Transparent cap

All patients suspected of having ingested foreign bodies underwent routine radiologic and laryngologic examination to exclude oropharyngeal foreign bodies. Patients also underwent computed tomography scan of the chest, depending on the clinical setting and at the discretion of the attending physician, before or after the endoscopic procedure. Endoscopic foreign body removal was initially attempted using a flexible endoscope (GIF-H260 or GIF-Q260; Olympus Optical Co., Ltd., Tokyo, Japan). All patients were examined by board-certified endoscopists under local pharyngeal anesthesia with lidocaine. Because of the risk of aspiration, patients were not placed under conscious sedation. Vital signs, including blood pressure, heart rate, and oxygen saturation, were monitored continually throughout the procedure. Accessory devices used to remove foreign bodies included standard biopsy forceps, rat-tooth forceps, alligator forceps (FG-47 L-1; Olympus), a retrieval basket (MTW Endoskopie, Wesel, Germany), or a snare (MTW Endoskopie). Additional protective measures, such as a latex protector hood (DIAGMED, Thirsk, England), an overtube (TS-12,140 or TS-13,140; Fujinon, Saitama, Japan), or a transparent cap (Olympus), were occasionally used to prevent damage to the gastrointestinal tract during the removal of sharp or pointed foreign bodies.

We performed colorectal ESD in the endoscopy room under intravenous anesthesia using midazolam and pethidine. Scopolamine was used as an antispasmodic drug; however, glucagon was substituted in patients with cardiac disease or benign prostatic hyperplasia. All ESD procedures were performed using therapeutic endoscopes (PCF-H290ZI or GIF-H290T; Olympus, Medical Systems Co., Tokyo, Japan) with a transparent cap (Olympus). Generally, PCF-H290ZI is used for colorectal ESD and GIF-H290T is often used for rectal lesions. We used a 1.5-mm DualKnife J (KD655Q; Olympus) to perform the mucosal incision or submucosal dissection. Using an electrosurgical generator (VIO 3; ERBE Elektromedizin, Tübingen, Germany), the endoCUT I mode (effect 1 or 2, duration 2, interval 2) was used for the mucosal incision, forced coagulation (effect 4.5 or 6.1) was used for the submucosal dissection, and spray coagulation (effect 1.2) was used for hemostasis. To elevate the submucosa, a local injection of 0.4% sodium hyaluronate (MucoUp^®; Boston Scientific, Tokyo, Japan) and a small amount of indigo carmine were administered.