Background: Endoscopic placement of self-expandable metallic stents for palliation of patients with malignant gastric outlet obstruction is safe and feasible.
Methods: Patients with malignant gastric outlet obstruction undergoing enteral stent insertion were identified from endoscopy databases. Duration of oral intake after stent insertion was calculated by using the log-rank test. Factors associated with duration of oral intake were assessed by using Cox multivariable regression analysis.
Results: A total of 176 patients (mean age 65  years) treated at 4 centers from 1996 to 2003 were identified. Obstruction was caused by cancer of the pancreas in 84, the stomach in 20, the bile duct in 15, the major duodenal papilla in 8, another primary site in 16, and metastases in 33. The site of obstruction was the duodenum in 125, the distal stomach in 17, the stomach and the duodenum in 18, and surgical anastomosis in 16 patients. Stent deployment was technically successful in 173. Complications occurred in 14 patients. Seventeen patients were lost to follow-up. Of the remaining 159 patients, 133 resumed oral intake for a median time of 146 days: 95% CI [65, 202]. On regression analysis, chemotherapy after stent placement was associated with prolonged duration of oral intake (hazard ratio 0.41: 95% CI [0.23, 0.72]).
Conclusions: After enteral stent insertion for malignant gastric outlet obstruction, 84% of patients resumed oral intake for a median time of 146 days. Chemotherapy after enteral stent insertion was independently associated with prolongation of oral intake.