Related Experiment Video
Updated: Jul 4, 2026

Modified Single-Loop Reconstruction for Pancreaticoduodenectomy
Published on: September 28, 2019
Case Report: Endoscopic geometric remodeling for angulation-type efferent loop stenosis after pancreaticoduodenectomy
Donghai Wu1, Shihua Ding1, Na Li1
1The First Affiliated Hospital of Shenzhen University, Shenzhen Second People's Hospital, Shenzhen, China.
Background:
Efferent loop stenosis (ELS) after pancreaticoduodenectomy (PD) with gastrojejunostomy reconstruction is an uncommon cause of delayed gastric emptying. Reoperation in the early postoperative period carries substantial risk. Endoscopic balloon dilation and stenting are often used for benign postoperative strictures, but when fixed angulation rather than a short concentric narrowing is the main problem, these approaches may be unsafe or ineffective.
Case Description:
A 51-year-old woman with a more than 2-year history of intermittent epigastric fullness and pain underwent laparoscopic PD with Child reconstruction for an approximately 2 cm pancreatic head mass. Imaging suggested a pancreatic neuroendocrine tumor and histopathology confirmed a grade 2 pancreatic neuroendocrine tumor (Ki-67 about 5%) with negative margins. The early postoperative course was initially stable. On postoperative day 7, she developed upper abdominal distension and high-output bilious nasogastric drainage (800-1,100 mL/day), and oral intake could not be advanced. Conservative treatment with nasogastric decompression, bowel rest, intravenous fluids, proton-pump inhibitor therapy, and prokinetic agents was continued for five days but did not improve the symptoms. A barium meal examination showed delayed gastric emptying with impaired flow into the efferent limb. Upper endoscopy demonstrated viscous bile pooling in the stomach and a sharply angulated efferent loop at the gastrojejunostomy, without a discrete anastomotic ring stricture. These findings were consistent with ELS caused mainly by loop angulation. Because stable coaxial positioning of a balloon catheter across the kinked segment was considered difficult and potentially unsafe, we chose a combined strategy. Endoscopic submucosal dissection (ESD) was used to partially remove a saddle-shaped mucosal ridge at the efferent-loop entrance. Clips were applied for hemostasis and to provide traction, straightening the efferent loop and improving alignment with the gastric lumen. A 20 mm × 80 mm fully covered self-expanding metal stent (FCSEMS) was then placed to maintain patency. After the procedure, the patient's abdominal distension improved, nasogastric drainage decreased, and the tube was removed on day 3. Oral intake was gradually resumed. At 1 month, a contrast study and repeat endoscopy showed good passage through a straightened efferent limb, and the FCSEMS was removed without complications. At 3 months, she reported good appetite, a weight gain of about 5 kg, and endoscopy showed a widely patent efferent loop without recurrent stenosis.
Conclusion:
ESD-assisted mucosal release with clip-based traction and temporary FCSEMS placement corrected angulation-type ELS early after PD in this patient. This endoscopic approach, which focuses on modifying the loop geometry rather than simple radial dilation, may be considered as an alternative to surgical revision in carefully selected cases where standard balloon dilation is not feasible.
