Early Feeding in Acute Pancreatitis in Children: A Randomized Controlled Trial

Abstract

BACKGROUND: Studies have increasingly challenged the traditional management of acute pancreatitis (AP) with bowel rest. However, these studies used a low-fat diet or transgastric feeding and only included adults. Aiming to generate higher-quality prospective pediatric data, we compared the traditional approach of fasting and intravenous fluids and early enteral feeding with standard diet or formula. METHODS: Randomized controlled trial of children (2–18 years) with mild-moderate AP. Patients were randomly assigned 1:1 to initial fasting and intravenous fluids or an immediate, unrestricted diet. Pain scores, blood measures, and cross-sectional imaging were recorded throughout admission and follow-up. The primary outcome was time to discharge, and secondary outcomes were clinical and biochemical resolution and local and systemic complication rates. RESULTS: Of 33 patients (17 [52%] boys, mean age of 11.5 [±4.8] years), 18 (55%) were randomly assigned to early feeding and 15 (45%) were randomly assigned to initial fasting. We recorded the median (interquartile range [IQR]) time to discharge (2.6 [IQR 2.0 to 4.0] vs 2.9 [IQR 1.8 to 5.6]; P = .95), reduction in serum lipase levels by day 2 (58% [IQR 2% to 85%] vs 48% [IQR 3% to 71%]; P = .65), and readmission rates (1 of 18 [6%] vs 2 of 15 [13%]; P = .22) between the early feeding and fasting cohorts, respectively. Immediate or delayed complication rates did not differ. Patients randomly assigned to early feeding had weight gain of 1.3 kg (IQR 0.29 to 3.6) at follow-up, compared with weight loss of 0.8 kg (IQR −2.1 to 0.7) in fasted patients (P = .028). CONCLUSIONS: This is the first randomized controlled trial in pediatric AP. There was no difference between early commencement of a standard oral diet and initial fast in any of the major outcome measures.