Efficacy of endoscopic management of leak after foregut surgery with endoscopic covered self-expanding metal stents (SEMS)

Anastomotic or staple-line leak after foregut surgery presents a formidable management challenge. In recent years, with advancement of endoscopy, self-expanding covered stents have been gaining popularity. In this study, we aimed to determine the safety and effectiveness of self-expanding covered stents in management of leak after foregut surgery. Consecutive patients who received a fully covered self-expandable metal stent (SEMS) due to an anastomotic leak after upper gastrointestinal surgery between 2009 and 2014 were retrospectively reviewed. Demographic data, stent placement and removal, clinical success, time to resolution, and complications were collected. Predictive factors for clinical success rate were assessed. A total of 20 consecutive patients underwent placement of fully covered SEMS for anastomotic leak, following esophagectomy (n = 5), esophageal diverticulectomy (n = 1), gastric sleeve (n = 4), gastric bypass (n = 3), partial gastrectomy (n = 4), and total gastrectomy (n = 3). All the stents were removed successfully, and clinical resolution was achieved in 18 patients (90 %) after a median of two (range 1-3) procedures and a mean of 6.2 weeks (range 0.4-14). Complications presented in 12 patients (60 %), including stent migration (n = 8), mucosal friability (n = 4), tissue integration (n = 2), and bleeding (n = 2). Two (10 %) patients' treatment was complicated by aorto-esophageal fistula formation resulting in one death. Demographic factors, comorbidities, and type of surgery were not predictive of clinical success rate or time to resolution. SEMS are effective tools for the management of leaks after foregut surgery. The biggest challenge with this approach is stent migration. Caution is warranted due to the risk of fatal complications such as aorto-esophageal fistula formation. No type of surgery or particular patient factor, including age, sex, BMI, albumin, history of radiation, malignancy, and comorbid diabetes or coronary artery disease, appeared to be correlated with success rate. Larger studies are needed to determine factors predictive of clinical success.