Fractures associated with low bone mass are prevalent among older adults. Incident hip, vertebral, and forearm fractures annually outnumber the three leading major disease events in the U.S., including heart attacks, strokes, and breast cancer diagnoses, combined1. Hip fractures in particular are extremely debilitating and costly, both economically and in human burden, with mortality at approximately 20-25% in the first year following fracture, and morbidity, including permanent disability or inability to walk independently, as high as 70%2.

Bone strength, and thus resistance to fracture, is determined primarily by bone size/mass and its microarchitecture. Bone mass measured as bone mineral density (BMD) using dual-energy absorptiometry (DXA) has been the gold-standard clinical tool for diagnosis of osteoporosis and fracture prediction for nearly three decades3. In women and men, both vertebral and hip fracture risk increases nonlinearly approximately two-fold for every standard deviation decrease in BMD with increasing age beyond early middle-age; with a steeper rise in risk after age 75-804. However, with increasing age, the lumbar spine is susceptible to scan interpretation errors in the presence of osteoarthritis, which may erroneously increase BMD; thereby lowering apparent fracture risk5. These arthritic changes often render the spine unreliable for diagnosis and tracking BMD changes over time, a problem that worsens with advancing age; and thus, can lead to underdiagnosing and treating the segment of the population that may benefit the most from early detection of bone loss.

In addition to bone loss, loss of muscle mass and strength, known as sarcopenia, often coincide with bone loss with advancing age6. Thus, muscle and bone share similar age-associated changes: both decline in mass and quality, which in part contribute to loss of strength often leading to falls and fractures. Accordingly, the term “osteosarcopenia” has recently been used to describe parallel changes in bone and muscle due to aging or disuse, with proposed measures of muscular strength and gait speed to define muscle weakness together with BMD to assess bone loss7.

The contribution of bone to the muscle-bone relationship has largely been limited to assessment of bone mass, which fails to capture the contribution of bone quality and microarchitecture to this relationship. While DXA is limited by its inability to directly assess microarchitecture of trabecular bone, a relatively new software tool called trabecular bone score (TBS), is gaining interest among clinicians and researchers. TBS provides an indirect assessment of the strength and quality of trabecular bone within lumbar vertebrae, approximately two thirds of which are comprised of trabecular bone8, thereby adding value to fracture risk assessment1. TBS uses gray scale textural analysis of DXA-derived images of the posterior-anterior (PA) view of the lumbar spine to provide an indirect, non-invasive assessment of bone microarchitecture5. TBS is expressed as a unitless score that correlates with trabecular bone strength determined by Quantitative Computed Tomography (QCT) and high resolution peripheral QCT (HRpQCT)9. Because TBS is determined from the same DXA image as that for lumbar spine BMD, it does not require additional scan time, patient burden, or radiation exposure. Another advantage of TBS is that it appears not to be influenced by arthritic changes within the lumbar spine, thus reducing or eliminating the risk of underestimation of fracture risk when using BMD alone in the presence of osteoarthritis within the lumbar spine region of interest10.

In studies measuring the utility of TBS in determining fracture risk, in conjunction with BMD, the prediction of vertebral fractures increased by as much as 19% in comparison to TBS or BMD alone11. Yet, TBS has been shown to correlate poorly with BMD in postmenopausal women with fragility fractures, and in white men over 40 years of age12,13. The low correlations between these two measures are evidence that BMD and TBS capture different aspects of bone health but complement each other to create a more complete assessment of fracture risk5. Given these differences between BMD and TBS, and previous evidence of associations among lean mass, muscular strength, physical function, and BMD5,9,11,12,13, we sought to evaluate TBS as a possible marker of osteosarcopenia. Thus, the purpose of the present study was to determine the relationship between indices of DXA-determined lean mass, muscular strength, and physical function, with bone quality determined by TBS.