Normal pancreatic parenchymal thickness by CT in healthy children.

Abstract

Chronic pancreatitis is increasingly recognized in the pediatric population. Atrophy is an important, but qualitative, finding of chronic pancreatitis. To transition to a quantitative measure that can specifically define atrophy requires knowledge of normal pancreatic parenchymal bulk in children. The purpose of this study was to define normal pancreatic thickness (linear measurements) at multiple anatomic locations in the pancreas of healthy children. This was an Institutional Review Board-approved, Health Insurance Portability and Accountability Act (HIPAA)-compliant retrospective study. Three reviewers measured the thickness of the pancreas at four locations (head, neck, body, tail) on computed tomography (CT) examinations performed with intravenous contrast in children without a history of pancreatic disease. Measurements were made on 140 examinations, evenly distributed among 7 age groups (3-16.99years) with 10 males and 10 females in each group. Agreement and bias between reviewers were assessed by intra-class correlation coefficients (ICC) and Bland-Altman analyses. Correlation with age, height and weight were assessed with Spearman's rho (ρ). The mean (for 3 readers) thicknesses of the head, neck, body and tail were 1.3-2cm, 0.7-1cm, 0.9-1.6cm and 1-1.6cm, respectively, depending on patient age. Measurement agreement between reviewers was fair to good (ICC: 0.52-0.7). Bias between reviewers ranged from 0 to 3mm. Pancreatic thickness was weakly to moderately correlated with age (ρ=0.39-0.52), height (ρ=0.44-0.61) and weight (ρ=0.51-0.64). We have defined normal ranges for thickness of the pancreas at four locations, and have shown that these measurements depend on patient age and size. These data may be useful to more objectively define pancreatic atrophy in children with suspected pancreatic disease.