Original scoring system for predicting postoperative morbidity after esophagectomy for esophageal cancer

Abstract

Purpose: Postoperative morbidity remains common after esophagectomy. There are currently few tools that can be used to estimate the incidence of complications. This study aimed to create a novel scoring system for predicting postoperative morbidity.

Methods: A total of 365 patients who underwent elective esophagectomy with two- or three-field lymphadenectomy for esophageal cancer were retrospectively analyzed. Patients were divided into development (n = 250) and validation datasets (n = 115). A scoring system was established by specifying the expected morbidity incidence values for all independent predictors in the development dataset, after which, the reproducibility was confirmed in the validation dataset.

Results: The incidence of any morbidity was 44.7 %. Male sex, severe emaciation (body mass index <18), a lower performance status (grade 1 or 2), operation ≥540 min and massive bleeding (blood loss/body weight ≥20) were independent predictors of postoperative morbidity. The total of all predictive scores was calculated. The incidence of any morbidity and of severe morbidity (Clavien-Dindo classification ≥IIIb) significantly correlated with the score after dividing the scores into three groups (≤8, 10 and 11 and ≥13). This finding was also confirmed in the validation dataset.

Conclusion: The current scoring system is considered to be useful for predicting postoperative morbidity after esophagectomy.