Table 2 Results of the complex perianal fistulas of cryptoglandular origin analysis

Minimal invasive techniques

 De la Portilla 2019 [31]

Double-blind randomized parallel groups phase III study

56

Platelet-rich plasma vs. fibrin plug

1+; 2.108

• 32 patients with platelet-rich plasma and 24 with fibrin plug were treated

• The healing rate was 71% in platelet-rich plasma versus 58.3% in patients with fibrin plug (p = 0.608)

• The complete healing rate was 48.4% versus 41.7%, respectively; partial healing rate was 22.6% versus 16.7%, respectively

• Patients receiving platelet-rich plasma had increased pain reduction, (p = 0.023) which caused an impact on improving quality of life in patients receiving platelet-rich plasma

• Data from abstract congress, paper published in: Stem Cells Transl Med. 2020;9(3):295–301

 Dozois 2019 [32]

Open phase I clinical trial

15

Mesenchymal stem cell-coated fistula plug derived from adipose tissue

3; 3.991

• Treatment-related short-term side effects were observed in 3 patients

• At 6 months, 3 patients had complete healing, 8 partial healing and 4 showed no clinical improvement

• Radiographic improvement was observed in 11 out of 15 patients. In 8 of these patients the fistula tract was smaller and reduced compared to baseline

 Di Visconte 2018 [33]

Retrospective analysis

31

Injection of porcine collagen paste versus rectal feed with flap

2+; 2.108

• Five patients (16%) in the rectal feed group and one (5%) in the pig collagen group experienced fecal incontinence after the operation

• Disease-free survival at age 2 was 65% in the rectal feed group and 52% in the pig collagen group (p = 0.659)

• Average satisfaction scores were 5 (range 1–10) in the rectal feed group and 7 (range 2–10) in the pig collagen group (p = 0.299)

 Marinello 2018 [34]

Retrospective observational study

10

OTSC device® (over-the-scope-clip)

3; 1.323

• Nine fistulas were of cryptoglandular origin and one patient had CD controlled with thiopurines

• 80% of patients had previous anal surgeries

• The healing rate of the fistula was 60% with an average follow-up of 15 months (range: 6–26 months)

• Three patients had clinical recurrence and one patient required removal of the clip for invalidating pain

• There was no appearance of fecal incontinence

 Schniewind 2018 [35]

Pilot study based on prospective case series

7

Endoluminal vacuum therapy (VAC) with polyuretan sponge

3; 3.991

• Four patients had fistula of cryptoglandular origin and 3 related to CD

• All patients tolerated therapy well and no AE were observed

• The closure of the fistula tract occurred within 4 weeks of the completion of vacuum therapy

• A patient with cryptoglandular fistula developed a recurrence within the 3-month follow-up

 De la Portilla 2017 [36]

Phase II clinical trial

36

Autologous platelet growth factors

3; 2.571

• 7 AE related to injected product or surgical procedure was detected in 4/36 patients

• 33.3% of patients (12/36) achieved complete fistula healing and 11.1% of patients (4/36) had partial healing

• In successfully healing patients, a reduction in gradual pain was detected through an AVS, from 5.625 to 0.125 after 1 year, p = 0.0438

• A significant improvement in Wexner score was observed in patients achieving total or partial fistula healing, from 3.0625 to 1.125 in a year, p = 0.0195

 Choi 2017 [37]

Phase I clinical trial

15

Stem cells derived from adipose tissue, two-dose comparison

2+; 2.721

• Nine out of 13 patients had fistula closure, 69.2%

• Six days follow-up patients kept closing at 6 months, 83.3%

• No AE were reported

 Nordholm 2017 [38]

Retrospective unique cohort study

35

Nitinol clip

3; 3.991

• The healing rate of fistula 1 year after the procedure was 54.3% (19 out of 35 patients)

• At the end of follow-up, 49% (17 out of 35) of patients had persistent closure of the fistula

• No deterioration of continence was observed

• The result of treatment was not found to be statistically associated with any clinical-pathological characteristics

 Giordano 2016 [39]

Prospective, multicenter observational study

30

Collagen permacol paste

3; 1.095

• Of the 28 patients with available data, 15 patients (54%) presented a fully curated fistula at 6 and 12 months

• More than 60% of patients were satisfied or very satisfied with the intervention

 Ratto 2016 [40]

Estudio observacional prospectivo

10

Dispositivo Curaseal AF

3; 3.665

• Two months after the intervention, healing and absence of discharge or abscess formation was achieved in 5 patients (50%)

• The final percentage of success was 70%, 7/10

 Prosst 2015 [41]

Pilot study

20

OTSC device® (over-the-scope-clip)

3; 1.095

• Within 6 months of surgery, 18 patients (90%) were considered cured because they had no clinical signs or symptoms of fistula

• In 13 of the 18 patients (72%), the clip caused no problems, while in 3 patients the clip spontaneously detached. In the remaining 2 patients the clip was removed due to discomfort and late wound healing

• There were no AE related to the clip, such as necrosis or ischemia pressure ulcers

• No case of fecal incontinence was observed during follow-up

 Borowski 2015 [42]

Prospective one-arm study

7

Treatment with stem cells derived from autologous adipose tissue

3; 1.33

• 71.4% of patients showed signs of fistula closure

• 57.1% had complete fistula closure at 46 months (average) follow-up

• There were no AE associated with the technique

• There were no new cases of incontinence

 Stamos 2015 [43]

Prospective observational study

93

Biosabsorbable synthetic anal fistula plug

3; 4.087

• Fistula healing rates at 6 and 12 months were 41% (95% CI 30–52%; total, n = 74) and 49% (95% CI 38–61%; total n = 73), respectively. Half of the patients in which prior treatment failed achieved healing

• At 6 months, Wexner’s average score improved significantly (p = 0.0003)

• At 12 months, 93% of patients had minimal or no pain

• AE included 11 infections/abscesses, 2 new fistulas and 8 total and 5 partial plug extrusions

• Fistula healed in 3 patients with partial extrusion

 Ozturk 2014 [44]

Pilot study

10

Autologous cartilage plug

3; 2.721

• Nine patients had cryptoglandular abscess, and one patient had Crohn’s disease

• The average follow-up time was 24 months (range 10–32 months). Of the ten patients, nine were treated for fistula without complications in the short term

• Two late recurrences were observed among the nine patients with successful operations

 Tan 2013 [45]

Retrospective observational study

30 anal fistula plugs in 26 patients

Anal fistula plug

3; 1.095

• Twenty-nine (96.7%) fistulas had a previously inserted seton with an average duration of 12 weeks before the anal fistula stopper procedure

• After a follow-up of 59 weeks (median) (range 13–97) 26 (86.7%) fistulas resorted

• Of these 26 cases, the time to failure was 8 weeks (range 2–54). Surgical interventions were needed in 20 of the 26 cases

 Almeida 2013 [46]

Retrospective observational study

51 (41 with complete follow up)

Biological anal fistula plug

3; –

• Twenty-three (56.1%) patients had complete healing while 18 (43.9%) patients failed the fistula plug procedure during the follow up period of 12 months

 Heydari 2013 [47]

Retrospective review

48 patients with 49 fistulas

Fistula plug made with biosorbable polymers

3; 3.991

• The overall healing rate in the series was 69.3% (34/49 fistulas, 33/48 patients)

• Eight patients (24.2%) had fistula healing within 3 months after surgery, 21 patients (63.6%) 6 months and 4 patients (12.1%) 12 months

• At 3 months, no patient had perineal pain or fecal incontinence

 Herreros 2012 [48]

Randomized, simple blind clinical trial

200

Treatment with stem cells derived from autologous adipose tissue

1+; 3.991

• Group A: 64 patients; 20 million stem cells; Group B: 60 patients, 20 million stem cells derived from adipose tissue plus fibrin glue; Group C: 59 patients; fibrin glue

• The intention to treat population comprised 183 patients out of 200. Of these, 165 (90.2%) completed the study

• The healing rates of fistula at week 12 were 26.56%, 38.33% and 15.25% in arms A, B and C, respectively (p = 0.01)

• A total of 61.75% (n = 113) patients received 2 doses of treatment. The healing of the fistula after the second dose was 39.1% (n = 25), 43.30% (n = 26) and 37.29% (n = 22) (p = 0.79). The healing time was similar in all groups

• The proportion of patients with fistula healed at week 12 who experienced reopening at week 24 to 26 was 25.00%, 14.29% and 11.11% (p > 0,5). These patients did not receive a second dose of treatment

• No statistically significant differences were found when comparing SF-36 (mental and physical health domains) at baseline and at 24 to 26 weeks

• Most patients suffered at least 1 AE (proctalgia, pain, perianal abscess, itching, swelling); 59 AE (90.8%), 49 (84.5%) 51 AE (85.0%) groups A, B and C, respectively (p = 0.51)

• Each year, the healing rates were 57.1%, 52.4% and 37.3% in groups A, B and C, respectively (p = 0.13)

 Ommer 2012 [49]

Retrospective observational study

40

Synthetic anal fistula cap (Gore BioA Fistula Plug®)

3; 1.17

• Six months after surgery, fistula healed in 20 patients (50%). Three fistulas healed after 7, 9 and 12 months

• The overall healing rate was 57.5%, dependent on the number of previous interventions

• In patients having only drainage of the abscess, success occurred in 63.6% (14/22) whereas in patients after one or more flap fistula reconstruction, the healing rate decreased slightly to 50% (9/18)

 Cintron 2012 [50]

Prospective follow-up study

72, 11 with CD

Inserting a swine intestinal submucosa plug

3; 2.635

• There was no difference in closing rates between primary and recurring fistula, 38% and 40%, respectively

• The overall patient success rate was 38%

• Patients with Crohn’s disease had a 50% closure (4/8)

• There were no intraoperative complications

• Four postoperative abscesses (4/73; 5) were reported

 Wilhem 2011 [51]

Pilot study

11

Use of the FiLaCLaser ® in fistula repair

3; 2.721

• Nine out of eleven patients had primary healing (81.8%)

• No side effects were reported during follow-up

 Van der Hagen 2011 [52]

Prospective randomized study

15

Mucous feed cap vs. fibrin sellant

1+; 1.806

• Three (20%) patients in the mucosal feed plug group had a recurrent fistula compared to 9 (60%) of the fibrin sellant group, (p = 0.03)

• No differences were reported between groups in quality of life or incontinence score

 Van der Hagen 2011 [53]

Pilot study

10

Platelet-rich autologous plasma

3; 1.095

• One patient (10%) presented a recurring fistula after 12 months

• Patients did not develop de novo incontinence

• No adverse events were reported

• Platelet-rich plasma may be a valid alternative in patients with cryptoglandular fistula

 El-Gazzaz 2010 [54]

Retrospective review of cases

33

Biologically absorbable anal fistula plug

3; 1.095

• Thirty-three patients underwent 49 plug inserts

61% of fistulas were of cryptoglandular origin and 39% related to CD

• Eight out of 32 patients (25%) (1 patient was loss of follow-up) had successful closure on all of their fistulas

• The success rate after the first intervention was 8/33 patients (24.2%)

• Fourteen patients underwent a second plug insertion, 2 patients (14.3% succeeded in closing the fistula)

• Two patients underwent a third attempt at closure with plug insertion; one had fistula closure

• The success rate for cryptogenic fistulas was 34.6% (9/26) vs 9.1% (2–22) of CD-related fistulas

• Closure was most successful in patients with the placement of a seton prior to the insertion of the plug, 28 vs 4 patients without seton, p = 0.05

• Fistula recurrence was observed in 24 patients (75%) with an average recurrence time of 56 days (range 7–251 days). Two patients had late recurrence at 6 and 8 months after insertion of the plug

 Lenisa 2010 [55]

Prospective observational study

60

Collagen plug

3

• Thirty-eight fistulas were recurrent

• The average operating time was 26 ± 10 min

• There were no serious complications, no sepsis, no cases of mortality

• The average recurrence time was 5.7 ± 1.7 months. Recurrences were observed in 24 patients, resulting in a rate of 60% success per patient

• The average reduction in the score in the continence rating was 0.6, 95% CI (1.3– −0.1); p = 0.01

• The overall success rate of recurrent patients was 72% without impaired continence

 Owen 2010 [56]

Retrospective observational study

32 patients with a total of 35 insertions

Cook Surgisis AFP Plug™

3; 1.355

• The overall healing rate at the end of the follow-up period (15 months; range 2–29 months) was 37% (13 out of 35 patients)

• By fistula type, the healing rate of CD-related fistulas (1 out of 3) was 33% and (11 out of 31), 35%, in cryptoglandular origin fistulas

 Queralto 2010 [57]

Prospective observational study

34

Synthetic glue

3; 1.14

• The healing rate in the first month was 67.6% (23 patients); the fistula was unable to close in 11 patients

• Of the 23 patients with cured fistula, all remained free of recurrence, with no disorders associated to continence during the 34 months of follow-up (median) (range 21–43 months)

 McGee 2010 [58]

Prospective observational study

41 patients with 42 fistula tracts

Anal fistula plug in fistula tracts > 4 cm

3; 3.991

• The complete healing was achieved in 18 out of 42 (43%) fistulas over an average follow-up period of 24.5 months

• There were no AEs or cases of incontinence during postoperative clinical visits

• The successful closure of the fistula was significantly associated with a longer length of the tract; fistulas larger than 4 cm were nearly threefold more likely to heal compared to shorter fistulas

Classic techniques

 El-Said 2019 [59]

Pilot study

32

Modification of the original Park technique extending the internal sphincterotomy

3; 1.841

• Two patients (6.25%) experienced recurrence and 5 (15.6%) developed complications (urinary retention, moderate postoperative bleeding, wound infection and new onset fecal incontinence (1 patient)

• Twenty-eight (87.5%) patients were completely satisfied with the procedure

• Quality of life showed a significant improvement at 6 months. All the physical and mental components of the SF-36 questionnaire showed a significant increase, except in the role of limitations due to emotional problems that showed a non-significant increase (p = 0.7)

 Dadou 2018 [60]

Retrospective case series study in a single center

76

Drainage seton

3; 3.991

• The average time for seton removal was 36.6 weeks (range, 6.0–188.0 weeks)

• The average follow-up was 63 months (range, 7–121 months)

• Fifty-six patients (73.7%) had full resolution of symptoms and 14 (18.4%) had a significant improvement in symptoms without the need for additional surgical treatment

• Six patients (7.9%) had persistent severe symptoms and five (7.1%) had a recurrence after seton removal

 Podetta 2018 [61]

Retrospective observational study

32

Mucosal advancement flap

3; 2.108

• Of the 121 patients (group A) treated with flap, 32 (26.4%) (group B) lodged appeals and required a second procedure

• Group B healing rate was 78.1%. Six patients in this group needed a second surgery with healing in all cases

• The complication rate was 9.4% in group B compared to 8.3% for patients in group A

• A slight continence deficit (Miller score 1, 2 and 4) was detected after the first forward flap in 3 patients. Miller’s score did not change after the next procedure

 Balciscueta 2017 [62]

Retrospective observational study

119

Change in resting pressure after full thickness rectal feed flap

3; 2.125

• Significant decrease in maximum postoperative resting pressure was reported (from 90.6 ± 31.9 to 45.2 ± 20 mmHg; p < 0.001)

• The minimum pressure values did not differ significantly (from 8.2 ± to 18.3 to 23.2 ± 13.5 mmHg; p = 0.1)

• Recurrence rate: 5.9%

• Same percentage of patients with anal continence before and after (76.5%)

 Boenicke 2017 [63]

Prospective observational study

61

Flap technique

3; 2.234

• The independent parameters for the failure of therapy in a multivariate data analysis were history of surgical drainage of abscesses [OR = 8.09, p = 0.048, 95% CI (0.98–64.96)], supraspherical fistula [OR = 6.83, p = 0.032, 95% CI (1.17–6.83)] and BMI [OR = 1.23, p = 0.017, 95% CI (1.03–1.46)]

• Low risk of incontinence

 Emile 2017 [64]

Case control retrospective study

251

Seton placement

2+; 1.84

Recurrences were observed in 26 patients (10.3%) after a mean duration of 12.2 ± 3.9 months after removal of seton

Previous recurrent fistulas (OR = 2.81, p = 0.02); supra-lift extension (OR = 3.19, p = 0.01), previous fistulas (OR = 3.36, p = 0.004) and horseshoe fistula (OR = 5.66, p = 0.009) were the most significant predictors of recurrence

Cases of fecal incontinence were detected as complications in 8 patients (3.2%) and seton infection in 16 patients (6.3%)

The female sex (OR = 15.2, p = 0.0003) and the horseshoe fistula (OR = 8.66, p = 0.01) were significant risk factors for fecal incontinence following the procedure

 Sugrue 2017 [65]

Retrospective review

462

Sphincter-sparing repair

3; 3.991

• 420 sphincter-sparing repairs (44%) resulted in healing and 283 (56%) resulted in non-healing with a median follow up of 9 (range, 1–125) months

• The median time to recurrence of fistula was 3 months (range, 0–75) with 79% and 91% of recurrences observed within 6 and 12 months. Patients treated with dermal feed flap, rectal feed flap or ligature of the intersphynteric tract procedure were less likely to have a recurrence than patients treated with a fistula plug or fibrin glue (p < 0.001)

 Herold 2016 [66]

Prospective and multicenter observational study

60

Synthetic Plug (GORE® BIO-A®)

3; 2.721

• Complete follow-up (12 months) of all the patients was reached

• Mean intervention time was 32 ± 10.2 min and mean duration of the hospital stay was 3.3 ± 1.8 days

• No intraoperative complications were observed in any patient. The healing rate after 4 weeks was 6% (3 out of 54 patients), and after 3 months it was 42% (18 out of 46 patients). The healing rate after 6 months of follow-up did not change and stayed just above 50%

• Treatment had no consequences on continence

• The rate of displacement of the plug was 10% (6 out of 60 patients) at 6 months of the operation

• 34% of patients (16 out of 47) required a new operation

 Seow-en 2016 [67]

Retrospective study of consecutive patients

41

Assisted video treatment

3; 2.721

• Primary healing rate was 70.7% with a median follow-up of 34 months

• Twelve patients resorted or did not heal and underwent a repeated procedure

• The secondary healing rate was 83% with two recurrences

• Overall, the stapling of the internal opening had a 22% recurrence rate, while the anorectal feed plug had a failure rate of 75%

• No recurrence was observed in six cases after using the over-the-scape clip

 Visscher 2016 [68]

Retrospective study

143

Fistulotomy and fistulectomy combined with mucous flap

2+; 1.095

• 27% of patients had recurrence

• The risk factors for recurrence were secondary pathway formation [HR: 2.4; 95% CI (1.2–51), p = 0.016] and previous fistula surgery [HR: 1.2; 95% CI (1.0–4.6), p = 0.041]

 Raslan 2016 [69]

Prospective descriptive study

51

Cutting seton

3; 0.808

• Recurrence rate was 9.8%

• Postoperative incontinence rate was 15.7% for gases and 5.9% for liquid depositions. There was no solid stool incontinence

• The cutting seton is a valid option for complex perianal fistulas, but in patients and previous perianal surgery, other surgical options are recommended

 Rosen 2015 [70]

Retrospective review

121

Cutting seton

3; 1.095

• The median time to healing was 3 months (range 1–18)

• 7.4% of patients required additional surgery but 98% had a complete fistula healing

• The incontinence rate decreased from 19 to 11.6% postoperatively

• Of 23 patients with pre-intervention incontinence, 17 (73.9%) resolved his symptoms

 Soliman 2015 [71]

Case series study

140

Cutting seton

3; 5.238

• Most patients, 111 (79.3%), had cryptoglandular fistulas and 14 patients (10%) had CD-related fistulas

• Of the 111 patients, 81 (73.0%) presented transsphynteric fistulas. After 35 months of follow-up (mean) (range, 2–83 months), 70 transsphynteric fistulas had healed (86.4%), 10 were still being treated (12.3%) and one patient was loss of follow-up (1.2%)

• Six patients developed recurrence (7.4%). Three “true” recurrences (3.7%) and three “de novo” fistulas (3.7%)

• No cases of incontinence were reported

 Lee 2015 [72]

Retrospective review

61

Advancement flap

3; 2.387

• Fifty-three (86.9%) surgeries developed successfully, while in 8 of them (13.1%) the procedure failed. Four of them underwent further surgery

• Of the 53 patients who had a successful procedure, 27 responded on the Wexner scale, 21 patients (77.8%) presented a score of 0 (perfect continence)

 Uribe 2015 [73]

Retrospective observational study

119

Tunneling fistulectomy (core out) and curettage

3; 2.387

• The “Core out” technique was performed in 78 patients (group I) and curettage in 41 (group II)

• The total recurrence rate was 5.88%, 5 of group I (6.4%) and 2 of group II (4.9%), without statistical significance (p = 0.74)

 Gottgens 2015 [74]

Retrospective observational study

537

Fistulotomy

3; 2.387

• 88 patients (16.4%) had recurrence which resulted in a primary healing rate of 83.6%

• Of the 88 patients with recurrence 40 healed, resulting in a secondary healing rate of 90.3%

• Kaplan–Meier’s analysis showed that 1 year healing rate was 0.70, 95% CI (0.33–0.89)

• Major incontinence defined as a Vaizey mean value greater than 6 was reported in 28% of patients; the average score was 4.67 (SD: 4.80)

• Only 26.3% of patients had a perfect continence state (Vaizey = 0)

• The SF-36 survey scores were no different from the general population

• It is not clearly specified whether they are simple or complex perianal fistulas, which can condition the results, the author refers that they are of high type

 Patton 2015 [75]

Retrospective observational study

59

Cutting seton

3; 1.355

• Mean follow-up time was 9.4 years (range 1.7–15.6 years)

• The majority of patients had a single seton (n = 56) and three patients had two setons

• Mean time from seton insertion to the time of follow-up where healing was noted (primary healing) was 17.7 months (median 11 months). Four patients (6.8%) developed recurrent fistulas. Three of the four patients underwent a second cut-off seton treatment, the fourth continuing to be treated

• The primary healing rate was 93% (55 cured) and the secondary rate was 98% (58)

• Seventy-eight percent of patients had normal continence or minor incontinence (St. Mark score 0–6), 13.5% moderate incontinence (score 7–12) and severe incontinence 8.5% (score > 12). 63% of patients had no changes or improved control

• St Mark’s continence scores showed a reverse correlation with FIQL (p < 0.001). Average FIQL scores were high and correlated significantly with continence

• The average patient satisfaction score was 9 out of 10

 Hirschburger 2014 [76]

Retrospective observational study

50

Fistulectomy with primary sphincter reconstruction

3; 2.387

• Fistula healing was obtained in 44 patients (88%) who, moreover, did not develop recurrence

• In 5 patients (10%), fistula healed, but they developed a recurrence during the observation period (average follow-up 22 months). In 1 of these patients (2%) the fistula didn’t close

• The score on the continence scale before and after the operation showed a slight decrease in continence in 3 patients. A patient with 2nd grade incontinence improved

• Pre-existing pain was significantly reduced with the intervention

 Ratto 2013 [77]

Prospective observational study

72

Fistulotomy after primary sphinteroplasty

3; 3.991

• Of the 72 patients, 12 (16.7%) had fistula recurrences and 29 patients (40.3%) required seton drainage following surgery

• Three patients had recurrence

• Eight patients (11.6%) without basal incontinence) reported spotting after defecation

• No factors related to surgical success were located

• Patients with recurrent fistula after previous surgery were five times more likely to be affected in continence

 Van Onkelen 2013 [78]

Case series

14

Reparation combined with abscess drainage

3; 3.991

• Healing was reported in 79% patients

• The 3 patients who were not cured at the first intervention were given a second, third or fourth intervention with 100% healing

• The median Rockwood Fecal Incontinence Severity Index incontinence score after the intervention was 0

 Wallin 2012 [79]

Retrospective review

93

Ligation of the intersphincteric fistula

3; 3.991

• The median follow-up time was 19 months (range, 4–55)

• Thirty-two patients (32%) had a history of previous surgery

• The healing success rate was 40% after the first ligature procedure

• The total success rate after ligature including patients previously treated with ligation of the intersphincteric fistula was 47% (44 out of 93)

• Patients with successful fistula closure reported an average CCFFIS score of 1.0 (± 2.3)

• No predictors were found for the successful closure of the fistula

 Abbas 2011 [80]

Retrospective review

169

Fistulotomy, advancement flap, and fistula plugging

2+; 13.265

• Failure of intervention: 15.6%, 15.6% and 7.3%, in fistulotomy, feed graft and fistula capping respectively

• The plugging had the highest failure rate (83.3%) compared to fistulotomy (10.1%) [OR: 44.3; 95 CI (8.9–221.0), p = 0.001]

• Transphincteric and suprasphinteric high fistulas were incontinence predictors compared to subcutaneous fistulas with OR of 22.9, 95% CI (2.2–242.0), p = 0.009] and 61.5; 95% CI (4.5–844.0), p = 0.002), respectively

• The only predictor of septic complications was plugging compared to fistulotomy [OR: 15.1; 95% CI (2.3–97.7), p = 0.004]

 Mitalas 2010 [81]

Retrospective observational study

278

Seton drainage prior to transanal advancement flap

3; 2.108

• The average healing time was 2.2 months

• In patients without preoperative seton drainage, the healing rate was 63%, whereas the healing rate was 67% in patients who underwent preoperative seton drainage

• The overall healing rate was 64%

• The preoperative drainage seton did not improve the result of repair with forward flap

 Roig 2010 [82]

Retrospective study

146

Endoanal advanced graft (Group A, n = 71) vs. immediate sphincter repair after fistulectomy (Group B, n = 75)

2+; 2.769

• After a mean follow up of 13 months (12–60), fistula persisted or recurred in 13 (18.3%) patients in Group A vs 8 (10.6%) in Group B (p = 0.19). Thirty-one (43.6%) patients in Group A vs 16 (21.3%) in Group B presented postoperative continence disturbances (p < 0.001)

• The average postoperative stay (SD) was 6.9 (2.4) days in Group A versus 5.9 (2.5) in Group B (p = 0.01)

• No changes were observed with the FIQL scale

• Group A patients had a significant reduction in maximum resting pressure after surgery