Comparative analysis of early and late recurrence rates in perianal fistula surgery

Introduction

Perianal fistula (PAF), one of the most harmful anorectal diseases, is defined as the formation of an abnormal, epithelialized tract located between the anal canal, the lower rectum and the perianal skin (1). With an incidence of 3.6%, PAF exhibits a prevalence of 12.3/100,000 in men and 5.6/100,000 in women, making them >2 times as common in men (2). This disease most commonly manifests between the ages of 30-50 (1-3). Despite being considered a benign infection, it leads to marked morbidity rates due to high recurrence rates of up to 50% and its potential to damage the anal sphincter (4).

The underlying pathogenesis typically involves a chronic inflammatory process and the development of acute anal abscesses, often stemming from the infection and obstruction of cryptoglandular structures associated with the anal canal mucosa. Following spontaneous or surgical drainage, the loss of intervening tissue leads to the fistulization of the anorectal canal to the skin (1-3). The drainage of a perianal abscess, in particular, increases the incidence of PAF by up to 30-60%. However, not all fistula formations can be explained by the same mechanism. In patients with recurrent or atypical morphology, systemic diseases with manifestations such as Crohn's disease must also be considered. It has been reported that ~25-30% of Crohn's disease cases develop PAF (3). Furthermore, tuberculosis, radiotherapy, trauma and malignant neoplasms represent other etiological factors that are prerequisites to the formation of fistulas (5).

In clinical practice, fistulas are most commonly described using the Parks classification (1-5), which defines them based on the association between the fistula tract and the internal and external sphincters. In addition, fistulas are categorized into two groups: Simple (superficial or intersphincteric) or complex (transsphincteric, suprasphincteric, extrasphincteric, those with multiple tracts or those associated with Crohn's disease). Practically, intersphincteric fistulas, accounting for between 30-45% of cases, are the most frequently encountered (6).

Patients with PAF typically present with symptoms that notably impair their quality of life, including pruritus, persistent discharge, pain, bleeding, incontinence of flatus or feces and recurrent infections (1,2). Diagnosis is primarily based on careful inspection, digital rectal examination and probing of the fistula tract. Proctosigmoidoscopy provides additional information regarding sphincter anatomy and the localization of the internal opening. For complex or recurrent fistulas, MRI is considered the most effective technique. Other modalities such as endoanal ultrasonography, fistulography and anal manometry may also be conducted (1,2,7).

Fistulectomy represents the most established surgical treatment. In ~70% of cases, intervention to excise the tract or provide fistulotomy drainage is sufficient. However, due to the risk of potential sphincter damage and the development of incontinence, there is a ongoing search for alternative treatments such as seton placement, ligation of the intersphincteric fistula tract (LIFT), video-assisted anal fistula treatment (VAAFT), fibrin glue application, Fistula-tract Laser Closure (FiLaC) and transanal opening of the intersphincteric space (TROPIS) (8).

The success of treatment is contingent upon the elimination of infection, preservation of the sphincter, alleviation of symptoms, closure of the fistula, avoidance of complications such as incontinence and prevention of recurrence. Notably, an unsuccessful treatment alone does not cause recurrence. Factors leading to incomplete treatment can include frequent changes of doctor or hospital, discontinuation of follow-up before complete healing is achieved, social and privacy considerations, habituation to low-tolerance symptoms, loss of hope in the treatment and increasing costs (9). Therefore, the present study primarily aimed to address the long-term healing rates of classical treatment methods, the challenges in diagnosis and follow-up and the issue of missing data due to various causes.

Materials and methods

Study population and design

In the Department of General Surgery, Sağlık Bilimleri Üniversitesi, Bursa Yüksek İhtisas Training and Research Hospital (Bursa, Turkiye) the records of 245 patients (43 female and 202 male) with a mean age of 44 years (range, 20-77 years) who were treated for primary or recurrent anal fistula between 2020 and 2023 were retrospectively analysed. The demographic information of the patients, prior surgical procedures, diagnostic investigations performed and surgical methods applied were recorded. Data from the early postoperative period and follow-ups at 1, 6 and 24 months were obtained from the hospital system, while 24-month follow-up data were acquired by contacting the patients directly. Patients who attended the 24-month outpatient follow-up and presented with no notable complications, active symptoms, discharge, open wounds or MRI evidence of active disease were classified as the recovered group (‘none’). Conversely, patients who were unreachable or noncompliant with scheduled visits, often due to perceived recovery, a change of provider or institution, disregard for mild symptoms or a preference for alternative treatments, were not excluded from the analysis. Instead, these patients were designated as a separate cohort and evaluated independently as the lost to follow-up (LTFU) group. The analysis was conducted in two stages: First, categorization was based on the type of operation performed and subsequently on the fistula type, followed by statistical examinations. Those who remained in the latest follow-up and those who were not LTFU were compared as two separate groups. Patients in the groups were evaluated within the framework of ethical approval number 2024-TBEK 2025/10-04 awarded by the Ethical Committee of Bursa Yüksek İhtisas Training and Research Hospital, and the reasons for loss to follow-up were examined. To analyze the reasons for loss to follow-up in the LTFU group, the follow-up data of all patients at the 1st month and 6th marks were also examined and compared.

Inclusion/exclusion criteria

The study included all patients aged 18 years or older who underwent surgical treatment for perianal fistula, had available 1st- and 6th-month follow-up data, and were contactable for evaluation at the 24th month, irrespective of whether they were classified as healed, unhealed or lost to follow-up. Patients were excluded from the present study if they met the following criteria: i) A history of anorectal malignancy; ii) prior surgery due to severe trauma resulting in disrupted primary anorectal anatomy or; iii) unwillingness to voluntarily participate in the present study for any reason, including those with incomplete data records.

Statistical analysis

Descriptive statistics for continuous variables are presented as the mean ± SD and median (minimum/maximum). Descriptive statistics for categorical variables are presented as frequencies and percentages. Variables collected from the patients were analyzed. For the comparison of categorical variables between groups, Pearson χ² test or Fisher's exact test were conducted. Statistical analysis was performed using SPSS statistics software (version 21; IBM Corp.). α=0.05 was considered to indicate a statistically significant difference.

Results

Patient demographics and baseline characteristics

Of the 245 patients analyzed, 43 (17.60%) were female and 202 (82.40%) were male. A total of 190 patients (77.60%) had no history of prior perianal surgery, while 55 (22.40%) had undergone previous surgery. MRI was performed on 221 patients, with anal fistula determined in 201 (90.95%) and not detected in 20 (9.05%). Intersphincteric fistula was the most common type, identified in 156 patients (63.70%). Endoscopic evaluation was conducted on 229 patients and an internal opening was not identified in 224 (97.82%) of them (Table I). No additional issues or complications were detected in any patient during the early postoperative period.

Table I Frequency distribution of patient demographic characteristics.

Table I

Frequency distribution of patient demographic characteristics.

Variable	n	%
Sex
Female	43	17.60
Male	202	82.40
Anorectal surgery
No history of prior surgery	190	77.60
Anal fissure and haemorrhoid	1	0.40
Anal fistula	38	15.50
Anal polyp	1	0.40
Haemorrhoid	5	2.00
Haemorrhoid and anal fistula	2	0.80
Pilonidal sinus	8	3.30
Pelvic MR
Not obtained	24	9.80
Anal fistula	201	82.00
No anal fistula	20	8.20
Fistula type
Intersphincteric	156	63.70
Superficial	31	12.70
Suprasphincteric	18	7.30
Transsphincteric	40	16.30
Endoscopic examination
Performed, internal opening identified	5	2.00
Performed, internal opening not identified	224	91.40
Not performed	16	6.60
Early complication
No development	245	100.00
1-month follow-up (complaint)
None	165	67.30
Yes	80	32.70
6-month follow up
None	137	55.90
Discharge	10	4.10
Fistulectomy	19	7.80
Incontinence	1	0.40
Re-seton	71	29.00
LTFU	7	2.90
24 month follow-up
None	83	33.90
Discharge	5	2.00
Abscess	2	0.80
Fistulectomy	3	1.20
Re-seton	30	12.20
LTFU	122	49.80
Surgical procedure
Fistulectomy	50	20.40
Seton	186	75.90
Other	9	3.70

[i] Pelvic MR imaging was successfully used in 82.00% of cases during the diagnostic phase, the fact that the internal opening could only be determined in 2% of cases during endoscopic examination reinforces the importance of imaging methods in clinical practice. The majority of the patients participating in the present study (82.40%) were male, revealing a significant difference in the sex distribution of anal fistulas. LTFU, lost to follow-up; MR, magnetic resonance.

Early postoperative outcomes at 1 month

At the 1-month follow-up, no problems or complaints were identified in 165 patients (67.30%), while the remaining 80 patients (32.70%) presented with various complaints. At the 6-month follow-up, 137 patients (55.90%) were followed up without issues, while 7 patients (2.90%) were LTFU for various reasons. Intervention or recurrence was observed in the remaining 101 patients (41.30%). Of these, the most common intervention was the application or renewal of a seton, performed on 71 patients (29.00%). At the 24-month follow-up, 83 patients (33.90%) exhibited uncomplicated healing, with problems being identified in 40 patients (16.20%). Among these, a seton was applied or renewed in 30 patients (12.20%). A total of 122 patients (49.80%) were LTFU for various reasons. When considering only the patients who completed follow-up, 83 out of 123 (67.48%) healed without complications (Table I). When evaluated based on the surgical procedure performed, a seton was applied in 186 patients (75.90%), fistulectomy in 50 patients (20.40%) and different procedures in 9 patients (3.70%).

At the 1-month follow-up, among patients who underwent fistulectomy, 35 (70.00%) healed without complications, while problems or complaints were recorded in 15 (30.00%). Among patients who received a seton, 124 (66.70%) healed without complications, whereas 62 (33.30%) developed various issues.

Mid-term outcomes and interventions at 6 months

At the 6-month follow-up, among the fistulectomy patients, 38 (76.00%) exhibited uncomplicated healing, while 12 (24.00%) experienced problems or recurrence. Among the seton patients, 93 (50.00%) progressed without complications, while the remaining 93 (50.00%) required further intervention or experienced problems. The most common intervention was the reapplication or revision of a seton, performed on 64 out of 186 patients (34.41%).

Long-term outcomes at 24 months

At the 24-month follow-up, the number of patients LTFU in the fistulectomy group increased to 32 (64.00%). Of the patients who continued follow-up, 14 out of 18 (77.78%) achieved healing, while problems or recurrence were detected in the remaining 4 out of 18 (22.22%). In the seton group, 86 patients (46.20%) were LTFU. Among those who were followed, 67 out of 100 (67.00%) exhibited uncomplicated healing, while 33 out of 100 (33.00%) required further interventions. The most frequent intervention was the application or renewal of a seton, performed on 26 out of 100 patients (26.00%; Table II).

Table II Distribution of χ² tests regarding recurrence rate and surgical procedure value at 1, 6 and 24 months for patients participating in the present study.

Table II

Distribution of χ² tests regarding recurrence rate and surgical procedure value at 1, 6 and 24 months for patients participating in the present study.

	Surgical procedure
Variable	Fistulectomy, n (%)	Seton, n (%)	Other, n (%)	P-value
1-month complaint				0.904
Yes	15 (30.00)	62 (33.30)	3 (33.30)
No	35 (70.00)	124 (66.70)	6 (66.70)
6-month complaint				0.001
None	38 (76.00)	93 (50.00)	6 (66.70)
Discharge	2 (4.00)	6 (3.20)	2 (22.20)
Fistulectomy	0 (0.00)	19 (10.20)	0 (0.00)
Incontinence	0 (0.00)	1 (0.50)	0 (0.00)
Re-seton	7 (14.00)	64 (34.40)	0 (0.00)
LTFU	3 (6.00)	3 (1.60)	1 (11.10)
24-month complaint				0.021
None	14 (28.00)	67 (36.00)	2 (22.20)
Discharge	0 (0.00)	5 (2.70)	0 (0.00)
Abscess	1 (2.00)	0 (0.00)	1 (11.10)
Fistulectomy	0 (0.00)	2 (1.10)	1 (11.10)
Re-seton	3 (6.00)	26 (14.00)	1 (11.10)
LTFU	32 (64.00)	86 (46.20)	4 (44.40)

[i] A lack of a significant difference in healing between surgical methods (including fistulectomy and seton) at 1-month post-surgery (P=0.904) suggests that early results are independent of the technique. When the follow-up period reached 6 and 24 months, the effect of the surgical procedure on the results became statistically significant (P=0.001 and P=0.021, respectively), proving the role of the method in long-term success. LTFU, lost to follow-up.

Outcomes stratified by fistula type

For superficial fistulas, 23 fistulectomies and 3 seton placements were conducted; for intersphincteric fistulas, 19 fistulectomies and 136 seton placements were conducted; for suprasphincteric fistulas, 5 fistulectomies and 13 seton placements were conducted; and for transsphincteric fistulas, 3 fistulectomies and 34 seton placements were conducted. At the 1-month follow-up, transsphincteric fistulas demonstrated the most uncomplicated healing rate at 29 out of 40 cases (72.50%), while the highest rate of complaints was observed in suprasphincteric fistulas at 7 out of 18 cases (38.90%). At the 6-month follow-up, superficial fistulas exhibited the highest healing rate at 24 out of 31 cases (77.40%), while transsphincteric fistulas were associated with the most problems at 22 out of 40 cases (55.00%). At the 24-month follow-up, after excluding patients LTFU, intersphincteric fistulas exhibited the most successful healing outcome at 62 out of 84 cases (73.80%), whereas the highest recurrence rate was detected in transsphincteric fistulas at 13 out of 24 cases (54.20%). Of these recurrences, a seton was reapplied or renewed in 9 patients (37.5%) (Table III).

Table III Distribution of χ² tests regarding recurrence rate and fistula type at 1, 6 and 24 months for patients participating in the present study.

Table III

Distribution of χ² tests regarding recurrence rate and fistula type at 1, 6 and 24 months for patients participating in the present study.

	Fistula type
Variable	Intersphincteric, n (%)	Superficial, n (%)	Suprasphincteric, n (%)	Transsphincteric, n (%)	P-value
1-month complaint
Yes	53 (34.00)	9 (29.00)	7 (38.90)	11 (27.50)	0.775
No	103 (66.00)	22 (71.00)	11 (61.10)	29 (72.50)
6-month complaint					0.001
None	86 (55.10)	24 (77.40)	9 (50.00)	18 (45.00)
Discharge	4 (2.60)	2 (6.50)	0 (0.00)	4 (10.00)
Fistulectomy	12 (7.70)	1 (3.20)	3 (16.70)	3 (7.50)
Incontinence	0 (0.00)	0 (0.00)	1 (5.60)	0 (0.00)
Re-seton	51 (32.70)	1 (3.20)	5 (27.80)	14 (35.00)
LTFU	3 (1.90)	3 (9.70)	0 (0.00)	1 (2.50)
24-month complaint					0.016
None	62 (39.70)	7 (22.60)	3 (16.70)	11 (27.50)
Discharge	3 (1.90)	0 (0.00)	0 (0.00)	2 (5.00)
Abscess	0 (0.00)	1 (3.20)	0 (0.00)	1 (2.50)
Fistulectomy	1 (0.60)	1 (3.20)	0 (0.00)	1 (2.50)
Re-seton	18 (11.50)	1 (3.20)	2 (11.10)	9 (22.50)
LTFU	72 (46.20)	21 (67.70)	13 (72.20)	16 (40.00)
24-month complaint					0.055
None	62 (73.80)	7 (70.00)	3 (60.00)	11 (45.80)
Discharge	3 (3.60)	0 (0.00)	0 (0.00)	2 (8.30)
Abscess	0 (0.00)	1 (10.00)	0 (0.00)	1 (4.20)
Fistulectomy	1 (1.20)	1 (10.00)	0 (0.00)	1 (4.20)
Re-seton	18 (21.40)	1 (10.00)	2 (40.00)	9 (37.50)

[i] Despite the anatomical type of fistula not having a decisive effect on the clinical picture at 1 month (P=0.775), this may change in the medium and long term. Intersphincteric fistulas were the most common type in the clinical picture, accounting for 63.70% of the entire group and represented the group with the highest follow-up rate at the end of 24 months. LTFU, lost to follow-up.

Characteristics of patients lost to follow-up

The 122 out of 245 patients (49.80%) LTFU were separated into a different group. It was observed that the female to male ratio was 23/99, close to the main group. A total of 19 patients (15.60%) had a history of prior anorectal surgery. Based on their follow-ups at 6 months, 87 patients (71.31%) had no complaints. A total of 72 patients (59.00%) had intersphincteric fistulas; of these, 5 (6.90%) had received a seton and 9 (12.50%) had undergone fistulectomy, 21 patients (17.20%) had superficial fistulas, with only 1 (4.80%) having undergone fistulectomy. A total of 13 patients (10.70%) had suprasphincteric fistulas, with 1 (7.70%) receiving a seton and 1 (7.70%) undergoing fistulectomy. A total of 16 patients (13.10%) had transsphincteric fistulas, with 3 (18.80%) receiving a seton and 3 (18.80%) undergoing sphincterotomy (Table IV). When looking at the ages, no significant difference was observed between the two groups (Table V). Overall, recurrence (seton + fistulectomy) of LTFU group in month 6 was recorded in 25 out of the 122 patients (20.50%).

Table IV χ² test results exhibiting the relapse rate at 24 months in the participating patients and the distribution of some variables.

Table IV

χ² test results exhibiting the relapse rate at 24 months in the participating patients and the distribution of some variables.

	24-month
Variable	Follow-up, n (%)	LTFU, n (%)	P-value
Sex			0.594a
Female	20 (16.30)	23 (18.90)
Male	103 (83.70)	99 (81.10)
Fistula type			0.019a*
Intersphincteric	84 (68.30)	72 (59.00)
Superficial	10 (8.10)	21 (17.20)
Suprasphincteric	5 (4.10)	13 (10.70)
Transsphincteric	24 (19.50)	16 (13.10)
Surgical procedure			0.082b
Fistulectomy	18 (14.60)	32 (26.20)
Seton	100 (81.30)	86 (70.50)
Other	5 (4.10)	4 (3.30)

[i] Categorical variables are presented as n (%). According to the results, unlike parameters such as age, sex and surgical method, only fistula type was found to have a statistically significant association with the follow-up discontinuation process (P=0.019).

[ii] ^aPearson χ² test and

[iii] ^bFishers exact test. ^*P<0.05. LTFU, lost to follow-up.

Table V Distribution of χ² test results regarding the age values of the groups at the 24th month among patients participating in this study.

Table V

Distribution of χ² test results regarding the age values of the groups at the 24th month among patients participating in this study.

24-month group	n	Mean ± SD	Median (minimum-maximum)	Mann Whitney U	P-value
Age				7439	0.908
Follow-up	123	44.67±12.79	44 (20-75)
LTFU	122	44.85±12.69	44 (20-77)

[i] SD, standard deviation; LTFU, lost to follow-up.

Handling and interpretation of missing data

Regarding the quantitative treatment of missing data resulting from LTFU, it should be noted that the present final results should not be considered conclusive, given that formal missing-data analyses, such as sensitivity analysis or data imputation, were not performed. The present results provide estimated ratios based solely on the available data analyzed.

Discussion

PAF is a chronic and recurrent anorectal disease characterized by a lack of a standard, uniform treatment algorithm, despite its prolonged and often difficult treatment modalities, lengthy clinical course, heterogeneous anatomical variations and a broad spectrum of therapeutic options. The disease has a natural tendency to become chronic and the high post-treatment recurrence rate can complicate the determination of an ideal management strategy (10). Although numerous treatment options such as fistulotomy, seton placement, the LIFT, FiLaC and VAAFT have been described, Sohrabi et al (8) reported notable disparities in their long-term healing rates. Complications arising from these methods, such as anal incontinence and the issue of recurrence remain subjects of ongoing debate in the literature.

A marked number of studies and meta-analyses exist in the literature concerning PAF, however, there is a notable scarcity of large-scale, international, prospectively designed randomized controlled trials that compare short- and long-term clinical outcomes. A primary reasons for this is hypothesized to be low patient recruitment rates and the frequent exclusion of cases that are not enrolled in studies from subsequent analysis. In the present study, rather than excluding patients LTFU from the analysis, they were categorized as a distinct group and reasons for their attrition were investigated.

In the present study, the male-to-female ratio was greater than 4:1. While the literature typically reports this ratio as 2-3 times higher in males, the notably lower proportion of women in the present cohort may be attributable to privacy concerns, which constitute a marked cultural factor in Turkey and lead to serious health consequences. Hoffman et al (11) reported that up to 17% of colorectal surgical patients request a sex-specific appointment, with 70% of these requests coming from women. The primary reason for the lower proportion of female patients in the present study is likely the scarcity of female surgeons in the clinic, which may create a barrier resulting in delayed reporting of perianal complaints or reduced healthcare-seeking behavior among women. However, prospective controlled studies in the literature that can establish causality on a regional basis for this issue are limited.

The recurrence rate of anorectal fistulas is high. It is clinically important to note that 22.4% of the patients included in the present study had undergone previous anorectal surgical interventions for various reasons. The fact that the majority of prior interventions were performed at external centers highlights the potential influence of systemic and individual factors on treatment outcomes. This cannot be explained solely by surgical technical proficiency; rather, it is hypothesized to involve multidimensional factors such as accessibility to healthcare, lack of standardization in patient education processes, levels of patient satisfaction and compliance and socioeconomic factors (11,12). Furthermore, insufficiently structured and clear communication between physicians and patients regarding treatment options and potential outcomes, combined with the limited digital literacy of patients and potential conceptual confusion regarding their condition, can lead to a discrepancy between treatment expectations and realistic clinical outcomes (13). For these and similar reasons, while they may contribute to high recurrence rates, as observed in the present study, it is believed that repeated consultations for the same condition are gradually decreasing.

MRI is an established technique for the preoperative evaluation of PAF. It can provide valuable information, particularly in complex, recurrent and Crohn's-associated fistulas and can facilitate surgical planning. However, while its routine necessity in simple primary fistulas remains debated and it possesses certain technical and practical limitations (14), a study by Garg et al (15) showed that even fistulas appearing simple are complex on MRI in 34.6% of cases. In the present study, the preoperative assessment success rate was found to be 91%, indicating that patients were thoroughly evaluated prior to surgery. This high rate reflects the adoption of a standardized imaging protocol and a multidisciplinary approach. In the diagnosis and differential diagnosis of PAF, MRI, owing to its multiplanar imaging capability and high soft-tissue contrast, can demonstrate with high accuracy the association with the anoderm, secondary tracts, sphincter involvement and the presence of abscesses (16). It is reported to be more useful compared with endoanal ultrasound, particularly in suprasphincteric and complex fistulas and is especially preferred in patients with Crohn's disease (3,17).

The disadvantages and limited use of MRI in the diagnosis and treatment planning of PAF stem from its cost and the fact that it may not be performed with a standardized protocol at every center (14). Particularly in cases requiring an anal coil, imaging artifacts can complicate MRI interpretation, potentially leading to misdiagnosis and incorrect treatment planning (18). In the present study, as the majority of pre- and postoperative MRIs were reported by a single center and a single radiologist, it is believed this consistency markedly contributed to treatment planning. Studies have suggested that the use of MRI may contribute more to planning and local success rather than having a notable impact on overall surgical healing (15-17). In a study by Buchanan et al (19), the limited therapeutic impact of MRI is discussed, reporting its contribution to therapeutic decision-making to be ~10%.

Despite utilizing rectosigmoidoscopy (RSS) in the present diagnostic management, an internal fistula opening was not identified in >90% of cases. A marked factor contributing to the wide variability of this rate in the literature is the practice of diagnosing a fistula based on the reporting of an external opening, even though diagnosis is possible through inspection alone. RSS may not provide notable added value for evaluating the internal opening (20). The identification of the internal opening in PAF can be a key factor in recurrences. For instance, in analyses associated with the TROPIS method, the internal opening was visualized in only 30% of cases. This indicates that in >70% of the cases, detection via RSS would not have been feasible. Given its importance in recurrence and the lower detection rate even with the TROPIS method, RSS is primarily used to detect other rectal pathologies rather than for the definitive diagnosis of PAF (21).

The average hospitalization for patients in the present study was 1 day, with almost all procedures performed under spinal anesthesia and no early complaints or pain reported. Ternent et al (22) indicated that simple anorectal procedures can be safely performed in an ambulatory setting, with 1-year recurrence rates comparable to those of in-hospital applications. Research directly investigating the association between factors such as postoperative pain control and early mobilization with recurrence appears to be limited. However, upon examining cases of recurrence, it becomes evident that disease-specific factors serve a more prominent role, with diabetes, fistula complexity and a prior history of abscess being the most notable among these factors (23). As expected, intersphincteric fistulas were the most prevalent type, ranking first in the present analysis. While rapid and high rates of healing are anticipated in the short-term follow-up, long-term monitoring reveals that recurrences are prominent within this group and they also constitute the majority of patients LTFU.

Issues and recurrences begin to emerge at the 1-month mark, with the present study identifying this rate at 33%. During this period, nearly all patients adhere to their follow-up appointments. The manifestation of recurrence in the early postoperative phase may stem from technical reasons, incomplete diagnosis or patient-associated factors. However, inadequate preoperative evaluation, where endoanal ultrasound and MRI may fail to detect complex fistulas or possible secondary tracts, can lead to incomplete surgery and early recurrence (24). A number of studies have reported more frequent recurrences in cases with insufficient preoperative imaging (24-26).

At month 1, there was found to be no significant difference between seton placement and fistulectomy. A seton generally facilitates staged healing through drainage, whereas fistulectomy is a more radical approach involving tissue excision. Both methods provide similar results in terms of early symptom resolution; however, they do not offer definitive information regarding permanent cure or recurrence risk. Long-term follow-up results (>6-12 months) are required to assess true healing and recurrence rates. Andreou et al (27) reported that early outcomes (<1-3 months), often evaluated based on pain and early wound healing, show similarities or no differences, serving as short-term indicators. By contrast, outcomes such as incontinence, recurrence or successful treatment have been reported to differ notably in the long term (>6-12 months).

At 6 months, true early recurrence or prolonged treatment was observed. While prolonged seton treatment was a primary reason, accounting for 32% of cases, the presence of a seton should not always be perceived as treatment failure. By contrast, studies often consider the seton as a planned part of the therapeutic pathway (28,29).

In complex and recurrent cases at 6 months, some seton applications may prove insufficient. Although numerous studies (23,30) have used the 6-month mark for early recurrence or healing assessment, recurrences can also be observed in the long term (12-24 months). Therefore, 6 months alone may not be a sufficiently long period to serve as a definitive predictor of ultimate healing or recurrence (27). Reasons for prolonged seton placement may include persistent local abscess, awaiting control of Crohn's disease with medical therapy, a planned staged surgical strategy and, in some cases, delays due to the quality of life of the patient or barriers to accessing healthcare services (3,9,29). Consequently, a 6-month period may be insufficient.

At 6 months, the healing rate of fistulectomy (76.00%) demonstrates superiority compared with that of seton placement (50.00%). For PAF treatment, the 6-month mark is often a suitable period for detecting early recurrences and initial treatment outcomes. However, a minimum follow-up of 12 months or more may provide a safer assessment of definitive healing or recurrence (31). According to the data obtained at 24 months, the healing rate was 67.5%. Based on the standard exclusion criteria (considering only followed patients), the net recurrence rate was determined to be 32.5%. Mei et al (4) reported cumulative recurrence rates of 20-44% at 24-month follow-ups after anal fistula surgery. However, while reported recurrence rates after anal fistula surgery vary widely from ~2.5-57.1%, pooled recurrence rates are generally reported ~19% (27), indicating a broad and heterogeneous distribution.

While the healing rate of fistulectomy remained nearly constant (77.80%), the healing rate for seton placement began to show greater improvement, increasing from 50.00% at 6 months to 67.00% at 24 months. This suggests that while fistulectomy healing is stable, healing rates in patients treated with a seton exhibit an increasing trend over time.

Seton placement appears to be more successful and more prominent in the management of complex and recurrent cases. In clinical practice, a seton is often preferable in high transsphincteric, complex or Crohn's-associated fistulas, as well as in cases of recurrence. Although initial healing rates in seton-treated patients may appear lower, it has been reported to be a strong predictor for long-term management of complex and recurrent fistulas (27,32).

In short-term follow-ups, transsphincteric fistulas show rapid healing, whereas in long-term follow-ups, the best healing outcomes are observed in intersphincteric fistulas. Concurrently, it was found that the rate of late recurrence increases in transsphincteric fistulas. Transsphincteric fistulas, particularly high variants, are known to be more complex procedures due to both surgical challenges and the risk of incontinence. For these types of fistulas, setons, LIFT, advancement flaps, fistula plugs, fibrin glue and staged approaches are commonly employed. In transsphincteric cases, seton applications, which provide drainage and symptom control in the early period, are frequently preferred; however, long-term recurrence rates have been reported to be higher compared with intersphincteric cases (21,33,34).

In clinical studies, missing data and patients LTFU represent the most notable methodological challenge affecting the accuracy and generalizability of results (35). If the attrition of patients LTFU is non-random, particularly in conditions such as anal fistula with high inherent recurrence risk, it can lead to an overestimation of healing rates. According to verbal information obtained during interviews conducted for the present study, a number of the patients who discontinued the follow-up process believed they have fully recovered and stopped attending their appointments. However, they more frequently consist of individuals experiencing ongoing symptoms, those at higher risk of recurrence (36) or those who have encountered treatment failure (37). Consequently, excluding these patients can result in a relative overestimation of healing rates and an underestimation of recurrence rates in a study. A thorough investigation into the reasons for attrition may provide more accurate insights into true healing and recurrence rates.

Numerous methods for handling missing data have been presented in the literature. Guidelines such as those from the International Council for Harmonisation (38) and the Consolidated Standards of Reporting Trials statement (39) recommend approaches like intention-to-treat analysis (where patients LTFU are assigned to the worst outcome category), sensitivity analyses and multiple imputation to evaluate such missing data (40,41).

In this context, it is key to analyze not only fistula type but also patient, surgeon, hospital and society-associated factors. A number of patients LTFU may indeed be fully healed and thus opt not to participate further. Alongside them are patients who have accepted a partial recovery and continue their lives with minimal, manageable symptoms. This underscores that patient-reported complaints may be more important compared with the mere presence of recurrence (42,43). For example, among patients presenting to the present polyclinic with recurrences, discharge and pruritus are less frequently reported, whereas perianal abscess and pain are more prominent concerns.

Based on available data and clinical experience gained in the outpatient setting, the following factors have been identified as prominent predictive elements: i) Change of hospital or doctor; ii) a lower number of female patients presenting due to privacy concerns, with a further decline in their follow-up attendance; iii) transition from a public hospital to a private institution or from a lower-tier to a higher-tier healthcare facility and; iv) influence of advertising, research exposure and social media, leading patients to seek newer, less invasive treatment methods.

With regard to a change in hospital or doctor, patients undergoing treatment for PAF who experience persistent complaints or recurrence may seek care at other centers, contributing to missing data. Those encountering post-treatment problems, such as ongoing complications, pain, suspected recurrence or dissatisfaction with their care, are more likely to present to different institutions. In additions, variations in follow-up protocols between centers, along with social and geographic factors, can lead to the loss of patient data (44). This constitutes a limitation of the present study.

With regard to a lower number of female patients presenting due to privacy concerns and further decline in follow-up, in certain sociocultural contexts where demographic, ethnic and religious factors are closely interrelated, the anorectal examination, a procedure that demands a high degree of privacy, may be associated with greater embarrassment and pose more notable challenges for female patients compared with their male patient counterparts. Consequently, influenced by pain, feelings of shame and socially prescribed sex roles, female patients may underreport their symptoms and consent only to acceptable therapeutic measures. Such dynamics can result in delayed diagnosis and deferred early intervention for various malignant conditions, extending beyond anorectal pathologies (45,46). Especially in diseases such as PAF, which impair social functioning and quality of life and are often kept secret, the diagnostic algorithm and fundamental physical examination practices remain standardized, notwithstanding cross-country differences in cultural norms (47).

From a sociological perspective, the higher rate of follow-up loss among female patients with proctological diseases can be attributed to factors such as neglecting self-care due to work and household burdens, reluctance to return for repeat anal examinations, unwillingness to be re-examined by a male doctor and lack of support from a spouse or partner (48).

Therefore, the lower number of female patients in the present study reflects not only a clinical epidemiological issue but also sociocultural and sex-based barriers to healthcare access. In this context, the more frequent loss of follow-up and records for female patients is a recognized limitation in the field of proctology.

With regard to a transition from a public to private hospital or from a lower-tier to a higher-tier facility, this transition can affect the prognosis of PAF patients both negatively and positively. Delays between the initial presentation and referral to a higher-tier center can lead to progression of the fistula, chronicity of abscesses or infections (49). Particularly in Crohn's-associated PAF, delays in initiating anti-TNF medication can worsen outcomes (50). As higher-tier centers predominantly manage more difficult and complex cases, their reported outcomes may appear less favorable. As is the case in the Department of General Surgery, Sağlık Bilimleri Üniversitesi, Bursa Yüksek İhtisas Training and Research Hospital, referrals of complex and challenging perianal diseases are received from numerous different provinces. Studies have reported that complex fistula is an independent predictor of recurrence (23,51,52).

Higher-tier centers are typically equipped with routine MRI, endoanal ultrasound, biologic therapies and advanced surgical techniques (LIFT, FLAP and complex reconstructions). Access to these resources facilitates timely treatment of the disease. However, inefficient utilization of these resources can conversely lead to delays in both treatment and follow-up.

With regard to the influence of advertising, research and social media in the search for newer, less invasive methods, in recent years, the proliferation of social media, the internet and health advertising has led to notable changes in the treatment-seeking behavior of patients diagnosed with PAF. Promotional content for surgical techniques, particularly VAAFT, FiLaC (laser), LIFT, plugs, fibrin glue and other minimally invasive methods, has markedly increased patient expectations for more comfortable, painless and sphincter-preserving treatments (53,54). Studies have indicated that social media use influences the treatment decisions of patients, an effect that is especially pronounced in chronic diseases. In conditions with a high tendency for recurrence like PAF, promises of minimal pain, shorter recovery times and organ preservation can make patients more inclined to switch doctors or hospitals (55,56). Proactively addressing and mitigating patient attrition should be recognized as an indispensable component of the therapeutic pathway. Furthermore, with ongoing advancements in technology, decisive steps forward must be taken. Rather than relying solely on conventional methods or social media-based communication, a robust, internet-anchored foundation for scientific research in surgical practice should be established.

Technological evolution in surgery has undergone a transformation with the advent of the Surgical Internet of Things (IoT), a network comprising interconnected biosensors and data processing units. In patients with perianal fistula, IoT-based postoperative monitoring enhances recurrence follow-up by enabling continuous, remote physiological assessment, thereby facilitating early detection of healing deviations or potential complications. The integration of telemedicine-based follow-up can notably reduce rates of LTFU by offering a convenient digital alternative to routine in-person visits, consequently improving patient adherence to long-term care protocols. This transition toward a smarter healthcare system supports a more personalized, efficient and proactive surgical care continuum (57). To this end, technologies such as 3D printing offer a paradigm shift: Instead of verbally describing the fistula anatomy of the patient, the concept, morphology, pathophysiology and treatment strategy can be visualized and understood through tangible, patient-specific models. This represents a realistic, scientifically grounded and intuitively comprehensible approach that aligns with the modern era of multimodal surgical education and innovation. This stands in contrast to speculative or anecdotal promotion, offering instead a verifiable and reproducible standard for advancing clinical understanding and patient engagement (58).

A large amount of health-related social media content may lack scientific accuracy; some studies have shown that the generally low level of digital health literacy in the public can be a negative factor that impacts both recurrence rates and a patients understanding of surgical risks (59,60). Social media and internet searches have become one of the most important resources determining the choice of surgeon and medical center by a patient. This process can disrupt treatment continuity for PAF patients, as individuals may present to multiple centers, initiate various treatments, making patient follow-up difficult and increasing missing data (53,54,56).

The most key methodological limitation of the present study is the high LTFU rate of 49.80% observed at 24 months, coupled with the descriptive approach employed to manage this incomplete data. Despite the importance of methods such as sensitivity analysis and multiple imputation being addressed, the lack of formal application of these techniques to the current dataset may limit the overall robustness of the present findings and introduce a degree of selection bias. Therefore, the present results, particularly those associated with long-term surgical success, should be interpreted with great caution and prudence, as potential trends rather than definitively proven effects.

Active telephone interviews with patients LTFU revealed that the reasons for attrition frequently included sociocultural differences, perceived well-being, physician- or hospital-associated factors, pursuit of new management strategies and relocation. However, the lack of access to objectively validated clinical outcomes for these patients increases the risk of uncertainty and speculative assumptions regarding long-term data. Furthermore, given that specific informed consent was not obtained for the structured qualitative data collection process for the present subgroup, the present findings are reported as contextual and exploratory observations rather than as formal qualitative data.

Despite the terminology used in the present study being standardized, it should be acknowledged that parameters such as ‘healing’ (defined as the clinical absence of symptoms and closure of the fistula tract) and ‘recurrence’ (defined as the reappearance of symptoms or reopening of the fistula tract after healing) are subject to inter-observer variability in clinical practice.

In conclusion, in the treatment of PAF, MRI, as an established method for diagnosis, demonstrates superiority compared with techniques such as RSS. Furthermore, long-term treatment outcomes are determined not solely by surgical skill or methodology, but by a complex and multifaceted network of factors. The lack of critical anatomical data, such as internal openings not detected during routine examination, socioeconomic constraints leading to missing data, cultural perspectives and the increasingly influential role of social media in shaping patient expectations and anxieties, markedly impact the treatment process.

Therefore, sustainable success in the management of PAF disease can only be achieved with a treatment strategy that makes the most effective use of technological capabilities MRI, understands the patient as a whole (physical, psychological and social context) and adopts a personalized, multidisciplinary approach. This strategy, by minimizing data loss and overcoming missing data and invisible barriers, will transform treatment from a purely technical procedure into a patient-centered and lasting solution. The present study demonstrated that long-term management outcomes for PAF are dependent on minimizing patient attrition. Accordingly, future clinical practice and research should aim to prioritize the development of more resilient and patient-centered follow-up systems, with particular focus on the specific needs of female patients, those with complex fistulae and socioeconomically vulnerable populations. Proactively addressing and mitigating patient attrition should be recognized as a key component of the therapeutic pathway.

Acknowledgements

Not applicable.

Funding

Funding: No funding was received.

Availability of data and materials

The data generated in the present study are included in the figures and/or tables of this article.

Authors' contributions

AT and BÖ were responsible for conceptualizing the present study. AT provided supervision. HH and ME were responsible for data collection and processing. AT and BÖ conducted analysis and/or interpretation. AT conducted the literature review. AT and HH wrote the manuscript. AT conducted critical review and controlled the references. BÖ was responsible for the stationery supplies used. AT and HH confirm the authenticity of all the raw data. All authors have read and approved the final version of the manuscript.

Ethics approval and consent to participate

Ethics approval was granted in October 2025 by Bursa Yüksek İhtisas Training and Research Hospital (approval no. 2024-TBEK 2025/10-04).

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References