We have read with interest the review by Lee et al.1 titled "Endoscopic treatments of endoscopic retrograde cholangiopancreatography-related duodenal perforations." Herein, we would like to describe the management of an analogous case from among our experience of 3,000 endoscopic retrograde cholangiopancreatographies (ERCPs) in 8 years.

An 84-year-old man underwent ERCP because of recurrent cholangitis with intrahepatic bile duct dilatation, as seen on MRCP. The patient underwent a Billroth II gastrojejunostomy for the treatment of gastroduodenal ulcer disease 30 years earlier. During ERCP, multiple attempts to introduce the scope were made because of difficulty in approaching the papilla. An endoscope-related perforation was visualized at the end of the afferent loop (Fig. 1). The diameter of the defect was approximately 20 mm, and we first contemplated nonoperative closure with an over-the-scope clip (OTSC) device.1 Despite the existence of published data2,3 reporting successful endoscopic closure of large duodenal perforations by OTSC, we felt that it was necessary to perform a computed tomography (CT) scan to rule out the presence of any substantial intraperitoneal fluid collection. The abdominal CT was performed 30 minutes after the recognition of the perforation, and demonstrated pneumoperitoneum with fluid collection around the afferent loop (Fig. 2). All fluids previously detected in the afferent loop had been carefully suctioned at the time of insertion, although type I duodenal perforations caused by the endoscope tend to be large with persistent fluid leaks in the retroperitoneal or intraperitoneal space.4 Therefore, we decided that, despite the patient's advanced age, an operation would be preferable to the application of an OTSC device in this case. Laparotomy revealed a peritoneal cavity full of bilious fluid. The patient underwent a thorough washout of the abdominal cavity, surgical closure of the defect, and drainage. Recovery was uneventful and he was discharged 10 days later.

On the basis of the findings of the present case, we would like to suggest that OTSC application could be considered an optimal treatment for duodenal perforations in inoperable patients or in patients who are not septic and have minimal peritoneal fluid collection. However, we believe that in cases involving a septic patient with intraperitoneal fluid collection, an endoscopic closure is probably unsuitable. Prompt surgical intervention with washout, closure of the perforation, and drainage is crucial for achieving recovery without sepsis or abscess formation.