7.3 Integrating TBS with Central DXA & FRAX
Key Takeaways
A valid TBS adds information associated with fracture risk. It does not change the DXA densitometric classification.
Use a validated TBS adjustment and document it when reporting FRAX probabilities. The result is most likely to affect care near a local intervention threshold.
In type 2 diabetes, fracture risk can be elevated despite relatively preserved BMD. TBS is one part of the clinical assessment.
ISCD advises against routinely monitoring or reporting TBS change during osteoporosis therapy.
Use TBS to answer a risk question
The central dual-energy X-ray absorptiometry (DXA) report, fracture history and clinical risk factors answer different parts of the assessment. Bone mineral density (BMD) supports densitometric classification. The Fracture Risk Assessment Tool (FRAX) estimates absolute fracture probability. An eligible lumbar trabecular bone score (TBS) adds information associated with fracture risk and can be used to adjust the estimate.
The 2023 International Society for Clinical Densitometry (ISCD) positions identify associations with hip, vertebral and major osteoporotic fracture in postmenopausal women and with hip and major osteoporotic fracture in men over age 50. TBS is also associated with major osteoporotic fracture risk in postmenopausal women with type 2 diabetes. These are population findings. A score cannot determine the cause of one patient's skeletal fragility or establish exactly how much of their risk comes from microarchitecture.
| Information | Role in the decision |
|---|---|
| Valid DXA T-scores | Establish the densitometric classification in the appropriate population |
| Prior low-trauma fracture | Changes risk and can independently support clinical treatment consideration |
| Conventional FRAX | Estimates 10-year hip and major osteoporotic fracture probability from the appropriate model and inputs |
| Valid lumbar TBS | May refine the estimate, particularly if a change would affect the management decision |
Adjust the probability using a validated implementation
Check whether TBS can be reported from L1–L4 under the applicable technical rules. Use a validated implementation of the TBS adjustment when an adjusted probability is appropriate. Document the FRAX model, positive clinical inputs, whether femoral neck BMD was included, and whether the reported estimates were adjusted for TBS. Do not substitute lumbar spine BMD into the standard femoral neck input or treat TBS as a second DXA T-score. The size and direction of the adjustment depend on the model and patient information, so a hand-applied universal multiplier or fixed reclassification rate is unsuitable.
A result near a local intervention threshold has more potential to change the decision than one far from it. A patient who already has a separate treatment indication, such as a qualifying hip or vertebral fracture under U.S. guidance, does not need a TBS adjustment to establish that indication. When several factors could modify FRAX, use clinical judgment and the applicable guidance. FRAXplus advises against stacking multiple post hoc corrections without evidence of their combined accuracy.
Worked threshold decision
Consider a 63-year-old postmenopausal woman with a valid femoral neck T-score of −1.9 and a spine T-score of −1.8. She is untreated and has no known hip or vertebral fracture. Suppose her documented U.S. FRAX output is 17.2% for major osteoporotic fracture and 2.8% for hip fracture. The report also supplies eligible TBS-adjusted probabilities of 18.2% and 3.1%, respectively. These are illustrative outputs supplied for interpretation.
Her hip BMD falls in the low bone mass range. The unadjusted estimate is below both U.S. Bone Health and Osteoporosis Foundation (BHOF) thresholds. The adjusted hip estimate meets the 3% criterion, while the adjusted major osteoporotic fracture estimate remains below 20%. Either qualifying probability suffices for treatment consideration in this setting. The TBS adjustment changes the risk discussion. It does not change her BMD classification or select a specific medication.
The increase from 2.8% to 3.1% is 0.3 percentage points. A hip fracture is included in the major osteoporotic fracture outcome, so adding the two probabilities would double count it. Explain each probability over the same 10-year horizon and why either can matter under the stated guideline.
Interpret diabetes and follow-up in context
In type 2 diabetes, fracture risk can be elevated even when BMD is relatively preserved. A lower eligible TBS can add risk information, but neither BMD nor TBS alone explains the full risk. Consider factors such as prior fracture, falls, and the clinical course of diabetes. When reviewing someone with another secondary contributor such as glucocorticoid exposure, use the applicable disease-specific assessment. A TBS number cannot replace that assessment.
ISCD advises against routinely monitoring or reporting TBS change during therapy. For treatment follow-up, assess fractures and other clinical changes and interpret technically comparable BMD with the applicable least significant change. A change in a serial TBS or FRAX result alone cannot establish therapeutic success or failure.
ISCD Adult Official Positions · FRAXplus adjustment guidance · Manitoba diabetes cohort · BHOF Clinician’s Guide
A 64-year-old postmenopausal woman with type 2 diabetes has a spine T-score of −0.9, a femoral neck T-score of −1.4 and a low, technically valid TBS. How should these findings be interpreted?
The discrepancy confirms a TBS software malfunction.
The patient must have osteopetrosis.
Type 2 diabetes can be associated with increased fracture risk despite relatively preserved BMD. The TBS can add risk information.
Subtract two standard deviations from the femoral neck T-score to establish osteoporosis.
In which case is an eligible TBS-adjusted FRAX estimate most likely to affect management?
A very low conventional estimate far below an applicable treatment threshold, with no additional risk concerns.
An established hip fragility fracture that already supports treatment consideration.
A healthy 30-year-old with no indications for TBS.
An untreated postmenopausal patient with a conventional probability close to a local treatment threshold.
A clinician asks whether routine serial TBS should be used to decide if teriparatide or alendronate is working. How should the DXA interpreter respond under the 2023 ISCD Adult Positions?
Routine TBS change monitoring and reporting are not recommended. Evaluate the patient clinically and use comparable BMD with the applicable LSC when indicated.
Use a fixed 3.5% TBS threshold for every patient and product.
A stable TBS confirms that any antiresorptive treatment is effective.
Replace serial BMD with TBS for all anabolic therapy patients.
Sections you finish are checked off in the contents.