Objective: To determine if early enteral feeding, in an intensive care unit (ICU) patient population, using a formula supplemented with arginine, dietary nucleotides, and fish oil (Impact), results in a shorter hospital stay and a reduced frequency of infectious complications, when compared with feeding a common use enteral formula (Osmolite.HN).
Design: A prospective, randomized, double-blind, multicenter trial.
Setting: ICUs in eight different hospitals.
Patients: Of 326 patients enrolled in the study, 296 patients were eligible for analysis. They were admitted to the ICU after an event such as trauma, surgery, or sepsis, and met a risk assessment screen (Acute Physiology and Chronic Health Evaluation II [APACHE II] score of > or = 10, or a Therapeutic Intervention Scoring System score of > or = 20) and study eligibility requirements. Patients were stratified by age (< 60 or > or = 60 yrs of age) and disease (septic or systemic inflammatory response syndrome).
Interventions: Patients were enrolled and full-strength tube feedings were initiated within 48 hrs of the study entry event. Enteral feedings were advanced to a target volume of 60 mL/hr by 96 hrs of the event. One hundred sixty-eight patients were randomized to receive the experimental formula, and 158 patients were randomized to receive the common use control formula.
Measurements and main results: Both groups tolerated early enteral feeding well, and the frequency of tube feeding-related complications was low. There were no significant differences in nitrogen balance between groups on study days 4 and 7. Patients receiving the experimental formula had a significant (p = .0001) increase in plasma arginine and ornithine concentrations by study day 7. Plasma fatty acid profiles demonstrated higher concentrations of linoleic acid (p < .01) in the patients receiving the common use formula and higher concentrations of eicosapentaenoic and docosahexaenoic acid (p < .01) in the patients receiving the experimental formula. The mortality rate was not different between the groups and was significantly (p < .001) lower than predicted by the admission severity scores in both feeding groups. In patients who received at least 821 mL/day of the experimental formula, the hospital median length of stay was reduced by 8 days (p < .05). In patients stratified as septic, the median length of hospital stay was reduced by 10 days (p < .05), along with a major reduction in the frequency of acquired infections (p < .01) in the patients who received the experimental formula. In the septic subgroup fed at least 821 mL/day, the median length of stay was reduced by 11.5 days, along with a major reduction in acquired infections (both p < .05) in the patients who received the experimental formula.
Conclusions: Early enteral feeding of the experimental formula was safe and well tolerated in ICU patients. In patients who received the experimental formula, particularly if they were septic on admission to the study, a substantial reduction in hospital length of stay was observed, along with a significant reduction in the frequency of acquired infections.