Close to the bone – osteoporosis images and biomarkers

18 February 2015



It rarely enters the limelight but osteoporosis is one of the greatest challenges facing modern healthcare. Dr Thomas Link, professor of musculoskeletal imaging at the University of California, San Francisco, tells Jack Wittels about the ongoing evolution of diagnostic imaging in the field and the latest research into new biomarkers.


Every three seconds, someone, somewhere, suffers an osteoporotic fracture. Older women are most at risk: one in ten aged 60 and a staggering two in three 90-year-olds are affected, according to the International Osteoporosis Foundation. In men, around 20% of those aged over 50 experience an osteoporotic fracture.

The problem is by no means limited to developing countries, either. In the EU, approximately 22.0 million women and 5.5 million men aged 50-84 have osteoporosis, according to 2010 figures. By 2025, the total is expected to grow by 23%. In the US, osteoporosis and low bone mass were estimated to affect 44 million men and women aged 50 and over in 2011 - around 14% of the total population.

With Western nations facing spiralling healthcare costs and an aging society, osteoporosis looks set to become an even bigger problem. Worse still, it is not receiving anything like the attention given to other major diseases from either the medical community or general public.

"Try comparing osteoporosis with something like cancer," says Professor Thomas Link, head of musculoskeletal imaging at the University of California, San Francisco (UCSF). "Cancer has such a devastating image. It makes people think, 'Oh my God, we have to invest so much money in this'. With osteoporosis, hip fractures happen a lot, and they're pretty devastating - up to 20% of elderly hip fracture patients die within a year, but people aren't so conscientious about it; it's seen as a disease of older people."

Link has worked at UCSF's Department of Radiology and Biomedical Imaging for over a decade. During that time, he's developed relationships with a number of referring physicians, including endocrinologists and rheumatologists, who regularly call to discuss osteoporosis-related issues and the latest literature. Patients have also shown a strong interest in the condition, but the key medical group, Link's fellow radiologists, remain frustratingly aloof.

"The numbers of radiologists interested in osteoporosis is always small. People see it as an unattractive subject to study because it's purely quantitative. The images aren't exciting; it's all about the numbers," he explains.

"Up to 20% of elderly hip fracture patients die within a year, but people aren’t so conscientious about it; it’s seen as a disease of older people."

Consequently, radiologists often fail to spot the telltale signs of osteoporosis early on - it's simply not on their diagnostic radar. Link gives the example of vertebral spine fractures: a strong indicator of future osteoporosis frequently missed by radiologists, often due to a lack of symptoms. Regular training, he says, is crucial to overcoming such mistakes.

Mass effect

Diagnosing osteoporosis traditionally involves measuring bone-mineral density (BMD) using dual X-ray absorptiometry (DXA). This indicates bone mass, which accounts for around 70% of overall bone strength. Highly reproducible and reliable, DXA analysis can predict future fracture risk and monitor patients undergoing osteoporosis therapy.

In 2008, the World Health Organization, in collaboration with Sheffield University, developed the FRAX system, an online tool combining BMD data of the femoral neck with clinical risk factors to calculate the ten-year probability of a major osteoporotic fracture - clinical spine, forearm, hip or shoulder. However, while FRAX certainly improves diagnosis, it is often an ineffective measure for patients already suffering from other conditions. Those with diabetes, for example, have high DXA BMD scores but also have a
high risk of fracture.

In recent years, the limitations of DXA BMD analysis have pushed radiologists to focus on a new area: bone quality. Defined as "bone architecture, turnover, damage accumulation (for example, microfractures) and mineralisation" by the National Institutes of Health, it offers fresh insight for physicians looking to measure fracture risk.

Among the emerging technology associated with this new approach, Link sees high-resolution peripheral quantitative computed tomography (HR-pQCT) as the most exciting. With a comparatively high signal:noise ratio and low effective radiation dose, it offers excellent images of trabecular bone architecture and the cortex - a dualism that's proven extremely effective.

"In the past, we always thought bone metabolism was best assessed through looking at trabecular bone, but over the past decade, we've learnt that cortical bone also provides independent and important information. Diabetic bone disease, for instance, is better quantified using measures of cortical bone than trabecular," says Link.

"It's currently only used as a research tool, though it's available at most major institutions with a focus on bone metabolism. It also requires a special machine, and can only be performed on the radius and distal tibia."

CT rethink

These limitations can be overcome by using multidetector CT - a standard clinical technique that gives access to central skeletal regions with high fragility-fracture risk, such as the spine and proximal femur. Good spatial resolution is also possible with this approach, though it comes with the downside of greater radiation exposure - DXA analysis involves effective patient doses of 0.001-0.050mSv, while the multidetector CT dose is approximately 3.000mSv.

Link is hopeful, however, that a relatively new research technique, known as biomechanical CT (BCT), will allow radiologists to use multidetector CTs more effectively and without the high radiation risk.

It is currently under development by one of his colleagues, Dr Tony Keaveny of Berkeley, also part of the University of California, and uses finite element modelling to analyse CT scans and calculate bone strength. BCT has already shown greater sensitivity than DXA in fracture risk assessment and, because it is effectively an add-on to existing CT scans, clinical implementation should be cheap.

"Potentially, with BCT, we could have a tool to calculate bone strength in every single patient who undergoes a CT. In older patients, this will be a terrific tool to diagnose increased fracture risk."

"It's an extremely interesting technique, and Dr Keaveny is now testing it on thousands of standard CT images," says Link. "Potentially, with BCT, we could have a tool to calculate bone strength in every single patient who undergoes a CT. In older patients, this will be a terrific tool to diagnose increased fracture risk and subsequently prevent fractures.

"We radiologists have a lot of imaging technologies at our disposal that we've not used adequately for assessment of bone strength or osteoporosis prediction," he adds.

Personalising osteoporosis

Link is also enthusiastic about recent developments in magnetic resonance spectroscopy (MRS). An effective measure of bone marrow adiposity, the technique lets researchers explore the recently proposed hypothesis that osteoporosis is essentially obesity of the bone, providing a quantitative assessment of water and fat levels.

"It's a very exciting technique," says Link. "We've already shown that unsaturated fats are a good way of separating out patients with and without fractures. For the moment though, MRI-based tools are too expensive for routine osteoporosis and fracture risk diagnosis. I think they'll mainly be useful in scientific studies for testing new treatments and for better characterising metabolic bone diseases."

In the longer term, Link sees osteoporosis diagnosis heading down a similar route to cancer treatment, with physicians taking a more genetics-based, personalised approach.

"We're now at a stage where every patient has far more individualised healthcare than 20 years ago, when everyone was treated in the same way," he explains.

"With cancer, the focus is increasingly on trying to understand patients' genetic codes, allowing physicians to tailor treatment accordingly. I believe osteoporosis is heading in a similar direction. Recently, the number of biopsies we've been performing has tripled - largely in the hunt for biomarkers to use for personalised medication."

Link is currently working with physicians in Vienna to better understand the connection between microRNAs, genetic markers, biomarkers and osteoporosis. So far, the ongoing analysis has revealed a correlation between a special subset of microRNAs and an abnormally high number of fractures in some patients.

Yet, despite his focus on the future of diagnostic techniques, Link still sees the more traditional approaches of awareness-raising and early prevention as the most effective ways to fight osteoporosis over the years to come.

"I'm frequently seeing older patients with devastated skeletons - there's so much loss of bone architecture and bone mass. Governments need to invest in prevention, not only to improve people's quality of life but also because, after hip fractures, patients commonly end up in nursing homes, which is expensive and a major challenge for our healthcare systems," he says.

"If there's no awareness, patients just slide into osteoporosis but, once it's diagnosed, we can do a lot. Everybody needs to understand the risks - and what we can do about it."

Dr Thomas Link is chief of the musculoskeletal imaging section and clinical director of the Musculoskeletal and Quantitative Imaging Research Group in the Department of Radiology and Biomedical Imaging at the University of California, San Francisco. His work focuses particularly on osteoporosis imaging and osteoarthritis.
A 3D-rendered illustration of osteoporotic bone. Scientists are hopeful that advanced imaging techniques could offer improved diagnosis and risk assessment.


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.