Background: To determine whether early parenteral nutrition (PN) (25% of goal energy within 48 hours of PICU admission) is associated with poorer outcomes in children receiving no enteral nutrition (EN).
Methods: Multicenter retrospective study of patients aged 1 month to 18 years who had a PICU length of stay (PLOS) >96 hours. We obtained weight, sex, pediatric index of mortality 2 score (PIM-2), PLOS, duration of mechanical ventilation (DMV), mortality data, and nutrition intake data. Logistic and mixed model regression analysis were used to compare data.
Results: 2069 patients (53.2% male, median age 6.61 years) received no EN in the first 4 days. Children receiving early PN were more likely to die than those who did not when adjusted for PIM-2, propensity score, and center (odds ratio = 2.10 [1.41-3.13], median [IQR]; P = 0.0003). The unadjusted PLOS (9.48 [5.94-18.19], and unadjusted DMV (6.73 [3.48-13.98]) for patients receiving early PN were both significantly longer than those who did not (6.75 [4.95-11.65]; P < 0.0001 and 4.9 [1.88-10.19]; P = 0.009, respectively). When adjusted for PIM-2, center, percentage of energy from protein, and age, the PLOS and DMV for those receiving early PN did not differ from those who did not (P = 0.14 and P = 0.76, respectively).
Conclusion: In children with PLOS >96 hours receiving no EN for 4 days, early PN is strongly associated with higher mortality but not with differences in PLOS or DMV.
Keywords: critical care; parenteral nutrition; pediatrics; propensity score.
© 2019 American Society for Parenteral and Enteral Nutrition.