Parenteral Nutrition in Critically Ill Children
Nilesh and Mehta, N Engl J Med 2016
Adequate nutrient delivery may help to offset the catabolic burden imposed by critical illness, preventing nutritional deterioration and improving outcomes. However, questions regarding the most appropriate dose of macronutrients, the route of delivery, and particularly the timing of supplemental parenteral nutrition in critically ill children remain unanswered. Enteral nutrition is preferable in patients with a functioning gut but may not always be feasible. Intestinal failure was incompatible with life until the 1960s, when the development of stable intravenous amino acid solutions and lipid emulsions allowed the administration of life-sustaining nourishment through the parenteral route. Parenteral nutrition was subsequently extended to critically ill patients when enteral nutrition was insufficient or contraindicated. As awareness of the unintended side effects of parenteral nutrition has increased, patient selection and the timing of administration are being investigated and the optimization of enteral nutrition emphasized.
The benefits of early parenteral nutrition during critical illness in children have been questioned, given the results of studies in adults. When enteral nutrient intake is insufficient in critically ill children, two distinct strategies are now considered: some medical centers initiate early parenteral nutrition to prevent macronutrient deficits, while others, concerned about side effects, delay supplementary parenteral nutrition for a week while providing enteral nutrition. Fivez et al. now address this area of uncertainty in the Journal by reporting the results of a three-center trial, Early versus Late Parenteral Nutrition in the Pediatric Intensive Care Unit (PEPaNIC).
In the trial, 1440 patients with insufficient enteral nutrition were randomly assigned to a control group in which parenteral nutrition was administered within 24 hours after admission or to an intervention group in which parenteral nutrition was withheld until day 8. All patients received early enteral nutrition followed by stepwise, standardized advancement to the prescribed goal. The sites used different methods to determine energy requirements, parenteral nutrition strategies, and targets for glycemic control. An absolute reduction in the risk of new infections (7.8 percentage points) and a reduced time to “readiness for discharge” from the intensive care unit (ICU) were recorded in the group receiving late parenteral nutrition as compared with that receiving early parenteral nutrition. The duration of mechanical ventilation and the odds of renal replacement were also lower in the late-parenteral-nutrition group. The internal validity of the study was high in terms of sample size, randomization, concealed allocation, adherence to the protocol, standardization of the intervention, and statistical analysis. To determine the external validity of the study, some design elements must be discussed — the population selected, the generalizability of the protocol, and the relevance of the outcomes.
More than 55% of the patients in the control group were discharged by day 4. These patients would not be considered candidates for parenteral nutrition within 24 hours after admission in most medical centers. More than 77% of patients in the group receiving late parenteral nutrition were discharged by day 8 without having received any parenteral nutrition. The equations used to determine energy requirements are unreliable.The true adequacy of energy delivered, and the potential caloric underfeeding and overfeeding, cannot be determined in the absence of the measurement of energy expenditure. The dose of enteral nutrient delivery beyond which the risks of supplemental parenteral nutrition offset its benefits is unknown. A threshold enteral delivery of less than 80% of the nutrient target triggered supplementation with parenteral nutrition in this trial, whereas enteral delivery of 66% of the nutrient target has been associated with improved outcomes in children receiving mechanical ventilation.The poor outcomes in the group receiving early parenteral nutrition may represent the risks associated with the parenteral route of delivery or the effects of caloric overfeeding. The device investigators used to calculate the risk of malnutrition, the Screening Tool for Risk on Nutritional Status and Growth (STRONGkids), is based on a simple questionnaire without anthropometric variables. The use of STRONGkids has not been validated in critically ill children and may not be a reliable indicator of severe malnutrition. According to z scores for body-mass index, only a small fraction of patients in this trial had moderate or severe malnutrition. Standard outcomes related to infection, such as ventilator-associated pneumonia and catheter-associated bloodstream infections, were not recorded. The intensive care team determining the pragmatic “readiness for discharge” outcome was not masked to patient randomization, which introduces the potential for bias. Despite the above limitations, the results of this landmark trial will change practice in centers that supplement insufficient enteral nutrition with parenteral nutrition on day 1 and will provide data that support the strategy of delayed parenteral nutrition practiced in other centers. Further examination is required to determine the best possible timing for the introduction of supplemental parenteral nutrition in severely malnourished children in the pediatric ICU.
Ideally, nutrition therapy for critically ill children must be individualized. The presupposition that a uniform approach would apply to all is too simplistic. On the basis of the results of the PEPaNIC trial, delaying parenteral nutrition, along with early initiation and stepwise advancement of enteral nutrition, seems prudent in patients who are not severely malnourished. Patience and substantial resources will be required to systematically address unresolved questions through randomized, controlled trials. As the tapestry of evidence is gradually woven, we must be circumspect in our interpretation of study results, invest in mechanistic and hypothesis-generating studies, and explore relevant long-term outcomes. The PEPaNIC trial represents an important step forward, and its results will help to recalibrate the safe application of parenteral nutrition in critically ill children.