- Design
- Retrospective multi-reader validation study, single centre, MRI 2015 to 2025
- Population
- 80 patients younger than 18 with a solitary bone lesion, read by five radiologists
- Primary outcome
- Inter-rater agreement and diagnostic accuracy of Bone-RADS and OT-RADS
- Effect
- Fleiss kappa 0.470 (95% CI 0.410 to 0.530) for Bone-RADS; 0.432 (0.389 to 0.475) for OT-RADS
This single-centre study had five radiologists of differing experience score MRIs of 80 children and adolescents (mean age about 12) with a solitary bone lesion, using both Bone-RADS and OT-RADS. Imaging dated from 2015 to 2025.
Agreement across all five readers was moderate for both systems: kappa 0.470 (95% CI 0.410 to 0.530) for Bone-RADS and 0.432 (95% CI 0.389 to 0.475) for OT-RADS. Accuracy for a binary benign-versus-concerning split ranged from moderate to good for Bone-RADS and moderate to excellent for OT-RADS. The authors attribute Bone-RADS's higher agreement to its algorithm, and note that OT-RADS demands more knowledge of individual diseases because it lacks a full lexicon and algorithm.
These systems were developed mainly with adult data. This study suggests they can be applied to children, but moderate agreement means the category is a prompt for discussion with the tumour team, not a final answer.
- Consider using a structured bone tumour system for paediatric MRI, but expect moderate reader agreement.
- Bone-RADS gave higher inter-reader agreement, likely because of its algorithm.
- Do not use the score as a substitute for multidisciplinary discussion or biopsy planning.
- Record the system and category in the report so the referring team can interpret it.
- Use consensus review for borderline categories.
Why it matters
A reporting system is only useful in practice if two radiologists reach the same category.
Don't overread it
A single-centre retrospective study of 80 children; it does not show that scoring changes management.
The statistics, in plain English
A kappa of 0.47 means agreement was better than chance but far from perfect. Kappa values from 0.41 to 0.60 are usually labelled moderate. With only 80 children and five readers, subgroup accuracy figures are imprecise.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for radiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free