Rationale: Early parenteral nutrition to supplement insufficient enteral feeding during intensive care (early PN) delays recovery as compared with withholding parenteral nutrition for 1 week (late PN).
Objectives: To assess whether deleterious effects of early PN relate to severity of illness or to the dose or type of macronutrients.
Methods: Secondary analyses of a randomized controlled trial (EPaNIC; n = 4,640) performed in seven intensive care units from three departments in two Belgian hospitals. In part 1, all patients were included to assess the effect of the randomized allocation to early PN or late PN in subgroups of patients with increasing-on-admission severity of illness. In part 2, observationally, the association of the amount and type of macronutrients with recovery was documented in those patient cohorts still present in intensive care on Days 3, 5, 7, 10, and 14.
Measurements and Main Results: The primary end point was time to live discharge from the intensive care unit. For part 1, a secondary end point, acquisition of new infections, was also analyzed. All statistical analyses were performed by univariable and adjusted multivariable methods. In none of the subgroups defined by type or severity of illness was a beneficial effect of early PN observed. The lowest dose of macronutrients was associated with the fastest recovery and any higher dose, administered parenterally or enterally, was associated with progressively more delayed recovery. The amount of proteins/amino acids rather than of glucose appeared to explain delayed recovery with early feeding.
Conclusions: Early combined parenteral/enteral nutrition delayed recovery irrespective of severity of critical illness. No dose or type of macronutrient was found to be associated with improved outcome.