The outcome of fistulotomy for anal fistula at 1 year: a prospective multicentre French study

Transcription

1 Original article doi: /codi The outcome of fistulotomy for anal fistula at 1 year: a prospective multicentre French study L. Abramowitz*, D. Soudan, M. Souffran, D. Bouchard, A. Castinel, J. M. Suduca**, G. Staumont**, F. Devulder, F. Pigot, R. Ganansia, M. Varastet and for the Groupe de Recherche en Proctologie de la Societe Nationale Francßaise de Colo-Proctologie and the Club de Reflexion des Cabinets et Groupe d Hepato-Gastroenterologie *Proctologie Medico-Chirurgicale, H^opital Bichat-Claude Bernard, AP-HP, Paris, France, Institut de Proctologie Leopold Bellan, H^opital Saint Joseph, Paris, France, Service de Proctologie, Clinique Saint Augustin, Nantes, France, Service de Proctologie, H^opital Bagatelle, Talence, France, Clinique Theodore Ducos, Bordeaux, France, **Service de Proctologie, Clinique St Jean Languedoc, Toulouse, France, Clinique Courlancy, Reims, France and ClinSearch, Bagneux, France Received 13 February 2015; accepted 3 July 2015; Accepted Article online 18 September 2015 Abstract Aim The study aimed to evaluate outcome at 1 year of one- and two-stage fistulotomy for anal fistula in a large group of patients. Method A prospective multicentre observational study was designed to include patients with anal fistula treated by one- or two-stage fistulotomy. Data were collected using a self-administered questionnaire before surgery, during healing and at 1 year after surgery. Results Group A (133 patients) with a low anal fistula underwent a one-stage fistulotomy. The median Wexner scores before and after surgery were 1.0 (0 11) and 2.0 (0 18) (P = 0.032) and the median Vaizey scores were 2.0 (0 14) and 3.0 (0 21) (P = 0.055). The Wexner scores and percentage of patients before and after fistulotomy were as follows: 0 5: 88%, 86%; 6 10: 10.7%, 10.7%; 11 15: 1.0%, 2.6%; and 16 20: 0%, 2%. Eighty-seven per cent of the patients were satisfied. Group B (62 patients) underwent two-stage fistulotomy for a high transsphincteric fistula. The Wexner scores and percentage of patients before the first stage and 1 year after the second stage were as follows: 0 5: 86%, 66%; 6 10: 4.5%, 20%; 11 15: 9%, 11%; and 16 20: 0%, 2%. The median Wexner scores before the first stage and after the second stage were 1 (0 14) vs 4 (0 19) (P < 0.001), and the median Vaizey scores were 1.5 (0 11) vs 4(0 20) (P < 0.001). Eighty-eight per cent of the patients were satisfied. Conclusion Low transsphincteric anal fistula can be treated by fistulotomy without clinically significant continence disturbance. Treating high transsphincteric anal fistulae with two-stage fistulotomy is followed by mild continence disturbance. Satisfaction rates were high. Keywords Anal fistula, seton, fistulotomy, anal incontinence What does this paper add to the literature? We report the results of a prospective evaluation of one- and two-stage fistulotomy in a multicentre study that assessed the outcome at 1 year using a self-administered questionnaire. There was a small rise in the continence scores after two-stage fistulotomy but patient satisfaction was high. Introduction Correspondence to: Laurent Abramowitz, Service de gastroenterologie/ proctologie, AP-HP, H^opital universitaire Bichat Claude Bernard, 46 rue Henry Huchard 75877, Paris Cedex 18, France. laurent.abramowitz@bch.aphp.fr Partial reports of these data have been presented at the JFHOD 2010 (25 28 March, Paris, France, oral presentation), the ASCRS 2010 (15 19 May, Minneapolis, Minnesota, USA, poster P44) and the ESCP 2010 (22 25 September, Sorrento, Italy, poster LTP84). Anal incontinence and relapse of sepsis can usually be avoided by fistulotomy for low anal fistula [1 3]. In the case of complex fistulae, one-stage fistulotomy can seriously impair anal continence. Impaired continence is usually but not only related to the extent of sphincter muscle division [4 6]. Reports of the results of fistulotomy in previous studies are confusing Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

2 One-year outcome of fistulotomy L. Abramowitz et al. because various approaches to treatment may have been used and a seton will often have been employed [7,8]. Sphincter-saving procedures are available, but their results have been somewhat disappointing in that their reported success in avoiding relapse of sepsis is very variable ranging from < 20% to over 70% [7]. Most studies that have evaluated the short- and long-term results of fistulotomy have come from single centres [6,9,10]. To avoid this, a multicentre study was designed to assess the results of one- and two-stage fistulotomy 1 year after treatment in tertiary care referral centres in France. Relapse, continence and quality of life were evaluated before and at 1 year after surgery using validated scoring systems and a visual analogue scale in a self-administered questionnaire. Method This was a prospective multicentre observational study. Nine gastroenterologists from Paris, Bordeaux, Talence, Toulouse, Reims and Nantes experienced in anal surgery were asked to participate. All the patients gave their written consent, and the study was approved by the Commission Nationale de l Informatique et des Libertes. Patients who were to have a planned one- or twostage fistulotomy were invited to participate in the study. Patients with inflammatory bowel disease or a fistula scheduled to be treated with more than a two-stage fistulotomy, e.g. a horseshoe fistula or an extrasphincteric fistula, or who were unable to complete the questionnaires were excluded. The surgical procedures carried out were standard. The fistula was evaluated under general anaesthesia according to the classification of Arnous et al. [11], which is similar to Parks system for transsphincteric fistula [12]. Patients with a low transsphincteric track were designated group A and were treated by fistulotomy with an electrocautery knife. If an abscess was present it was initially drained and the fistula track was treated according to its level in the anal canal. In group B patients, a loose non-cutting seton was inserted into the primary track and any secondary tracks were laid open. The external sphincter was partially divided during the first stage according to the treatment commonly used in France [11,13,14]. From the external opening, the cephalad part of the external sphincter was divided with an electrocautery knife until the track was in a lower position, and a loose seton was inserted. The internal sphincter was not divided at this time if it was not necessary to do so. The second operation was performed after the first had healed and it included a fistulotomy of the remaining track drained by the loose seton. Data collection included the following: patient demographics, obstetric history and previous treatment for any proctological diseases. Before surgery, during the healing period and 1 year after surgery, the patients completed a standardized anonymous questionnaire on the presence and severity of proctological symptoms (pain, soiling, anal discomfort, constipation, diarrhoea and continence), their health-related quality of life [15] and the degree of satisfaction after surgery using a visual analogue scale ranging from 0 (none) to 10 (maximal). Constipation was assessed using the Knowles Eckersley Scott Symptom (KESS) questionnaire [16]. Continence was assessed with the Vaizey scale [17] and the Wexner scale [18]. Data were collected before surgery, during the healing period and at 1 year after surgery using a standardized anonymous questionnaire. Healing was assessed by a surgeon. In group B, the staged fistulotomy was not carried out until the wound created by the first procedure had healed and data were collected after each of the two stages. Patients in both groups were sent a questionnaire and a return pre-paid envelope 1 year after surgery. Statistical analysis Statistical analysis was performed using SAS version 9.2, Bagneux, France. Proportions were calculated using the number of available data as the denominator. Pre- and post-surgical variables were compared using the Wilcoxon signed-rank test for paired samples. Group comparisons at a given time point were performed using the chi-squared test or Fisher s exact test as applicable. Statistical significance was set at the 0.05 level. Results Before surgery There were 133 patients in group A including 107 men (80.5%) and 62 in group B including 37 (58.6%) men in the study from January 2007 to June The difference in proportions of men in each group was statistically significant (P = 0.02). The median age in groups A and B was 48 (21 83) years and 43 (24 78) years (Table 1). Chronic diarrhoea was present in 10 (7.5%) and six (10%) patients in groups A and B, and constipation in seven (5.26%) and five (8.2%) patients. Dyschesia was present in six (4.51%) and no patients. A history of previous anal surgery was present in 22 (13.6%) and 11 (17%) of the patients in groups A and B, respectively. Six (4.5%) patients in group A and four (6.4%) in group B had had a previous abscess incised or had undergone fistula surgery (P > 0.5). 280 Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

3 L. Abramowitz et al. One-year outcome of fistulotomy Table 1 Patients characteristics. Before fistulotomy, n = 133, group A Before seton insertion for staged fistulotomy, n = 62, group B P value Age, median (range), years 48.4 (21 83) 43.3 (24 78) Male gender (%) 107 (80.5) 37 (58.6) Constipation (%) 7 (5.26) 5 (8.2) Dyschesia (%) 6 (4.51) 0 (0) Diarrhoea (%) 10 (7.52) 6 (10) Past anal surgery (%) 22 (13.6) 11 (17) Abscess or fistula (%) 6 (4.5) 4 (6.4) Haemorrhoidectomy, pexy, warts, fissurectomy (%) 16 (12) 7 (11.2) Perineal surgery (%) 5 (3.1) 1 (1.5) Cholecystectomy (%) 4 (3.01) 1 (1.64) Colectomy (%) 2 (1.50) 1 (1.64) Small bowel resection (%) 1 (0.75) 1 (1.5) Diabetes (%) 5 (3.76) 3 (4.6) HIV positive* (%) 2 (1.50) 0 (0) Previous treatment Antibiotics (%) 7 (4.3) 1 (1.6) Single abscess incision (%) 17 (12.8) 1 (1.6) Surgery for fistula (%) 4 (4.3) 3 (4.8) Obstetric history (%) 26/27 17/32 Vaginal delivery (% of women) (%) 25 (96) 16 (64) Perineal tear, forceps, episiotomy (%) 11 (40) 4 (16) Fisher s exact test was used to compare baseline characteristics between group A and group B. P significant if < *Median CD4 count: 515 cells/mm 3 (range ). Surgery The operation note was available for all patients in both groups. In group A, a low intersphincteric fistula was present in 12 (9%) patients. The lower third of the external sphincter was affected in all the other cases. An abscess with a diameter greater than 2 cm was present in 13 (8.3%) patients. It was necessary to lay open the intersphincteric blind track more than 1 cm above the dentate line in 15 (9.5%) patients. In group B, an abscess was present in 25 (43.1%) patients and the median interval between the two stages was 8 (3 12) weeks. Early results Of the group A patients, 128 (96%) were followed until the fistula healed. One patient had an immediate haemorrhage that required reoperation to ensure haemostasis. The median healing time was 8 (3 16) weeks. All of the 62 group B patients were followed after surgery. Early complications included bleeding (n = 1), local infection (n = 1), hyperalgesia (n = 1) and urinary retention (n = 1). The median healing time was 9 weeks (4 12) after the second-stage fistulotomy. Two per cent of the patients before the procedure and 9% after the first stage had a Wexner score of more than 10. One-year follow-up Relapse of sepsis, anal discomfort and KESS score Sepsis that required a further operation occurred in three patients (one in group A and two in group B). The incidence of anal discomfort in group A and B patients was 5.75 (before) vs 0.1 (after) (P < 0.001) and 4 vs 0.3 (P < 0.001). The KESS score changed from 4 (before) to 3 (after) (P = 0.028) in group A and was 4 and 4 in group B. Anal incontinence Continence was assessed by both the Vaizey and the Wexner scales before and 1 year after fistulotomy in 93/133 (70%) of the patients in group A. A Wexner score of > 5 was reported in 11 (8.2%) patients before surgery and in 13 (13%) at 1 year (Table 2). The median Vaizey continence score at 1 year and at baseline was 3 (0 21) vs 2(0 14) (P = 0.055) and 2 (0 18) vs 1(0 11) (P = 0.032) using the Wexner scale. The Wexner score did not change or increased to 5 points in 54 (69%) patients, increased to > 5 points in six (6%) and decreased in 23 (25%). Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

4 One-year outcome of fistulotomy L. Abramowitz et al. Table 2 Wexner score 1 year after fistulotomy (group A). Wexner score Before surgery N (%) 82 (88) 10 (10.7) 1 (1) 0 (0) One year after surgery N (%) 80 (86) 10 (10.7) 2 (2.6) 1 (1) Table 3 Wexner score 1 year after staged fistulotomy (group B). Wexner score Before surgery N (%) 38 (86) 2 (4.5) 4 (9) 0 (0) One year after second stage N (%) 29 (66) 9 (20) 5 (11) 1 (2) Continence was evaluated in 55/62 group B patients after the first-stage fistulotomy and in 44/55 1 year after the second stage. The median Vaizey continence scores 1 year after the second stage and at baseline were 4(0 20) and 1.5 (0 11) (P < 0.001) and 4 (0 19) and 1 (0 14) (P < 0.001) using the Wexner scale. After the second-stage fistulotomy, the Wexner score did not change or increased to 5 points in 36 (82%) patients, increased to > 5 points in three (7%) and decreased in five (11%) (Table 3). For both group A and group B patients, female gender and having diarrhoea were risk factors for a Wexner score greater than 5 at 1 year (P < 0.05). Obstetric trauma, the number of vaginal deliveries, dyschesia, cholecystectomy, perineal surgery, HIV status and diabetes mellitus were not identified as risk factors for incontinence at 1 year (Table 4). Patient quality of life and satisfaction The patients quality of life assessed by the SF-36 [15] improved significantly at 1 year after surgery compared with baseline in both groups (P < 0.001). At 1 year, the mean subscores in each group were approximately the same as those reported in the general population [15]. In group A 83 (86%) patients stated they were satisfied (n = 25, 26%) or very satisfied (n = 58, 60.4%) at 1 year, and 89 (92.7%) indicated that they would undergo fistulotomy again. In group B, 38 (88%) patients were satisfied (n = 13, 30.2%) or very satisfied (n = 25, 58.1%) at 1 year, and 37 (93%) indicated that they would undergo staged fistulotomy again. Discussion This prospective multicentre French study evaluated the results of one- and two-stage fistulotomy for transsphincteric anal fistula. Most proctologists in France practise staged fistulotomy from the cephalad to the caudal part of the external sphincter without division of the internal sphincter as a first stage unless division is necessary for treating a high or complex fistula [11,14,19,20]. The study is the first prospective multicentre evaluation of fistulotomy. Immediate and early complications were very uncommon with only one postoperative haemorrhage occurring in 257 procedures. Sepsis was successfully treated in almost every patient with only one (0.7%) group A patient and two (3%) in group B needing an additional procedure for continuing sepsis during the year following the fistulotomy. These data are in agreement with the relapse rates (0 29%) reported in recent reviews [9,21]. In a survey of specialist surgeons, relapse of sepsis occurred in 1.1% and 7.5% of 88 low and 53 high transsphincteric fistulae during a 19-month duration of follow-up [22]. Several studies have found that the recurrence rate of sepsis is significantly influenced by the type of surgeon [4,21]. Moreover an audit has shown that, although specialist surgeons performed more staged fistulotomies than general surgeons (44.1% vs 10.0%), fewer recurrences occurred among those treated by the specialist (9.7% vs 30.0%) [23]. Other factors that play an important role in recurrence include the presence of complex fistula tracks or horseshoe extensions and failure of the surgeon to find the internal opening [2,4,11,22]. Such cases, however, were excluded from the present study. It is difficult or impossible to make comparisons of rates of continence disturbance between studies based on the current literature because of the heterogeneity of definitions and methods of data collection. The reported incidence of incontinence after fistulotomy varies from 2% to 82% [1,21,24]. Because the Wexner and the Vaizey scales are validated measures of incontinence that provide more information and are more sensitive, they were preferred to the simple description of loss of flatus, liquid or solid stool [8]. An improvement in the Wexner score has not been reported previously and is therefore somewhat surprising. The improvement may have been due to draining of an associated abscess. This possibility should be considered when collecting information for the purpose of evaluating the results of fistulotomy. It is also noteworthy that the preoperative Wexner score was more than 5 in around 10% of patients in each group, a figure similar to that of the general French population [25]. A preoperative continence disturbance 282 Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

5 L. Abramowitz et al. One-year outcome of fistulotomy Table 4 Proportion of patients with worsened anal incontinence 1 year after fistula surgery in selected subgroups that are potentially at risk. Score < 5 before and Score < 5 before surgery and > 5 Total 1 year after surgery 1 year after surgery N N (%) N (%) P value Age, years (77) 14 (77) 1.00 > (23) 4 (22) Gender Male (75) 9 (50) Female (25) 9 (50) Constipation No (95) 16 (94) 1.00 Yes 7 6 (5) 1 (6) Dyschesia No (97) 16 (94) Yes 4 3 (3) 1 (6) Diarrhoea No (92) 12 (70) Yes 14 9 (8) 5 (30) Perineal surgery No (98) 17 (100) 1.00 Yes 2 2 (2) 0 (0) Cholecystectomy No (98) 17 (100) 1.00 Yes 2 2 (2) 0 (0) Diabetes No (94) 17 (100) Yes 7 7 (6) 0 (0) HIV positive No (99) 16 (94) Yes 2 1 (1) 1 (6) Vaginal deliveries (n) (32) 3 (33) (18) 2 (22) (29) 1 (11) (18) 0 (0) (4) 1 (11) (0) 2 (22) Episiotomy No (36) 4 (44) Yes (25) 1 (11) Unknown (40) 4 (44) Perineal tears No (59) 05 (55) Yes (22) 01 (11) Unknown (18) 03 (33) Forceps No (45) 05 (56) Yes 3 3 (10) 0 (0) Unknown (44) 4 (44) N, number of available data. Percentages are n/n of patients. Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

6 One-year outcome of fistulotomy L. Abramowitz et al. has been identified as a significant risk factor for postoperative incontinence in around 5% of patients [22]. The absence of any change in the Vaizey scale in group A after fistulotomy contrasts with the slight difference in the Wexner scores, but this is unlikely to be clinically relevant since the change was only one or two points. The increase by more than five points in the small number of group A patients is in agreement with previous reviews and guidelines [1 3,26]. In a study of 51 patients with a low fistula operated on by a senior surgeon, the proportion of fistulotomy-induced continence disturbance for flatus was 30%, for soft stool 4% and for hard stool 2%. Eight per cent needed to wear a pad, according to St Mark s incontinence score [8]. In our study, it is noteworthy that each of the two fistulotomy stages decreased the continence scores of the patients in group B. A Wexner score of over 10, which is usually associated with an alteration in the quality of life [27], was reported in 2% before surgery, 9% after the first stage and 13% after the second stage, as a consequence of sphincter division. Although Williams et al. [28] reported minor incontinence in 54% of patients after a two-staged fistulotomy, the stepwise worsening with each staged procedure in the present study is the first time this observation has been reported. The overall mean score of the patients in group B remained at the moderate figure of 4/20, but it is impossible to compare our findings to those of other publications. The only other risk factors, in addition to fistula track anatomy and the surgical treatment, found to be significant in the present study were diarrhoea and female gender, both of which were associated with an increase in the Wexner score of more than 5 points. Female gender is already a recognized risk factor [4]. Improvement of sepsis with maintained continence is related to the quality of life [10,27]. This was demonstrated in the present study with improvement of the SF-36 index at 1 year to approximate the values seen in the general French population [15]. The visual analogue scale assessment of discomfort was improved despite differences in continence disturbance, suggesting that the relief of symptoms of anal fistula is more important to the patient than small disturbances in continence. It is known that the surgeon and the patient may have a different opinion on which symptoms are most troublesome. Thus in a study examining this question continence and leakage were regarded by surgeons to be more important to independent activity and pain, whereas patients considered these to be more important than continence [29]. This may explain why differences in continence were not correlated with quality of life score in the present study. Incontinence and reoccurrence assessed by SF-12 or the Gastrointestinal Quality of Life Index both impair quality of life [25,30], but this was not significant for Wexner scores < 9 [27]. A survey on postoperative satisfaction among 624 patients [31] found a higher dissatisfaction rate (61% vs 24%) in patients with recurrence of the fistula but, because a greater proportion of patients with anal incontinence were dissatisfied (84% vs 33%), dissatisfaction was attributed more to incontinence. The symptoms of anal fistula vary widely and satisfaction and quality of life will depend on several factors including the anatomy of the fistula and its treatment in addition to gender, age, parity, consistency of stools, sphincter preservation and previous anal surgery [32]. Each fistula should be regarded as unique and its pretreatment characteristics can lead to different recommendations for treatment based on the aim to deal with sepsis while preserving continence. More patients are willing to accept an increased risk of recurrence in order to minimize any risk of continence disturbance but, when the success rate of a sphincter-preserving technique is as low as 30%, more will be willing to accept the risk of a degree of impairment of continence in exchange for successful treatment of the fistula [33]. The availability of sphincter-saving procedures provides patients with more choice so that each can receive the best tailored treatment. The opportunity for sphincter-preserving treatment resulted in an inevitable bias in our study in that some patients opted for such treatment rather than fistulotomy. The use of these new procedures was not common, however, during the period of this prospective multicentre study. Another limitation may be that the patients were treated in specialist centres and not in general units. This multicentre evaluation of everyday practice has shown that fistulotomy is effective in preventing relapse of sepsis at 1 year while inducing only a very mild degree of continence disturbance in patients with a simple fistula. A two-staged fistulotomy was effective in curing the fistula but was associated with some degree of incontinence in about 10% of patients. In both cases, quality of life is improved and patient satisfaction high. Acknowledgements The authors wish to acknowledge the late Doctor Philippe Guyot (H^opital Universitaire Edouard Herriot, Lyon), who contributed to the concept and design of the study, and Doctors Paul Preziosi and Asmaa Zkik of ClinSearch, Bagneux, who analysed the data. The Laboratoires Pierre Fabre and ClinSearch funded part of this work. 284 Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,

7 L. Abramowitz et al. One-year outcome of fistulotomy Conflicts of interest The authors have no conflicts of interest to declare concerning this paper. References 1 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: Ommer A, Herold A, Berg E, Furst A, Sailer M, Schiedeck T. Cryptoglandular anal fistulas. Dtsch Arztebl Int 2011; 108: Williams JG, Farrands PA, Williams AB et al. The treatment of anal fistula: ACPGBI position statement. Colorectal Dis 2007; 9(Suppl 4): Garcia-Aguilar J, Belmonte C, Wong WD, Goldberg SM, Madoff RD. Anal fistula surgery. Factors associated with recurrence and incontinence. Dis Colon Rectum 1996; 39: Lunniss PJ, Kamm MA, Phillips RK. Factors affecting continence after surgery for anal fistula. Br J Surg 1994; 81: Garces-Albir M, Garcia-Botello SA, Esclapez-Valero P et al. Quantifying the extent of fistulotomy. How much sphincter can we safely divide? A three-dimensional endosonographic study. Int J Colorectal Dis 2012; 27: Blumetti J, Abcarian A, Quinteros F, Chaudhry V, Prasad L, Abcarian H. Evolution of treatment of fistula in ano. World J Surg 2012; 36: Atkin GK, Martins J, Tozer P, Ranchod P, Phillips RK. For many high anal fistulas, lay open is still a good option. Tech Coloproctol 2011; 15: Roig JV, Jordan J, Garcia-Armengol J, Esclapez P, Solana A. Changes in anorectal morphologic and functional parameters after fistula-in-ano surgery. Dis Colon Rectum 2009; 52: Cavanaugh M, Hyman N, Osler T. Fecal incontinence severity index after fistulotomy: a predictor of quality of life. Dis Colon Rectum 2002; 45: Arnous J, Parnaud E, Denis J. Quelques reflexions sur les abces et les fistules a l anus (a propos de 3000 interventions) [Abscesses and fistula of the anus (with regard to 3000 operations)]. Rev Prat 1972; 22: Parks AG, Gordon PH, Hardcastle JD. A classification of fistula-in-ano. Br J Surg 1976; 63: Ramanujam PS, Prasad ML, Abcarian H. The role of seton in fistulotomy of the anus. Surg Gynecol Obstet 1983; 157: Pearl RK, Andrews JR, Orsay CP et al. Role of the seton in the management of anorectal fistulas. Dis Colon Rectum 1993; 36: 573 7; discussion Perneger TV, Leplege A, Etter JF, Rougemont A. Validation of a French-language version of the MOS 36-Item Short Form Health Survey (SF-36) in young healthy adults. J Clin Epidemiol 1995; 48: Knowles CH, Eccersley AJ, Scott SM, Walker SM, Reeves B, Lunniss PJ. Linear discriminant analysis of symptoms in patients with chronic constipation: validation of a new scoring system (KESS). Dis Colon Rectum 2000; 43: Vaizey CJ, Carapeti E, Cahill JA, Kamm MA. Prospective comparison of faecal incontinence grading systems. Gut 1999; 44: Jorge JM, Wexner SD. Etiology and management of fecal incontinence. Dis Colon Rectum 1993; 36: Ramanujam PS, Prasad ML, Abcarian H, Tan AB. Perianal abscesses and fistulas. A study of 1023 patients. Dis Colon Rectum 1984; 27: Vial M, Pares D, Pera M, Grande L. Faecal incontinence after seton treatment for anal fistulae with and without surgical division of internal anal sphincter: a systematic review. Colorectal Dis 2010; 12: Roig JV, Garcia-Armengol J. Tratamiento de las fistulas de ano complejas de causa criptoglandular. Aun se requiere un cirujano con experiencia? [Treatment of complex cryptoglandular anal fistulas. Does it still require an experienced surgeon?]. Cir Esp 2013; 91: Jordan J, Roig JV, Garcia-Armengol J, Garcia-Granero E, Solana A, Lledo S. Risk factors for recurrence and incontinence after anal fistula surgery. Colorectal Dis 2010; 12: Nwaejike N, Gilliland R. Surgery for fistula-in-ano: an audit of practice of colorectal and general surgeons. Colorectal Dis 2007; 9: Cirocchi R, Santoro A, Trastulli S et al. Meta-analysis of fibrin glue versus surgery for treatment of fistula-in-ano. Ann Ital Chir 2010; 81: Damon H, Guye O, Seigneurin A et al. Prevalence of anal incontinence in adults and impact on quality-of-life. Gastroenterol Clin Biol 2006; 30: Whiteford MH, Kilkenny J III, Hyman N et al. Practice parameters for the treatment of perianal abscess and fistulain-ano (revised). Dis Colon Rectum 2005; 48: Rothbarth J, Bemelman W, Meijerink W, Stiggelbout A, Buyze-Westerweel M, Delemarre J. What is the impact of fecal incontinence on quality of life? Dis Colon Rectum 2001; 44: Williams JG, MacLeod CA, Rothenberger DA, Goldberg SM. Seton treatment of high anal fistulae. Br J Surg 1991; 78: Wong S, Solomon M, Crowe P, Ooi K. Cure, continence and quality of life after treatment for fistula-in-ano. ANZ J Surg 2008; 78: Seneviratne SA, Samarasekera DN, Kotalawala W. Quality of life following surgery for recurrent fistula-in-ano. Tech Coloproctol 2009; 13: Garcia-Aguilar J, Davey CS, Le CT, Lowry AC, Rothenberger DA. Patient satisfaction after surgical treatment for fistula-in-ano. Dis Colon Rectum 2000; 43: Nicholls J. Anal fistula. Colorectal Dis 2012; 14: Ellis CN. Sphincter-preserving fistula management: what patients want. Dis Colon Rectum 2010; 53: Colorectal Disease ª 2015 The Association of Coloproctology of Great Britain and Ireland. 18,