X-rays for Soft-tissue Imaging
The ability to harness x-ray characteristics as they interact with matter, other than the absorption of conventional radiography, has been a subject of emerging interest in the development of both experimentally and clinically based imaging technology. The basis for this interest has been the desire to render images of soft tissues and hard tissues simultaneously for medical and biologic applications. Because soft tissues do not possess the density variations necessary to be detected through conventional radiographic means, other sources of x-ray contrast are now an object of research.
Absorption is the primary source of contrast in conventional radiography, but it is not the only interaction that x-rays have with matter; x-rays can also be refracted and scattered. Both refraction and scatter are properties that describe angular deviation of x-rays as they traverse matter. Scattering falls into two categories: ultra-small angle scattering, in which the x-rays are randomly refracted by multiple structures within the sample, and small to wide-angle scattering, in which the x-rays are redirected through either Compton or elastic scattering. The small to wide-angle scattering degrades the image quality in conventional radiography.
The amount of x-ray refraction within a biologic specimen depends on three factors: the angle of the interface between two adjacent regions, the difference between the refractive indices, and the x-ray energy.1 Typical angular changes to the x-ray beam from biologic tissues are on the order of tenths of microradians (comparable to 1 mm at 10 km). Until recently, it was not possible to resolve these subtle angular changes in the x-ray beam.
Analyzer-based imaging (ABI, also referred to as diffractionenhanced imaging) employs precision x-ray optics to extract the refraction information from the x-ray beam. The properties of perfect crystal optics are exploited to convert small angular deviations in the transmitted x-ray beam into largeintensity variations in the image.2 Because the x-ray optics constitute, in essence, a notch filter affecting both angular range and delivered energy, only x-rays within a very narrow energy and angular range will be transmitted through the system. Thus, almost all small to wideangle scattered x-rays are removed from the transmitted beam, and image clarity is improved relative to conventional radiography,2 in which scatter is generally an unwanted property. Both in removing scattered x-rays and in extracting refraction information, ABI reveals information already present in the x-ray beam but not currently being used to generate image contrast.
The original ABI studies were performed at facilities such as the National Synchrotron Light Source at Brookhaven National Laboratory in Upton, NY, because synchrotrons are very bright, collimated x-ray sources. Early studies have shown significant contrast-tonoise ratio gains over conventional radiography for imaging applications such as mammography in both planar3 and CT modes.4 These studies have provided the basis on which non-synchrotron ABI units (using a conventional x-ray tube) are now being developed so that they may eventually be used in the laboratory setting and in the clinic.
However, the synchrotron studies themselves also tell us what the capabilities of ABI really are and, therefore, what we should strive for from a system using a conventional x-ray source. For instance, we have found that ABI based on refraction information allows the visualization of soft tissues in intact specimens such as cadaveric knee (Figure 1), finger joints, and whole foot/ankle specimens while simultaneously producing high contrast in the calcified tissues.5-7 The contrast for articular cartilage, including changes characteristic of osteoarthritis, is quite impressive. The significance of this lies in the ability to simultaneously follow cartilage and subchondral bone changes through degenerative stages. The x-ray refraction contrast at borders, such as between tendons and other surrounding soft tissue, and between the menisci and articular cartilage, is also apparent.
The capabilities described in synchrotron studies are being developed in ABI systems that use a conventional x-ray source at high energies, in which the radiation dose is lower. These systems provide simultaneous imaging of hard and soft tissues and should find much appeal in the longitudinal study of diseases such as osteoarthritis and in the image-based assessment of treatment strategies.
Carol Muehleman PhD
Dean Connor PhD
New Tough Gels for Cartilage Replacement
For many years, it has been common to grow chondrocytes in threedimensional (3D) gel culture. The reason is that chondrocytes grown in 3D culture retain more of their normal behavior compared with those grown in a flat dish. This makes sense teleologically because chondrocytes are normally resident in a 3D collagenous gel, namely, cartilage.
Although chondrocytes are grown in a 3D gel and retain their behavior as chondrocytes, manufactured gels cannot be used to replace cartilage lost to arthritis because the gels are too weak to sustain the normal mechanical loading in an articular joint. Normal stresses in the knee, for example, are more than ten times the crush strength of an ordinary culture gel. What is needed is a gel with strength similar to that of healthy articular cartilage.
A new class of recently invented hydrogels (ie, polymer gels filled with water) is extremely strong and tough.8-16 This new design, called an interpenetrating network hydrogel, is soft, yet it has a very great tolerance for stress and strain. The properties of these hydrogels approach those required for artificial cartilage,15-17 and they are compatible with cell culture.
The key feature of interpenetrating gels is a strong but easily compressed fibrous network, which is filled with a weak but viscous and difficult-tocompress “gel” that traps water in the fiber matrix. This feature is, of course, reminiscent of the structure of cartilage, ligament, tendon, and other soft connective tissue. As a result of the similarity between these manufactured gels and natural cartilage, they are sometimes called “biomimetic” gels—that is, of a structure that mimics the natural tissue.
The mathematical models for an interpenetrating gel might also be a key to understanding bone toughness. If the soft and weak glycosaminoglycans in cartilage are replaced by a fragile but stiff mineral matrix, then the resulting “calcified cartilage” (or bone, because the model is so simple) is an interpenetrating network composite. Any microcrack in the mineral is immediately bridged and supported by the collagen matrix, resulting in a toughened material. This is similar to the model of Yeni and Fyhrie,18 who analyzed bone as a collagen-mineral composite with bridging at the microscopic level.
Publications on these materials are appearing at an accelerating rate, and mathematical models for their remarkable mechanical properties now exist. Artificial interpenetrating network polymers may well become a regular part of medical practice for both space-filling and mechanical load-bearing applications.
David P. Fyhrie PhD
Toward an Objective Measurement of Articular Fracture Injury Severity
The severity of an articular fracture, in particular the energy causing the injury, determines to a significant extent whether the joint progresses to disabling posttraumatic osteoarthritis (PTOA). However, fracture severity has not been amenable to objective measurement. Current articular fracture classification systems do not completely capture or quantify the severity of injury; fracture classification also is subject to poor interobserver reliability. Treatment decisions by clinicians must therefore be made based on subjective assessments of the injury severity. This approach is in contrast to decision-making done in many other clinical scenarios, in which objective measures are a fundamental part of evaluating and treating patients. For instance, physiologic and anatomic measurements (eg, ejection fraction and percent damage of the myocardium) are considered routine in assessing a patient with a myocardial infarction. These objective measurements of the severity of the cardiac injury are critically important guides to clinical management.
A novel CT-based method has been developed to quantify the severity of articular injury. Fracture mechanics theory predicts that the fracture energy is directly proportional to the interfragmentary surface area liberated by a fracture, a theory that we have corroborated in bench-top studies of bone fracture.19 Clinically, this means that fracture comminution is proportional to fracture energy. Clinicians observe a radiograph of a severely comminuted fracture and call the injury a “high-energy fracture;” alternatively, a noncomminuted fracture, based on its radiographic appearance, is called a “lowenergy fracture.” What clinicians observe in the high-energy case is greater fragment surface area. Thus, what the clinician sees on radiographs can now be quantitatively assessed using CT-based image analysis techniques.
These analysis techniques have been refined to measure not only fragment surface area (ie, comminution) but also fracture fragment displacement.20 The bone density is accounted for through a correction based on CT scan Hounsfield units. With an articular fracture of the distal tibia used as a clinical model, these metrics have been shown to closely correlate with PTOA on radiographs at a minimum of 2 years after injury.21 In addition, this research has suggested the concept of an energy threshold, above which all ankles develop radiographic osteoarthritis and below which ankles maintain their articular surface. This finding raises the possibility that the information present on the CT scan at the time of injury can be used to predict joints that are highly likely to develop PTOA despite current treatment, thereby setting the stage to consider alternative treatments soon after injury.
Currently, clinicians make treatment choices for high-energy articular fractures by intuition and clinical experience. For example, instead of reduction and fixation of a fracture, joint arthrodesis might be chosen as a treatment alternative by the clinician, based on his or her subjective assessment of the severity of articular injury. Changing from this paradigm of reliance on subjective judgment to decisions based on objective measures would be an important step in improving clinical decision making. Such a change will be even more important in the future, with the advent of biologic interventions to help preserve damaged articular surfaces. Choosing new interventions clinically, or designing clinical research studies to assess their effectiveness, requires that the degree of damage to the articular surface be meaningfully stratified. This research to measure fracture energy makes objective stratification of injury severity a reality.
In determining fracture energy from clinical CT scans, the fragment surface area is calculated, for example, slice by slice across the area of the distal tibia.20 Both limbs are scanned so that the surface area on the noninjured contralateral side can be subtracted from that of the injured side, resulting in the liberated surface area caused by the injury. When corrected for bone density and multiplied by a scalar material property, the liberated surface area provides the energy of injury. Fragment dispersal is calculated through a volumetric envelope of the injured versus noninjured side. Further work with texture analysis is under way to expedite the technique and to minimize or eliminate the need for operator intervention.
In summary, an objective articular fracture severity metric based on CT data obtained as part of routine clinical care has shown tremendous promise as an aid to clinical research and to patient care. This metric has been shown to correlate with the propensity of tibial plafond fractures to develop PTOA,21 and preliminary data suggest an energy threshold above which PTOA is practically inevitable.21 This technique is being refined for use in other anatomic areas (eg, tibial plateau). In addition to the authors' institution, it will soon start to be used in two additional trauma centers. Further work is under way to speed the calculations of these objective measures, to translate their use from the experimental realm, and to realize their potential as a valuable tool for clinical research and, eventually, clinical care.
J. L. Marsh MD
Don Anderson PhD
1. Connor DM, Sayers D, Sumner DR, Zhong Z: Identification of fatigue damage in cortical bone by diffraction enhanced imaging. Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
2. Chapman D, Thomlinson W, Johnston RE, et al: Diffraction enhanced x-ray imaging. Phys Med Biol
3. Hasnah MO, Zhong Z, Oltulu O, et al: Diffraction enhanced imaging contrast mechanisms in breast cancer specimens. Med Phys
4. Bravin A, Keyriläinen J, Fernández M, et al: High-resolution CT by diffractionenhanced x-ray imaging: Mapping of breast tissue samples and comparison with their histo-pathology. Phys Med Biol
5. Li J, Zhong Z, Lidtke R, et al: Radiography of soft tissue of the foot and ankle with diffraction enhanced imaging. J Anat
6. Li J, Zhong Z, Connor D, Mollenhauer J, Muehleman C: Phase-sensitive X-ray imaging of synovial joints. Osteoarthritis Cartilage
2009; Mar 24: [Epub ahead of print].
7. Muehleman C, Li J, Zhong Z, Brankov JG, Wernick MN: Multiple-image radiography for human soft tissue. J Anat
8. Calvert P: Hydrogels for soft machines. Advanced Materials
9. Chen YM, Gong JP, Tanaka M, et al: Tuning of cell proliferation on tough gels by critical charge effect. J Biomed Mater Res A
10. Weng LH, Gouldstone A, Wu YH, Chen WL: Mechanically strong double network photocrosslinked hydrogels from N,N-dimethylacrylamide and glycidyl methacrylated hyaluronan. Biomaterials
11. Tominaga T, Tirumala VR, Lee S, Lin EK, Gong JP, Wu WL: Thermodynamic interactions in double-network hydrogels. J Phys Chem B
12. Tominaga T, Kurokawa T, Furukawa H, Osada Y, Gong JP: Friction of a soft hydrogel on rough solid substrates. Soft Matter
13. Tanabe Y, Yasuda K, Azuma C, et al: Biological responses of novel hightoughness double network hydrogels in muscle and the subcutaneous tissues. J Mater Sci Mater Med
14. Myung D, Koh WU, Ko JM, et al: Biomimetic strain hardening in interpenetrating polymer network hydrogels. Polymer
15. Yasuda K, Ping Gong J, Katsuyama Y, et al: Biomechanical properties of hightoughness double network hydrogels. Biomaterials
16. Yang-Ho N, Katsuyama Y, Kuwabara R, et al: Toughening of hydrogels with double network structure. e-Journal of Surface Science and Nanotechnology
17. Azuma C, Yasuda K, Tanabe Y, et al: Biodegradation of high-toughness double network hydrogels as potential materials for artificial cartilage. J Biomed Mater Res A
18. Yeni YN, Fyhrie DP: A rate-dependent microcrack-bridging model that can explain the strain rate dependency of cortical bone apparent yield strength. J Biomech
19. Beardsley CL, Anderson DD, Marsh JL, Brown TD: Interfragmentary surface area as an index of comminution severity in cortical bone impact. J Orthop Res
20. Anderson DD, Mosqueda TV, Thomas TP, Hermanson EL, Brown TD, Marsh JL: Quantifying tibial plafond fracture severity: Absorbed energy and fragment displacement agree with clinical rank ordering. J Orthop Res
21. Thomas TP, Anderson DD, Mosqueda TV, et al: Objective CT-based metrics of articular fracture severity to assess risk for post-traumatic osteoarthritis. J Orthop Trauma
2009 (provisionally accepted).