Endoscopic Management of a Hypopharyngeal‐Subglottic Fistula: Case Series and Literature Review

Abstract

A rare complication of prolonged intubation or laryngeal radiation is destruction of the posterior cricoid cartilage with resultant hypopharyngeal‐subglottic fistula leading to aspiration and inability to consume a regular diet. We describe a step‐by‐step endoscopic approach to the closure of a hypopharyngeal‐subglottic fistula. This technique successfully closed the fistula in two patients and allowed the safe resumption of an oral diet.

1. Introduction

A rare complication of prolonged intubation, laryngeal trauma, radiation, or surgery is destruction of the posterior cricoid cartilage with resultant fistula through the cricoid, known as a hypopharyngeal‐subglottic fistula (HSF) (Figure 1). This leads to aspiration, inability to consume a regular diet, and often gastrostomy‐tube dependence [1, 2, 3, 4]. HSF closure allows patients to resume a regular diet and improves quality of life [3].

FIGURE 1.

Hypopharyngeal‐subglottic fistula. Sagittal view of the airway demonstrating location of a hypopharyngeal‐subglottic fistula.

A related but distinct clinical entity is a tracheoesophageal fistula (TEF), which is lower in the aerodigestive tract. Many endoscopic and open surgical techniques exist for closure of the relatively more common TEF. On the contrary, few reports have described the management of HSF, and most procedures involve open approaches including lateral pharyngotomy [1, 2, 3, 5]. One study showed promising results using an endoscopic intervention with porcine small intestine submucosal (SIS) plug [4].

This study aims to add to the existing literature surrounding endoscopic management of a HSF. We present a minimally invasive, endoscopic operative technique utilized successfully for two patients who presented with dysphagia and aspiration secondary to a HSF.

2. Case Series

2.1. Patient Histories

Patient 1 was a 65‐year‐old male with right‐sided glottic cancer treated with radiotherapy 13 years prior who presented with dysphagia to liquids. Patient 2 was a 66‐year‐old male with quadriplegia secondary to gunshot wound years prior and progressive respiratory failure requiring tracheostomy following prolonged intubation. After tracheostomy, he developed recurrent aspiration pneumonia. Modified barium swallow (MBS) revealed a HFS for both patients, which was confirmed on endoscopic evaluation (Figure 2). Each HSF measured less than 1 cm.

FIGURE 2.

Patient cases. (A) Modified barium swallow (MBS) study showing presence of hypopharyngeal‐subglottic fistula (HSF) for Patient 1 along with hypopharyngeal view of fistula. (B) MBS demonstrating HSF for Patient 2, and endoscopic images showing the subglottic and hypopharyngeal sides of the fistula.

2.2. Endoscopic Technique

This was performed under general anesthesia with rigid endoscopic exposure alternating between the hypopharynx and subglottis for visualization and manipulation of materials used during each step. There were six key components to the surgery. Schematics and Patient 2's operative case are depicted in Figures 3 and 4.

1. 14‐gauge angiocath pre‐threaded with 2–0 vicryl suture passed from the hypopharynx into the subglottis through fistula.

2. The suture end in the hypopharynx was held externally through the oral cavity while the suture end in the subglottis was grasped and pulled out through the tracheostomy stoma (neck). Of note, Patient 1 did not have a tracheostomy tube, so the subglottic suture end was pulled out via the oral cavity.

3. Angiocath removed while leaving suture in place.

4. The hypopharyngeal side of the suture was passed through a Keith needle. The Keith needle with attached suture was threaded through the graft or tracheoesophageal puncture (TEP) occluder via multiple passes. The Keith needle was then removed from the suture, and the suture was tied to secure it to the graft or occluder.

5. Subglottic end of suture grasped externally. Under telescopic guidance, the graft/occluder was gently pulled by the suture from the hypopharynx and into the fistula until visualized on the subglottic side.

6. The suture was cut from the subglottic side of the graft/occluder and subsequently removed from the airway. This left the graft/occluder in position transiting the fistula.

FIGURE 3.

Initial operative steps. (A) Under rigid endoscopic exposure of the hypopharynx a 14‐gauge angiocath with pre‐threaded 2–0 vicryl suture is passed from the hypopharynx and into the subglottis via the fistula. (B) Suture ends are held externally. (C) Angiocath removed leaving suture in place transiting fistula.

FIGURE 4.

Subsequent operative steps. (A) Tracheoesophageal puncture (TEP) occluder threaded onto hypopharyngeal side of suture with Keith needle. (B) Occluder guided into position from hypopharynx and into the subglottis using the suture. Smaller flange of implant with suture visualized at hypopharyngeal opening before transiting fistula (top). Larger flange of implant now flush with hypopharyngeal opening (bottom). (C) Final implant position in subglottis after suture cut and removed. Smaller flange flush with subglottic opening.

2.3. Patient Outcomes

Patient 1 had their fistula plugged with a temporalis fascia graft. A 2 by 2 cm graft was harvested and rolled into a cylindrical shape with one side broader (hypopharyngeal) and the other tapered (subglottic). A single 4‐0 vicryl suture was tied onto the broader end to prevent the graft from unraveling during placement. Before the graft was pulled into position, the hypopharyngeal and subglottic sides of the fistula were gently cauterized. Then, as outlined in Step 5, the graft was pulled into place to occlude the fistula. Additional 4‐0 vicryl sutures were placed endoscopically to try and close some of the posterior cricoid mucosa over the subglottic side of the fistula to prevent graft dislodgement.

Swallow study obtained 2 months later for Patient 1 demonstrated a sealed fistula, and the patient tolerated a regular diet without any issues. Unfortunately, the patient was lost to follow‐up thereafter, so we do not have longer‐term data to report.

For Patient 2, we elected mechanical occlusion with a size 16 Fr/10 mm TEP occluder. A TEP occluder is similar in size and structure to a TEP, but instead of having a one‐way valve in the middle, it is completely closed off. The TEP occluder was chosen for this patient after a multidisciplinary discussion with our speech‐language pathology colleagues, as they use this device with a high rate of success for laryngectomy patients who develop a leak through their TEP site. Therefore, we felt this device may provide a durable solution to seal this patient's fistula. MBS the day after surgery showed the fistula was closed (Figure 5A). He resumed a regular diet in conjunction with G‐tube feeds, given underlying oropharyngeal dysphagia.

FIGURE 5.

Post‐operative imaging for Patient 2. (A) Modified barium swallow (MBS) study on post‐operative day 1 shows no evidence of leak through the fistula. (B) MBS 6 weeks post‐operatively shows very small amount of contrast entering the fistula. Sagittal view of neck computed tomography scan shows tracheoesophageal puncture (TEP) occluder in position transiting fistula.

Due to other medical comorbidities, Patient 2 was frequently admitted to the hospital. During an admission 6 weeks post‐operatively, an esophagram ordered by his medical team showed a very small amount of contrast transiting the fistula around (not through) the inferior aspect of the TEP occluder (Figure 5B). We offered to exchange the occluder to obtain a better seal, but since he was doing well clinically, the patient elected observation. Importantly, he had not been admitted due to aspiration pneumonia for over 9 months since the TEP occluder was placed.

3. Discussion

HSF should be considered for patients with aspiration following prolonged intubation, neck trauma, and laryngeal radiation or surgery [2, 5]. Diagnosis begins with MBS, and rigid laryngoscopy demonstrates the extent of destruction and fistula formation [4]. Cricoid chondronecrosis can also be seen on neck computed tomography scan [4].

Given the rare incidence, only a few reports document the management of HSF, mainly outlining open surgical interventions. Three studies described a lateral pharyngotomy approach with an interpositioned muscle flap to manage four patients with fistulas measuring up to 5 cm [1, 2, 3]. Another case of HSF, caused by penetrating trauma through the thyroid cartilage, utilized the trauma‐induced laryngofissure to access the fistula for open repair [5].

As a standard for HSF management does not exist, it is desirable to minimize risks of surgical intervention. Endoscopic approaches are less invasive and represent an emerging research topic for TEFs. Success with minimal complications or recurrence has been reported. Similar approaches for HSFs remain scarce, with only one study reporting endoscopic repair [4].

Dessard et al. describe a patient with a TEF secondary to cricoid necrosis (anatomically a HSF) following prolonged intubation in the presence of a nasogastric tube [4]. After failure to repair the defect with a strap muscle flap, an endoscopic approach using a porcine SIS plug was pursued. Using a guidewire, the plug was pulled from the hypopharynx and into the fistula without any fixation. The excess SIS plug in the airway required resection [4]. The patient resumed a regular diet 1 week post‐operatively. Of note, the authors indicate this plug is limited by fistula size and works best for long, narrow fistulas. This could lead to issues with plug stability depending on the fistula's dimensions.

Our endoscopic approach displays similarity to that reported by Dessard's group but with some important differences [4]. First, we guide the plug into the fistula with suture as opposed to guidewire, which provides more flexibility and introduces less trauma. There is also the advantage of customizing the plug to fistula size. The large surface area of temporalis fascia allows for a wide range of graft sizes to be utilized. TEP occluders, which are traditionally used for laryngectomized patients, come in various sizes and may increase the customizability for different HSFs. Although the approach described in this study carries key differences compared to Dessard et al., we do not have enough long‐term data to determine which procedure has a more durable outcome [4].

Certainly limitations exist for this study, principally the small sample size and shorter follow‐up period. Further studies will focus on the longevity of this intervention.

4. Conclusion

A HSF represents a rare complication of prolonged intubation, laryngeal radiation, trauma, or surgery that significantly impacts a patient's quality of life due to aspiration and the inability to consume a regular diet. The literature on this clinical entity, especially as it relates to endoscopic management, is quite limited. We feel the operative technique and case series presented here add important information to the literature regarding the endoscopic management of HSFs.

Disclosure

This work was presented as a poster presentation at the Fall Voice Conference, Phoenix, AZ, USA, October 24–26, 2024.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1 Supporting Information.

Kessler L., Jaamour D., Lakpa K. L., Bowen A., Francis D. O., and Stein A. P., “Endoscopic Management of a Hypopharyngeal‐Subglottic Fistula: Case Series and Literature Review,” The Laryngoscope 135, no. 11 (2025): 4312–4316, 10.1002/lary.32320.