- Design
- Systematic review and bivariate diagnostic meta-analysis
- Population
- 9,214 participants in 42 studies, mostly degenerative spine populations
- Primary outcome
- Accuracy of CT HU and MRI VBQ for osteopenia and osteoporosis vs DXA or QCT
- Effect
- Osteoporosis AUC 0.86 (HU) vs 0.79 (VBQ); threshold 104.6 HU
A diagnostic meta-analysis pooled 42 studies of 9,214 people, largely with degenerative spinal disease, comparing lumbar CT Hounsfield units (HU) and MRI vertebral bone quality (VBQ) scores against DXA or quantitative CT.
For osteoporosis, pooled AUC was 0.86 for HU and 0.79 for VBQ; for osteopenia, 0.87 and 0.75. Threshold synthesis gave pooled cut-offs of 104.6 HU for osteoporosis and 138.9 HU for osteopenia. Validation analyses consistently favoured HU; VBQ varied more.
Most older patients having abdominal or spinal CT never have DXA. A trabecular HU measurement costs a few seconds and no extra radiation, and could flag low bone density that would otherwise go unrecognised.
The heterogeneity across studies means the thresholds are guides rather than diagnostic cut-offs, and DXA remains the reference test for treatment decisions.
- Consider measuring trabecular HU on lumbar vertebrae in older patients having CT for another reason.
- Values near or below about 105 HU warrant a line in the report suggesting DXA.
- Prefer CT HU over MRI VBQ scores when both are available.
- Agree a standard measurement method across your department, since technique varies.
Why it matters
Opportunistic bone density assessment uses images already acquired to reach patients who would never otherwise be tested.
Don't overread it
Thresholds came from heterogeneous studies in mainly spinal surgery populations and have not been validated as a replacement for DXA.
The statistics, in plain English
An AUC of 0.86 means that, given one person with osteoporosis and one without, the HU value ranks them correctly 86% of the time. A pooled threshold is an average across studies with different scanners and methods, so a single local cut-off may perform differently.
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