Endoscopic magnetic gastroenteric anastomosis for palliation of malignant gastric outlet obstruction: a prospective multicenter study

Abstract

Background: Palliation of malignant gastric outlet obstruction remains challenging. Although there are 2 established treatment options, ie, surgical gastrojejunostomy and endoscopic duodenal stent insertion, there is an ongoing search for a technique that would combine the safety and rapid effect of duodenal stent placement with the long-term efficacy and low reintervention rate of a surgical gastrojejunostomy.

Objective: To investigate the safety and success rate of endoscopic creation of a gastroenteric anastomosis formed by magnetic compression and stent placement.

Design: Prospective, multicenter cohort study.

Setting: Four referral centers.

Patients: The expected number of patients with symptomatic malignant gastric outlet obstruction to be included at the participating hospitals during a year was 40. Because of a serious adverse device event, the study was terminated after inclusion of 18 patients.

Intervention: Creation of an endoscopic gastroenteric anastomosis by using the Cook Magnetic Anastomosis Device with transanastomotic deployment of a self-expandable stent.

Main outcome measurements: Primary endpoints were safety and success rate associated with the creation of an endoscopic gastrojejunostomy by using a magnetic anastomotic device with transanastomotic deployment of a self-expandable stent.

Results: Because of a serious adverse event, the study was terminated prematurely. A success rate of 66.7% (12 of 18 patients) was achieved; 1 serious adverse event (stent perforation) occurred leading to the death of the patient. Three patients (25%) experienced an adverse device effect (stent migration).

Limitations: Small sample size, lack of a control group.

Conclusion: Endoscopic creation of a gastroenteric anastomosis by magnetic compression is feasible and safe; however, the necessity of a stent led to serious morbidity and even mortality in this study. The current system can therefore not be recommended for clinical use.