Bone Densitometry (DXA)

On this page
  1. Direct answer
  2. What you must remember
  3. Scanning and interpreting a postmenopausal referral
  4. Where students slip
  5. Frequently asked questions
  6. Related topics

Direct answer

Dual energy X-ray absorptiometry (DXA) diagnoses osteoporosis by passing two X-ray energies (typically 70 and 140 kVp by switching, or gadolinium filtration producing 44 and 100 keV effective beams) through bone and soft tissue and solving for areal bone mineral density in grams per square centimetre. The WHO classification anchors to the T-score — standard deviations below young-adult mean — at the lumbar spine (L1-L4) and proximal femur: normal above −1, low bone mass between −1 and −2.5, osteoporosis at or below −2.5, severe with a fragility fracture. Precision is the technique's lifeblood — identical positioning visit to visit, daily phantom scanning, ISCD exclusion rules — and radiation dose is a few microsieverts, the lowest in diagnostic imaging.

What you must remember

  • T-score versus Z-score: the T-score compares with young-adult mean and applies to postmenopausal women and men over 50; the Z-score compares with age-matched peers in younger groups, where below −2 is "low for age", never "osteoporosis".
  • WHO thresholds: normal above −1; low bone mass −1 to −2.5; osteoporosis −2.5 or lower at spine, total hip or femoral neck; a low-trauma hip or vertebral fracture diagnoses osteoporosis regardless of the number.
  • Sites and rules: PA spine L1-L4, excluding vertebrae distorted by fracture or sclerosis (at least two usable); femoral neck and total hip — the femoral neck is the validated FRAX site; forearm when spine and hip are unusable.
  • Technique points: supine centring, knees over a block to flatten the lumbar lordosis, femoral rotation of 15-25 degrees with the foot strapped, replicated exactly on follow-up.
  • Confounding artefacts: aortic calcification, endplate sclerosis, osteophytes, vertebroplasty cement, barium and recent radiopharmaceuticals spuriously elevate BMD — defer scanning about 24-72 hours after nuclear medicine.
  • Quality control: daily phantom scan plotted on control charts, precision commonly within about 1% coefficient of variation; the least significant change defines whether follow-up change is real.
  • Indications worth memorising: women 65 and over, men 70 and over, fragility fracture after 50, early menopause, glucocorticoids for 3 months or more, hyperthyroidism, hyperparathyroidism, chronic kidney or liver disease.

Scanning and interpreting a postmenopausal referral

A 63-year-old with a wrist fracture from a standing-height fall is referred. No metal, zip-free gown, no barium or nuclear scan this week. For the PA spine she lies centred with knees over the block; the scan captures mid-T12 to the sacrum and analysis boxes frame L1-L4 with intervertebral lines at the endplates. For the left hip the leg rotates inward with the foot strapped — the femoral neck must not be foreshortened, the lesser trochanter barely visible. The machine reports BMD, T-scores and Z-scores per region.

Interpretation is a discipline of exclusions. If L2 is compressed and L3 sclerotic, exclude both and diagnose from L1 and L4, or shift weight to the hip; if every lumbar level reads above the hip suspiciously, look for a calcified aorta on the scan image. A femoral neck T-score of −2.7 with the radius fracture seals the diagnosis and feeds FRAX (femoral neck BMD by convention). When she returns on treatment in two years, only a change exceeding the least significant change — roughly 2.8-5.5% at the spine — counts as real, which is why follow-up runs on the same scanner with the same positioning.

Where students slip

The recurring error is quoting the T-score for a 38-year-old with amenorrhoea — Z-score territory — and labelling her osteoporotic when ISCD wording is "low bone density for chronological age". The second is arithmetic-blind faith: BMD is areal, not volumetric, so a taller patient's bone reads denser from size alone. Viva examiners probe why the leg is rotated for the hip — to bring the femoral neck parallel to the scan plane so true neck width is measured. And the dose question must be answered with confidence: a spine and hip study delivers 1-10 microsieverts, comparable to a day of background radiation.

Frequently asked questions

What T-score defines osteoporosis on DXA?

A T-score of −2.5 or lower at lumbar spine, total hip or femoral neck; a fragility fracture diagnoses osteoporosis independently.

When is the Z-score used instead of the T-score?

In premenopausal women, men under 50 and children — compared with age-matched controls, with below −2 reported as below the expected range for age.

Which conditions spuriously elevate DXA bone density?

Aortic calcification, osteophytes and sclerosis, compression fractures, vertebroplasty cement, retained barium and recent radiopharmaceuticals.

How is DXA quality control performed daily?

A manufacturer's spine phantom is scanned before the first patient and plotted on control charts, staying within about 1% coefficient of variation.

Why must follow-up DXA scans be performed identically?

Treatment decisions rely on changes exceeding the least significant change; different scanners or positions manufacture or mask change.

Practise this in the PrepElephant app

Question banks, previous-year questions, mock tests and revision tools — for Bone Densitometry (DXA) and Allied Health Radiology and Imaging Technology. Free to start.

Get the free app WhatsApp