I now have all needed material. Here is the complete, exam-focused MSc Radiology answer — structured exactly as you'd need it for a written or viva exam.
DEXA (DXA) — Complete MSc Radiology Exam Answer
✦ STEP 1 — Definition & Full Name
Full name: Dual-Energy X-ray Absorptiometry
Abbreviation: DXA (international standard) / DEXA (older term, still widely used)
Definition:
DXA is a non-invasive, low-dose imaging technique that uses two X-ray beams of different photon energies to measure Bone Mineral Density (BMD) and body composition by exploiting the differential attenuation of X-rays through bone and soft tissue.
It is the WHO-endorsed gold standard for:
- Diagnosis of osteoporosis and osteopenia
- Fracture risk assessment
- Monitoring response to treatment
- Body composition analysis
✦ STEP 2 — Historical Background (Exam Context)
| Era | Technique | Limitation |
|---|
| 1960s | Single Photon Absorptiometry (SPA) | Radioactive source (¹²⁵I); peripheral sites only |
| 1970s | Dual Photon Absorptiometry (DPA) | Radioactive source (¹⁵³Gd); slow; source decays → imprecise |
| 1980s | DXA introduced | X-ray tube replaces isotope → stable, fast, precise |
| 1990s–now | Fan-beam DXA | Sub-minute scans; vertebral fracture assessment added |
Key advance: Replacing radioactive isotopes with an X-ray generator eliminated source decay, reduced scan time from 20–40 minutes to under 5 minutes, and dramatically improved precision.
✦ STEP 3 — Physical Principle (The Core Exam Topic)
3.1 The Beer-Lambert Law (Foundation)
X-ray attenuation through matter:
$$I = I_0 \cdot e^{-\mu x}$$
where μ = linear attenuation coefficient (tissue-specific), x = path length through tissue.
At any single energy, it is impossible to separate attenuation due to bone from attenuation due to overlying soft tissue — because you have one equation with two unknowns.
3.2 Why Two Energies Solve This
By using two different X-ray energies (H = high, L = low), the system creates two simultaneous equations with two unknowns (bone mineral + soft tissue), which can be solved mathematically:
$$\mu_{H,\text{tissue}} \cdot x_{\text{tissue}} + \mu_{H,\text{bone}} \cdot x_{\text{bone}} = \ln(I_{0H}/I_H)$$
$$\mu_{L,\text{tissue}} \cdot x_{\text{tissue}} + \mu_{L,\text{bone}} \cdot x_{\text{bone}} = \ln(I_{0L}/I_L)$$
The ratio of attenuation at the two energies (R-value) is unique for each tissue type — the software uses this ratio pixel-by-pixel to map pure bone mineral content independent of soft tissue thickness or composition.
3.3 Why Bone Attenuates More at Low Energy
- Bone mineral is primarily calcium hydroxyapatite [Ca₁₀(PO₄)₆(OH)₂]
- Calcium has a high atomic number (Z = 20) → strong photoelectric absorption at low kVp
- At low energy (~40–70 kVp): bone–soft tissue contrast is high
- At high energy (~100–140 kVp): both attenuate similarly → contrast is low
- The difference in these two attenuation values encodes the bone mineral content
3.4 Methods of Producing Two Energies
| Method | Mechanism | Used In |
|---|
| K-edge filtration | Cerium (Ce) or samarium (Sm) filter with K-edge at ~40 keV shapes the beam into two distinct energy peaks from a single exposure | Older/some current fan-beam systems |
| Voltage switching | X-ray tube rapidly alternates between high kVp (~100–140) and low kVp (~40–70) pulses; detector reads each separately | Most modern fan-beam systems |
✦ STEP 4 — Equipment Design
4.1 System Types
| Generation | Beam Type | Scan Method | Scan Time | Precision |
|---|
| 1st gen | Pencil beam | Point-by-point rectilinear scan | 20–30 min | Moderate |
| 2nd gen | Narrow fan beam | Sweep with small detector array | 5–10 min | Good |
| Current standard | Wide fan beam | Single sweep with wide detector array | 30 sec – 3 min | Excellent |
4.2 Machine Architecture
X-ray Tube (kVp alternating or filtered)
↓
Beam collimation
↓
Patient on table (supine)
↓
C-arm sweeps over patient
↓
Scintillation detector array (below table)
↓
ADC + Computer software
↓
BMD map → T/Z scores → printed report
4.3 Fan-Beam vs Pencil-Beam
| Feature | Pencil Beam | Fan Beam |
|---|
| Scan time | Long | Short |
| Spatial resolution | Lower | Higher |
| Motion artifact | More susceptible | Less susceptible |
| VFA possible? | No | Yes |
| Magnification artifact | Negligible | Present (diverging beam) |
| Radiation dose | Slightly lower | Slightly higher (still very low) |
✦ STEP 5 — Scan Sites and Positioning
5.1 Standard Primary Sites
| Site | Vertebrae / Region | Bone Type | Clinical Relevance |
|---|
| Lumbar spine | L1–L4 (AP/PA) | Rich trabecular bone | Sensitive to early metabolic change; high precision |
| Proximal femur | Femoral neck + total hip | Cortical + trabecular | Best predictor of hip fracture; WHO classification site |
| Distal 1/3 radius | Non-dominant arm | Predominantly cortical | Used when spine/hip not measurable |
5.2 Proximal Femur Sub-Regions (Know All Four)
| Label | Region | Note |
|---|
| A | Femoral neck | Narrow ROI; primary fracture prediction site |
| B | Ward's triangle | Intersection of 3 trabecular groups; lowest BMD; no longer used for diagnosis |
| C | Greater trochanter | Cancellous-rich |
| D | Intertrochanteric / shaft | Cortical-dominant |
| — | Total hip | Sum of above; most reproducible; preferred for monitoring |
5.3 Positioning Protocol (Exam-Critical)
Lumbar spine:
- Supine, arms folded on chest
- Hips and knees flexed ~90° over a foam positioning block → flattens lumbar lordosis → reduces posterior element (spinous process, facets) overlap with vertebral bodies → accurate anterior body measurement
Proximal femur:
- Supine, leg extended
- Foot internally rotated 15–25° using a foot-holder → brings femoral neck parallel to table → removes anteversion artifact → accurate neck measurement
- Slight abduction of the hip
Forearm:
- Seated beside table
- Non-dominant arm extended, palm down
- Positioning device immobilizes arm
- ROI at 33% (one-third) site from wrist
✦ STEP 6 — Output Measurements (What DXA Produces)
6.1 Primary Outputs
| Output | Unit | Formula |
|---|
| BMC (Bone Mineral Content) | grams (g) | Directly measured |
| Area | cm² | Projected 2D area of bone |
| BMD (Bone Mineral Density) | g/cm² | BMC ÷ Area |
Critical exam point — DXA limitation: DXA measures areal BMD (2D projection), not true volumetric BMD (g/cm³). A larger bone will appear denser than a smaller bone with identical volumetric density. This is the size artifact — a key weakness of DXA.
6.2 T-Score
$$\text{T-score} = \frac{\text{Patient BMD} - \text{Young adult mean BMD}}{\text{SD of young adult reference population}}$$
- Compares to peak bone mass (healthy young adults, ~30 years)
- Used for diagnosis in postmenopausal women and men ≥ 50 years
WHO Classification (T-score):
| T-score | Diagnosis |
|---|
| > −1.0 | Normal |
| −1.0 to −2.5 | Osteopenia (low bone mass) |
| ≤ −2.5 | Osteoporosis |
| ≤ −2.5 + fragility fracture | Severe (established) osteoporosis |
6.3 Z-Score
$$\text{Z-score} = \frac{\text{Patient BMD} - \text{Age-matched mean BMD}}{\text{SD of age-matched reference population}}$$
- Compares to same age, sex, ethnicity
- Z-score ≤ −2.0 = "below expected range for age" → investigate for secondary cause of bone loss
- Used in: premenopausal women, men < 50 years, children (Z-score only in children)
6.4 Graphical DXA Report Output
The color-coded BMD vs. age reference graph is standard on all DXA reports:
- 🟢 Green zone — T-score > −1.0 (Normal)
- 🟡 Yellow zone — T-score −1.0 to −2.5 (Osteopenia)
- 🔴 Red zone — T-score < −2.5 (Osteoporosis)
The patient's data point is plotted, and the trend line shows expected age-related BMD decline.
6.5 Full DXA Report — Osteoporosis Example
The full table shows Area (cm²), BMC (g), BMD (g/cm²), T-score, Peak Reference (PR%), Z-score, and Age-Matched (AM%) for each ROI — all elements required in a complete DXA report.
✦ STEP 7 — Indications for DXA
7.1 Clinical Indications
| Population / Condition | Indication |
|---|
| Women ≥ 65 years | Universal screening |
| Men ≥ 70 years | Universal screening |
| Postmenopausal women < 65 with risk factors | Selective screening |
| Men 50–69 with risk factors | Selective screening |
| Fragility fracture (any age) | Diagnostic and severity assessment |
| Long-term glucocorticoid use (≥ 3 months) | Glucocorticoid-induced osteoporosis (GIO) monitoring |
| Rheumatoid arthritis, IBD, CKD, malabsorption | Secondary osteoporosis |
| Androgen deprivation therapy / aromatase inhibitors | Drug-induced bone loss monitoring |
| Hyperparathyroidism, hypogonadism | Metabolic bone disease |
| Children with osteogenesis imperfecta or chronic illness | Pediatric DXA (spine + whole body; Z-scores only) |
| On osteoporosis treatment | Monitoring every 1–2 years |
| Body composition assessment | Sarcopenia, obesity surgery, oncology, sports medicine |
7.2 Risk Factors That Lower the Screening Age
- Family history of hip fracture
- Current smoker
- Excessive alcohol (> 3 units/day)
- Low body weight (BMI < 19 kg/m²)
- Immobilization
- History of hypogonadism or early menopause (< 45 years)
- Chronic renal or liver disease
✦ STEP 8 — FRAX Tool (Must Know for MSc Level)
Fracture Risk Assessment Tool (FRAX) — WHO-endorsed:
- Calculates 10-year probability of:
- Major osteoporotic fracture (spine, hip, wrist, humerus)
- Hip fracture alone
- Inputs: 11 clinical risk factors ± femoral neck BMD
Treatment threshold (general guidelines):
| Threshold | Action |
|---|
| 10-year hip fracture ≥ 3% | Consider pharmacotherapy |
| 10-year major osteoporotic fracture ≥ 20% | Recommend pharmacotherapy |
FRAX integrates DXA BMD data with clinical risk to guide treatment decisions beyond T-score alone — patients in the osteopenic range (T −1.0 to −2.5) may still warrant treatment if FRAX risk is high.
✦ STEP 9 — Patient Preparation
| Requirement | Detail |
|---|
| Contrast media | Avoid barium/IV contrast for 7–10 days (residual attenuates beam falsely) |
| Nuclear medicine | Wait 3–7 days after radioisotope administration |
| Metal objects | Remove all jewellery, belt, metal zips, underwire bras |
| Calcium supplements | Hold 24 hours before scan (some protocols) |
| Pregnancy | Absolute contraindication |
| Weight limit | Standard table ~204 kg; bariatric scanners available |
| Prior fractures/hardware | Note — influences ROI selection and interpretation |
| Clothing | Light, no metal; gown if required |
| History | Document height, weight, ethnicity, medications, fracture history |
✦ STEP 10 — Radiation Dose (Key Advantage of DXA)
| Examination | Effective Dose |
|---|
| DXA lumbar spine | 1–3 µSv |
| DXA hip | 1–5 µSv |
| VFA (lateral spine, fan-beam) | 3–6 µSv |
| Whole body DXA | 3–10 µSv |
| Chest X-ray (comparison) | ~20 µSv |
| Daily background radiation | ~8 µSv/day |
| CT abdomen (comparison) | ~5,000–10,000 µSv |
Exam point: A DXA lumbar spine scan delivers less radiation than a single day of natural background exposure — supporting its use for population screening and serial monitoring.
✦ STEP 11 — Vertebral Fracture Assessment (VFA)
A major additional capability of fan-beam DXA systems — performed in the same session as BMD scanning.
What it does:
- Lateral image of spine T4–L4 using the DXA C-arm
- Detects vertebral deformities and morphometric fractures that are often clinically silent (two-thirds of vertebral fractures are asymptomatic)
- Far lower dose than conventional lateral spine X-ray (~3–6 µSv vs. ~600 µSv)
Why it matters clinically:
- A vertebral fracture independently multiplies fracture risk 5× for another vertebral fracture and 2–3× for hip fracture
- Changes FRAX output and may directly initiate treatment
Genant Semi-Quantitative Grading Scale:
| Grade | Description | Anterior Height Loss |
|---|
| 0 | Normal | None |
| 1 | Mild deformity | 20–25% |
| 2 | Moderate deformity | 25–40% |
| 3 | Severe deformity | > 40% |
2025 Guideline (PMID: 41338753): The International Working Group on DXA Best Practices now recommends VFA should be routinely integrated into standard clinical DXA assessment — not just as an optional add-on.
✦ STEP 12 — Body Composition by DXA
Whole-body DXA provides 3-compartment model:
Total Body Mass
│
├── Bone Mineral Content (BMC in grams)
├── Lean Soft Tissue (kg) — muscle, organs
└── Fat Mass (kg) — subcutaneous + visceral
Clinical applications:
- Sarcopenia — low appendicular lean mass index (ALMI) diagnosis
- Obesity — visceral fat estimation; android vs gynoid distribution
- Cancer/HIV — cachexia monitoring
- Sports science — athletic body composition tracking
- Bariatric surgery — pre/post body composition changes
- Pediatrics — growth and development studies
ISCD Sarcopenia thresholds:
- Men: Appendicular Lean Mass Index (ALMI) < 7.0 kg/m²
- Women: ALMI < 5.5 kg/m²
✦ STEP 13 — Artifacts & Sources of Error (High-Yield Exam Topic)
Falsely ELEVATED BMD (Overestimation)
| Artifact | Mechanism | Site Affected |
|---|
| Vertebral osteophytes | Added mineralization in scan field | Lumbar spine |
| Aortic calcification | Calcified aorta overlies lumbar vertebrae | Lumbar spine |
| Vertebral compression fracture | Smaller area but preserved mineral → higher g/cm² | Lumbar spine |
| Scoliosis | Vertebral rotation → abnormal ROI | Lumbar spine |
| Spinal metallic hardware | Extreme hyperattenuation | Lumbar spine / hip |
| Hip prosthesis | Metal artifact | Proximal femur |
| Residual barium / contrast | Beam attenuation in GI tract | Any site |
| Obesity (fan-beam) | Magnification artifact | Any site |
| Metal in clothing | Zips, hooks, buttons | Any site |
Falsely REDUCED BMD (Underestimation)
| Artifact | Mechanism |
|---|
| Laminectomy / spinal surgery | Removed posterior elements reduce total mineral measured |
| Poor foot rotation (hip) | Femoral neck not parallel → foreshortening |
| Patient rotation (spine) | Asymmetric vertebral projection |
| Incorrect ROI placement | Excludes bone or includes non-bone tissue |
Precision / Reproducibility Errors
| Error Source | Impact |
|---|
| Repositioning variability | Most common source of scan-to-scan variability |
| ROI placement variation | Manual editing inconsistencies |
| Machine drift | Detected by daily phantom calibration |
| Software version change | Different algorithms → different values; must document |
| Different machines | Cannot directly compare values without cross-calibration |
✦ STEP 14 — Precision, LSC and Monitoring
When using DXA to monitor treatment response, a change in BMD is clinically significant only if it exceeds the Least Significant Change (LSC):
$$\text{LSC} = 2.77 \times \text{CV%} \text{ (coefficient of variation)}$$
- Typical spine CV: ~1% → LSC ≈ 2.77%
- Typical hip CV: ~1.5% → LSC ≈ 4.15%
- If measured change < LSC → could be measurement noise
- If change ≥ LSC → true biological change (95% confidence)
Rules for valid serial monitoring:
- Same scanner every time
- Same operator (technologist) for positioning
- Same ROI placement
- Document software version used
✦ STEP 15 — Quality Assurance (QA)
| QA Measure | Frequency | Purpose |
|---|
| Spine phantom scan | Daily | Detect machine drift and calibration error |
| Precision study | Per technologist (30 patients × 2 scans) | Calculate in-house LSC |
| Cross-calibration | When changing machines | Ensure serial data continuity |
| CV% monitoring | Ongoing | Flag technologist-specific precision errors |
| ISCD/ECTS accreditation | Periodic | International quality benchmarking |
✦ STEP 16 — Limitations of DXA
| Limitation | Why It Matters |
|---|
| Areal not volumetric BMD | Size bias — larger bones score higher even with same true density |
| 2D projection | Cannot separate cortical from trabecular compartments |
| Artifacts inflate lumbar BMD | Degenerative spine changes are almost universal in elderly → hip BMD more reliable |
| Bone quality not assessed | Measures mass, not microarchitecture or collagen quality |
| Population-specific databases needed | Non-Caucasian reference data less well established |
| Most fractures in osteopenic range | Large osteopenic population means many fractures occur above the −2.5 threshold |
| Cannot detect all fractures | Clinical fragility fracture diagnosis independent of BMD |
✦ STEP 17 — Comparison with Other Bone Density Techniques
| Technique | Measures | Radiation | Advantage | Disadvantage |
|---|
| DXA | Areal BMD (g/cm²) | Very low (1–5 µSv) | Gold standard; fast; low dose | 2D; size bias; no microarchitecture |
| QCT | Volumetric BMD (mg/cm³) | Higher (~1000–3000 µSv) | True 3D; separates cortical/trabecular | Higher dose; expensive; less standardized |
| pQCT / HR-pQCT | Microarchitecture | Low–moderate | Trabecular number, thickness, connectivity | Research tool; radius/tibia only |
| QUS | Broadband ultrasound attenuation (BUA) | None | Portable; no radiation; cheap | Cannot diagnose WHO osteoporosis; poor precision |
| Radiogrammetry | Cortical thickness (hand X-ray) | Minimal | Simple; plain film | Peripheral only; crude measure |
| MRI | Trabecular microarchitecture | None | No radiation; structural detail | Expensive; not routine; research mostly |
✦ STEP 18 — Treatment Context (DXA Triggers Treatment Decisions)
After DXA diagnosis, pharmacological treatment is initiated based on:
| Criterion | Action |
|---|
| T-score ≤ −2.5 | Initiate treatment |
| Fragility fracture (any BMD) | Initiate treatment |
| Osteopenia + FRAX hip fracture ≥ 3% | Consider treatment |
| Osteopenia + FRAX major fracture ≥ 20% | Recommend treatment |
First-line agents: Bisphosphonates (alendronate, risedronate, zoledronic acid) or denosumab
High-risk agents: Teriparatide, abaloparatide, romosozumab (T ≤ −2.5 with fractures)
Monitoring on treatment: DXA every 1–2 years until stable, then every 2–3 years
✦ STEP 19 — Current Guidelines (Most Recent — Exam Booster)
| Guideline | Key Message |
|---|
| Slart et al. 2025 (EJNMMI, [PMID: 39316095]) | Technical QC, positioning, ROI analysis, reference range selection, and VFA integration are all critical for valid DXA reporting |
| ISCD 2023 | Official positions on scan sites, reporting standards, pediatric DXA, body composition thresholds |
| IOF/ESCEO | FRAX integration; VFA in routine practice |
| VFA Working Group 2025 ([PMID: 41338753]) | VFA should be standard, not optional, in all DXA assessments |
✦ STEP 20 — One-Page Exam Rapid Revision Table
| Feature | Key Fact |
|---|
| Full name | Dual-Energy X-ray Absorptiometry |
| Gold standard for | BMD measurement, osteoporosis diagnosis |
| Two X-ray energies | Low ~40–70 kVp + High ~100–140 kVp |
| Two energy methods | K-edge filtration OR voltage switching |
| Primary output | BMD in g/cm² (areal, NOT volumetric) |
| Primary scan sites | L1–L4 spine, Proximal femur, 1/3 distal radius |
| T-score cutoff — osteoporosis | ≤ −2.5 (WHO) |
| T-score — used in | Postmenopausal women; men ≥ 50 |
| Z-score — used in | Premenopausal; men < 50; all children |
| Z-score cutoff — secondary cause | ≤ −2.0 |
| Radiation dose (spine DXA) | ~1–3 µSv (< 1 day background radiation) |
| Most common artifact | Osteophytes → falsely elevate lumbar BMD |
| Positioning key for hip | Foot internally rotated 15–25° |
| Positioning key for spine | Knees flexed 90° over block |
| LSC formula | 2.77 × CV% |
| Daily QA | Spine phantom scan |
| VFA grading scale | Genant semi-quantitative (Grade 0–3) |
| Body composition model | 3-compartment: bone + lean + fat |
| Size artifact | Areal BMD overestimates large bones |
| Monitoring frequency | Every 1–2 years on treatment |
| Latest guideline PMID | 39316095 (Slart 2025); 41338753 (VFA 2025) |
Sources:
- Lippincott Illustrated Reviews: Pharmacology, pp. 898–899
- Goldman-Cecil Medicine, p. 3995
- Washington Manual of Medical Therapeutics, p. 5728
- Grainger & Allison's Diagnostic Radiology, p. 813–820
- Slart RHJA et al. Eur J Nucl Med Mol Imaging 2025 [PMID: 39316095]
- Lems WF et al. Mayo Clin Proc 2025 — VFA Best Practices [PMID: 41338753]