PURPOSE: Treatment of blunt injury of the pancreas in children remains controversial. Some prefer nonoperative treatment, whereas others prefer operative management in selected cases. This report reviews the treatment of patients with blunt pancreatic trauma admitted to a level I pediatric trauma center in The Netherlands.
METHODS: Medical records of all children less than 15 years with blunt pancreatic trauma admitted to the University Medical Center St Radboud in the period 1975 to 2003 were retrospectively analyzed.
RESULTS: Thirty-four children were included, age 3 to 14 years. Most injuries were because of bicycle accidents (58%). On admission, amylase was raised in 90% of the patients. Five patients had pancreatic duct injuries identified by imaging (endoscopic retrograde cholangiopancreaticography was used once, magnetic resonance cholangiopancreaticography twice) or at surgery. Thirty-one children were initially managed nonoperatively. Pancreatic surgery was performed in 3 children (1 Roux-Y, 2 drainage only). Mean hospital stay was 29 days in the operative group and 24 days in the nonoperative group. Fluid collections developed in 2 operated patients. Both resolved spontaneously. In 14 of the 31 nonoperated patients, a pseudocyst developed. Only 6 of these needed secondary intervention. Of these, 3 were drained percutaneously. There was no mortality and no long-term morbidity in both groups.
CONCLUSIONS: Nonoperative management of pancreatic injury in children has good clinical outcome. Only 10% need secondary surgery. In 50%, pseudocysts develop of which half can be managed nonoperatively. The reliability of computed tomographic scan grading is of limited value to decide whether to operate primarily. There is little to gain with ERCP and stenting. The place of MRCP as a noninvasive diagnostic tool remains to be determined.