How do you interpret a Dual-Energy X-ray Absorptiometry (DXA) scan?

How to Interpret DXA Scans

Interpret DXA scans using a systematic approach that evaluates T-scores for diagnosis in postmenopausal women and men over 50, assesses technical quality using the PARED acronym (Positioning, Artifacts, Regions of interest, Edge detection, Demographics), and reports BMD values from the lumbar spine (L1-L4), total hip, and femoral neck to guide fracture risk assessment and treatment decisions. 1

Understanding T-scores and Z-scores

T-score Interpretation (Primary Diagnostic Tool)

• T-scores compare an individual's BMD to a healthy young adult at peak bone mass, expressed as standard deviations from this reference 2, 3
• Use T-scores for postmenopausal women and men over 50 years as the primary diagnostic measurement 2
• Diagnostic thresholds:
  • Normal: T-score ≥ -1.0 2
  • Osteopenia (low bone mass): T-score between -1.0 and -2.5 2
  • Osteoporosis: T-score ≤ -2.5 2
  • Severe/established osteoporosis: T-score ≤ -2.5 plus one or more fragility fractures 2

Z-score Interpretation (Age-Matched Comparison)

• Z-scores compare BMD to people of the same age and sex, expressed as standard deviations 2, 3
• Use Z-scores for children, adolescents, premenopausal women, and men under 50 years 2
• Z-score ≤ -2.0 is considered "below the expected range for age" and warrants evaluation for secondary causes of bone loss 2

Systematic Interpretation Using PARED Approach

Follow the PARED acronym to systematically evaluate every DXA scan for technical quality before making diagnostic decisions 1:

P - Positioning

• Verify correct patient positioning by checking that the body and limbs are aligned to the midline, with minimal rotation 1
• Confirm consistent positioning between baseline and follow-up scans to ensure valid comparisons 1
• Document any positioning limitations due to patient-specific physical constraints 1

A - Artifacts

• Identify and document artifacts such as surgical hardware, calcifications, or external objects that may falsely elevate BMD 1
• Degenerative changes in the spine can falsely elevate BMD measurements, requiring exclusion of affected vertebrae 2
• Report technical quality limitations that may affect interpretation 1

R - Regions of Interest (ROI)

• Measure BMD at the lumbar spine (L1-L4), total hip, and femoral neck as standard sites 1
• Use the lowest T-score from any measured site for diagnostic classification 2
• Exclude vertebrae with structural abnormalities (fractures, severe degenerative changes, surgical hardware) that would artificially elevate BMD 1
• Do not use a single vertebral body for diagnosis or monitoring; precision worsens with fewer than 4 vertebrae 1

E - Edge Detection

• Verify that automated edge detection correctly identifies bone boundaries 1
• Manually adjust ROI boundaries when automated analysis is inaccurate 1

D - Demographics

• Confirm patient demographics (age, sex, ethnicity, menopausal status) are correctly entered, as these affect reference database selection 1
• Document height and weight at every scan, as changes may affect BMD accuracy 1

Clinical Decision-Making Based on Results

Diagnostic Classification

• Base diagnosis on the lowest T-score from lumbar spine, femoral neck, total hip, or 33% radius 2
• An osteoporotic fracture supersedes any DXA measurement and establishes the diagnosis of osteoporosis regardless of T-score 2

Treatment Recommendations

• Recommend treatment for all postmenopausal women and men >50 years with T-score ≤ -2.5 2
• For patients with osteopenia (T-score -1.0 to -2.5), use FRAX or other fracture risk assessment tools to determine treatment need 2
• Recommend treatment when FRAX shows 10-year probability of hip fracture ≥3% or major osteoporotic fracture ≥20% 2

Monitoring Changes Over Time

Proper Comparison Methodology

• Evaluate changes using absolute BMD values (g/cm²), NOT T-scores or Z-scores, as T-scores can change due to aging of the reference population 2
• Calculate the Least Significant Change (LSC) for your specific DXA facility to determine if BMD changes are real versus measurement variability 1, 2
• Only consider changes that meet or exceed the LSC as clinically significant 2

Follow-up Scanning Best Practices

• Return patients to the same DXA machine for follow-up scans whenever possible to minimize variability 2
• Maintain consistent patient positioning between baseline and follow-up 1
• Typical follow-up interval is 2-3 years, though this may be shortened based on clinical circumstances 1

Essential Components of DXA Report

Every DXA report must include 1:

• Patient demographics (name, age, sex, weight, height, menopausal status)
• Indication for testing
• Scanner manufacturer and model
• Technical quality assessment and any limitations
• Skeletal sites scanned with laterality if applicable
• BMD in g/cm² for each site with corresponding T-scores
• Comments on degenerative changes or artifacts
• Risk factors for fracture
• FRAX scores when applicable (10-year hip and major osteoporotic fracture risk)
• Diagnostic conclusion (normal, osteopenia, or osteoporosis)
• Recommendations for evaluation of secondary causes if indicated
• Treatment recommendations based on guidelines
• Follow-up DXA timing recommendation

Critical Pitfalls to Avoid

Technical Pitfalls

• Avoid diagnosing osteoporosis based on spine BMD alone in elderly patients with extensive degenerative changes, as this falsely elevates measurements 2
• Do not compare T-scores between different skeletal sites or manufacturers without proper cross-calibration 3
• Recognize that extreme body sizes (BMI <20 or >40 kg/m²) may require alternative methods such as quantitative CT 2

Interpretation Pitfalls

• Do not use peripheral DXA (heel, forearm) results with WHO T-score criteria established for central DXA 4
• Avoid monitoring treatment response with T-scores; use absolute BMD values instead 2
• Do not diagnose osteoporosis in premenopausal women or men <50 years using T-scores; use Z-scores and clinical context 2

Additional Assessments

Vertebral Fracture Assessment (VFA)

• Perform VFA in patients ≥50 years with T-score <-1.0, historical height loss ≥4 cm, or prospective height loss ≥2 cm 1
• VFA can identify unrecognized vertebral fractures that establish osteoporosis diagnosis regardless of BMD 1

Full Femur Imaging (FFI)

• Use FFI as a screening tool for atypical femur fractures in patients on long-term bisphosphonates 1
• Report focal lateral cortical thickening with transverse lucent line as HIGH likelihood for atypical fracture, requiring urgent consultation 1