PS01.167: MUMELE STUDY: MULTICENTER STUDY ON INCIDENCE OF MEDIASTINAL LEAKS AFTER ESOPHAGECTOMY

Abstract Background Mediastinal anastomotic leak (ML) represent one of the most feared complication of esophageal resection. The incidence of ML, and of the associated mortality rate and the treatment strategy are variously reported. A standard strategy for diagnosis and treatment is difficult to establish Methods In order to evaluate the incidence, predictive factors, treatment and mortality of ML in 7 Italian surgical centers with interest in esophageal surgery (5 high volume centers) a retrospective study including all esophagectomies (E) with intrathoracic esophagogastric anastomosis performed in a 3 year period (2014–2017), was planned. ML were defined according to the classification proposed by the Esophagectomy Complications Consensus Group. Results The data of 501 E were collected. Overall incidence of ML was 11.8%. Surgical approach significantly influenced the rate of ML: leakage rate was highest for totally minimally invasive (TMIE) and lowest for hybrid esophagectomy (HE) (respectively 20 and 9%). No other predictive factor was found. Overall 30 and 90 day (d) mortality rate (M) were respectively 1,4% and 3,2%; 30 and 90 d M for leaks were respectively 5% and 15,3%; 90 d M for TMIE and HE were 5,9% and 1,8% respectively. Endoscopy was the first line treatment in 49% of leaks, with a need for retreatment in 17,2% of cases. Surgery was needed globally in 44,1% of ML. Endoscopic treatment appeared to have the lowest M (6,9%). Removal of the gastric tube with stoma formation was necessary in 8 cases (13.6%). Conclusion The incidence of mediastinal leaks after esophagectomy in the MUMELE study is high mainly in TMIE group. General and specific (leak) mortality rate is however low. Early ‘aggressive’ treatment of severe leaks is mandatory, with no hesitation to redo surgery if first attempt of conservative management fails. Disclosure All authors have declared no conflicts of interest.