Comparative Study of Predictive Mortality Scores in Esophagectomy with Three-Field Lymph Node Dissection in Patients with Esophageal Cancer

Background/Aim: Esophagectomy is still the best therapeutic option for curing resectable esophageal cancer (EC). Radical surgical resection with three-field lymphadenectomy (3FLD) is a potentially curative treatment option. We compared the predictive accuracy of 5 different scores in patients with EC who underwent 3FLD. Methods: Five years' worth of medical records in a single institution were analyzed (January 2010 to January 2015) from 311 patients who underwent esophagectomy for EC. We selected 191 in whom 3FLD was performed. Mortality was calculated based on 5 predictive scores. Outcomes measures were intraoperative mortality, 30-day mortality, and 1- and 2-year mortality after surgery. Results: Intraoperative mortality and 30-day mortality after surgery was 0%; 1 and 2-year mortality were 19.8 and 31.4%, respectively. The area under the curve showed poor discriminatory power for all 5 scores (<0.7). In one-way analysis of variance, for 1 year mortality, Portsmouth-Physiological and Operative Severity Score for mortality (P-Possum) was significant (p = 0.0424); in a multivariable analysis for 2-year mortality, P-Possum (p < 0.0001) remained significant. Conclusion: There is no accurate prognosis score for esophagectomy in patients who undergo high-risk procedures like 3FLD. New scores are needed to predict the mortality after 3FLD with good discriminatory power. Independent factors affect survival and may function as the baseline for obtaining a new accurate mortality score. (C) 2018 S. Karger AG, Basel