- Design
- Registry analysis with unsupervised cluster analysis
- Population
- 904,209 routine head CTs (864,182 adult, 40,027 paediatric), 2015–2021
- Primary outcome
- Size-adjusted dose across acquisition clusters
- Effect
- More than two-fold, up to 3.4-fold, dose variation; adult helical DLP 372–998 mGy·cm
Using an international CT dose registry, this study grouped 864,182 adult and 40,027 paediatric routine head CTs performed between 2015 and 2021 into clusters by tube current, voltage, collimation, scan length and pitch, then compared size-adjusted dose between clusters.
In every group (adults and children, helical and axial), mean size-adjusted dose differed more than two-fold between clusters, and up to 3.4-fold. For adult helical scans, size-adjusted dose-length product ranged from 372 to 998 mGy·cm. The lowest-dose clusters generally used lower tube current and voltage, though several different combinations reached low doses.
The registry does not measure image quality or diagnostic accuracy, so it cannot say that every low-dose protocol was adequate. But the scale of variation for a standard examination suggests many protocols give more radiation than needed. It was published in September 2026.
- Compare your routine head CT protocol's size-adjusted dose with national or registry references.
- Review tube current and voltage settings first; the lowest-dose clusters generally used lower values of both.
- Pay particular attention to paediatric head CT protocols.
- Check image quality with your neuroradiologists before lowering dose.
- Audit dose regularly rather than relying on vendor defaults.
Why it matters
Head CT is among the most frequent scans, so protocol choice multiplies across millions of patients.
Don't overread it
The registry measured dose, not image quality, so it cannot show every low-dose protocol was diagnostically adequate.
The statistics, in plain English
Size-adjusted dose-length product accounts for head size, so comparisons are fairer between patients. A relative dose of 2.7 means the highest-dose cluster gave about 2.7 times the radiation of the lowest for the same type of scan.
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