Duodenal perforation is a rare but serious complication of digestive endoscopy, associated with significant morbidity and mortality.

Traditionally treated surgically, its management now tends towards endoscopic approaches thanks to the development of advanced closure devices. Nevertheless, specific data on duodenal perforations remain limited and prognostic factors poorly defined.

We conducted a single-centre retrospective study including all patients who presented with iatrogenic duodenal perforation of type Stapfer I or II during diagnostic or therapeutic endoscopy between November 2004 and December 2023. Demographic, clinical, endoscopic and evolutionary data were collected. The primary endpoint was the clinical success of endoscopic treatment, defined as complete healing without the need for surgery. Secondary endpoints included initial technical success, clinical success after a single procedure, and the impact of the time to diagnosis (immediate vs. delayed) and the type of endoscopy (diagnostic vs. therapeutic).

Thirty-five patients (mean age 65.5 ± 12.6 years; 51.4% women) were included. Perforation occurred during ERCP (40%), EUS (31.4%),

EGD (20%) or duodenoscopy (8.6%). The mechanism was related to a therapeutic manoeuvre in 68.6% of cases, and the most common location was the second duodenum (62.9%), of which 72.7% were periampullary. Perforation was identified intra-procedurally in 82.9% of cases. The initial technical success rate of endoscopic treatment was 90.3%, reaching 100% after repeat procedures. Overall clinical success was achieved in 71.4% of cases, including 60% after a single session, with a median healing time of 5 days [2–107]. Salvage surgery was necessary in 28.6% of patients. Overall mortality was 8.6%. Immediate diagnosis was associated with better outcomes: more frequent clinical success after one procedure (69.0% vs. 16.7%; p = 0.028), shorter hospital stay (12.0 ± 14.1 vs. 79.2 ± 60.9 days; p = 0.001) and fewer intensive care admissions (10.3% vs. 50%; p = 0.049). Perforations occurring during endoscopy were associated with a higher rate of reoperation (34.7% vs. 16.7%; p = 0.001) and a longer hospital stay (12.0 ± 14.1 vs. 79.2 ± 60.9 days; p = 0.001).

0.001) and less need for intensive care (10.3% vs. 50%; p = 0.049). Perforations occurring during therapeutic endoscopy required more endoscopic procedures (3.3 vs. 1.3; p = 0.048).

Endoscopy enables effective management of iatrogenic duodenal perforations, with clinical success in over 70% of cases and a reduction in the need for surgery. Immediate recognition of the perforation is a major prognostic factor, significantly improving the outcome. The development of advanced closure devices reinforces the role of the endoscopic approach, but comparative studies are needed to define the optimal strategy and choice of equipment.