Introduction: Nutritional support in severe acute pancreatitis has been controversial according to outcomes. Our objective is to describe results of length of hospital stay, costs and complications according to nutritional support (enteral, parenteral and mixed) in patients with severe acute pancreatitis. Methods: An observational descriptive study (Case Series) was made in a sample of 45 patients with severe acute pancreatitis who were admitted to Hospital Pablo Tobon Uribe between 1999 and 2004. Statistical analysis was made by the Chi2 test for proportions comparison, the Kuskall Wallis test and the Mann Whitney’s U for comparison between three and two groups respectively. Results: The principal cause of pancreatitis was gallstones (49%). Enteral nutrition was the main nutritional support (48,8%). Comparison between total parenteral nutrition and enteral nutrition showed statistically significant difference in hyperglycemia (p=0.0001) and length of hospital stay (p=0.047). There were no statistically significant differences in general complications (p=0.053), or infectious complications (p=0.136). Mixed nutrition had similar results as total parenteral nutrition in hyperglycemia, costs and length of hospital stay. No mortality was registered in this study. Conclusion: With EN there was better control of glycemia, and a reduced length of hospital stay. We also describe the importance of mixed nutrition in patients unable to use just enteral feeding.