The optimal timing of enteral nutrition (EN) initiation in predicted severe acute gallstone pancreatitis (SAGP) and its influence on disease outcomes are not well known. We conducted a retrospective study of patients with predicted SAGP treated with endoscopic retrograde cholangiopancreatography and EN. The patients were classified into two groups according to the timing of EN initiation after admission: within 48 h, and more than 48 h. The primary outcome was in-hospital mortality. The secondary outcomes were length of hospital stay, need for intensive care admission, need for surgical intervention, improvements in blood test results after 7-10 days of EN, incidence of pancreatic necrosis and infection, and hospital care costs. The microbiological profiles of infectious complications were also evaluated. Of the 98 patients, 31 and 67 started EN within 48 h, and more than 48 h after admission, respectively. Early EN was associated with a decrease in in-hospital mortality (0 vs. 11.9%; p=0.045), length of hospital stay (median:18 vs. 27 days; p=0.001), need for intensive care admission (3.2% vs. 20.9%; p=0.032), and hospital care costs (median:9,289 vs. 13,518 US$; p=0.007), compared to delayed EN. Moreover, early EN for 7-10 days had more beneficial effects on blood test results than delayed EN, including total protein (p=0.03) and CRP (p=0.006) levels. However, the need for surgical intervention and incidence of pancreatic necrosis did not differ between the two groups. In our study, Gram-negative bacteria were the main responsible pathogens (50.5%). Infection with multidrug-resistant organisms (MDRO) was found in 19.4% of the patients. The most common MDRO was MDR Enterococcus faecium. Early EN was not superior in reducing incidence of infected pancreatic necrosis, bacteremia, polymicrobial infection, or MDROs. In patients with predicted SAGP, early EN is associated with a decrease in in-hospital mortality, length of hospital stay, need of intensive care admission, and hospital care costs, compared to delayed EN. There are no significant benefits of early EN in reducing the rate of infection-related complications. Further studies with larger sample sizes are warranted.