The lack of a preoperative CCF-FI score makes it difficult to determine whether the reported incontinence was a result of the procedure, a preexisting condition, or simply age. The issue of incontinence requires further prospective study.
Total fistulectomy with simple closure of the internal opening is effective for the long-term closure of complex cryptoglandular fistulas. However, this procedure may affect continence despite its sphincter-sparing quality, and a less invasive approach may be preferable as an initial treatment. Nonetheless, the high success rate of total fistulectomy with simple closure of the internal opening in patients with posterior transsphincteric or suprasphincteric fistulas renders this procedure a reasonable option in this subgroup of patients with complex fistulas. Ultimately, the effect of any sphincter-sparing approach requires further prospective study.
1. Westerterp M, Volkers NA, Poolman RW, van Tets WF. Anal fistulotomy between Skylla and Charybdis. Colorectal Dis. 2003;5:549–551
2. Lindsey I, Jones OM, Smilgin-Humphreys MM, Cunningham C, Mortensen NJ. Patterns of fecal incontinence
after anal surgery. Dis Colon Rectum. 2004;47:1643–1649
3. Steele SR, Kumar R, Feingold DL, Rafferty JL, Buie WD. Practice parameters for the management of perianal abscess and fistula-in-ano. Dis Colon Rectum. 2011;54:1465–1474
4. Bleier JI, Moloo H. Current management of cryptoglandular fistula-in-ano. World J Gastroenterol. 2011;17:3286–3291
5. Ritchie RD, Sackier JM, Hodde JP. Incontinence rates after cutting seton treatment for anal fistula. Colorectal Dis. 2009;11:564–571
6. Stone JM, Goldberg SM. The endorectal advancement flap procedure. Int J Colorectal Dis. 1990;5:232–235
7. Johnson EK, Gaw JU, Armstrong DN. Efficacy of anal fistula plug vs. fibrin glue in closure of anorectal fistulas. Dis Colon Rectum. 2006;49:371–376
8. Bleier JI, Moloo H, Goldberg SM. Ligation of the intersphincteric fistula tract: an effective new technique for complex fistulas. Dis Colon Rectum. 2010;53:43–46
9. Soltani A, Kaiser AM. Endorectal advancement flap for cryptoglandular or Crohn’s fistula-in-ano. Dis Colon Rectum. 2010;53:486–495
10. Loungnarath R, Dietz DW, Mutch MG, et al. Fibrin glue treatment of complex anal fistulas has low success rate. Dis Colon Rectum. 2004;47:432–436
11. Garg P, Song J, Bhatia A, Kalia H, Menon GR. The efficacy of anal fistula plug in fistula-in-ano: a systematic review. Colorectal Dis. 2010;12:965–970
12. Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-in-ano. Br J Surg. 1976;63:1–12
13. Jorge JM, Wexner SD. Etiology and management of fecal incontinence
. Dis Colon Rectum. 1993;36:77–97
14. Schouten WR, Zimmerman DD, Briel JW. Transanal advancement flap repair of transsphincteric fistulas. Dis Colon Rectum. 1999;42:1419–1423
15. Thomson WH, Fowler AL. Direct appositional (no flap) closure of deep anal fistula. Colorectal Dis. 2004;6:32–36
16. Gustafsson UM, Graf W. Excision of anal fistula with closure of the internal opening: functional and manometric results. Dis Colon Rectum. 2002;45:1672–1678
17. Athanasiadis S, Helmes C, Yazigi R, Kohler A. The direct closure of the internal fistula opening without advancement flap for transsphincteric fistulas-in-ano. Dis Colon Rectum. 2004;47:1174–1180
18. Sainio P. Fistula-in-ano in a defined population. Incidence and epidemiological aspects. Ann Chir Gynaecol. 1984;73:219–224
19. Champagne BJ, O’Connor LM, Ferguson M, et al. Efficacy of anal fistula plug in closure of cryptoglandular fistulas: long-term follow-up. Dis Colon Rectum. 2006;49:1817–1821
20. Tyler KM, Aarons CB, Sentovich SM. Successful sphincter-sparing surgery for all anal fistulas. Dis Colon Rectum. 2007;50:1535–1539