Covered self-expanding stent treatment for anastomotic leakage: outcomes in esophagogastric and esophagojejunal anastomoses.

BACKGROUND: Anastomotic leakage of esophagogastric and esophagojejunal anastomoses is a severe complication after esophagectomy and gastrectomy associated with a high mortality. We conducted this non-randomized observational study to evaluate the outcomes and clinical effectiveness of covered self-expanding stents (CSESs) in treating esophageal anastomotic leakage. METHODS: From 2002 to 2013, consecutive patients with anastomotic leakage after esophagogastrostomy or esophagojejunostomy who received CSESs were analyzed concerning leakage characteristics, leakage sealing rate, success and failure rates of CSES treatment, stent-related complications, and mortality. RESULTS: In 35 patients, anastomotic leakage originating from 5 cervical esophagogastrostomies, 6 thoracic esophagogastrostomies, 12 mediastinal esophagojejunostomies and 12 abdominal esophagojejunostomies were treated with 48 CSESs (16 fully CSES, 32 partially CSES). Of 35 patients, 24 received one stent, 9 received two consecutive stents, and 2 received three consecutive stents. Stent-related complications occurred in 71 % of patients (25/35). The most frequent complications were leakage persistence (44 %) and stent dislocation (19 %). Sealing of the anastomotic leakage was achieved in 24 (69 %) patients after a median (range) stenting time of 19 (1-78) days. Sealing rates differed significantly with 20 % (cervical esophagogastrostomies), 50 % (thoracic esophagogastrostomies), 92 % (mediastinal esophagojejunostomies) and 67 % (abdominal esophagojejunostomies) of patients (p = 0.023). Moreover, clinical success rates differed among these groups (60 vs. 67 vs. 92 vs. 58 %; p = 0.247). Clinical failure of stent treatment was more likely to be recognized in early postoperative leakage (median postoperative day 3 vs. 8; p = 0.098) compared with successful treatment, whereas no difference for clinical success rates was found comparing leakage ≤ 10 versus >10 mm (68 vs. 64 %; p = 0.479). CONCLUSION: CSESs are an effective treatment for anastomotic leakage in patients with esophagogastrostomies and esophagojejunostomies. Best results can be achieved in patients with anastomotic leakages following mediastinal esophagojejunostomy, and in leakages occurring after the very early postoperative phase.