Introduction

Anal fistula (AF), one of the most common causes of proctological surgery [1], is a condition that can have impact on patients’ anorectal function and quality of life (QoL) [2].

The classification of AFs into “simple” or “complex” has the greatest practical and surgical significance. Usually, the majority of simple AFs are considered to have “low” tracts. However, the definition of low fistula has changed over time, with a trend towards lowering the percentage of the external anal sphincter (EAS) crossed by the fistula tract [3]. According to several guidelines [4,5,6], an AF is defined “simple” when the tract is intersphincteric, or low transsphincteric (crossing < 30% of the EAS). Instead, AFs are defined as complex in cases of––high transphincteric tract (crossing > 30% of the EAS); in patients considered at risk for postoperative fecal incontinence (anterior fistula in women, recurrent fistula, or pre-existent fecal incontinence) even though with low transphincteric tract; suprasphincteric or extrasphincteric tracts; and in AFs with multiple tracts in a horseshoe fashion or those associated with inflammatory bowel disease (IBD), radiation, malignancy, tuberculosis, or chronic diarrhea [4,5,6].

Surgical treatment of AFs is therefore usually based on the amount of the sphincters involved, and, based on this concept, anal fistulotomy is considered the gold standard to treat simple AFs. This procedure, however, may have side effects such as deformities and esthetic alterations [7], together with detrimental effects on continence and on patient satisfaction [2, 4,5,6,7,8].

For these reasons, several minimally invasive techniques have been developed, even if their adoption (mainly in simple AFs) is limited by a higher failure rate. They also tend to be more expensive and are rarely used in real practice scenarios [9]. Reflecting this, guidelines do not offer specific indications regarding the clinical application of these techniques in simple AFs [4,5,6].

The aim of this systematic review was to assess the surgical treatment of simple AFs by sphincter-cutting and sphincter-sparing techniques, and specifically—(1) peri-operative features and morbidity, (2) clinical results in terms of efficacy, (3) the risk of postoperative continence impairment and impact of surgery on patients’ QoL.

Materials and methods

Literature review and eligibility criteria

This review was carried out according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses Statement (PRISMA) guidelines [10]. A literature research was performed using PubMed, Cochrane, and Google Scholar. “Simple anal fistula”, “low anal fistula”, “intersphincteric fistula”, “low transphincteric fistula”, “fistulotomy” were the search terms used. Studies were included if they provided any number of cases analyzing any surgical treatment for simple AFs as defined by commonly adopted guidelines [4,5,6]. Prospective, retrospective, observational studies, and randomized clinical trials were included, while reviews, meta-analyses, trial proposals, thesis articles, technical notes, commentaries, letters, and meeting abstracts were excluded. The time range covered was 1990–April 2020, and only articles written in English were selected. Additional articles responding to the inclusion criteria were extrapolated from the bibliography of relevant material via backward citation tracking.

All articles concerning complex, recto-vaginal or ano-vaginal, tuberculosis- and IBD-related AFs were excluded, as well as any study where data on simple AFs could not be extrapolated.

Database research was performed by three authors individually (FL, AP, LF) and the results were then discussed and merged by a working group. Article inclusion, when in doubt, was decided on a per-case basis after discussion.

Data extraction

Data from eligible literature was thus extracted and inserted in tables using SPSS® version 21.0 for Windows® software (SPSS, Chicago, IL, USA), including publication data (author, year of publication, study type), type of intervention, characteristics of participants (number of patients, mean age, male-to-female ratio), perioperative details, and other outcomes (operating time, hospital stay, mean healing time, complications, recurrence, and/or success rates, continence impairment, pre- and postoperative anorectal manometry, QoL scores).

Data extraction was performed by two reviewers (AP, LF) and independently assessed by another (FL) for completeness and accuracy. Surgical procedures were summarized as sphincter-cutting procedures (fistulotomy, fistulectomy, and cutting setons) or sphincter-sparing techniques [glues/pastes, laser, flap, ligation of intersphinctericv fistula tract (LIFT), etc.].

Risk of bias assessment

A risk of bias and quality assessment was performed for each article. For randomized studies, the Jadad scale was used (1–5 points, 1 = poor and 5 = excellent) [11], while for non-randomized studies, the Methodological Index for Non-Randomized Studies (MINORS) Scale for comparative (0–24 points, 0 = poor and 24 = excellent) or non-comparative (0–16 points, 0 = poor and 16 = excellent) studies was applied [12].

Data reporting and statistical analysis

Descriptive statistics have been reported as absolute frequencies and percentages for qualitative data; quantitative variables have been described as mean value (standard deviation) or median (range), based on availability. For means, the weighted averages were calculated as follows: (single study average × study cohort size) 1, 2, …, n/pooled cohort size. This was done to minimize the effect of the different cohort sizes of the studies on the calculated averages and to provide an overall value for the outcome measures evaluated.

Results

Study selection and risk of bias

The search returned 456 records of interest. After removal of duplicated records, 437 were screened; after title and abstract evaluation, 343 were excluded according to the inclusion criteria. Finally, 94 full text articles were assessed for eligibility; however, 28 of them were excluded, mainly because of the impossibility of isolating data on patients affected by simple AFs from mixed case reports. Therefore, a total of 66 articles [2, 8, 13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76] were found to be eligible (Fig. 1). The publication dates of the articles range from 1994 to 2020. Among the included studies, 28 were prospective studies [21,22,23,24,25, 29, 38,39,40, 42, 44, 48, 51, 53, 56, 58,59,60,61,62, 64, 65, 68, 70, 71, 74,75,76], 19 were retrospective series [2, 8, 13,14,15, 18, 26,27,28, 33, 43, 45, 57, 63, 66, 67, 69, 72, 73], and 19 were randomized clinical trials (RCT) [16, 17, 19, 20, 30,31,32, 34,35,36,37, 41, 46, 47, 49, 50, 52, 54, 55] (Tables 1–2). The quality of the studies was generally low with a consistent risk of bias; the median score of the Jadad Scale for RCT was 3 (1–5), and only 2 studies had the highest possible score [52, 55]; the median MINORS score for non-comparative studies was 12 (3–16), with only one study that could be regarded as excellent [44], while the median MINORS score for comparative studies was 17 (8–21) (Tables 1, 2). Risk of bias of the selected studies could be attributed mainly to a retrospective design, difficulty or impossibility of patients’ and operators’ blinding, small sample size, short follow-up, heterogeneity of the analyzed variables, absence of uniform definition of the main outcomes (success rate, continence impairment).

Fig. 1
figure 1

PRISMA flow diagram of the included studies

Full size image
Table 1 Patient and study characteristics—sphincter-cutting procedures
Full size table
Table 2 Patient and study characteristics—sphincter-sparing techniques
Full size table

Patient characteristics and surgical procedures

A total of 4883 patients (weighted average age: 41.3 years; M:F ratio 6:1) underwent a sphincter-cutting procedure, which was usually fistulotomy or fistulectomy (Table 1). Main technical variations reported were marsupialization [16, 30, 32, 36, 40, 52] or primary sphincteroplasty [19, 35, 47, 56, 57].

Sphincter-sparing techniques were adopted to treat 602 patients (weighted average age: 36.2 years; M:F ratio 4:1) with a simple AF (Table 2). Among those, glues/pastes (fibrin glue, Permacol® collagen paste, and cyanoacrylate glue) were the most frequently analyzed procedures with ten records [20, 28, 58, 60, 61, 64, 65, 68, 71, 74]. LIFT and the laser closure of fistula tracts were reported in five [54, 62, 66, 69, 76] and four studies [59, 67, 70, 72], respectively. Other procedures adopted are detailed in Table 2.

Perioperative details

When reported, the weighted average duration of the sphincter-cutting procedures was 21.9 (8.0–43.0) minutes, and the weighted average duration of hospital stay was 3.1 (0–13.0) days. The weighted average healing time was 41.0 (8.0–183.0) days (Table 3). The most frequent complication reported was wound infection (123 cases, 6%), followed by bleeding (53 cases, 2.9%) and urinary retention (40 cases, 2.6%) (Table 3).

Table 3 Perioperative details—sphincter-cutting procedures
Full size table

The overall weighted average operation time of sphincter-sparing procedures was 34.5 (19.0–52.5) minutes, with a weighted average postoperative hospital stay of 0.8 (0–1.5) days. Only 3 studies reported healing time [50, 59, 62]; the weighted average was 15.1 (7.7–28.0) days (Table 4). The morbidity rate was very low, with a total of 6 complications registered (Table 4).

Table 4 Perioperative details—sphincter-sparing techniques
Full size table

Success rate and continence status

After a weighted average follow-up of 14.7 (1–77) months, the weighted mean success rate after a sphincter-cutting procedure was 93.7% (61.0–100%), while any postoperative continence impairment was reported in 12.7% of patients (0–45.7%) (Table 5).

Table 5 Results—sphincter-cutting procedures
Full size table

Overall, sphincter-sparing techniques reached a weighted average success rate of 77.7% (25.0–100%) after a weighted average follow-up of 13.2 (2.3–71.0) months. No study reported any postoperative continence deterioration, with the exception of a retrospective study reporting minor incontinence in 1 out of 9 patients (11.1%) with a simple AF and treated with a sphincter-saving technique [28] (Table 6).

Table 6 Results—sphincter-sparing procedures
Full size table

Only four studies reported anorectal manometry data in patients affected by simple AFs—unfortunately, differences in the instruments and units of measurement adopted (mmHg, cmH2O or kPa) made it impossible to pool the manometric results. In 3 studies, resting and squeeze pressures did not change [17, 20, 35], while a prospective study reported a significant reduction of postoperative resting and squeeze pressures [21].

QoL and patient satisfaction

Seven studies [2, 38, 41, 44, 51, 71, 76] evaluated the effects of surgery on patients’ QoL and satisfaction, even if the data for simple AFs could not be extrapolated for two of them [51, 71]. The postoperative onset of fecal incontinence reduced patients’ QoL in a retrospective series [2], while it had no significant effect in another prospective study [44]; the recurrence of the disease had a negative impact on QoL in a recent prospective study [76]; finally, two reports [38, 41] stated that patient satisfaction after surgery for a simple AF was high or very high in 86.4% and 90.6% of patients, respectively.

Discussion

Surgical treatment of simple AFs is usually considered “simple” by definition. However, over time, the definition of “simple” AFs has led to a reduction in the percentage of the sphincters that is involved, mainly due to the feared risk of postoperative continence impairment [3]. Moreover, it must be considered that the lack of an univocal definition of "simple" fistula can make it difficult to pool the results of the different studies available. However, the selection of studies in this review was performed considering the definition of “simple” fistula provided by the most important international guidelines [4,5,6]. Only a few of the studies analyzed reported the adoption of imaging techniques (magnetic resonance imaging or endoanal ultrasound) to define the diagnosis, although this probably reflects the infrequent use of these techniques in this kind of anal fistula.

From this systematic review, it emerged that fistulotomy/fistulectomy is by far the most suitable surgical intervention to treat simple AFs. The use of these procedures was constant over the years, and they provide a very high overall success rate (Table 5). However, it should be noted that many of the studies analyzed had a short follow-up, and the healing rate seems to decrease in some studies with a long follow-up. Van der Hagen et al. [24] stated that the recurrence rate after fistulotomy for low AFs gradually increased over time, being 7%, 16%, and 39% after 12, 24, and 72 months of follow-up, respectively. The same study underlined that in more than half of the cases, the recurrence occurred in a different location from the previous fistula tract. Therefore, the authors hypothesized that the recurrence in patients with a simple AF was "more likely a matter of patient disease than a failure of the treatment" [24]. Another large retrospective series on 537 patients showed that the healing rate at the 5-year follow-up was about 83% [8], while a study by van Koperen et al. failed to identify a significant risk factor for fistula recurrence [26].

Concerning the sphincter-sparing procedures, the pooled healing rate calculated in this review was 77.7% (Table 6); however, it should be considered that small sample sizes, short follow-up, and the heterogeneity of the evaluated procedures do not allow us to draw definitive conclusions. LIFT is an attractive recently developed procedure which has proven to be effective (91% success rate) also in a study on simple AFs with a long mean follow-up (71 months) [69]. However, other investigations on this topic are needed to really assess the efficacy of this procedure specifically to treat simple AFs.

Our study confirmed that fistulotomy is a simple and quick procedure (mean operation time: 21 min), with an acceptable mean wound healing time (41 days), although a certain amount of morbidity has been registered (Table 3). In contrast, sphincter-sparing techniques are sometimes more technically demanding and therefore operating time can be longer, even if postoperative complications are almost absent (Table 4). Several new sphincter-sparing techniques have been developed, mainly to reduce the most feared complication of fistulotomy, which is postoperative fecal incontinence. In fact, even this systematic review has confirmed that this complication is almost absent if one of the above-mentioned procedures is used (Table 6). On the other hand, techniques that are not sphincter-sparing could cause continence impairment in approximately 13% of patients with simple AFs (Table 5). Moreover, studies with longer follow-up showed that the incontinence rate after fistulotomy could be higher [2, 8, 21, 44], much more than expected for such a "simple" operation. A study by Visscher et al. [2] reported about 25% of continence impairment (mainly minor) in simple AF patients, with a significant reduction of QoL. Similarly, a high postoperative incontinence rate (about 45%) emerged from a study on 537 patients [8].

There is still debate about how to reduce or contain the risk of fecal incontinence even in simple AFs. Some studies suggest using preoperative anorectal manometry to evaluate baseline anorectal function. Chang and Lin [21] analyzed 45 patients with low intersphincteric fistula with anorectal manometry performed at baseline and at least 6 months after surgery. They found that maximum anal resting pressure significantly decreased, and a lower preoperative anal resting pressure was the only independent predictive factor of postoperative continence disturbances. Similarly, Toyonaga et al. [77] recommend avoiding a fistulotomy in patients with intersphincteric fistulas and with a preoperative low anal squeeze pressure at the anorectal manometry. Therefore, a sphincter-sparing procedure in this kind of patient could be advisable. A technical variation of lay-open fistulotomy to reduce postoperative fecal incontinence seems to be immediate sphincter reconstruction, both for simple and complex fistulas [78].

Some issues regarding the length of hospital stay (3.1 days in the fistulotomy group and 0.8 in the "sphincter-sparing" group) must be considered—the "sphincter-cutting" group includes a relevant number of studies that were carried out decades ago (1990–2000), when protocols on shorter hospital stay were not yet fully implemented. Furthermore, the series with a longer hospital stay were often conducted in specific geographic areas (mainly Asia). We could therefore hypothesize that in both cases, the length of stay was due to hospital requirements. However, we believe that today, in most centers, it is possible to perform both "sphincter-cutting" and "sphincter-sparing" procedures in a day-hospital setting.

Strengths and limitations

This systematic review pooled a large number of patients undergoing surgery for simple AF and analyzed clinically relevant outcomes of different kinds of procedures. As far as we could gather from the literature, an analysis of this size regarding the treatment of simple AFs has never been carried out. The interrogation of multiple search databases allowed us to collect articles from countries with different ranges of income and cultural impact of the disease providing an extensive coverage of both population and procedure types.

The implementation of quality assessment allowed more accurate quantification of selection bias and partially moderated the inhomogeneity of reports.

Nonetheless, a number of limitations of the articles included in this systematic review have surfaced. A large number of studies are low quality and many report a small sample size and/or short follow-up. Additionally, a substantial heterogeneity of the examined procedures, concerning mainly sphincter-sparing techniques was observed. Finally, it must be emphasized that in many of the studies analyzed the continence impairment assessment was performed without the adoption of validated incontinence scores.

Conclusions

Surgical treatment of simple AFs by sphincter-cutting procedures provides excellent cure rates, even if a certain morbidity should be expected. Postoperative continence impairment is not a negligible risk, which could have a detrimental effect on both patients’ QoL and satisfaction. The adoption of sphincter-sparing procedures could be useful in selected patients, and this should be better evaluated in future prospective studies with adequately long follow-up.