Bone Microarchitecture Independently Predicts Fractures

December 27, 2018

By Will Boggs MD

NEW YORK (Reuters Health) - Bone microarchitecture, as measured by high-resolution peripheral quantitative CT (HR-pQCT) independently predicts fracture risk in older women and men, according to the Bone Microarchitecture International Consortium (BoMIC) study.

"This study provides the groundwork for the development in the future of new approaches to screening for fracture risk," said Dr. Elizabeth J. Samelson from Hebrew SeniorLife and Beth Israel Deaconess Medical Center, Harvard Medical School, in Boston.

"Use of HR-pQCT or other imaging modalities to be developed in the future offer promising improvements in screening strategies to ultimately reduce the tremendous public health burden of osteoporosis," she told Reuters Health by email.

Bone-mineral density (BMD) assessed by dual-energy x-ray absorptiometry (DXA) is the clinical standard for estimating fracture risk, but most adults who sustained a fracture have T scores greater than -2.5, the threshold for diagnosing osteoporosis.

Dr. Samelson and colleagues investigated the association between HR-pQCT bone indices and incident fracture in more than 7,200 individuals (mean baseline age, 69 years) followed for a mean of 4.63 years.

During this time, 11% of the participants had incident fractures, with femoral neck T scores greater than -2.5 found in 86% of these people.

In unadjusted analyses, nearly all HR-pQCT bone measures were worse in individuals who sustained an incident fracture than in those who did not.

After adjustment for sex, age, height and weight, the strongest association with incident fracture was seen for failure load at the distal tibia and distal radius, the researchers report in The Lancet Diabetes & Endocrinology, online November 28.

Failure load retained the strongest association with fracture after adjustment for femoral neck areal BMD (aBMD), and most associations between HR-pQCT parameters and incident fractures remained significant albeit attenuated.

Results were similar after adjusting for Fracture Risk Assessment Tool (FRAX) score.

"The consistency of our findings was the most interesting and surprising" result, Dr. Samelson said. "Specifically, the ability of bone microarchitecture to predict fracture was independent of both BMD and FRAX, and this finding held true for both older women and men, for those with T-scores considered normal, osteopenic, and osteoporotic, and for different fracture outcomes."

The best set of predictors for fracture (cortical density, trabecular number and trabecular thickness at the distal radius plus cortical area at the tibia) improved the ability to predict major osteoporotic fracture beyond femoral neck aBMD at the radius but not at the tibia.

"It is important for clinicians to continue to follow clinical guidelines that rely on DXA BMD and FRAX for screening and monitoring treatment for fracture reduction," Dr. Samelson said. "However, low BMD is not the only cause of skeletal fragility. In fact, most individuals who fracture do not have BMD T-scores that meet diagnostic criteria for osteoporosis. By identifying bone microarchitecture as an independent risk factor for fracture, this study highlights opportunities to develop new strategies to evaluate characteristics of bone fragility, beyond bone density, that can ultimately improve risk stratification and provide new targets for therapy."

Dr. Elizabeth Shane from Vagelos College of Physicians and Surgeons at Columbia University, in New York City, who wrote an accompanying editorial, told Reuters Health by email, "HR-pQCT and finite element analysis of HR-pQCT scans predict fracture better than our currently available tools and could help decide which patients require treatment. These newer tools can provide additional information about bone quality not captured by DXA and should be incorporated into observational and interventional research on osteoporosis."

"However, few HR-pQCT machines are available, and osteoporosis and fractures affect millions who don't have access to them," she said. "More research is needed on whether DXA of the ultradistal radius outperforms DXA of the hip and spine in fracture prediction."

"HR-pQCT would fit best into quantifying fracture risk of a person who did not have osteoporosis by DXA," Dr. Shane added. "As noted, the majority of fractures occur in people without osteoporosis, who have 'osteopenia or low bone mass.' That is the group of people for whom better tools are needed to determine who needs to initiate therapy to prevent fractures."

Dr. Tuan V. Nguyen from Garvan Institute of Medical Research, in Darlinghurst, Australia, who recently reviewed fracture-risk assessment at the individual level, told Reuters Health by email, "I am a bit surprised that the HR-pQCT measures (e.g., cortical thickness, cortical bone area, cortical bone density, trabecular thickness, trabecular numbers, trabecular bone density, bone failure load) do not provide much more information about fracture risk that a simple DXA bone-mineral density measurement. Indeed, the 1 to 2 percentage points improvement in the AUC value (a measure of fracture discrimination) over and above that provided by traditional bone-mineral density measurement attributed to HR-pQCT was not clinically significant."

"Given the cost associated with HR-pQCT measurements and their small incremental predictive value (over and above areal bone-mineral-density measurement), I think more evidence of cost-effectiveness is needed in order to make decision concerning the routine use of HR-pQCT in clinical practice," he said.

"Bone architecture does matter to bone fracture," said Dr. Nguyen, who was not involved in the research. "The present study reiterates that when it comes to fracture risk assessment, a measurement of bone mineral density or 'T-scores' alone is probably not enough; other factors, such as bone architecture, fall, lifestyle factors, and comorbidities, need to take into account to improve the accuracy of fracture prediction for an individual."


Lancet Diabetes Endocrinol 2018.

(c) Copyright Thomson Reuters 2018. Click For Restrictions -

For more Annals of Long-Term Care articles, visit the homepage

To view the Annals of Long-Term Care print issue, click here