Stanislaus Reimer, Johan F. Lock, Sven Flemming, Alexander Weich, Anna Widder, Lars Plaßmeier, Anna Döring, Ilona Hering, Mohammed K. Hankir, Alexander Meining, Christoph-Thomas Germer, Kaja Groneberg, Florian Seyfried

• Background Endoscopic vacuum therapy (EVT) is an evidence-based option to treat anastomotic leakages of the upper gastrointestinal (GI) tract, but the technical challenges and clinical outcomes of patients with large defects remain poorly described. Methods All patients with leakages of the upper GI tract that were treated with endoscopic negative pressure therapy at our institution from 2012–2021 were analyzed. Patients with large defects (>30 mm) as an indicator of complex treatment were compared to patients with smaller defects (controlBackground Endoscopic vacuum therapy (EVT) is an evidence-based option to treat anastomotic leakages of the upper gastrointestinal (GI) tract, but the technical challenges and clinical outcomes of patients with large defects remain poorly described. Methods All patients with leakages of the upper GI tract that were treated with endoscopic negative pressure therapy at our institution from 2012–2021 were analyzed. Patients with large defects (>30 mm) as an indicator of complex treatment were compared to patients with smaller defects (control group). Results Ninety-two patients with postoperative anastomotic or staplerline leakages were identified, of whom 20 (21.7%) had large defects. Compared to the control group, these patients required prolonged therapy (42 vs. 14 days, p < 0.001) and hospital stay (63 vs. 26 days, p < 0.001) and developed significantly more septic complications (40 vs. 17.6%, p = 0.027.) which often necessitated additional endoscopic and/or surgical/interventional treatments (45 vs. 17.4%, p = 0.007.) Nevertheless, a resolution of leakages was achieved in 80% of patients with large defects, which was similar compared to the control group (p = 0.42). Multiple leakages, especially on the opposite side, along with other local unfavorable conditions, such as foreign material mass, limited access to the defect or extensive necrosis occurred significantly more often in cases with large defects (p < 0.001). Conclusions Overall, our study confirms that EVT for leakages even from large defects of the upper GI tract is feasible in most cases but comes with significant technical challenges.…