Which Regions of Spine and Hip Should Be Used to Assess Osteoporosis and/or Risk of Fracture?

Dennis Black, PhD


March 17, 2004


I understand that the gold standard for bone mineral density (BMD) testing is central dual-energy x-ray absorptiometry at the hip and spine. However, what about the more-specific reports that list T-scores for parts of the hip and spine, eg, GE Lunar Prodigy readings of neck, upper neck, Wards area, trochanter, L1, L2, etc? Are these numbers reliable, or should we go by the AP spine (L1-4) and dual-femur (total mean) scores instead?

Frank Spence, MD

Response from Dennis Black, PhD

There is general agreement among experts with respect to which regions/subregions of hip and spine in BMD assessment should be used to assess osteoporosis and/or risk of fracture, although there are also some areas of disagreement.

First, the easy part: areas of agreement. At the spine, the individual vertebrae are unreliable and should not be considered. Therefore, the overall mean (L1 to L3 or L1 to L4, depending on machine) T-score should be used in diagnosis.

At the hip, the "total hip" (an average across the whole hip) is generally recommended as a reliable measurement. There is agreement that Ward's triangle should be disregarded; the Ward's triangle definition differs drastically from machine to machine. Also, the trochanteric and intertrochanteric regions are generally not used. There is some disagreement as to whether the femoral neck should be considered, but most clinicians do measure the femoral neck.

The International Society for Clinical Densitometry recommends consideration of 3 measurements: total spine, total hip, and femoral neck.

The following seems like a reasonable compromise: take the lowest of the 3 T-scores for the total spine, total hip, and femoral neck. The use of more measurements runs the risk of overdiagnosis or diagnosing/treating a patient whose fracture risk is only moderate.

In fact, almost all studies of the relationship of BMD-to-fracture risk have looked only at a single measure (eg, total spine, total hip, or femoral neck). There is little research to definitively judge whether "lowest of the 3" is a better predictor of fracture risk than just a single measurement. In the absence of definitive studies, however, 'the lowest of the 3' seems a reasonable clinical strategy.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.