Physical barriers or mechanical closure – how to seal the deal for upper GI fistulas?

10.1055/a-2675-1616

Once it was the exclusive realm of surgeons to close pathological perforations or the occasional injudicious endoscopic intervention, but closure of gastrointestinal (GI) mural defects is now commonplace for the endoscopist. A dizzying array of tools have become available to the interventional endoscopist to either close or occlude defects, including mechanical clipping devices, endoscopic suturing applications, endoluminal stents, and vacuum suction therapies [1].

“One would have to be very concerned about leaving such a device in place for years, owing to the risk of local tissue trauma.”

Complete mechanical closure of an acute defect, either deliberately or inadvertently created, will usually seal the deal; however, a fistula, which is a mature tract linking two epithelialized structures, represents a more difficult problem. In adults, the commonest causes of fistulas in the upper GI tract are postoperative or postendoscopic intervention, malignancy, or radiotherapy. Less commonly, Crohn’s disease, infections, foreign bodies, and corrosive substances can lead to such problems [2]. Whatever the cause, fistulas to the tracheobronchial tree can be life-changing and life-shortening. Attempts to close such defects are challenging, with perhaps the most successful approach being endoscopic suturing, although this is associated with modest long-term closure rates [3] [4].

In this edition of Endoscopy, Drs. Qian and colleagues present an endoscopic device for occluding refractory GI to tracheobronchial fistulas [5]. The technique is wonderful in its simplicity – a nitinol dumbbell-shaped device, loaded onto a catheter, is passed through the defect and the overlying sheath is withdrawn, so the device resumes its original shape, with nitinol discs occluding each side of the defect. Sounds familiar? Yes, these are much the same as lumen-apposing metal stents (LAMSs), with the exception that the central waist is occluded and the discs have an anticorrosion coating and silicone membrane, presumably to maintain their integrity over time.

The authors have compared two devices with similar shape but subtly different designs. The first has two umbrella-shaped discs (double umbrella [DU]) and the second has a mushroom-shaped disc on the GI side (mushroom umbrella [MU]) that is proposed to cause less mucosal abrasion. Similar devices have been subject to previous case reports and the concept is an adaptation of cardiac septal defect closure devices, which have been reported in the GI literature for fistula closure since before 2012 [6] [7].

The study reports 86 consecutive patients (36 DU, 50 MU) with upper GI to airway fistulas treated between September 2020 and June 2024, with the design being switched midway through the study period. All patients had established chronic airway fistulas from the esophagus, intrathoracic stomach, or surgical anastomosis, with a median duration of 5.5 (DU) and 12 months (MU). The underlying etiology included the usual array of causes, although cases with active malignant disease were notably absent. These were refractory cases – patients had had a median of two prior attempts at closure – although only 64% and 42%, respectively, had had prior endoscopic suturing.

Owing to the elegant simplicity of the design, technical success was 100%, with a procedure time of 26 minutes and median hospital stay of 3 days, with few acute adverse events. Follow-up was for a median of 31 and 18 months, respectively, with the commonest side effect being a foreign body sensation, experienced by a third of patients. Sustained occlusion (12 months) was achieved in 65.5% and 82.1% of patients, respectively, the difference primarily being due to increased esophageal wall injury in the DU cases, which required removal of the device. Notwithstanding the innate bias from the study design, the mushroom-shaped device does appear to offer a preferable upgrade.

Given that this was a group of refractory fistulas, this is a remarkable report and a potential step-change for these patients. The treatment is simple, safe, and achieves good outcomes. Is this too good to be true or is there a drawback? The main concern here is that the reason that many of the patients have a foreign body sensation is simple – there is indeed a permanent foreign body present in their upper GI tract and airway. The question therefore arises about the safety and longevity of such a design in benign disease. There does not appear to be any suggestion that these devices would promote long-term healing allowing for removal. As a result, one would have to be very concerned about leaving such a device in place for years, owing to the risk of local tissue trauma.

It therefore appears that these devices could fulfil an important role in the treatment of patients with limited life expectancy or provide short-term symptom control prior to more definitive therapy. In addition, they could potentially be used as a simple occlusion device for a nonfistulating defect where external drainage of sepsis has been achieved. There may also be a role for these devices in the closure of malignant fistulas, although this would require testing in the clinical setting first; however, long-term open-ended use in benign disease would require further evaluation. This is a welcome new option that hopefully will have greater availability in the not-too-distant future. It has the benefit of simplicity and short-term safety but, as with all interventional endoscopy, case selection, consideration of the expanding available options, and long-term planning will all be paramount.

Publication History

Article published online:
02 September 2025

• References

• 1 Papaefthymiou A, Norton B, Telese A. et al. Endoscopic suturing and clipping devices for defects in the GI tract. Best Pract Res Clin Gastroenterol 2024; 70: 1-8
• 2 Bhurwal A, Mutneja H, Tawadross A. et al. Gastrointestinal fistula endoscopic closure techniques. Ann Gastroenterol 2020; 33: 1-9
• 3 Papaefthymiou A, Aslam N, Norton B. et al. Endoscopic suturing for defect closure in the upper gastrointestinal tract: A retrospective cohort study. Gastrointest Disord 2025; 7: 1-13
• 4 Mukewar S, Kumar N, Catalano M. Safety and efficacy of fistula closure by endoscopic suturing: a multi-center study. Endoscopy 2016; 48: 1023-1028
• 5 Qian H, Sang H, Wang Y. et al. Endoscopic management of refractory gastrointestinal–tracheobronchial fistulas with two novel occluders: a comparative cohort study. Endoscopy 2025;
• 6 Zhu C, Li L, Wang Y. et al. Endoscopic closure of tracheoesophageal fistula with a novel dumbbell-shaped occluder. Endoscopy 2022; 54: E334-E335
• 7 Banerjee S, Barth B, Bhat Y. et al. Endoscopic closure devices. Communication from the ASGE Technology Committee. Gastrointest Endosc 2012; 76: 244-251