DXA scan for osteoporosis
Currently the standardised method for screening for and diagnosing generalised osteoporosis.
Principle
- Measures bone mineral density (BMD) using dual-energy X-ray absorptiometry (DXA).
- Two X-ray energies are used to distinguish bone (calcium-rich) from soft tissue in the region of interest.
- The measured BMD is compared with a reference population of the same sex (and usually race) to generate T- and Z-scores.
Diagnostic criteria
Based on T-score (standard deviation from young adult peak bone mass, same sex ± race):
-
Normal: T-score ≥ −1.0
-
Osteopenia: −1.0 > T-score > −2.5
-
Osteoporosis: T-score ≤ −2.5
-
Severe (established) osteoporosis: T-score ≤ −2.5 plus a typical insufficiency fracture / fragility fracture
Since these criteria were developed for primary generalised osteoporosis, they are applied mainly to specific risk groups according to guidelines (e.g. postmenopausal women and older men).
For younger patients (often with secondary generalised osteoporosis), the Z-score (deviation from normal BMD compared with the same race, sex, and age) is used instead.
- Z-score < −2.0 is reported as “below the expected range for age”, but this alone is not a formal diagnosis of osteoporosis and must be interpreted in clinical context.
Where to measure?
Most centres measure lumbar spine and hip, and use the lowest T-score of these sites for diagnosis.
Spine
- Primary site because generalised primary osteoporosis predominantly affects the axial skeleton.
- Use L1–L4 vertebrae.
When to exclude a specific lumbar level? → Artefacts that corrupt the BMD measurement, e.g.: - Abdominal vascular calcification
- Marked lumbar degenerative change (large osteophytes, disc degeneration)
- Post-operative change: spinal fixation, vertebroplasty, etc.
- A vertebra whose T-score differs by >1.0 from the mean of the adjacent two levels (often falsely high from degenerative changes)
When to exclude spine measurement entirely? - When fewer than two contiguous vertebrae remain usable (cannot calculate a reliable L1–L4 mean).
Also assess vertebral morphology:
- Look for vertebral deformity / collapse → a vertebral insufficiency fracture both confirms severe osteoporosis (when T-score ≤ −2.5) and indicates a higher risk of future insufficiency fractures, especially of the hip.
Hip
- Primary alternative site when spine measurement is unusable or unreliable.
- Left hip is measured by convention, but either hip can be used; if both are measured, the lower T-score is reported.
When to avoid a given hip side? → Artefacts or pathology affecting BMD:
- Severe hip osteoarthritis or deformity
- Post-operative changes (arthroplasty, ORIF hardware, etc.)
Distal 33% non-dominant radius
- Optional site, but emphasised in certain secondary generalised osteoporosis where appendicular skeleton involvement predominates (e.g. hypercortisolism, hyperparathyroidism, some CKD/mineral bone disorders).