Diabetes-specific enteral nutrition formula in hyperglycemic, mechanically ventilated, critically ill patients: a prospective, open-label, blind-randomized, multicenter study

Introduction: Although standard enteral nutrition is universally accepted, the use of disease-specific formulas for hyperglycemic patients is still controversial. This study examines whether a high-protein diabetes-specific formula reduces insulin needs, improves glycemic control and reduces ICU-acquired infection in critically ill, hyperglycemic patients on mechanical ventilation (MV). Methods: This was a prospective, open-label, randomized (web-based, blinded) study conducted at nine Spanish ICUs. The patient groups established according to the high-protein formula received were: group A, newgeneration diabetes-specific formula; group B, standard control formula; group C, control diabetes-specific formula. Inclusion criteria were: expected enteral nutrition >= 5 days, MV, baseline glucose > 126 mg/dL on admission or > 200 mg/dL in the first 48 h. Exclusion criteria were: APACHE II <= 10, insulin-dependent diabetes, renal or hepatic failure, treatment with corticosteroids, immunosuppressants or lipid-lowering drugs and body mass index >= 40 kg/m(2). The targeted glucose level was 110-150 mg/dL. Glycemic variability was calculated as the standard deviation, glycemic lability index and coefficient of variation. Acquired infections were recorded using published consensus criteria for critically ill patients. Data analysis was on an intention-to-treat basis. Results: Over a 2-year period, 157 patients were consecutively enrolled (A 52, B 53 and C 52). Compared with the standard control formula, the new formula gave rise to lower insulin requirement (19.1 +/- 13.1 vs. 23.7 +/- 40.1 IU/day, p < 0.05), plasma glucose (138.6 +/- 39.1 vs. 146.1 +/- 49.9 mg/dL, p < 0.01) and capillary blood glucose (146.1 +/- 45.8 vs. 155.3 +/- 63.6 mg/dL, p < 0.001). Compared with the control diabetes-specific formula, only capillary glucose levels were significantly reduced (146.1 +/- 45.8 vs. 150.1 +/- 41.9, p < 0.01). Both specific formulas reduced capillary glucose on ICU day 1 (p < 0.01), glucose variability in the first week (p < 0.05), and incidences of ventilator-associated tracheobronchitis (p < 0.01) or pneumonia (p < 0.05) compared with the standard formula. No effects of the nutrition formula were produced on hospital stay or mortality. Conclusions: In these high-risk ICU patients, both diabetes-specific formulas lowered insulin requirements, improved glycemic control and reduced the risk of acquired infections relative to the standard formula. Compared with the control-specific formula, the new-generation formula also improved capillary glycemia.