Diagnosis of chronic inflammatory bowel disease (IBD) in children requires noninvasive, atraumatic diagnostic tools that depict localization and acuity of inflammation and yield only a low radiation dose. This retrospective analysis evaluates the diagnostic potential of FDG-PET. Twenty-six consecutive FDG-PET scans of 23 patients (age: 2-16, years, 14 M, 9 F) with suspected IBD were analyzed in this retrospective study. Results were compared to endoscopic, histologic, and abdominal ultrasound (US) finding. In these examinations, presence of inflammation was evaluated in each patient in 8 bowel segments (score 1-4). Standardized uptake values (SUVs) for FDG-PET were measured for all segments. Sensitivity, specificity, and accuracy were calculated using histology as the standard of reference on a segment-based analysis (pathologic if inflammation score > or = 3 or SUV(max)/SUV(liver)>1.2). With histology as the standard of reference, FDG-PET showed a sensitivity/specificity/accuracy of 98%/68%/8%/3 as compared to endoscopy (90%/75%/82%) and US (56%/92%/75%). For the small bowel, FDG-PET was even more reliable (100%/86%/90%). Because of its high sensitivity and accuracy,FDG-PET is an excellent, noninvasive diagnostic tool for IBD. Depicting inflammation in the whole bowel, while being not traumatic, it is attractive for use especially in children. FDG-PET is especially reliable for the small bowel and can inform application of topical therapy.