SAN FRANCISCO — Investigational bone imaging techniques hold the promise of providing clinically useful information about bone structure in three dimensions unattainable with dual-energy x-ray absorptiometry, Mary L. Bouxsein, Ph.D., said at a meeting on osteoporosis sponsored by the University of California, San Francisco.
While dual-energy x-ray absorptiometry (DXA) measurements show moderate to strong correlations with whole bone strength, the technique cannot distinguish specific attributes of three-dimensional geometry, cortical versus cancellous density, trabecular architecture, or intrinsic properties of the bone matrix.
However, there are some exciting newer techniques, although they are currently in the research phase, said Dr. Bouxsein of Harvard Medical School, Boston.
She highlighted the advantages and limitations of five promising novel imaging techniques:
▸ Hip strength analysis (HSA). This technique uses image data from 2-D DXA to derive 3-D geometry. Developed in the early 1980s, HSA uses data from the attenuation profile of the x-ray beam to calculate such things as cortical thickness and bone strength. But HSA makes several assumptions, most notably that there is constant mineral density in the bone and that the neck and shaft of bones are circular.
“This makes a lot of sense in measurements of long bone, and that was where it was first developed,” Dr. Bouxsein said. “Now where I think the challenge comes in … applying this exact same technique to measurement of the femoral neck. … It would be a big challenge to extract properties of the cortex using this direct method.”
One attraction of HSA is that it requires no new data collection, since researchers can reanalyze old DXA scans with this new technique. However, there is much more attention currently being focused on techniques that are truly three-dimensional.
▸ Quantitative computed tomography (QCT). With QCT, standard CT images are made with a bone-density “phantom” in the viewing area. This allows a quantitative measure of bone density in the final image and gives a 3-D view of bone geometry as well as an isolated look at the trabecular and cortical compartments.
The technique's precision, currently at about 2%–6%, is a bit worse than DXA, Dr. Bouxsein said. The radiation dose is higher than DXA, although still low enough to be acceptable for longitudinal studies.
The resolution is on the order of 300 mcm by 1 mm, somewhat too low to resolve individual trabecular elements, which measure about 100–300 mcm in the adult spine.
Studies have shown that QCT does show age- and treatment-related effects and is useful at both axial and peripheral sites. On the other hand, there are no prospective fracture data (although some are coming), reference data are limited, analysis methods are not yet standardized, and marrow fat can influence QCT measurements, with an increase in marrow fat resulting in an apparent decrease in trabecular bone mineral density (BMD).
▸ High-resolution peripheral quantitative CT (pQCT). This technique has become available within the past 2 years. It uses specialized equipment and can measure the peripheral skeleton, including the distal radius and distal tibia.
It has a voxel size of just 82 mcm
The technique's precision is quite good, with short-term reproducibility of 0.7%–1.3% for density and 0.9%–5.1% for microarchitecture. Because it's a peripheral technique, the patient receives a relatively low dose of radiation, and because it uses dedicated equipment, measurements are standardized. Using pQCT, researchers have been able to discriminate among osteopenic women with and without a history of fragility fracture, even when their BMDs are quite similar.
On the other hand, pQCT can't be used to measure central sites, and access is quite limited, with only about 10 machines in the world, including three in the United States. Clinical studies have so far been limited.
▸ High-resolution MRI (HR-MRI). This technique for trabecular bone uses standard clinical scanners and allows physicians to conduct virtual bone biopsies. The technique discriminates patients with a history of fragility fracture from controls, uses nonionizing radiation, and may be useful for monitoring response to treatment.
At the moment, HR-MRI can only be used at peripheral sites, although there is some potential for its use at the hip, Dr. Bouxsein noted.
Its resolution is a bit too low, at just about 100–300 mcm, but that is expected to improve with higher magnetic-field strengths.
Its short-term reproducibility, meanwhile, has been measured at a disappointing 3%–8%. Finally, the HR-MRI technique has been the subject of only limited clinical studies.
▸ Finite element analysis (FEA). Dr. Bouxsein said she was particularly excited about this technique, which is a standard engineering technique that has been used for decades to measure the mechanical properties of airliners and other complex structures. FEA for bone can provide multiple strength metrics.