227. CAUSES OF IN-HOSPITAL MORTALITY FOLLOWING TRANSTHORACIC ESOPHAGECTOMY FOR CANCER

Abstract Background Rates of peri-operative mortality following esophagectomy for cancer have markedly decreased over the last few decades, with a 90-day mortality rate below 5% considered a benchmark for quality of care. Although studies have focused on mortality rates, little is known about the causality. The aim of this study was to perform a multicenter, in-depth analysis of peri-operative mortality following esophagectomy for cancer. Methods Data was obtained multicenter, from prospective databases, between January 2010 – June 2020. All patients with a lethal postoperative course (Clavien-Dindo V) following elective transthoracic esophagectomy were included. Data collection included baseline characteristics, preoperative comorbidities, surgical procedures, postoperative complications and their management. In each participating center, two surgeons independently reassessed patient work-up and the management of intra-operative and postoperative complications to finally classify the management of each section as adequate, non-adequate or undetermined. Statistics encompassed descriptive analysis. Results One-hundred and twenty-one out of 3899 patients died in-hospital following oesophagectomy; a mortality rate of 3.1%. Patients deceased on a median of 32 days after surgery (IQR: 18-60). In preoperative work up: 24 patients (19.7%) had a Charlson-Comorbidity-Index (CCI) of 5 of higher; with cardiac comorbidity reported most often, in 39% of patients. Reassessment revealed work-up was not adequate in two patients (1.4%). During surgery, 6 patients (4.6%) suffered complications, all managed adequately in reassessment. Following surgery, a total of 343 major complications were identified in the 121 patients (mean 2.4 ± 1.2) with anastomotic leakage (AL) reported most often, in 65 patients (53.7%). AL was reported as cause of death in 44 patients (36.5 %). In eight patients (6.6 %) postoperative management was reassessed to be non-adequate, in these cases recognition and initiation of treatment for AL (n=7) or conduit necrosis (n=1) was delayed. Conclustion In contrast to other published series, reassessment of this homogenous patient cohort with a lethal clinical course in expert centers revealed only a low rate of preventable mortality with respect to the preoperative patient selection and postoperative complication management. However, anastomotic leakage was identified to be the leading cause of death following esophagectomy contributing to a significantly prolonged clinical course and lethal outcome. As the development of surgical procedures with complex anastomotic techniques is inevitably associated with an increasing leakage rate expressed by prolonged learning curves, modification of complication management might be considered in order to reduce overall death rates.