Congenital pyriform sinus fistula (CPSF) is a rare branchiogenous disease of the neck, which is known as the third or fourth branchial cleft anomaly. According to the direction of fistula, third and fourth branchial cleft can be distinguished. Third branchial cleft originates from the base of the pyriform fossa, then passes through the thyrohyoid membrane, and is located above the superior laryngeal nerve (1). Fourth branchial cleft originates from the tip of the pyriform fossa and passes through the cricothyroid membrane, located below the superior laryngeal nerve (2). As the clinical manifestations and treatment methods for third and fourth branchial cleft are the same, they are collectively referred to pyriform sinus fistula. The primary clinical manifestations of CPSF include recurrent neck swelling, neck abscess, suppurative thyroiditis or subacute thyroiditis (3). Due to its rarity and close association with thyroid gland, it is often overlooked and misdiagnosed as thyroiditis (4), which brings diagnostic and therapeutic challenges. Approximately 90% of the fistula occurs on the left side, with occasional instances on the right side and bilaterally (5).

For the treatment of CPSF in acute infection stage (AIS), conventional approaches include antibiotic treatment, incision and drainage, or puncture drainage. Over the years, traditional open surgery was commonly used to remove the entire fistulous tract (3, 6), but in recent years, with the advancement of minimally invasive techniques, endoscopic cauterization of the internal opening has become increasingly popular, including Radiofrequency Ablation (RA), laser coagulation, and others (5, 7).

In this study, we collected data from patients with CPSF and evaluated the indications and outcomes of open surgical excision (OSE) vs. RA. The aim of the present study was to comparatively evaluate patients with CPSF regarding the indications and outcomes of treatment with OSE vs. endoscopic RA, and the advantages of both therapeutic procedures were also analyzed.

CPSF was initially reported by Sandborn in 1972, accounting for 3%–10% of all pharyngeal lesions (8). It is a rare congenital anomaly which originates from the incomplete embryonic development of the third or fourth branchial arch. Research has indicated that there is no prominent difference of the incidence rate between males and females (5). In this study, the male-to-female ratio was 1:1.1, which indicated no gender dominance. The predominance of CPSF occurring on the left side (over 90%) is likely due to the degeneration or absence of the right ultimobranchial body, or the asymmetrical development of blood vessels in the fourth branchial arch (3).

Children with CPSF often receive initial diagnoses of recurrent neck swelling neck abscess, neck mass, suppurative thyroiditis or subacute thyroiditis (3). Clinically, recurrent thyroiditis frequently occurs because the thyroid is likely to be invaded by bacteria through the fistula between the thyroid and internal opening of CPSF (9). In this series, the misdiagnosis rate in previous treatment before surgery reached 21.7%. For some patients who experienced repeated neck infections before receiving an accurate diagnosis, it is common for them to undergo excessive antibiotic treatment and multiple I&D or puncture drainage (5). Apart from the aforementioned inaccurate diagnoses, CPSF has been misidentified as mumps, esophageal fistula, lymphadenopathy, lymphangioma, second branchial fistula, tuberculosis (6), dermoid cyst, thyroglossal cyst, ectopic thyroid, diverticular rupture (10), neck cellulitis (11). In clinical practice, suppurative thyroiditis is a rare occurrence in children. In cases where suppurative thyroiditis is observed in children, it is crucial to maintain a high level of vigilance for infection caused by CPSF. Therefore, we strongly advise conducting the following examinations for children exhibiting recurrent neck swelling, neck abscess, neck mass, suppurative thyroiditis, or subacute thyroiditis, particularly on the left side, along with repeated drainage and delayed wound healing. These symptoms raise a high suspicion of CPSF in such cases.

The gold standard for diagnosing CPSF is the identification of internal opening of the sinus tract under suspension laryngoscope. However, this examination is not routinely performed due to the requirement for general anesthesia. Besides, fiberoptic laryngoscopy is recommended by some experts to locate the internal opening of the CPSF. Nevertheless, it is reported that the positive rate of fiberoptic laryngoscopy is relatively low, similarly to that in this study (56.9%), possibly due to the difficulty of complete anatomical exposure and unsatisfactory cooperation from children (3–5). Ultrasound is the preferred choice to repetitively detect the infammatory process and neck fistula in children, which is respectively shown as a hypoechoic mass or cord-like tubular structures (4, 5). However, ultrasound has limitations in assessing the depth and extent of CPSF. Instead, CT can be used to understand the anatomy and vital adjacent structures better (5). Researches have shown that enhanced CT estimates the length of the fistula and the positional relation between the fistula and adjacent structures like blood vessel and thyroid tissue (5). Based on our experience, we recommend that ultrasound, enhanced CT, fiberoptic laryngoscopy are the preferred diagnostic tools for CPSF. Depending on the available diagnostic equipment and the varying levels of expertise among radiologists and surgeons, these methods can be used individually or in combination to assist in the diagnosis.

For the treatment of CPSF in AIS, the conventional approaches are antibiotic treatment, I&D or puncture drainage. This study shows that 82 cases underwent I&D. One patient endured 7 infections and 5 I&D procedures prior to receive radical surgery. This highlights that conservative treatment following infection still poses a significant risk of reinfection.

The typical surgical method of CPSF is OSE in NIS, which should be performed 2–3 months after controlling the acute infection. Nevertheless, when patients suffer from frequently repeated infections, the surgery can be expedited to 1 month after the acute infection (4). Owing to the importance of the thyroid in the growth of children, we recommend that the thyroid tissue should be preserved during surgery even if the thyroid is invaded, as it does not pose a risk of recurrence (3). Theoretically, complete excision of the tract is considered as the most ideal and reliable treatment to prevent a recurrence. However, complete resection of the whole tract and high ligation of internal fistula are difficult and challenging. The small size and deep location of the fistula, as well as its proximity to the carotid sheath and important nerves, make the procedure difficult (3, 6). The normal anatomy here is prone to be damaged by recurrent infections and previous treatments, such as I&D and failed radical operations (4, 12). To accurately locate the fistula, several methods can be applied, including endoscopy-assisted methylene blue injection, intubation through the internal opening and light guided procedure (13, 14). In this data, the complication rate and the recurrence rate associated with OSE were both low (6.3%, 0.8%). Typically, the recurrence rate was lower than the rate of other reports (6, 15). The low recurrence rate and low incidence of vocal cord paralysis observed in this study can be attributed to the surgeon's precise understanding of the anatomical structures surrounding the fistula. By meticulously exploring the deep extent of the fistula, directly elevating the thyroid cartilage plate to expose the internal fistula marked with methylene blue through hypopharyngeal route, and performing accurate high-level ligation, the surgeon ensured the complete resection of the fistula.

With the development of endoscopic technique and popularization of minimally invasive surgery, endoscopic surgeries have been widely employed to treat CPSF, such as electrocauterization, chemocauterization, laser coagulation, fibrin glue and so on (7, 16–19). The objective of endoscopic cauterization is to induce scar tissue formation at the internal opening of CPSF, enabling adhesion and closure, thereby avoiding pharyngeal secretions, pathogen and food debris into the fistula. Nevertheless, research has revealed significant issues associated with these approaches, including high recurrent rate and severe complications (5, 12). In this study, we utilized endoscopic RA to avoid the inconvenience, inaccuracy and insecurity of other endoscopic techniques. Some experts believe that this technology can be considered as the primary treatment option for CPSF (12, 20).

Endoscopic RA technology utilizes a system of radiofrequency bipolar electrical current through a medium of normal saline, which generates a plasma field of sodium ions. Molecular bonds can be broken by these this plasma layer, leading to molecular dissociation and tissue ablation. One notable advantage of this technique is its ability to significantly reduce thermal damage to the sinus tract and adjacent structures by saline flushing and suction at lower temperatures (40℃–70℃), which is completely different from thermal injury caused by heat signatures (400℃–1,000℃) of electrocauterization and laser coagulation (5). As a minimally invasive and transoral surgery, this technology can avoid cervical scar and postoperative pain arising from external cervical incision. While some researchers suggest that endoscopic RA should not be performed during the acute or subacute stage of CPSF (4), our previous researches have shown that RA in combination with I&D of the abscess is a promising method for CPSF in AIS (5, 12). Abbas et al. (21) also reported a successful case of using RA in combination with I&D for a girl with a history of recurrent thyroid abscesses secondary to CPSF. Here is the analysis of transient hoarseness following RA (13.6%). During the ablation of the internal fistula opening using a plasma knife, the knife head is positioned on the medial aspect of the thyroid cartilage, causing thermal injury to the surrounding tissues, including the recurrent laryngeal nerve. Thus, it is critical to pay attention to the direction of the knife head during surgery to minimize damage to the recurrent laryngeal nerve. Therefore, based on our clinical experience during the 5 years, we consider the infectious period as an appropriate time for RA surgery, as it can effectively prevent the need for readmission and additional medication.

Some researchers have focused on the comparison between OSE and endoscopic cauterization. Hwang et al. (22) compared open surgery with endoscopic trichloroacetic acid chemocauterization, of which the recurrence rates were not significantly different (35.7% vs. 46.1%). Pu et al. (12) found that the success rate and postoperative temporary hoarseness rate in RA group were similar to those in surgical excision group (100% vs. 100%, 11.1% vs. 6.5%), while the risk of postoperative swelling and pain was higher in RA group (11.1% vs. 0%). There was no recurrence in both groups. In Chen's research (6), longer hospital stay was found in surgical excision group (10.50 ± 3.93 vs. 5.02 ± 3.30 days), while neck infection rate in this group was obviously lower than that in RA group (0% vs. 8.3%). Recurrence rates in these groups were not significantly different. In this data, we also found no significant disparities in postoperative complications and recurrence rates between OSE and RA (6.3% vs. 14.4%, 2.1% vs. 5.9% respectively). The duration of hospitalization was longer for patients who underwent OSE (Table 2). However, in comparison to OSE, the gastric tube retention time after RA was longer (3 days vs. 14 days). This extended duration aims to allow the internal fistula opening to fully heal after cauterization and prevent any food-related stimulation.

Therefore, in our experience, both OSE and RA are primary treatment options for CPSF, the selection of the surgical methods should be based on by individual conditions and clinical expertise of surgeons. Here is the summary of the indications for these two procedures: 1.

RA is the preferred approach for pediatric CPSF, owing to its minimally invasive nature, shorter hospital stay, lower surgical complexity, and ease of operation. It is a safe method that avoids extensive damage. Moreover, RA can be performed even during the acute infection phase. For cases with abscess formation, simultaneous I&D can be performed, significantly reducing the treatment duration and eliminating the necessity for additional medication. Nonetheless, patients should be informed in advance about the potential complications following RA, such as temporary hoarseness and acute infection.

The limitations of this study are as follows. The presence of epithelial remnants and the persistence of a closed cavity following RA may give rise to long-term issues. Additionally, the retrospective nature of this study is another limitation. In consequence, it is essential to conduct a large-scale prospective clinical study with long-term follow-up.

CPSF is a rare congenital anomaly, over 90% of the fistula occurs on the left side, with occasional instances on the right side and bilaterally. Due to its rarity, CPSF should be considered in the differential diagnosis in children with recurrent neck infections, suppurative thyroiditis or subacute thyroiditis, accompanied by repeated drainage and long-delayed healing. Diagnostic tools such as ultrasound, enhanced CT and fiberoptic laryngoscopy are the preferred choices to aid preoperative diagnosis. Both OSE and RA are recognized as first-line treatment options for CPSF, which show no significant differences in prognosis, except hospitalization duration and the gastric tube retention time. Nevertheless, the indications for open surgical procedures and RA differ, which are influenced by factors such as the inflammatory stage, specific typing, previous treatments, and the surgeon's expertise. The selection of surgical approach should be carefully determined based on individual circumstances.