Background: Traditionally, ultrasound (US)-guided bowel mass biopsies are avoided in favour of endoscopic or surgical biopsies. However, endoscopy cannot easily reach lesions between the duodenojejunal flexure and the terminal ileum and lesions not involving the mucosa may not be accessible via an endoscopic route. Objective: The aim of this study was to report our technique and to assess the diagnostic accuracy and safety of US-guided biopsy of bowel masses in children. Materials and methods: We conducted a 14-year retrospective review of US-guided bowel mass biopsies at a single paediatric hospital. Results: Twenty US-guided bowel mass biopsies were performed in 19 patients (median age: 6 years and 6 months, range: 22 months–17 years, median weight: 22 kg, range: 10.2–48.4 kg). For 14 biopsies, there was no other lesion that could potentially be biopsied. A percutaneous coaxial technique was used for 19 biopsies and a transanal non-coaxial biopsy was performed in 1. A median of 9 (range: 2–15) cores of tissue was obtained at each biopsy. The technical success rate and adequacy of diagnostic yield were 100%. The most common diagnosis was lymphoma, which occurred in 16 biopsies. Three biopsies contained mucosa. There was one complication out of 20 biopsies (5%, 95% confidence interval 0–15%): a self-limiting, post biopsy pyrexia. Nineteen procedures were accompanied by a bone marrow aspirate and/or trephine within 2 weeks of the bowel biopsy, only one of which was diagnostic. Conclusion: US-guided bowel mass biopsy can be performed safely in children, with a high diagnostic yield and low complication rate.