You are here

Cone-like resection, fistulectomy and mucosal rectal sleeve partial endorectal pull-through in paediatric Crohn's disease with perianal complex fistula

Digestive and Liver Disease



Perianal abscesses and fistulae have been reported in approximately 15% of patients with paediatric Crohn's disease and they are associated with poor quality of life. Several surgical techniques were proposed for the treatment of perianal Crohn's disease, characterized by an elevated incidence of failure, incontinence, and relapse.

Aim of our study was to present the technical details and results of our surgical technique in case of recurrent, persistent, complex perianal ano-rectal destroying Crohn's disease not responding to medical treatment.


Data of patients who underwent surgical treatment (cone-like resection, fistulectomy, sphincter reconstruction, endorectal advancement sleeve flaps like in Soave endorectal pull-through) for complicated high-level trans, inter or suprasphincteric fistulae between January 2009 and June 2014 were retrospectively reviewed.


20 surgical procedures were performed in 11 patients (males 72.7%) with transsphincteric (n = 5), intersphincteric (n = 4) and suprasphincteric (n = 2) fistulae. Three patients needed a second treatment. Two patients needed more than 2 surgeries and one temporary colostomy. No patient presented anal incontinence at 15 months’ median follow-up.


Although several procedures may be required to obtain a complete remission of perianal lesions, in our series the proposed surgical technique seemed effective and safe, preserving anal continence in all treated cases and reducing the need of faecal diversion.

Keywords: Complex fistula, Cone-like resection, Paediatric, Perianal Crohn's.

1. Introduction

Paediatric Crohn's disease (PCD) accounts for 20–25% of total diagnosis of this inflammatory bowel disease, with an incidence ranging from 0.2 to 9.5:100.000 children in the United States.

Perianal disease has been reported in 8–15% of paediatric patients[1] and [2].

Complex fistula can have transsphincteric, suprasphincteric, intrasphincteric and extrasphincteric perianal localization [3] and represents a challenge for paediatric surgeons.

In recent years, magnetic resonance imaging (MRI) has become an important instrument to evaluate complex fistula severity and pelvic anatomy, providing indications for the correct type of surgical intervention. However, many false negative results and poor sensitivity of this tool have been reported[4] and [5].

Different approaches have been described for the treatment of complex fistula in children, including simple drainage, mobilization of tissue flaps, seton placement, fistulotomy, anus-sparing proctocolectomy, and defunctioning ileostomy[6], [7], and [8]. However, the risk of complications remained high, with more than one procedure for recurrence in 29–50% of cases[8], [9], and [10].

Aim of this study was to report on the surgical technique used in our institution, describing surgical details and main results.

2. Materials and methods

2.1. Study population

All patients with diagnosis of perianal PCD admitted to our Institute for complex fistula (defined according to Bell criteria) with recurrent/persistent, anorectal involvement not responding to medical treatment, between January 2009 and June 2014 were prospectively included in a database.

Patients with simple fistula (subcutaneous fistula) were excluded from the study.

This study was performed according to national ethical guidelines and informed consent was obtained for surgical treatment and data collection from parents or guardians. Data including demographics, previous surgical and medical treatments, surgical details of interventions, and clinical follow-up were retrospectively analyzed.

All patients were studied with pre-operative MRI. All patients were continent before surgical procedures.

2.2. Cone-like resection technique (CLR)

Peri-operative antibiotic prophylaxis with metronidazole was administered. Patients were placed in the lithotomy position under general anaesthesia without preoperative bowel preparation. A probe was inserted through the fistula to measure the distance from its internal opening. The fistula tract was completely mobilized en-bloc with the granulation tissue reaching the normal fatty tissue near the pelvic floor. A cone-like excision of skin and perianal tissue was performed with the cone base in the perineum including the anal canal if affected, and the cone apex in the rectal wall where the fistula opened (Fig 1 and Fig 2).


Fig. 1 Illustration of cone-like resection and rectal sleeve. IS, internal sphincter; ES, external sphincter; IF, intersphincteric fistula; TF, transsphincteric fistula; SS, suprasphinteric fistula; R, rectum; RS, rectal sleeve; CLFR, cone-like fistulectomy resection.


Fig. 2 The chronic Crohn's granulation tissue involves skin, fatty tissue, anal sphincter and rectal wall. In selected cases, when there is recurrence despite adequate medical treatment, therefore the inflamed tissue should be completely removed before complete destruction of sphincter activity due to risk of sepsis. The blue arrow indicates the previously placed seton.

Exposure of levator ani was needed to completely remove the affected tissue, including also rectal wall and anal sphincters if involved.

The second step of this surgical approach was to recreate the anal canal. The rectal sleeve was prepared proximally to the internal opening of the rectum. We used Soave endorectal pull-through (ERPT), pulling the normal rectal mucosa to the anal skin and suturing the sphincters to recreate anal ring normal shape. The perianal skin was always left open in order to reduce infection risks. No more than two areas were treated simultaneously (Fig 2 and Fig 3).


Fig. 3 Cone-like resection includes removal of ano-rectal canal and perianal tissue (skin and subcutaneous) macroscopically involved by Crohn's granulomatosis reaching the normal muscle of the perineum (levator ani, blue arrow). The second step includes restoration of normal continuity of the ano-rectal canal and sphincter activity (yellow arrow). The proximal rectal mucosa is pulled down to the perianal skin. The anal ring is recreated by suturing the normal muscle. The skin is left open. The aspect of resection is cone-like, the apex is in the rectum and the base is the skin of the perineum.

2.3. Endpoints

The primary endpoint was clinical recurrence, defined symptomatic recurrence requiring surgical treatment.

Secondary endpoints were definition of 30-day post-operative complication rate using Clavien-Dindo classification [11] , assessment of post-operative faecal incontinence using Yamataka score [12] , and analysis of post-operative pain using Faces, Legs, Activity, Cry, and Consolability (FLACC) score [13] .

Follow-up was performed with clinical evaluation.

2.4. Statistical analysis

Continuous variables are reported as means and standard deviation or median and range and were compared using Student'sttests; categorical variables are reported asn(%) and were compared using Chi-squared tests or Fisher's test. Possible risk factors as age at presentation, type of previous surgical treatment, pre-operative or post-operative medical treatment were analyzed with statistical significance defined asp < 0.05.

3. Results

From January 2009 to June 2014, 11 patients were treated (72.7% males, median age at surgery 12 years, range: 5–19 years).

In two patients perianal disease appeared after diagnosis of Crohn's disease (after 1 and 8 years), during immunosuppressive therapy (azathioprine) and on mesalazine. In the other cases, Crohn's disease was diagnosed during the evaluation of perianal disease. Location was ileocaecal (n = 7), ileocolic (n = 3), and panenteric (n = 1).

At preoperative endoscopy, rectal inflammation was present only in 2 cases (18.2%). One girl, previously followed for syndromic diarrhoea by home parenteral nutrition, presented with perianal rectal Crohn's-like disease involvement. Median Perineal Disease Activity Index (PDAI) was 5 (range: 3–12).

Diagnosis and classification of perineal disease were made by the surgeon with evaluation under anaesthesia with pelvic MRI.

Four subjects were initially treated at another hospital (3 with simple drainage and seton placement, 1 with fistulotomy). One patient presented a gluteus abscess and drainage with seton placement was performed along with prolonged antibiotic treatment. After abscess resolution, complex fistula persisted. The other 6 patients underwent simple drainage before CLR (54.5%). In 2 cases, patients were receiving medical treatment with biologics, azathioprine and thalidomide (18.2%); in the other 9 cases only antibiotics were administered.

The locations of the fistulae are shown in Table 1 ; transsphincteric location was present in 5 patients (45%).

Table 1 Overall patient population characteristics (n = 11).

Characteristic N (%)
Males 8 (72.7)
Median age (years) 12 (range 5–19)
Transsphincteric fistula 5 (45.4%)
Intersphincteric fistula 4 (36.4%)
Suprasphincteric fistula 2 (18.2%)
Median PDAI 5 (range 3–12)
Previous surgery
 Seton 4 (36.4%)
 Abscess drainage 6 (54.5%)
 Fistulotomy 1 (9.1%)
Median operative time (min) 40 (range: 20–80)
Median hospital stay (days) 4 (range: 3–7)
Median FLACC score 0
Median number of procedures 1 (range: 1–5)
Complications 0

PDAI, Perineal Disease Activity Index; FLACC, Faces, Legs, Activity, Cry and Consolability score.

Median surgery duration was 40 min (range 20–80 min) and median hospital stay was 4 days (range 3–7 days; Table 1 ).

Anti-TNF therapy was started in all patients within 10 days post-operatively, for a minimum of 12 months. No significant adverse effects were observed. Clinical and endoscopic follow-up was performed, post-operative MRI was performed in complex cases with more than two recurrences requiring repeated surgery or to rule out abscess. Step down to thiopurines was performed only in patients with sustained clinical and endoscopic remission. Median follow-up was 15 months (range 11–56 months).

A total of 20 CLRs with rectal sleeve were performed. In 6 cases complete remission was obtained after the first operation (54.5%). The remaining 5 required subsequent surgeries for relapses or new localizations. Three patients needed a second intervention, one a third procedure. One patient needed five treatments (Supplementary Table S1) and required colostomy for the recurrence of complex fistula, despite biological treatment. Colostomy was closed after 12 months and the patient underwent clinical follow-up.

The median number of procedures to obtain fistula healing was 1 (range: 1–5).

Evidence of wound healing by second intention was provided in the first month of follow-up ( Fig. 4 ), and at the end of surgical treatment all eleven patients healed with completerestitutio ad integrum. No other minor recurrences were observed.


Fig. 4 Healing by second intention with normal stool retention, sensory function, and continence. The cosmetic aspect is not objectively evaluated but scar retraction is limited and anal shape is normal.

Age at surgery, absence of previous surgical treatments, type of medical treatment before surgical procedure were not statistically related with recurrence. Post-operative pain was easily controlled with elastomeric pump (chirocaine plus clonidine) for the first two days and with non-steroidal anti-inflammatory drugs on day three every 8 h. Daily FLACC score was 0 for all patients.

No major complications nor anal incontinence were observed ( Table 1 ).

4. Discussion

Complex fistula is a debilitating condition for paediatric patients. Several treatments have been proposed, however the risk of recurrence remains high with a long history of medical and surgical procedures.

Very few studies on major surgical treatments for paediatric complex fistula are available in the literature. Current NASPHGAN guidelines [6] recommend surgeons not to perform advancement flaps or major surgery for high risk of failure, suggesting colostomy or ileostomy in case of severe or recurrent perianal Crohn's disease, especially in case of refractory infectious complications (such as recurrent abscess). Fistulectomy and other major surgical procedures did not gain popularity in the treatment of complex fistula for the risk of sphincter injury and incontinence.

In our opinion, it is better to completely remove perianal chronic inflammatory tissue, obviously only in case of persistent, recurrent complex disease not responding to medical treatment.

The purpose of this study was to use cone-like resection to reach complete removal of granulation tissue (fistulectomy) and recreate the ano-rectal canal. Sphincteric activity was restored using ERPT with rectal sleeve as main surgical technique in complex fistula in order to reach primary healing, low recurrence rate, and minimal risk of sphincter injury when sphincteric section was necessary to remove all inflammatory tissue. ERPT allows the restoration of nearly normal perineal shape and limited scar retractions.

In our series of 11 patients with complex fistula, cone-like fistulectomy with rectal sleeve was a safe and well-tolerated procedure. Fistula healing rate was 54.5% with no case of faecal incontinence after the first surgical procedure and 100% after maximum 5 procedures.

In accordance to Arroyo et al. [14] our series confirms that fistulectomy with sphincteric surgery is a procedure with limited risk of faecal incontinence.

The role of faecal diversion for complex fistula remains unclear in the literature. In adult patients, the reported incidence is about 31% [15] , while in children the incidence of faecal diversion was reported in few articles, and in the largest series, 23% of cases had defunctioning ileostomy with 38% of stoma-related complications [7] .

Though faecal diversion is an accepted major invasive treatment for complex fistula, in addition to stoma complications the child's quality of life must also be considered. In adult patients some studies on quality of life of have been published[16], [17], and [18], to our knowledge, there is only one reported study on this topic in the paediatric population [19] .

In our series, CLR was used as first major surgical procedure and faecal diversion was associated with fistulectomy only in one patient (9%) with recurrent complex fistula and high risk of sepsis. The introduction of biologic agents has dramatically changed the therapeutic strategy for IBD in children. The first evidence-based practical guidelines on medical management in paediatric-onset Crohn's disease have recently been published. Among the recommended biologics, anti-TNF agents are the treatment of choice in active perianal fistulising disease in combination with appropriate surgical intervention [20] .

Combined anti-TNF therapy and surgery showed improved healing and lower recurrence of fistulae compared with surgery alone in paediatric patients [21] . Combined treatment was also successful in adult patients with faster and prolonger fistula healing as described by Sciaudone et al. [22] .

Hukkinen reported a 70% healing rate in 13 patients and setons were kept for 8 months [23] . In our series, all patients recovered from complex fistula and wounds healed by second intention within one month.

NASPHGAN guidelines also suggested seton placement for treatment of paediatric complex fistula but as reported by Langer et al.[6] and [7], in some cases up to 7 placements with multiple anaesthesia procedures can be required. Furthermore, reported healing rate reported is low.

CLR is characterized by a low number of surgical interventions and consequently of anaesthesia procedures in children. CLR also allows an easy post-operative pain management: morphine is not required and patients may be discharged only with non-steroidal anti-inflammatory drugs.

In our experience, CLR is a safe and well-tolerated technique with high primary healing and low recurrence rates, without risk of sphincter injury and faecal incontinence when performed by experienced surgeons. In some cases multiple procedures are required, and adequate medical treatment with biologics is needed to consolidate remission. CLR could be considered as a primary major surgical technique in children with complex fistula, in association with biologics, reserving enterostomy only for very difficult non-responder cases.

However, as reported by Pellino et al. [24] , other surgical treatments like fibrin glue or adipose tissue-derived stem cell injection have shown promising preliminary results and further studies are required to improve surgical outcomes for the treatment of complex fistula.

Conflict of interest

None declared.


We thank Anna Capurro for her help in revising the manuscript.

Appendix A. Supplementary data

The following are the supplementary data to this article:

Download file

Table S1 Patients description with demographics and surgical outcomes.


  • [1] J. Markowitz, F. Daum, H. Aiges, et al. Perianal disease in children and adolescents with Crohn's disease. Gastroenterologia. 1984;86:829-833
  • [2] S. Kugathasan, R.H. Judd, R.G. Hoffmann, et al. Epidemiologic and clinical characteristics of children with newly diagnosed inflammatory bowel disease in Wisconsin: a statewide population-based study. Journal of Pediatrics. 2003;143:525-531 Crossref
  • [3] A.G. Parks, P.H. Gordon, J.D. Hardcastle. A classification of fistula-in-ano. British Journal of Surgery. 1976;63:1-12 Crossref
  • [4] B. Essary, J. Kim, S. Anupindi, et al. Pelvic MRI in children with Crohn's disease and suspected perianal involvement. Pediatric Radiology. 2007;37:201-208 Crossref
  • [5] P. Toma, C. Granata, G. Magnano, et al. CT and MRI of paediatric Crohn's disease. Pediatric Radiology. 2007;37:1083-1092 Crossref
  • [6] E.F. De Zoeten, B.A. Pasternak, P. Mattei, et al. Diagnosis and treatment of perianal Crohn's disease: NASPGHAN clinical report and consensus statement. Journal of Pediatric Gastroenterology and Nutrition. 2013;57:401-412 Crossref
  • [7] N.M. Seemann, A. Elkadri, T.D. Walters, et al. The role of surgery for children with perianal Crohn's disease. Journal of Pediatric Surgery. 2015;50:140-143
  • [8] S.A. Strong. Perianal Crohn's disease. Seminars in Pediatric Surgery. 2007;16:185-193 Crossref
  • [9] S.S. Short, M.C. Dubinsky, S. Rabizadeh, et al. Distinct phenotypes of children with perianal perforating Crohn's disease. Journal of Pediatric Surgery. 2013;48:1301-1305 Crossref
  • [10] D.J. Keljo, J. Markowitz, C. Langton, et al. Course and treatment of perianal disease in children newly diagnosed with Crohn's disease. Inflammatory Bowel Diseases. 2009;15:383-387 Crossref
  • [11] P.A. Clavien, J. Barkun, M.L. de Oliveira, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Annals of Surgery. 2009;250:187-196 Crossref
  • [12] T. Ochi, T. Okazaki, G. Miyano, et al. A comparison of clinical protocols for assessing postoperative fecal continence in anorectal malformation. Pediatric Surgery International. 2012;28:1-4 Crossref
  • [13] S. Nilsson, B. Finnström, E. Kokinsky. The FLACC behavioral scale for procedural pain assessment in children aged 5-16 years. Paediatric Anaesthesia. 2008;18:767-774 Crossref
  • [14] A. Arroyo, J. Perez-Legaz, P. Moya, et al. Fistulotomy and sphincter reconstruction in the treatment of complex fistula-in-ano: long-term clinical and manometric results. Annals of Surgery. 2012;255:935-939 Crossref
  • [15] M.K. Hong, A. Craig Lynch, S. Bell, et al. Faecal diversion in the management of perianal Crohn's disease. Colorectal Disease. 2011;13:171-176 Crossref
  • [16] S.H. Lim, S.W. Chan, J.H. Lai, et al. Journal of Advanced Nursing. 2014;10.1111/jan.12595
  • [17] L. Prieto, H. Thorsen, K. Juul. Development and validation of a quality of life questionnaire for patients with colostomy or ileostomy. Health and Quality of Life Outcomes. 2005;3:62 Crossref
  • [18] B. Person, R. Ifargan, J. Lachter, et al. The impact of preoperative stoma site marking on the incidence of complications, quality of life, and patient's independence. Diseases of the Colon and Rectum. 2012;55:783-787 Crossref
  • [19] L. Bray, C. Sanders. Preparing children and young people for stoma surgery. Paediatric Nursing. 2006;18:33-37 Crossref
  • [20] F.M. Ruemmele, G. Veres, K.L. Kolho, et al. Consensus guidelines of ECCO/ESPGHAN on the medical management of Paediatric Crohn's disease. Journal of Crohn's and Colitis. 2014;8:1179-1207
  • [21] A. Barabino, E. Castellano, P. Gandullia, et al. A girl with severe fistulizing Crohn's disease. Digestive and Liver Disease. 2000;32:792-794 Crossref
  • [22] G. Sciaudone, C. Di Stazio, P. Limongelli, et al. Treatment of complex perianal fistulas in Crohn's disease: infliximab, surgery or combined approach. Canadian Journal of Surgery. 2010;53:299-304
  • [23] M. Hukkinen, M.P. Pakarinen, M. Piekkala, et al. Treatment of complex perianal fistulas with seton and infliximab in adolescents with Crohn's disease. Journal of Crohn's and Colitis. 2014;8:756-762
  • [24] G. Pellino, F. Selvaggi. Surgical treatment of perianal fistulizing Crohn's disease: from lay-open to cell-based therapy—an overview. Scientific World Journal. 2014;2014:146281 10.1155/2014/146281


a Giannina Gaslini Institute, Genoa, Italy

b DINOGMI, University of Genoa, Italy

lowast Corresponding author at: Paediatric Surgery Unit, Istituto Giannina Gaslini, Largo G. Gaslini 5, 16147 Genoa, Italy. Tel.: +39 010 56362217; fax: +39 010 3075092.