Which bones are ideal for dual‑energy X‑ray absorptiometry (DXA) scanning and how many anatomical sites should be measured?

DXA Scanning: Ideal Skeletal Sites and Number Required

The standard DXA examination must measure three skeletal sites: lumbar spine (L1-L4), total hip, and femoral neck—these three measurements together provide the diagnostic framework for osteoporosis assessment. 1

Core Measurement Sites (Required for Every Standard DXA)

Lumbar Spine (L1-L4)

• The L1-L4 region of the lumbar spine is mandatory for every DXA examination and serves as one of the two core skeletal sites alongside hip measurements. 1, 2
• All four lumbar vertebrae (L1 through L4) should be measured and included in the analysis as the default region of interest. 3
• Up to two vertebral levels may be excluded from the L1-L4 analysis when structural artifacts are present (fractures, severe facet joint osteoarthritis, surgical hardware, or degenerative changes that would falsely elevate BMD). 1, 2

Hip (Two Regions Required)

• Both the total hip and femoral neck must be measured because they provide complementary diagnostic information for fracture risk assessment. 1, 2
• The femoral neck is designated as the reference site for epidemiological studies and serves as a critical diagnostic region. 1
• Either hip (left or right) may be measured, though the non-dominant hip is generally preferred; if one hip contains artifacts or prosthetic hardware, scan the contralateral side. 1
• The lowest T-score among total hip, femoral neck, or trochanter determines the diagnostic classification at the hip. 3

Diagnostic Classification Based on These Sites

A T-score ≤ -2.5 at any of the three core sites (lumbar spine L1-L4, femoral neck, or total hip) establishes the diagnosis of osteoporosis, with the lowest T-score among all measured sites determining the final classification. 1, 2

Conditional Fourth Site: Forearm (One-Third Radius)

When to Add Forearm Measurement

The one-third (33%) radius of the non-dominant forearm should be measured only when lumbar spine or hip measurements are compromised or unobtainable. 1, 2

Specific Indications for Forearm DXA:

• Presence of artifacts, fractures, or orthopedic hardware (hip replacements, spinal fusion, extensive degenerative changes) that prevent reliable interpretation of spine or hip measurements. 1
• Patient exceeds the weight limit of the scanner table or cannot be positioned comfortably for spine/hip scanning. 1
• Hyperparathyroidism diagnosis, where forearm measurement should always be performed because cortical bone loss at this site is particularly relevant. 1
• Large discordance (more than 1 T-score unit) between lumbar spine and hip scans, where forearm measurement may help clarify the diagnosis. 1

Important Caveat for Forearm Measurement:

Do not measure the radius in patients with chronic arthritides such as rheumatoid arthritis, because local juxta-articular bone loss results in falsely low BMD that does not reflect the overall skeleton. 1

Sites to Avoid

• Ward's area should not be used for diagnostic purposes and should not be included in default printouts. 3
• Lateral spine BMD should not be used to diagnose osteoporosis. 3
• Calcaneus and other peripheral sites measured by portable DXA are not recommended for standard diagnostic evaluation, as they show inferior discrimination of fractures compared to central DXA sites. 4, 5

Evidence Supporting This Approach

Hip BMD (particularly femoral neck) demonstrates superior ability to identify subjects with osteoporosis and vertebral fractures compared to lumbar spine or forearm measurements in elderly populations. 6 Research shows that femoral neck BMD has the highest odds ratio for identifying subjects with vertebral fractures and the best predictability using receiver operating characteristic curves. 6

Both lumbar spine and proximal femur BMD are superior to all forearm sites for discriminating between normal and osteoporotic patients, with fracture thresholds at spine and hip being 1 standard deviation below the normal mean, compared to at or only 0.3-0.6 SD below mean for forearm sites. 7

Measurements at appendicular skeleton sites (forearm) using either DXA or peripheral quantitative CT are poorly associated with axial bone mass and have inferior ability to discriminate persons with fractures compared to central DXA measurements. 5

Summary Algorithm

1. Always measure: Lumbar spine (L1-L4) + Total hip + Femoral neck = 3 sites minimum 1, 2
2. Add forearm (one-third radius) only if: 1
   • Central sites have artifacts/hardware
   • Patient has hyperparathyroidism
   • Weight/positioning limitations exist
   • Large discordance between spine and hip (>1 T-score unit)
3. Exclude forearm if: Rheumatoid arthritis or other chronic arthritides present 1
4. Use lowest T-score from any measured site for diagnosis 1, 2