A PROSPECTIVE, RANDOMIZED, DOUBLE-BLIND, CONTROLLED CLINICAL-TRIAL OFENTERAL IMMUNONUTRITION IN THE CRITICALLY ILL

Objective: To assess the effects of enteral immunonutrition (IMN) on h ospital mortality and length of stay in a heterogeneous group of criti cally ill patients. Design: Prospective, randomized, double-blind, con trolled clinical trial with an a priori subgroup analysis according to the volume of feed delivered in the first 72 hrs of intensive care un it (ICU) admission. Setting: A 13-bed adult general ICU in a London te aching hospital. Patients: A total of 398 patients were enrolled and d ata from 390 patients (IMN = 193, control = 197) were used for an inte ntion-to-treat analysis. There were 369 patients (IMN = 184, control = 185) who actually received some enteral nutrition, of whom 101 patien ts (IMN = 50, control = 51) received >2.5 L within 72 hrs of ICU admis sion, This latter group was defined as the successful ''early enteral nutrition'' group. Interventions: Within 48 hrs of ICU admission, pati ents were randomized to receive either the IMN impact(R) (Novartis Nut rition), an enteral feed supplemented with arginine, purine nucleotide s and omega-3 fatty acids, or an isocaloric, isonitrogenous control en teral feed. Measurements and Results: There was no significant differe nce in hospital mortality rate between the two groups on an intention to-treat analysis (Impact group 48%, control group 44%) nor in any oth er predefined subgroup analysis, However, patients randomized to recei ve the IMN had higher Acute Physiology and Chronic Health Evaluation I I scores (20.1 +/- 7.1 vs. 18.7 +/- 7.1 [p=.07] intention-to-treat [n = 390]; 20.1 +/- 7.2 vs. 18.5 +/- 7.1 [p=.04] received feed [n = 369]) . Of the 101 patients achieving early enteral nutrition, those patient s fed with the IMN had a significant reduction in their requirement fo r mechanical ventilation compared with controls (median duration of ve ntilation 6.0 and 10.5 days, respectively, p = .007) with an associate d reduction in the length of hospital slay (medians 15.5 and 20 days, respectively, p = .03). Conclusion: While the administration of entera l IMN to a general, critically ill population did not affect mortality , those patients in whom it was possible to achieve early enteral nutr ition with Impact had a significant reduction in the morbidity of thei r critical illness.