Endoscopic management of refractory gastrointestinal–tracheobronchial fistulas with two novel occluders: a comparative cohort study

Background: Refractory gastrointestinal (GI)–tracheobronchial fistulas are challenging to manage. This study compared the outcomes of two novel occluders specifically designed for endoscopic occlusion of GI–tracheobronchial fistulas. Methods: In this retrospective cohort study, 86 patients with refractory GI–tracheobronchial fistulas who underwent endoscopic occlusion using either a double umbrella-shaped (DU; n = 36) or mushroom umbrella-shaped (MU; n = 50) occluder were included. The primary outcome was sustained occlusion. Secondary outcomes included technical success, adverse events (AEs), occluder-related discomfort, and percentage weight gain. Results: The DU occluder was associated with a higher risk of occlusion failure compared with the MU occluder (hazard ratio [HR] 2.67, 95%CI 1.24–5.73; P = 0.01), with a 12-month cumulative sustained occlusion probability of 65.5% (95%CI 51.3%–83.5%) versus 82.1% (95%CI 71.4%–94.3%), respectively. Esophageal wall injury requiring occluder removal occurred in six DU patients, but in none of the MU group. Technical success was 100% in both groups. All procedure-related AEs were mild, occurring in five DU (13.9%) and three MU patients (6.0%). Conclusions: The MU occluder showed greater potential for sustained occlusion than the DU occluder and may be a viable option for refractory GI–tracheobronchial fistulas, although further prospective studies are needed.