Esophageal cancer represents a high-risk group of patients. This study determines the association of artificial nutrition with morbidity, mortality, and survival and studies clinical situations that determine the choice between enteral (EN) and parenteral support (PN). This retrospective single-center study compared 2 periods: 1) treatment centered in surgical process with discretionary demand of support, and 2) elective therapeutic and nutritional interventions were systematized. Risks factors that determined use of PN and survival were included in 4 multivariate regression models: 2 logistic, 1 multinomial, and a survival Cox analysis. Significance determined with 95% confidence interval (CI) of 95%; inclusion criteria was P < 0.1. During an 11-yr period, 175 patients were studied. Artificial nutrition consisted of 45 jejunostomy EN, 28 PN, and 102 both. Risk factors that conditioned PN were first period (OR: 2.41; 95% CI: 1.13-5.14), stay in intensive care unit (ICU) >3 days (OR: 1.70; 95% CI: 0.93-3.71), and surgical reintervention (OR: 3.83; 95% CI: 0.94-16.95). Risk factors associated with mortality were first period (OR: 22.7; 95% CI: 2.31-172.05), respiratory infection (OR: 11.23; 95% CI: 2.33-55.5) and coloplasty surgery (OR: 13.16; 95% CI: 2.11-83.33). Longer survival was associated with second period (OR: 2.36; 95% CI: 1.38-4.05) and lower neoplasm staging (OR: 1.43; 95% CI: 1.21-1.69). A multidisciplinary management that includes nutritional support of esophagectomized patients is 1 of the factors that improves survival. Protocol implies greater use of EN; PN remains an important nutritional therapy.