Introduction

A perianal fistula is an abnormal tract connection between the anal canal and the surrounding skin of the perineum. It is hypothesized that it forms through an intersphincteric cryptoglandular infection leading to the formation of an abscess which can subsequently lead to the formation of a fistulous tract1. In European countries, the overall prevalence of a perianal fistula is 18.37 per 100,000 people2. Perianal fistulas exert a substantial negative impact on patients through a burden of symptoms including: Fistula-related pain, a significant source of impaired wellbeing, which often coincides with the initial diagnosis of the fistula; Fistula-related discharge; Fistula-related restricted mobility, and Fistula-related fatigue3. Although clinical examination can confirm diagnosis, fistulas may require medical imaging for the preoperative identification of the complex tract shape, number, internal orifices, and Park’s classification which plays a crucial role in effective management4.

According to the tract location relative to the internal and external anal sphincter, Park’s classification of the perianal fistula include: 1-Intersphincteric, which is confined to the intersphincteric plane, 2-Trans-sphinecteric, where the track passes radially through the external anal sphincter 3-Suprasphicteric, where the track passes upward within the intersphincteric plane over the puborectalis muscles and descends through the levator muscles and the ischiorectal fossa, and 4-Extrasphicteric, in which the course is completely outside the external anal sphincter5. Perianal fistula can be intersphincteric (simple fistula), transsphicteric, suprasphincteric (complex fistula), or extrasphincteric (horseshoe fistula)6. Diagnosis and classification of perianal fistulas mainly depend on medical imaging.

Imaging of the rectum and surrounding tissues is difficult and technically challenging, and conventional X-ray imaging and computed tomography (CT) barium studies offer limited information about the extension of the perianal fistulas7. Preoperative pelvic magnetic resonance imaging (MRI), and endoanal ultrasonography (EAUS) are the first line imaging modalities for the evaluation of the perianal fistulas with better results and prognosis after surgery8. Conventional X-ray fistulography gradually lost its diagnostic value due to several of disadvantages including 1-the anorectal muscles cannot be imaged, 2- determination of the IO is often difficult, 3-it is an invasive technique with the risk of dissemination of sepsis, and 4-it can give a wrong impression of the presence of an extrasphincteric tract9. The inability of conventional X-ray and CT to show the fistula in relation to normal structures in a single image10, the CT radiation hazards11, the limited tolerance of the patient, and the operator and experience dependency of the EAUS12, the high cost of MRI13, in addition to the limited availability everywhere of MRI. All the above limitations create the need for an imaging modality that is effective, possible, and tolerable by the patient and a widely available imaging modality for the diagnosis of the perianal fistulas to plan for optimal management. The aim of this study was to describe the low-type intersphincteric perianal fistulas (number, site, number of EO, IO, and length of fistular track) using the TCUS. Characters of the intersphincteric perianal fistula including the length and distance of IO from the anal verge are critical points in surgical planning. The study was done in Hadramout region of Yemen which is a peripheral region in a developing country with a paucity of modern imaging methods to diagnose perianal fistulas such as endoanal ultrasound and MRI. The most available imaging method is TCUS which proved to be an effective diagnostic method for the common low-type perianal fistula.

Materials and methods

Study design and setting

A retrospective study was conducted with the electronic records of patients clinically diagnosed with perianal fistulas and diagnosed between April 2017 and December 2022. The study was conducted at Alsafwa Consultative Medical Center (ACMC) in Almukalla city, Hadhramout, Republic of Yemen.

Study sample

The study involved 581 low-type intersphincteric perianal fistulas from the reports of 549 patients who underwent TCUS for clinically diagnosed perianal fistula.

Inclusion criteria

Any patient diagnosed with a low-type intersphincteric perianal fistula (Fig 1), regardless of their age, gender or the number of fistulas. Based on the location of the IO, perianal fistulas classified into low and high types. Low-type perianal fistulas are those with IO in lower regions of the external sphincteric muscle. High-type perianal fistulas are those with IO located near the deep part of the upper external sphincteric muscles5.

Fig. 1
figure 1

The diagram shows the anatomy of the anus (A), anal canal (from the skin to the dentate line (D), rectum (R), internal anal sphincter (IS), external anal sphincter (ES), and the different types of the perianal fistula including the inter-sphincteric fistula (1), trans-sphincteric fistula (2), supra-sphincteric fistula (3), and extra-sphincteric fistula (4).

Full size image

Exclusion criteria

Patients with pilonidal sinus, perianal abscess, anal fissure, high-type perianal fistulas, and perianal fistula without an external opening were excluded from the study.

Ultrasound imaging procedure

All patients underwent perianal TCUS that were performed by a highly qualified radiologist with 12 years postdoctoral experience in general ultrasound imaging. A superficial linear transducer of 7.5 MHz, Mindray DC30 ultrasound machine was used to assess the perianal region in all the patients, and a deep curved probe of 3.5 MHz was used to follow the long tracts of some fistulas.

Scanning was performed in the knee-chest position or in left lateral decubitus position. The ultrasound probe was wrapped with a latex cover after applying a contact gel to its surface. TCUS scanning of the perianal region was performed in the sagittal, coronal, and axial planes. The probe was placed over the anal canal and the EO of the tract in all patients and ultrasound images was obtained in different planes as necessary. The distance of the IO of the fistula from the anal verge was described, the length of the tract was evaluated to the EO, and any ramifications in the tract was described14.

In the ultrasound images, the fistular tract appears as a tubular hypoechoic structure starting from the anal canal and extended to the perianal skin. A perianal abscess appears as a hypoechoic fluid-filled cavity in the perianal region15.

Statistical analysis

The data were analyzed using the Statistical Package for the Social Sciences (SPSS) version 27 (IBM, Armonk, NY). I. Inferential statistics was used, categorial variables are presented as frequency and percentage, and the continuous variables as means ± standard deviation and median. The Shapiro-Wilk, and Kolmogorov–Smirnov tests of normality revealed non-normality distribution of continuous categories length and distance of fistula (p < 0.001), for this reason the non-parametric Kruskal-Wallis test was used to assess the differences of length in different age group and length and distance in the four perianal quadrants. Furthermore, for length of fistula as there was significant differences in different categories of perianal quadrants, the Dunn-Bonferroni-Tests was used to compare the group in pairs to assess which were significantly differs. Adjusted p-values ≤ 0.05 consider statistically significant for differences.

To generate Fig. 2, the Sketchbook (https://www.sketchbook.com/) was used. To enhance the resolution of our ultrasound images, the Snapedit.app (https://snapedit.app/ar/change-sky/upload) was used for Figures 3, 4, and 5.

Fig. 2
figure 2

The diagram indicates the perianal region in the knee-chest position divided in clockwise direction into four quadrants, and 12 h. LAQ, left anterior quadrant, LPQ, left posterior quadrant, RPQ, right posterior quadrant, and RAQ, right anterior quadrant.

Full size image
Fig. 3
figure 3

An ultrasound image was taken by using a superficial linear probe (7.5 MHz) showing low type right sided perianal fistula, the right image shows the tract of the fistula (Thick white line), external opening (EO), Internal opening (IO), of the tract, around the anus (A), and anal canal (AC). Left image is the same clear image.

Full size image
Fig. 4
figure 4

An ultrasound image was taken by using a superficial linear probe (7.5 MHz) showing the low type right sided branching perianal fistula, the right image shows the tract of the fistula (Thick white line), Internal opening (IO), first external opening (EO1), and second external opening (EO2) of the tract, around the anus (A), and anal canal (AC). The left image is the same clear image.

Full size image
Fig. 5
figure 5

An ultrasound image was taken by using a superficial linear probe (7.5 MHz) showing a low type left sided perianal abscess. The right image displays the abscess (Filled with white color), Internal opening (IO) around the anus (A), and anal canal (AC). The left image is the same clear image.

Full size image

Results

A total of 581 intersphincteric perianal fistulas from 549 patients were included. Males represented the vast majority of the sample (83.97%), with 16.03% females. Shapiro–Wilk and Kolmogorov–Smirnov tests of normality revealed non-normality distribution of patients ages (p < 0.001) with mean age was 36.14 ± 13.37 years (range from 1 to 80 years). Kruskal–Wallis test results showed significant difference in perianal fistula distribution in different age groups (p < 0.001) with the majority were in the young adult age group, aged 21 to 40 years (56.47%), followed by the group aged 41 to 60 years (31.88%). However, the test of homogeneity of variances (Levene statistics) showed normal distribution of perianal fistulas within age groups (p = 0.063). The patients predominantly had one fistula (n = 518, 94.35%), see Table 1.

Table 1 Patients’ Characteristics.
Full size table

The fistulas almost always had one IO (99.65%), and the majority had one EO (79.10%). The fistula EO was located mainly in the LAQ (31.5%), followed by the LPQ (29.1%). The most frequent sites of the EO around the anus were 6 o’clock (14.3%) and 1 o’clock (13.6%) (Fig. 2&3). The IO was often 11–20 mm above the anal verge (58.5%) or ≤ 10 mm (27.4%). The length of the tract of the fistula was 21–30 mm (30.1%), 11–20 mm (27.4%), and 31–40 mm (25.0%), see Table 2.

Table 2 Characteristics of the low-type intersphincteric perianal fistulas.
Full size table

The Kruskal–Wallis test results indicated a significant variation in the length of the fistular tract between the different perianal quadrants (p < 0.001), see Table 3.

Table 3 Comparisons the length of the fistular tracts in different perianal quadrants.
Full size table

The Dunn-Bonferroni Post Hoc tests for comparisons of pairwise the length of the perianal fistular tract in the different perianal quadrants shows that the pairwise group comparisons of (LAQ)1–3—(LPQ)4–6, (LPQ) 4–6—(RPQ)7–9 and (LPQ) 4–6—(RAQ)10–12 demonstrated an adjusted p-value < 0.001), so these groups were significantly different in pairs, see Table 4.

Table 4 Pairwise comparisons of fistula length measurement in different perianal quadrants Quadrant.
Full size table

Table 5 displays the mean distance of the fistular IO from the anal verge in the different perianal quadrants. The Kruskal–Wallis test shows no significant variation in the IO distance from the anal verge between the different perianal quadrants (p = 0.945).

Table 5 Comparison the distance of the internal openings of the fistulas to the anal verge in different perianal quadrants.
Full size table

From our work, we selected TCUS images for low-type intersphincteric perianal fistulas from different patients using a 7.5 MHz superficial linear probe (Figs. 3, 4), and an image of the perianal abscess using the same linear probe (Fig. 5).

Discussion

An optimal diagnosis and imaging of a perianal fistula is essential to plan the optimal management and achieve a satisfactory outcome of the surgery for the health and comfort of the patient. This article reports the age and gender distribution, the number, site, and length, of the common low-type of perianal fistulas using TCUS imaging easily available, and without any radiation hazards. We found that 69.26% of the patients were in the young adult (21–40 years) age group and the average age of the patients was 36.14 ± 13.37-years. This finding is compatible with a previous study which reported that most fistulas occur between 20 and 40 years of age with the average age of diagnosis 38 years16. In line with our results, the results of another study with 120 patients with a mean age of 39 years17.

Our results indicated that males are affected more than female (83.97% vs 16.03%). These results are compatible with the results in literature from 1969 to 1978, the mean incidence of perianal fistulas in the population of Helsinki was 12.3 and 5.6 per 100,000 for male and females, respectively18. Sanchez-Haro et al. reported that males were affected more frequently than females (70% vs 30%) among Spain inhabitants which was consistent with our results19. Our results are in line with a previous study in Saudi Arabia which reported an 80.1%:19.9% male: female rate, although it was high perianal fistulas20. Contradictory to our results, another study reported that the mean age of patients with perianal fistula was 44.3 ± 12.1 years old and the male-to-female ratio was 16:121. The predominant involvement of males, as explained by Shindhe et al., are repeated perianal infections, infection from a hair follicle, infected sweat or sebaceous gland, Crohn’s disease, etc.22.

The classification of perianal fistula as low or high is more practical than other classifications, with implications for the treatment. A low perianal fistula is defined as a fistula with a tract located in the lower third of the external anal sphincter, and a high perianal fistula as a fistula in which the tract runs through the upper two thirds of the external anal sphincter muscle. Low fistulas can be managed safely by fistulectomy23. In our study, 85.9% of the intersphincteric perianal fistulas were within 20mm, 10.8% were within 21-30mm, and 3.3% were within 31 to 50mm distance from the anal verge. This was explained by Amato et al. who reported that the source of most cases of perianal sepsis is a non-specific cryptoglandular infection starting in the intersphincteric space24. Fortunately, the gold standard surgical treatment for a low anal fistula fistulotomy with an 80–100% success rate, with the exception of up to 62% fecal incontinence rate25. The current study confirms that TCUS determination of the length, EO and IO of perianal fistula has a critical role in going to further investigation in high-type perianal fistulas, and surgical planning and improving surgical outcomes in low-type.

Our results showed that the most common site of the EO of the intersphincteric perianal fistula was at 6 o’clock. This result is compatible with a previous study who reported that the IO of the perianal fistulas mostly was at the middle posterior wall of the anal canal (6 o’clock)26. In line with our results, a study conducted by Chauhan et al. reported that 78% of the perianal fistulas were posterior27. A perianal fistula is considered high when its IO lies above the dentate line28. The anal verge and the dentate line are two reliable landmarks for distance measurements inside the rectum. The median distance from the anal verge to the dentate line is 20mm29. Our results showed that 85.9% of the perianal fistulas IOs were less than 20mm above the anal verge which is below the dentate line. Recently, proposed comprehensive novel template was suggested for optimal treatment planning using anal endosonography and MRI. The template includes the following: Fistula type according to Park’s classification; fistula clockwise position location according to the AC; fistula height; differentiation between simple and complex by the presence of secondary extensions in the later; location and patency of the IO; description of a residual abscess type and location according to the clockwise position; morphology of the anal sphincters; and schematic drawing of the anal canal30. In the current study, ANOVA test revealed significant variation in length of perianal fistulas which was 33.73 ± 14.48mm, 32.12 ± 12.46mm, 29.85 ± 13.19mm, and 25.04 ± 12.32mm in RAQ, LAQ, RPQ and LPQ respectively. These results explained the results of Bakir et al. who found that the accuracy of Goodsell’s rules to predict the IO was more accurate in posterior fistulas (73%), which were shorter, than in anterior fistulas (52.4%)31. Shorter length of the posterior than in anterior fistulas is a significant point for success of surgery which may be more challenging in posterior midline than in lateral fistulas32.

Perianal fistula impairs the quality of life, and the patients may accept the risk of postoperative incontinence with fistulotomy33. A previous metanalysis evaluated the efficacy of fistulectomy compared to fistulotomy and reported that the two surgical methods have the same low rate of fecal incontinence with no significant difference in rate of fistula recurrence in low perianal fistula34. In the Hadhramout region, the only treatment method is surgery and most patient with a perianal fistula accept surgical treatment with good outcomes in most patients. The prognosis for the perianal fistula varies according to the etiology. With sphincter preserving surgery, perianal fistulas of cryptoglandular origin have 80% and 60% healing rates in simple and complex fistulas, respectively35. The challenge in management of perianal fistulas is to determine the course of the fistular tract between the IO and EO14. Short fistular tract and the short distance of IO from the anal verge as in the LPQ make diagnosis and surgery easier.

Ultimately, it is necessary to review the factors increasing the risk of perianal fistula formation, including obesity, diabetes mellitus, smoking, sedentary life, excessive intake of greasy and spicy food, and prolonged sitting on the toilet for defecation36. Sexually transmitted diseases are a risk factor, especially in immunosuppressed patients37. In total, 40%-60% of patients treated for an acute perianal abscess will eventually develop a perianal fistula26. After drainage of the perianal abscess, good management by in-hospital dressing with regular washing of the wound until closure of the abscess pouch decreases the risk of a perianal fistula6. Mocanu et al. reported that the incision and drainage of anorectal abscesses followed by an empiric 5 to 10-day course of antibiotics may avoid the morbidity of perianal fistula38. More studies about the methods of avoiding perianal sepsis to prevent abscess formation and subsequent perianal fistula are recommended.

Limitations

The study focused on description of low-type intersphincteric perianal fistulas using ultrasound imaging. Due to the retrospective nature of the study, no available clinical information about the patients especially the causes of perianal fistulas. This study was limited to the low intersphincteric fistula as the most common type of perianal fistula. High perianal fistulas were excluded due to decrease efficacy of ultrasound imaging in high fistulas where an MRI is preferred. TCUS is also limited in that it is less effective in high perianal fistulas. A previous study reported that EAUS was a relatively effective method in detecting perianal fistulas which may be more sensitive than MRI in detecting transsphincteric and intersphincteric fistulas. However, MRI was higher than endorectal ultrasonography in detecting suprasphincteric perianal fistulas39. TCUS is also limited by that it demands a high level of knowledge and skills, as it is operator dependent, and machine setting is necessary to achieve high image quality and avoid pitfalls of artifacts40, especially in the complex structure of the perineum region.

Future guidelines

Further studies about the possibility of using deep ultrasound transducers to diagnose high perianal fistulas is needed to solve the problem of the paucity of EAUS and MRI in rural regions of developing countries.

Conclusions

The study demonstrated that, the majority of patients diagnosed with low type intersphincteric perianal fistulas were young adult males. These fistulas typically feature a single tract with a solitary external and internal opening. The combination of a solitary tract, along with the short length and minimal distance of the internal opening from the anal verge, enhances surgical outcomes and reduces the risk of complications and recurrence associated with perianal fistulas. TCUS, when performed by an experienced operator, can be effectively utilized for the diagnosis and surgical planning of low-type perianal fistulas, with the offer of that it is a non-invasive, well-tolerated, and radiation-free imaging method.