13 - Secondary Osteoarthritis

Authors: Moskowitz, Roland W.; Altman, Roy D.; Hochberg, Marc C.; Buckwalter, Joseph A.; GoldberG, Victor M.

Title: Osteoarthritis: Diagnosis and Medical/Surgical Management, 4th Edition

Copyright 2007 Lippincott Williams & Wilkins

> Table of Contents > II - General Aspects of Diagnosis > 9 - Magnetic Resonance Imaging

function show_scrollbar() {}

9

Magnetic Resonance Imaging

Charles G. Peterfy

Julie C. DiCarlo

Manish Kothari

Imaging Osteoarthritis with Magnetic Resonance Imaging

For more than two decades, magnetic resonance imaging (MRI) has been the imaging method of choice for evaluating internal derangements of the knee and other joints. Despite this, however, MRI has thus far played only a minor role in the study or management of osteoarthritis (OA). The main reason for this discrepancy has been the lack of effective structure-modifying therapies for OA. In the absence of therapy, clinicians have little need for methods of identifying patients who are most appropriate for the therapy or for determining how well the therapy worked. However, new insights into the pathophysiology of OA, coupled with advances in molecular engineering and drug discovery, have generated a number of new treatment strategies and raised the possibility of long-term control of this disorder. With this development has come a new demand for better ways of monitoring disease progression and treatment response in patients with OA. Noninvasive imaging techniques, particularly MRI, have drawn considerable attention in this regard. This interest has been intensified by the growing acceptance of structure modification and repair as an independent therapeutic objective in arthritis. Underlying this treatment strategy is the classic disease-illness debate: must therapies that effectively slow or prevent structural abnormalities in arthritis necessarily show an immediate parallel improvement in clinical symptoms and function, as long as they ultimately yield clinical benefits for the patient. Elucidating the structural determinants of the clinical features in arthritis has, accordingly, become a key objective for academia as well as the pharmaceutical industry.

MRI is ideally suited for imaging arthritic joints. Not only is it superior to most other modalities in delineating the anatomy, but also it is capable of quantifying a variety of compositional and functional parameters of articular tissues relevant to the degenerative process and OA. Moreover, because MRI is nondestructive and free of ionizing radiation, multiple parameters can be analyzed in the same region of tissue, and frequent serial examinations can be performed on even asymptomatic patients.

Accordingly, it is anticipated that MRI will play an increasingly important role in the study of OA and its treatment and that the demand for expertise and experience in evaluating the disease with this technology will increase commensurately. This chapter reviews the current state-of-the-art for MRI of OA and points to areas from where future advances are most likely to come.

Magnetic Resonance Imaging Technique

The clarity and detail with which MRI depicts cross-sectional anatomy makes interpretation of the images appear deceptively simple. In reality, MRI is a highly sophisticated technology, and some background knowledge is essential to understand the findings, as well as to critically assess conclusions drawn from investigations that employ this technology. The following brief review of basic MRI principles and terminology will aid in understanding the remainder of the chapter and help investigators outside the discipline of Radiology to take better advantage of the growing number of published studies that use MRI. For the interested reader, there are several excellent books and articles that delve deeper into MRI physics and its applications in medicine.1,2,3,4,5

Basic Principles of Magnetic Resonance Imaging

MR imaging is based on the response of certain atomic nuclei to the presence of a magnetic field (Fig. 9-1). A number of different nuclei (for example, 23Na, 13C, 19F, and 1H) can be used to generate MR images. Hydrogen nuclei (or protons)

P.168


are the most abundant within biological tissue and are therefore the most feasible for clinical imaging. When the tissue is placed within a strong magnetic field in the bore of an MR imaging magnet, these nuclei show a net tendency to align their nuclear magnetic moments along the direction of the static magnetic field. This alignment creates what is referred to as longitudinal magnetization (Fig. 9-2). Exposure of these protons to a second dynamic magnetic field (a radio frequency, or RF field, usually called B1) that is rotating and perpendicular to the original static field of the magnet torques the protons 90 degrees away from the stronger static field (Fig. 9-3A). This process is known as excitation. The protons that now point in this direction make up what is referred to as transverse magnetization. The spins have a resonant frequency intrinsically tied to the strength of the main magnetic field, by the gyromagnetic ratio:

= B0

Figure 9-1 The nuclear magnetic moment. Spinning (precessing) anatomic nuclei ( spins ) generate small local magnetic fields analogous to the spinning planets. The magnitude of the magnetic moment depends on the rate of precession, or frequency, of the nucleus. The vector sum of individual magnetic moments for a pool of hydrogen nuclei ( protons ) in fat or water is the essential parameter measured in clinical MRI. (Courtesy of Synarc, Inc.)

Where is the spin's gyromagnetic ratio, B0 is the strength of the main static magnetic field (for example, 1.5 T), and is the proton, or spin resonant frequency in that field. This resonant frequency relationship means that after the spins are tipped into the transverse plane by the B1 pulse, they precess about the longitudinal axis along B0. When the RF (B1) tip-down pulse is turned off, the spins continue to precess. They act as tiny bar magnets, creating their own rotating magnetic field. This changing magnetic field induces a signal across the terminals of the same coil that was used to create the field, because the resonant frequency is the same (Fig. 9-3b). This signal is then used to generate the MR images by computerized Fourier transformation.

Figure 9-2 Longitudinal magnetization. A, Protons placed within the strong magnetic field B0 (Large open arrow) in the bore of a MRI magnet tend to align their magnetic moments (small arrow) parallel or anti-parallel with this large magnetic field. B, Protons have a slight affinity for parallel alignment, creating a net magnetic moment M0. The magnitude of this net longitudinal magnetic moment and therefore the maximal signal that could be generated during imaging varies directly with the field strength B0 of the MRI magnet. (Courtesy of Synarc, Inc.)

MR imaging uses three types of coils. The main magnetic field, B0, is created by a superconducting magnet enclosed in a cylindrical cryostat. These magnets must be both strong (field strengths range between 0.2 T and 11 T) and very uniform (within 1 part per million in the imaging volume) in order to precisely set spin frequency. Whole-body scanners are currently limited to a maximum strength of 3 T to 4 T, with most clinical systems ranging from 0.5 T to 3 T. The second coil that creates the RF B1 field is a birdcage-shaped coil that sits permanently within the main B0 coil. Smaller birdcage coils that fit certain volumes, such as the head or extremities, can be placed in the bore as a substitute. Even smaller ring-shaped coils of 10 mm to 100 mm in diameter can be placed directly on the region to be imaged. There is a great advantage in using the smallest possible receive coil, because these coils are not as sensitive to tissues outside the volume of interest, which contribute significantly to image noise. The greatest gain in image Signal-to-noise ratio (SNR) is then achieved by starting with the most appropriate RF coil. The third type of coil used in MR is the gradient coil, of which there is one for each of the three axes. These coils create much smaller magnetic fields that point in the same direction of B0, but vary as a function of distance to the coil. The strength of the linearly varying magnetic fields created by these coils is changed rapidly during an MR exam so that only spins in certain locations have frequencies in the range of those to which the receive coil is sensitive. The gradients force spins to move away from each other in frequency and then return to the same frequency to be in phase. This process is referred to as gradient echo formation. Both higher gradient amplitudes and faster gradient switching rates achieve echoes more quickly and result in shorter scan times.

The two main types of echoes in MR imaging are the gradient echoes (GREs) previously mentioned and spin

P.169


echoes (SEs). SEs rephase protons with a 180 degree RF refocusing pulse. This pulse reverses phase position of the spins, flipping the fastest precessing spins behind the slowest precessing ones. After the phase reversal provided by the refocusing pulse, the fastest-moving spins continue at the same precession speeds and catch up so that all spins refocus to a coherent signal. A useful analogy to this process is a track race, halfway through which, the race is reversed so that all runners finish together at the starting line, assuming they maintain constant running speed.

Figure 9-3 RF excitation of transverse magnetization. A, The net proton magnetization M0 that is longitudinally aligned with the high magnetic field B0 (large open arrow) in the MRI magnet bore will realign (resonate) with a second relatively smaller magnetic field B1 (small open arrow) if this new field is tuned to the proton precessional frequency. Since this resonant frequency is in the same range as radio waves, this second field is called a radio-frequency (RF) pulse. B, The RF pulse can be played for a specific duration given the pulse amplitude to produce a full 90 rotation of the net magnetization M0, called the flip angle. This realigned (flipped) magnetic moment (gray vector arrow) will continue to rotate transversely once the RF pulse is turned off to induce an alternating current (by Faraday's Law) in the wire of the receiver coil placed near the patient. This induced current is the basis for the MR image. (Courtesy of Synarc, Inc.)

The time it takes for echo formation is called the echo time, or TE. The total time it takes for both an RF excitation and a gradient echo formation/signal acquisition is the repetition time, or TR. TR can be as short as a few milliseconds or as long as a few seconds. The number of repetition times to form an image depends on the imaging method and whether 2D or 3D images are reconstructed. The length of the TR depends on the way the gradients refocus spins after allowing or forcing them to dephase.

Figure 9-4 T1 T2 relaxation. When the rotating 90 RF pulse is turned off, the transversely oriented magnetic moment re-aligns with the static field of the magnet B0. A, This recovery of longitudinal magnetization is called T1 relaxation, and the parameter, T1, is a measure of the rate of this recovery. If the 90 RF pulse is repeated before longitudinal magnetization has fully recovered, only this smaller longitudinal component is flipped into the transverse plane and the image signal is correspondingly lower these protons are said to be partially saturated. B, As the longitudinal magnetization re-grows, the transverse component decays. This is called T2 relaxation. (Courtesy of Synarc, Inc.)

Echoes create the signal for MR images, and TE/TR selection is the mechanism by which contrast is generated between different tissues. As soon as the RF pulse is turned off after excitation, the protons slowly return to their original alignment with the static main field of the magnet (Fig. 9-4). This process of recovering longitudinal magnetization and decaying transverse magnetization is called relaxation. T1 and T2 are the time constants with which this occurs. T1 is the time necessary for the longitudinal magnetization to recover, and T2 is the time for the transverse magnetization to decay. Both of these vary from tissue to tissue, depending

P.170


on the microenvironments of the different proton populations. T1 and T2 are therefore tissue characteristics that allow varying MR acquisition timing to generate contrast between tissues. Table 9-1 gives approximate values of different tissues in the knee at a main field strength of 1.5 T.

TABLE 9-1 APPROXIMATE T1 AND T2 RELAXATION TIMES OF TISSUES IN THE KNEE AT 1.5 T

Tissue T1 (ms) T2 (ms) T1/T2
Cartilage* 800 30 26.7
Fat 260 80 3.25
Synovial fluid** 2500 200 12.5
Muscle 870 50 17.4
*Measured values for cartilage range between 700 and 1100 ms for T1 and 20 and 60 ms for T2.
**Measured values for synovial fluid range between 1400 and 3000 ms for T1 and 200 and 900 ms for T2.

T1 relaxation, for example, occurs rapidly in fat, while water (abundant in muscle) shows slow T1 relaxation (Fig. 9-5A). T1 also varies slightly with the magnetic field strength so that relaxation of the longitudinal magnetization back to equilibrium is somewhat shorter at lower main field strengths. Under conditions of rapid RF pulsing, slow T1 substances such as water are not given sufficient time to recover between the pulses. Thus, there is little longitudinal magnetization available to be tipped again to create signal, and these substances therefore exhibit low signal intensity. Shorter T1 substances such as fat need less time for longitudinal regrowth and show higher signal intensity (Fig. 9-5A). Short TR sequences therefore generate contrast (relative signal intensity difference) among tissues on the basis of differences in T1 and are accordingly referred to as T1-weighted (Fig. 9-6).

Image contrast is also influenced by T2 relaxation. While the longitudinal magnetization is regrowing after the RF pulse is turned off, the transverse magnetization is slowly decaying. Although not intuitive, the rate of T2 relaxation is not necessarily coupled to the rate of T1 relaxation, other than that the time for T2 relaxation is always shorter than that for T1. Although T1 depends partly on the strength of the main static field, T2 remains constant across all field strengths. As T2 relaxation occurs, the transverse magnetization, and therefore signal, decrease. So, although shorter T1 species are brighter on T1-weighted images, the longer T2 species tissue is brightest on T2-weighted images (Fig. 9-5B). Freely mobile water protons (such as in synovial fluid) show slow T2 relaxation and therefore retain signal over time, whereas constrained or bound water protons (such as by collagen or proteoglycan) show rapid T2 relaxation and signal decay (Fig. 9-5B, Fig. 9-6).

Figure 9-5 Effect of TR/TE on signal intensity. Repetition time (TR) is the time between successive RF pulses in an imaging sequence. Typically, 192 to 256 repetitions are necessary to generate an MR image. If the TR is less than five times the T1 of a substance, there is insufficient time for complete recovery of longitudinal, or aligned magnetization and signal intensity after subsequent excitations is decreased. A, As TR is shortened, tissues with longer T1 relaxation times (e.g., muscle) begin to lose signal first, while tissues with shorter T1 relaxation times (e.g., fat) retain signal until the TR is very short. Short-TR sequences thus generate T1 contrast among tissues and are called T1-weighted. Echo time (TE) is the time between the initiating RF pulse and the point at which spins are refocused either by a 180 rephasing RF pulse (spin-echo) or gradient reversal waveform (gradient-echo). Substances with longer T2 relaxation times (e.g., fat) retain more signal intensity on long-TE(T2-weighted) sequences. B, As TE is lengthened, signal from tissues with shorter and longer T2 relaxation times undergo more decay, changing the contrast. Note that for each value of TE and TR, the contrast between any two tissues is proportional to the difference in signal level, or transverse M0, between the two. In the case of T1-weighted sequences, reductions in the aligned M0 translate into reductions in the transverse M0 after subsequent RF excitations.) (Courtesy of Synarc, Inc.)

In addition to the effects of neighboring protons on each other (T2 relaxation), heterogeneities in the static magnetic field or off-resonance caused by a chemical shift (as with the 220 Hz difference in fat) cause protons to dephase and lose additional transverse magnetization strength. This is noted as T2' relaxation. The combined effects of proton

P.171


dephasing and T2 signal loss result in an overall faster decay in transverse magnetization called T2*, which is defined as:

1   1   1
- = - + -
T2   T2   T2

Signal lost to fixed magnetic heterogeneity, but not that lost to T2 relaxation, can be recovered using the spin echoes already mentioned.

Figure 9-6 T1, T2, and PD-weighted MRI. A, Sagittal T1-weighted spin-echo image of a knee depicts structures that contain fat (short T1) with high signal intensity, and structures that contain water (long T1) with low signal intensity. The small differences in T1 relaxation time among synovial fluid, articular cartilage, and muscle do not generate substantial contrast among these structures on this image. It is difficult, therefore, to delineate the entire articular cartilage surface in this slice. B, T2-weighted fast spin-echo image of the same knee depicts synovial fluid (long T2) with higher signal intensity. Water in articular cartilage and muscle is relatively bound (short T2); these structures therefore show low signal intensity. The dynamic range is much improved with the use of fat suppression, hence the dark marrow in bone. High intrinsic contrast between cartilage and synovial fluid makes this technique useful for delineating the articular surface. C, Proton-density weighted, fat-suppressed fast spin-echo image of the same knee uses a shorter echo time, making cartilage signal brighter. Bone marrow edema is visible in all three images (long arrow) and a small meniscal tear is best visualized in the PD-FSE image (short arrowhead). (Courtesy of Synarc, Inc.)

Local perturbations of the magnetic field typically arise at interfaces between substances that differ considerably in magnetic susceptibility (the degree to which a substance magnetizes in the presence of a magnetic field), such as between soft tissue and gas, metal, or heavy calcification. Severe T2* at these sites is referred to as magnetic susceptibility effect. Spin-echo refocusing RF pulses correct for fixed magnetic heterogeneities and therefore can provide

P.172


images with true T2 contrast. The gradient-echo technique, which relies solely on pre-winding and rewinding gradient waveforms before and after signal acquisition, is faster than spin-echo. However, it does not correct for off-resonance effects and therefore provides only T2*-weighted images. These images are highly vulnerable to magnetic susceptibility effects, such as those caused by metallic prostheses, and can result in large signal voids in the vicinities of metal implants. Note that while SNR increases with higher field strength, artifacts from magnetic susceptibility differences become more severe.

Finally, diffusion of protons (for example, water) within a specimen during the acquisition of an MR image results in loss of phase coherence among the protons and therefore a loss in signal. This effect is usually insignificant in conventional MRI but can be augmented with the use of strong magnetic field gradients such as those employed in MR microimaging. Water diffusivity is thus an additional tissue parameter measurable with MRI.6,7

Both T1-weighting (short TR) and T2-weighting (long TE) involve discarding MR signal. If these effects are eliminated, signal intensity reflects only the proton density. Accordingly, long-TR/ short-TE images are often referred to as proton density-weighted. However, even the shortest finite TE attainable is too long to completely escape T2 relaxation, and extremely long TRs (> 2500 ms) are not practical for imaging in vivo. Therefore, even so-called proton density-weighted images contain some T1 and T2 contrast (Fig. 9-6, Fig. 9-14).

Another consequence of the relaxation that occurs from TR to TR is that during the first few repetitions, the signal will be different strengths. Because the magnetization has not fully recovered after the first TR, the second signal acquisition has less available signal to be tipped into the transverse plane. The third acquisition will differ from the second, and so forth. Eventually, however, these differences from TR to TR become smaller and smaller, and the signal is said to be in the steady state. The effort to shorten this transient time to maximize useful imaging time, as well as methods of manipulating spins at the end of each TR to reach this condition more quickly, has been an active area of imaging research.8,9,10.

Figure 9-7 Augmenting T1 contrast with fat suppression. A, Sagittal, T1-weighted spin-echo image of a knee acquired with a TR of 500 ms, at the short end of the usual range (500 ms to 700 ms) depicts the articular cartilage (arrows) with a slightly higher signal intensity than the adjacent synovial fluid. Contrast between cartilage and water is greater on this shorter-TR image than on the conventional T1-weighted image shown in Fig. 9-6A (TR = 600 ms), but is still overshadowed by the greater T1 contrast between fat and other tissues in the image. B, The same sequence repeated with fat suppression generates greater contrast between articular cartilage (arrows) and synovial fluid as their pixel intensities are rescaled across a broader range of gray scale values. The same effect can be achieved with water-selective excitation. (Courtesy of Synarc, Inc.)

Subtle T1 contrast (for example, between articular cartilage and synovial fluid) is usually overshadowed on T1-weighted images by the far greater difference in signal intensity that exists between fat and most other tissues, because the presence of fat increases the dynamic range of the resulting images. However, by selectively suppressing the signal intensity of fat, it is possible to expand the scale of image intensities across smaller differences in T1 and thus to augment residual T1 contrast (Fig. 9-7). Another application of fat suppression is to increase contrast between fat and other substances, such as methemoglobin and gadolinium (Gd)-containing contrast material, which also show rapid T1 relaxation. The most widely used technique for fat suppression is based on the chemical shift phenomenon: Because the frequency of protons in fat differs from that of protons in water, the magnetization of fat

P.173


(or water) can be selectively suppressed by a specifically tuned RF pulse at the beginning of the sequence (Fig. 9-8). This RF pulse, which is centered on the fat frequency instead of the water frequency, prematurely tips down the fat spins so that when the sequence RF tip-down pulse is played out, there is no longitudinal magnetization in fat available to produce a signal.

A similar technique can also be used to suppress the signal of water indirectly, through a mechanism called magnetization transfer. In this case, direct suppression of tightly constrained protons in macromolecules such as collagen, which are thermodynamically coupled to freely mobile protons in bulk water, evokes a transfer of magnetization from the water proton pool to the macromolecular pool to maintain equilibrium. This manifests as a loss of longitudinal magnetization and therefore signal intensity from water in proportion to the relative concentrations of the two proton pools in the tissue and the specific rate constant for the equilibrium reaction. Because collagen (unlike fat) is strongly coupled to water in this way, cartilage and muscle exhibit pronounced magnetization-transfer effects.11,12,13,14

Magnetization-transfer techniques are therefore useful for imaging the articular cartilage and could potentially be used to quantify the collagen content of this tissue.

The two most important parameters for describing the extent of tissue coverage are image resolution and field of view (FOV). Both of these parameters depend on the strength of the gradients, on gradient switching speed, and on how the gradient waveforms are played out during the acquisition. Finer resolution requires an increase in the number of frequencies that must be sampled and therefore longer gradient waveforms. Images that have larger FOV (that is, are zoomed out to show more anatomy) require frequencies to be more precisely resolved, which means that gradient waveforms must be lower in amplitude and of longer time duration. So, both resolution and FOV are also dependent on the amount of time available within a single TR to keep scan times reasonable, patient motion reduced, and contrast/signal available despite relaxation effects.

Figure 9-8 Frequency-selective fat suppression. The chemical-shift phenomenon separates the resonant frequencies of water and fat (by 220 Hz at a magnetic field strength of 1.5 T). This allows the longitudinal magnetization of either of these proton pools to be selectively suppressed by an RF pulse tuned to the correct resonant frequency. Since the resonant frequency and the magnitude of the chemical shift both depend on magnetic field strength, this method of fat suppression is dependent on the homogeneity of the static magnetic field and is not feasible at very low field strengths.

In 2D imaging, a slice is selected during excitation, and the other two spatial axes are localized down and across the image plane during acquisition. Alternatively, 3D data sets can be acquired by exciting all spins and playing localizer gradients on all three axes during acquisition. In either of these cases, dimensions of the individual volume elements, or voxels, comprising it define the spatial resolution of an MR image. Voxel size is determined by multiplying the slice thickness by the size of the in-plane subdivisions of the image, the pixels (picture elements). Pixel size, in turn, is determined by dividing the FOV by the image matrix, which most commonly ranges between 256 128 and 256 256 for knee imaging. The key point of pixel size is the smaller the pixel, the finer the spatial resolution. Typical sizes for FOV in knee imaging are around 14 cm.

All signals within a single voxel are averaged. Therefore, if an interface with high signal intensity on one side and low signal intensity on the other side passes through the middle of a voxel, then the interface is depicted as an intermediate signal intensity band the width of the voxel (Fig. 9-9). This effect is known as partial-volume signal averaging. However, as voxel size decreases, so does SNR. Accordingly, high-resolution imaging requires sufficient SNR to support the spatial resolution. SNR can be increased by shortening TE (less T2 decay), increasing TR (more T1 recovery), imaging at higher field strength (greater longitudinal magnetization), or utilizing specialized coils that reduce noise (small surface coils, quadrature coils, or phased arrays of small coils).15,16 Specialized sequences, such as those that fully refocus spin dephasing from T2* effects, also provide greater SNR.

Figure 9-9 Partial volume averaging. The smallest element of an MR image is the individual voxel (pixel size slice thickness). Different signal intensities within a single voxel are averaged. This effect is most noticeable at high contrast interfaces as shown in the magnified view of the femoral cartilage on this sagittal, fat-suppressed, T1-weighted gradient-echo image of a knee. (Courtesy of Synarc, Inc.)

P.174


Imaging Articular Cartilage

Magnetic Resonance Imaging Appearance of Articular Cartilage: Contrast Mechanisms

The signal behavior of articular cartilage on MRI reflects the complex biochemistry and histology of this tissue. The high water content (proton density) of articular cartilage forms the basis for MR signal. Water content in this tissue depends on the delicate balance between the swelling pressure of the aggregated proteoglycans and the counter resistance of the fibrous collagen matrix. But, in general terms, changes in cartilage proton density tend to be relatively small (typically <20%). Because the water constitutes approximately 70% of the weight of normal articular cartilage, proton density itself offers little scope for generating image contrast between cartilage and adjacent synovial fluid. However, this fundamental MRI signal in cartilage is modulated by a number of processes, including T1 relaxation, T2 relaxation, magnetization transfer, water diffusion, magnetic susceptibility, and interactions with contrast agents. These processes provide many different mechanisms for delineating cartilage morphology and probing its composition.

For comparison, Table 9-1 gives the T1 and T2 values of tissues in the knee. The T1 of articular cartilage at 1.5 T is approximately 800 ms. This time is much shorter than the T1 of adjacent synovial fluid, (2500 ms) but still longer than the T1 of subarticular marrow fat (260 ms). The gray scale on a conventional T1-weighted SE image is then so dominated by fat that the contrast between articular cartilage and adjacent synovial fluid is normally difficult to appreciate (Fig. 9-6). Intrinsic T1-contrast can be augmented slightly by shortening TR, but a more powerful approach is to suppress the fat signal or selectively excite protons in water, and rescale the smaller residual T1 contrast across the image. This generates images in which articular cartilage is depicted as an isolated high signal intensity band in sharp contrast with adjacent low signal intensity joint fluid, and nulled fat in adipose tissue (for example, Hoffa's fat pad) and bone.12,17 Fat suppression also eliminates chemical-shift artifacts that distort the cartilage bone interface and complicate dimensional measurements.

Figure 9-10 T2 Relaxation of normal adult articular cartilage. T2 map generated from multislice, multi-echo (11 echoes: TE = 9, 18, 99 ms) spin-echo images acquired at 3 T shows increasing T2 toward the articular surface. (Courtesy of B. J. Dardzinski, Ph.D. University of Cincinnati College of Medicine.)

T2 relaxation is another tissue characteristic that can be harnessed to image the articular cartilage. Fibrillar collagen in the articular cartilage immobilizes tissue water protons and promotes dipole-dipole interactions among them, increasing T2 relaxation and therefore signal decay. The T2 of normal articular cartilage increases from approximately 30 ms in the deep radial zone to 70 ms in the transitional zone18 (Fig. 9-10). Above the transitional zone, the superficial tangential zone shows extremely rapid T2 relaxation because of its densely matted collagen fibers. This radial heterogeneity of T2 gives articular cartilage a laminar appearance on all but extremely short-TE images.19 The pattern of T2 variation can be explained, to some extent, by the heterogeneous distribution of collagen in this tissue, but is also affected by the orientation of collagen fibrils relative to the static magnetic field (B0). T2 anisotropy in cartilage manifests as decreased signal decay in regions where the collagen fibrils are oriented at 55 to B0.19,20,21,22 This so-called magic-angle phenomenon is responsible for areas of mildly elevated signal intensity in the radial zone of appropriately oriented cartilage segments on intermediate-TE images (Fig. 9-11). It is also one explanation for the slower T2 seen in the transitional zone. Collagen fibrils in this zone are slightly sparser than in the radial zone, but more importantly they are also highly disorganized. Accordingly, a significant proportion of the fibrils in the transitional zone are angled at 55 to B0regardless of the orientation of the knee in the magnet. With sufficiently long TE (<80 ms), normal articular cartilage appears diffusely low in signal intensity even in regions normally affected by this magic-angle phenomenon.

Figure 9-11 Magic-angle phenomenon in articular cartilage. High-resolution spin-echo image of the patellar cartilage shows low signal intensity due to T2 relaxation in the radial zone of the central portion of the cartilage, where collagen is aligned with the static magnetic field (B0.) Increased signal intensity (arrow) indicative of prolonged T2 can be seen in areas where the collagen is oriented at approximately 55 relative to B0. (Courtesy of D. Goodwin, and J. Dunn, Dartmouth Medical School.)

Superimposed upon these histological and biochemical causes of laminar appearance in articular cartilage are patterns created by truncation artifacts.23,24 This manifests as one or several thin horizontal bands of low signal intensity midway through the cartilage on short-TE images. Truncation artifacts are less common on high-resolution images, but usually present on fat-suppressed 3D spoiled gradient-recalled (SPGR) images generated with most clinical protocols.

Long-TE images provide high contrast between articular cartilage and adjacent synovial fluid, but poor contrast between cartilage and bone. Shorter-TE images improve

P.175


cartilage bone contrast, but are vulnerable to magic-angle effects. Fast SE (FSE) combines T2 effects with magnetization transfer to decrease signal intensity in articular cartilage.25,26 Signal loss due to magnetization transfer results from equilibration of longitudinal magnetization between nonsaturated freely mobile protons in water and saturated restricted protons in macromolecules, such as collagen, that have been excited off the resonant frequency of free water during multi-slice imaging.11,12,14,25,26,27 The effect is exaggerated with FSE imaging because of the multiple 180 RF pulses used with this technique. Accordingly, intermediate TE (~40 ms) FSE images show relatively low signal intensity in articular cartilage while preserving high signal intensity in synovial fluid and subjacent bone marrow to delineate the articular cartilage with high contrast (Fig. 9-13). Both intermediate and long TE FSE images offer relatively good morphological delineation of articular cartilage in less time than is required for high-resolution fat-suppressed 3D-GRE images. The choice of which TE to use depends on the objectives of the imaging and how they relate to the range of normal and pathological T2 heterogeneity found in articular cartilage.

Figure 9-12 Comparison of FSE and SPGR to DEFT and three variants of SSFP. Axial patellofemoral water images from a normal volunteer using A, FSE (TR = 1800 ms), B, FSE (TR = 3200 ms), C, SPGR, D, DEFT, E, LC-SSFP, F, FEMR, and G, FS-SSFP. Fat images using LC-SSFP H, and FEMR I, require no additional scan time. Water images from DEFT D, and the SSFP-based techniques (e.g., FS-SSFP, LC-SSFP, and FEMR) all demonstrate both bright cartilage and excellent contrast with synovial fluid. (

Hargreaves, B.A., Gold, G.E., Beaulieu, C.F., et al. Comparison of new sequences for high-resolution cartilage imaging. Magn Reson Med 49:700 709, 2003.

)

Magnetic Resonance Pulse Sequences for Imaging Articular Cartilage Morphology

Two pulse sequences are the clinical workhorses of knee imaging: T1-weighted spoiled GRE, referred to as SPGR or FLASH (fast low-angle shot) and T2-weighted fast spin-echo (FSE), or turbo spin-echo (TSE). SPGR provides 3D acquisition in reasonable scan times and is currently the most available clinical option for quantitative measurements of cartilage volume.28 However, it does not provide strong cartilage-synovial fluid contrast and is susceptible to T2* dephasing. FSE is much more robust to off-resonance, and its contrast is not weighted by T2* However, the method requires a longer repetition interval, and therefore scan times for 3D acquisition would be prohibitively long.

Fat-suppressed, T1-weighted 3D SPGR is easy to use and widely available, and has become a popular MRI technique for delineating articular cartilage morphology.12,17,29,30,31,32 However, the sequence provides poor cartilage-synovial fluid contrast, making depiction of cartilage surface defects difficult. Driven equilibrium (DE, DEFT, or FR [Fast Recovery] as in FR-SE) produces higher cartilage-synovial fluid contrast than either FSE or SPGR (Fig. 9-12), making it a good choice for imaging cartilage surface defects. The contrast is based on tissue T2/T1, which makes synovial fluid brighter. Although DEFT provides shorter scan times and allows for 3D imaging, other sequences with higher SNR efficiency are better choices for cartilage volume measurement.

Figure 9-13 Fast spin echo imaging of cartilage. Sagittal T2-weighted fast spin echo image of the knee shows high contrast between the low signal intensity articular cartilage (white arrow) and adjacent high signal intensity synovial fluid (black arrow) and intermediate signal intensity subchondral marrow fat (f). (Courtesy of Synarc, Inc.)

P.176


Figure 9-14 Cartilage contrast with various pulse sequences. A, Sagittal fat-suppressed T1-weighted 3D GRE image depicting articular cartilage as a high-signal structure in sharp contrast against adjacent low-signal bone, marrow fat, intra-articular adipose, fluid, ligaments, and menisci. B, Sagittal 3D DESS image showing partial-thickness cartilage defect (arrow) over posterior lateral tibia. Note the similarities in contrast properties of fat-suppressed DESS with those of fat-suppressed FSE. C, Sagittal fat-suppressed IW 2D FSE shows a loose body (arrow) in the patellofemoral compartment. D, Sagittal T2-weighted 2D FSE image without FS shows a partial-thickness defect (arrow) of the lateral femoral cartilage adjacent to the posterior horn of the meniscus. E, Sagittal fat-suppressed T2-weighted 2D FSE image of a different knee shows a partial-thickness cartilage defect (arrow) in a similar location. (From

Peterfy CG, Gold G, Eckstein F, et al. MRI protocols for whole-organ assessment of the knee in osteoarthritis. Osteoarthritis Cartilage, 14 Suppl A:A95-111, 2006

)

P.177


SPGR sequences are susceptible to T2* signal decay because the signal is the average of spins that are dephasing, or falling out of step with the main-field resonant frequency. Balanced steady-state free-precession (bSSFP, SSFP, also FIESTA) or true fast imaging with steady-state free precession (True-FISP) pulse sequences are able to fully refocus spins each acquisition, using symmetric gradient waveforms. This results in a stronger signal and higher SNR images. The number of off-resonant frequencies that can be refocused at high signal strength is inversely proportional to the sequence TR, so imaging times are also inherently shorter. The concept behind SSFP is not new,33,34,35 but recent hardware advances resulting in faster gradient switching times has made implementation of SSFP at short TR lengths possible. Figure 9-12 shows a comparison of FSE and SPGR to three variants of SSFP, each with different implementations of fat suppression. As faster hardware becomes more ubiquitous in the clinical setting, 3D fat-suppressed SSFP (FS-SSFP) is a frontrunner to replace 3D FS-SPGR with higher-SNR images.36 Alternatively, the higher SNR available could be used to improve resolution, which could further improve cartilage segmentation and surface rendering, adding both sensitivity and specificity to volume and thickness measurements. For imaging cartilage lesions, the frontrunner is another variant, called dual-echo steady-state (DESS) imaging. DESS uses a second gradient echo separated by a refocusing pulse. This lengthens the acquisition window, but results in an image with higher T2* weighting, and has been shown to be superior for detecting superficial cartilage lesions.37 Figure 9-14 shows a comparison of DESS to other GRE and SE sequences.

Figure 9-15 Projection-Reconstruction Spiral Imaging of Cartilage. A Water frequency image (TE = 200 microseconds, 0.2-mm in-plane resolution, 8-min scan time). B Spectra from the voxels indicated in A, showing increasing peak area and decreasing width towards articular surface. This indicates increasing water density and T2-relaxation times (Reprinted with permission from the American Journal of Roentgenology: G. Gold, D. Thedens, J. Pauly, K. Fechner, G. Bergman, C. Beaulieu, and A. Macovski. MR Imaging of Articular Cartilage of the Knee: New Methods Using Ultra-Short TE's, AJR Am J Roentgenol 170(5):1223 1226, 1998.) C Gradient-recalled echo image from a patient with osteochondral allografts. Metal artifact obscures the articular cartilage. D Spectral maximum intensity image (SMIP) of projection-reconstruction spiral imaging of the same slice shows reduced artifacts.(From

Gold GE, Bergman AG, Pauly JM, et al. Magnetic resonance imaging of knee cartilage repair. Top Magn Reson Imaging 9(6):377 392, 1998.

)

More recently, projection-reconstruction (PR) based techniques have been developed to image the articular cartilage with ultra-short TE (<0.2 ms) and even greater contrast, fewer chemical shift effects, and lower vulnerability to magnetic susceptibility artifacts (Fig. 9-15).38 Geometric artifacts that arise from the non-grid sampling inherent to this method should be considered, but some other advantages of this technique include the potential for spectroscopic determination of water content and T2. Vastly undersampled isotropic projection (VIPR) is one variation of these techniques that has been shown to provide high cartilage-synovial fluid contrast and clear demarcation of cartilage defects.39 Kijowski et al. have

P.178


demonstrated a variant of VIPR using radial SSFP imaging. The image in Figure 9-16 depicts a large patellar cartilage defect along with adjacent subchondral bone marrow edema using this technique.

Figure 9-16 Axial VIPR-SSFP image of the knee shows a large defect within the articular cartilage of the lateral facet of the patella (black arrow) with adjacent subchondral bone marrow edema (white arrowhead). (Courtesy of Richard Kijowski, M.D., University of Wisconsin-Madison.)

There are several new methods of suppressing or separating out fat signal that have been developed for SSFP sequences. In addition to the use of traditional fat saturation pulses (FS-SSFP) or the use of spectrally selective RF pulses that only excite protons at the water frequency, the newer steady-state methods can be broken into two categories. These are: 1) those that create steady-state signals with low or suppressed fat signal, and 2) fat-water separation methods, which provide simultaneous water and fat images with some additional post-imaging reconstruction. The first category includes techniques such as fluctuating equilibrium magnetic resonance (FEMR)40 and oscillating SSFP.41,42 These techniques have lower SNR efficiency than FS-SSFP, and are generally less robust than FS-SSFP, so they may be better suited to other applications. The second category includes steady-state fat/water separation methods such as linear combination SSFP (LC-SSFP)43 and phase-sensitive SSFP (PS-SSFP).44,45 LC-SSFP requires two acquisitions and is more sensitive to patient motion. PS-SSFP is less sensitive to patient motion, because it images the water-fat difference and requires only one acquisition.44 It is faster than FS-SSFP and provides excellent cartilage delineation, which makes it a good choice for 3D imaging for volume and thickness assessment.46 With this method, each voxel is categorized as water or fat based on the majority of voxel tissue content. This method works well in cartilage, but would have difficulty in imaging bone edema, because the marrow fat would contribute partial volume errors to the water-fat separation. Figure 9-17 shows a comparison of PS-SSFP with Proton-density fast spin-echo (PD-FSE) and fat- suppressed, T2-weighted FSE in articular cartilage.

In addition to the methods specific to SSFP, there have been recent improvements to Dixon fat-water separation, which uses multiple acquisitions with different echo times to reconstruct water and fat images. Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) is a variation of the Dixon technique and can be used with FSE, SPGR, and SSFP.47,48 IDEAL is very promising for cartilage imaging because it has higher SNR efficiency in both cartilage and synovial fluid, with both SPGR and SSFP.49,50 Figure 9-18 shows a comparison of the IDEAL and fat saturation pulse methods with SPGR cartilage imaging.

Magnetic Resonance Pulse Sequences for Articular Cartilage Functional Imaging

In addition to delineating cartilage morphology, T2 relaxation can be used to probe the status of the collagen matrix in articular cartilage. This is because as the collagen network breaks down, tissue water in articular cartilage becomes more fluid and correspondingly less affected by T2 relaxation. Consistent with this, foci of high signal intensity are often seen within the cartilage of knees of patients with OA on T2-weighted images (Fig. 9-19). These signal abnormalities have been reported to correspond to arthroscopically demonstrable abnormalities.51,52 However, they have also been observed in cartilage that appeared normal by arthroscopy.52,53,54 This raises questions about the sensitivity of arthroscopy for assessing articular cartilage integrity, at least in very early disease.

A careful assessment of the sensitivity and specificity of subjective evaluations of T2 abnormalities in articular cartilage using images attainable with conventional MRI hardware and software and histological assessment as the gold standard has yet to be reported. Moreover, most studies that have looked at T2 abnormalities in cartilage have provided only cross-sectional information. Longitudinal data describing the natural history of this potential marker of cartilage matrix integrity and its association with subsequent cartilage loss and joint failure are scant. In one study,55 however, 5 (33%) of 15 meniscal surgery patients followed over 3 years postsurgery developed a total of six T2 lesions in otherwise normal-appearing articular cartilage. Two of these lesions progressed to focal cartilage defects during the study (Fig. 9-20), while three persisted and one regressed. Interestingly, the four lesions that did not progress were in patients who had undergone meniscal repair, while the lesions that progressed were in patients who had meniscal resection. Accordingly, abnormal T2 may identify cartilage at risk of future loss.

Water diffusion in cartilage also contributes to signal loss on T2-weighted MR images. This is because water molecules that have changed positions during a portion of the MRI acquisition can no longer be rephased properly and so do not contribute maximally to the net signal. This loss of phase coherence is proportional to the distance traveled by the diffusing water protons and is therefore worse on long TE images. The presence of proteoglycans, particularly

P.179


chondroitin sulfate, in normal cartilage inhibits water diffusion and keeps this effect relatively small, although with very strong gradients and specialized phase-sensitive pulse sequences, water diffusion can be demonstrated and even quantified in normal articular cartilage.7 With cartilage degeneration and proteoglycan loss, however, water diffusion has been shown to increase considerably. Accordingly, diffusion may play a more significant role in cartilage signal modulation in osteoarthritic joints.

Figure 9-17 Phase-sensitive fat/water separation. A, PD-FSE, B, PS-SSFP (water), and C, T2 FS-FSE sagittal images of the knee, comparing phase-sensitive fat/water imaging with fat-suppressed FSE. (Courtesy of Shreyas S. Vasanawala, M.D., Ph.D., Stanford University.)

Burstein et al.7 showed that treatment of a bovine cartilage sample with trypsin (for proteoglycan removal) resulted in a 20% increase in the measured rate of diffusion. They also showed that a 35% compression of a bovine cartilage sample corresponded with a 19% reduction in the rate of diffusion. Diffusion-weighted imaging of cartilage has also recently been demonstrated in vivo. Gold et al.56 were able to measure diffusion rates for water in cartilage using an in-plane resolution of 1.3 1.7 mm that were consistent with values determined in vitro at high resolution by Xia et al.6 Increases in the available gradient strength on clinical systems will be required to fully evaluate the clinical utility of diffusion-weighted imaging for OA. Using a local extremity gradient coil designed to improve the sensitivity and spatial resolution of imaging the knee with MRI, Frank et al.57 were able to achieve a spatial resolution of the 350 m x 350 m in-plane with a slice-thickness of 5 mm. Further advances in local gradient coils and improvements in system gradients will greatly aid the study of cartilage diffusion.

Proteoglycan loss also reduces cartilage hydration and therefore proton density. Since proteoglycan loss usually accompanies collagen loss, prolonged T2 associated with the collagen loss can be offset by T2 shortening due to increased diffusion and decreased tissue hydration of cartilage.

P.180


In addition to effects on water diffusion and tissue hydration, loss of proteoglycan from cartilage matrix results in decreased 23Na-ion concentration through the associated decrease in fixed negative charge density. Estimation of in vivo 23Na concentration of cartilage by 23Na Nuclear magnetic resonance (NMR) has been proposed as a means to provide an early marker for proteoglycan loss.58,59,60,61,62

Figure 9-18 Iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) fat/water separation. Sagittal images of the knee, comparing IDEAL fat/water separation with fat suppression. Images are 16-cm FOV, acquired using A, FS-SPGR, and B, IDEAL-SPGR (water image). (Courtesy of Garry E. Gold, M.D., Stanford University.)

Despite a high natural abundance in biological systems, the signal from 23Na is approximately 10% of the 1H signal due to a lower NMR sensitivity than protons. NMR sensitivity is defined as 3I(I+1), where is the gyromagnetic ratio and I is the spin.63 The NMR signal is directly proportional to the sensitivity of the nuclei. 23Na imaging is at an initial disadvantage because of these basic differences in the NMR properties of the two nuclei.

Figure 9-19 Patterns of abnormal cartilage signal. A, Coronal FSE image shows focal high signal in the cartilage over the lateral tibial plateau (arrow). B, In a different knee, a sagittal T2-weighted FS-FSE image shows a focus of increased signal (arrow) in cartilage over the lateral tibial plateau. Note the subchondral bone changes immediately beneath this region. (Courtesy of Synarc, Inc.)

The transverse relaxation time (T2) of 23Na for cartilage exhibits a bi-exponential behavior, with a fast T2 component between 0.7 and 2.3 ms and a slow T2 component between 8 and 12 ms.64 The in vivo longitudinal relaxation time (T1) of 23Na ranges between 14 and 20 ms.64 Rapid

P.181


transverse relaxation times make imaging more difficult due to the rapid loss of signal during the echo time. 23Na imaging is aided by a relatively short T1, which allows rapid signal averaging to partially overcome the poor sensitivity and short transverse relaxation times. Spatial resolution is generally the major concern in 23Na imaging due to the reduced signal strength. Clinical feasibility of 23Na imaging was first demonstrated in 1988.65,66 Granot65 acquired in vivo sodium images from various tissue structures (including knees) by employing a 3D sequence with short repetition (45 ms) and gradient-echo times (6 ms), concluding that sodium imaging of body organs is clinically feasible.

Figure 9-20 Progression of T2 lesions in articular cartilage. Serial sagittal T2-weighted fast spin-echo images show a focal T2 lesion (arrow) in the femoral cartilage adjacent to the posterior horn of the lateral meniscus at baseline A, Follow-up imaging nine months later B, shows a partial-thickness (Grade 2.0) defect at that exact location. (

Peterfy, CG. Scratching the surface: articular cartilage disorders in the knee. MRI Clin N Am. 8(2):409 430, 2000.

)

Several groups have shown in vitro studies that enzymatic degradation of proteoglycans leads to changes in 23Na relaxation rates.58,59,61,67 Reddy et al.64 demonstrated that 23Na MRI can differentiate between regions of proteoglycan depletion from healthy cartilage when imaging in vitro bovine patella. In addition, they also obtained 23Na images from a healthy volunteer with a 4T MRI scanner at an in-plane resolution of 1.25 2.5 mm and a slice thickness of 4 mm. 23Na imaging has also been shown to be sensitive to the mechanical deformation of cartilage. Shapiro et al.68 found that during recovery after exercise (50 deep knee bends), a 15% decrease in the thickness of the lateral facet of the subject's patella cartilage resulted in a 20% reduction in 23Na signal intensity. A possible cause for the loss in signal was attributed to the expulsion of saline from the cartilage during compression. An in vitro comparison of normal versus PG-depleted cartilage showed both specimens exhibited a decrease in T1 and T2 during compression.69

23Na imaging has been shown to have great potential in characterizing the physiological and mechanical state of cartilage. The major limiting factor to wide clinical usage of these techniques is the available signal strength on the standard 1.5 T system. All of the studies described earlier were performed on systems ranging from 1.5 to 4 T. Improvements in RF coil sensitivity,70 stronger gradients for shorter echo times, and greater clinical access to high field systems are prerequisites for 23Na imaging of cartilage to move from the research environment to the clinical setting.

Another interesting marker of cartilage matrix integrity is Gd-DTPA2- uptake.71,72,73 Under normal circumstances, anionic Gd-DTPA2- introduced into the synovial fluid (either by i.v. or direct intra-articular injection) is repelled by the negatively charged proteoglycans in normal cartilage. However, in areas of decreased glycosaminoglycan (GAG) content where the fixed negative charge density of cartilage is reduced, Gd-DTPA2- can diffuse into the cartilage and enhance T1 relaxation. These areas are depicted as conspicuous foci of high signal intensity in the otherwise low signal intensity cartilage on inversion recovery images. Cartilage T1 values correlate almost linearly with proteoglycan content in the range normally found in cartilage. However, quantifying T1 can be time consuming and impractical for clinical studies. Further work is necessary to establish the optimal method for acquiring this imaging data. Additional studies are also needed to define the relationship between this marker of proteoglycan matrix damage and elevated T2 as a marker of collagen matrix damage (Fig. 9-21). Whether one precedes the other and exactly how predictive each of these are alone or in combination for subsequent cartilage loss, the development of other structural features of OA, and ultimately for clinical manifestations of OA, have yet to be established. In addition to the use of Gd-DTPA2-, proteoglycan content of

P.182


cartilage can be probed with cationic contrast agents such as manganese74,75 or, as discussed earlier, by imaging sodium instead of hydrogen.76 Finally, there is currently investigation to determine if T1p imaging provides information that correlates with proteoglycan depletion. This method tips protons to the transverse plane, applies a longer, low-power spin-lock RF pulse, then tips the spins back to the B0 axis to prepare them for imaging.77 This technique has been shown to produce a unique contrast, referred to as T1p relaxation contrast,78 although finding a method that separates this from T2 effect is difficult due to the nature of T2 decay.79

Monitoring Changes in Articular Cartilage with Magnetic Resonance Imaging

Morphological markers of articular cartilage include both quantitative measures, such as thickness and volume, and semiquantitative measures, that grade cartilage integrity by a variety of scoring methods. Intermediate-TE and long-TE FSE images are usually adequate for most current clinical applications and in circumstances when lengthier high-resolution techniques are not justified (Fig. 9-12, Fig. 9-13). However, thinly partitioned, 3D SPGR images with selective fat suppression or water excitation are preferable for delineating cartilage morphology. Advantages of this latter technique include greater contrast, higher resolution, wide availability, ease of use, stable performance, no chemical shift artifact, and reasonable acquisition time (7 to 10 min). Disadvantages include longer acquisition times than those required for FSE imaging and vulnerability to magnetic susceptibility and metallic artifacts. These artifacts range from mild distortions arising near small postoperative metallic fragments or gas bubbles introduced into the joint by vacuum phenomenon, to severe distortions caused by metallic implants or other orthopedic hardware following tibial plateau fracture or cruciate ligament repair. Failure of fat suppression due to regional field heterogeneities is generally not a problem because of the cylindrical shape of the knee, but can arise if the knee is bent or if the patella protrudes excessively. Typically, however, failed fat suppression in the region of the patella usually involves the marrow and superficial soft tissues, but does not reach the articular cartilage.

Figure 9-21 Imaging cartilage matrix damage. A, Sagittal inversion-recovery image of a knee following i.v. administration of Gd-DTPA shows a region of high signal intensity (arrow) in the patellar cartilage indicative of abnormal uptake of anionic Gd-DTPA2-, and therefore, local proteoglycan depletion. Cartilage in the trochlear groove (arrowhead) shows low signal intensity indicative of repulsion of Gd-DTPA2- by negatively charged proteoglycans. B, Fat-suppressed, T2-weighted image of the same knee prior to Gd-DTPA2- injection shows a smaller focus of increased signal intensity (arrow) in the same location indicative of local collagen matrix loss. This is associated with subarticular marrow edema in the patella. (Courtesy of Synarc, Inc.)

Several studies have evaluated the diagnostic accuracy of fat-suppressed 3D SPGR for identifying areas of cartilage loss in the knee. In a comparison of 3D SPGR with and without fat suppression T2*-weighted GRE, and conventional T1-weighted, proton density-weighted and T2-weighted SE sequences in ten elderly cadaver knees, Recht et al.29 found fat-suppressed, 3D SPGR (flip angle = 60 degree, TE = 10 ms, voxel size = 469 m x 938 m x 1500 m) to have the greatest sensitivity (96%) and specificity (95%) for demonstrating patellofemoral cartilage lesions visible on pathological sections. Disler et al.32 similarly showed the same technique in vivo to have 93% sensitivity and 94% specificity for arthroscopically visible cartilage lesions.

Most scoring methods reported thus far simply count articular cartilage defects and grade them according to the depth of the cartilage loss (for example, 0 = normal-thickness, 1 = superficial fraying or isolated signal

P.183


abnormality, 2 = partial-thickness loss, 3 = full-thickness loss) (Fig. 9-20). Various more complex schemes, which take into account different patterns of cartilage involvement and the distribution of these changes in the knee, have been developed recently.80 However, the full validity of any of these schemes has not yet been thoroughly established. There is considerable face validity to the link between cartilage loss and clinical outcomes in OA, but the amount of cartilage loss that is clinically relevant has not yet been determined. The issue is complicated by the multifactorial nature of joint failure and the oversimplification that monostructural models suffer. Nevertheless, cartilage loss is currently the most broadly accepted metric of structural progression in OA. Unresolved issues of surrogate validity not withstanding, semiquantitative scoring of cartilage loss can be relatively precise and resolve progression in one year. In a recent study of 29 patients with OA in whom the articular cartilage was scored in 15 locations in the knee using a seven-point scale, the intraclass correlation coefficient between two specially trained radiologists was 0.99.80 A subsequent examination of 30 subjects from an ongoing cohort study of 3,075 elderly men and women imaged with a 15-min MRI protocol (T2-weighted FSE) found similar inter-reader precision for femorotibial cartilage using the same scoring method (ICC = 0.91).81

Figure 9-22 Example of cartilage segmentation performed by using fluctuating equilibrium MR imaging in a healthy 32-year-old male volunteer. Cartilage surfaces on the femur (red), patella (yellow), and tibia (blue) are all well seen. (

Gold GE, Hargreaves, BA, Vasanawala, SS, Webb, JD, Shimakawa, AS, Brittain, JH, Beaulieu CF: Articular cartilage of the knee: evaluation with fluctuating equilibrium MR imaging initial experience in healthy volunteers. Radiology 238:712 718, 2006.

)

Aside from semiquantitative scoring, a number of quantitative markers of cartilage morphology have been developed, including cartilage volume. This measurement can be derived from segmented images of the articular cartilage on fat-suppressed 3D SPGR or SSFP images using any of a variety of image analysis tools currently available (Fig. 9-22). A number of studies have validated the technical accuracy of these methods and established the precision error to range from 2% to 4% coefficient of variation (SD/mean volume)14,82,83, (Fig. 9-23). In one investigation, 16 elderly women with OA of the knee were imaged with MRI at yearly intervals for 2 years. The mean annual rate of cartilage loss was determined to be -6.7% 5.2% for the femur, -6.33% 4.3% for the tibia, and -3.4% 2.9% for the patella based on linear regression of the three time points.84 Eckstein et al.85 summarized the results of many different evaluations, concluding that annual changes of cartilage volume in most knee compartments in patients with OA are on the order of -4% to -6%. This range is greater than the expected precision error, providing strong evidence that the results are clinically significant.

Figure 9-23 Technical accuracy of volumetric quantification of cartilage with MRI. The graph depicts cartilage volumes determined from fat-suppressed, T1-weighted 3D gradient-echo images (open circles) and magnetization transfer subtraction images (closed circles) plotted against volumes measured directly by water displacement. A total of 12 cartilage plates (six patellar, three tibial, three femoral) from six knees were included. Line represents theoretical 100% accuracy. (Modified from

Peterfy C, van Dijke, CF, Janzen, DL, et al. Quantification of articular cartilage in the knee with pulsed saturation transfer and fat-suppressed MR imaging. Radiology 192:485 491, 1994.

)

Limitations of cartilage volume quantification include assumptions used to model cartilage volume change over time. For practical reasons, a linear model is usually the only feasible assumption for most clinical trials and epidemiological studies involving four or fewer time points. More complicated models (quadratic, and so on) may turn out to be more accurate, but until careful natural history studies have refined these models, curve-fitting challenges limit their use in most studies. Regardless, measurement precision for cartilage volume change combines errors related both to the measurement technique and the cartilage loss model used.

Other limitations of cartilage volume as a marker of disease severity and structural progression include insensitivity to small focal defects. These are more easily identified by semiquantitative scoring, or by regional cartilage volume mapping.86 Measurement precision and therefore statistical power decreases as the subdivisions get smaller. Accordingly, the tradeoff between sensitivity and measurement precision must be carefully balanced. One highly refined method of depicting regional variations in cartilage quantity is thickness mapping.87,88 As intuitive as cartilage

P.184


thickness may seem, however, questions remain as to whether the minimum, maximum, or average thickness is the most relevant, how to deal with multiple lesions, and to what extent the location of a lesion (weight bearing, non-weight bearing) is important.

Perhaps the greatest limitation of all markers of cartilage morphology, however, is their fundamentally irreversible nature and relatively slow responsiveness. Regardless of how precisely change in cartilage morphology can be measured, its rate of change cannot be driven any faster than the disease process itself. For a solution to this problem, one must look upstream to earlier stages in the disease process of cartilage degeneration. Accordingly, there has been a great deal of interest in developing MRI markers of cartilage composition.

MRI in markers of cartilage composition relate principally to the collagen matrix or constituent proteoglycans. The most promising markers of collagen matrix integrity include T2 relaxation and magnetization transfer coefficient. Markers of proteoglycan integrity include water diffusion, Gd-DTPA2- uptake, T1p, and 23Na concentration.

As discussed above, disruption of the fibrillar organization of collagen or actual decrease in collagen content reduces T2 relaxation and increases signal intensity on T2-weighted images. Areas of elevated signal in otherwise low signal-intensity cartilage on long-TE MR images therefore represent foci of chondromalacia. While several studies have verified this relationship between T2 relaxation and fibrillar collagen in cartilage, none have meticulously established the diagnostic accuracy (e.g., area under ROC curve, with histological verification) of subjective readings using MRI acquisition techniques that are applicable to multicenter studies or generalizable to clinical use. More importantly, the validity of cartilage T2 as a biomarker of matrix integrity depends on its predictive power for subsequent cartilage loss. Although there is considerable face validity to this model and some anecdotal longitudinal evidence to support it, further prospective validation is needed. If this hypothesis is indeed true, then abnormal cartilage T2 may identify cartilage at risk of future loss and thereby identify patients in need of aggressive therapy, hopefully before the point of no return. In addition to subjective evaluations of focal signal abnormalities in articular cartilage, regional changes in T2 relaxation can be quantified and monitored over time with multi-echo SE imaging.16,18 Limitations of this approach include technical tradeoffs between image acquisition time and the number of echoes, spatial resolution, and the attainable SNR. Further validation and performance characterization of cartilage T2 are clearly needed.

Significantly less work has been done with magnetization transfer as a marker of collagen integrity in articular cartilage. Theoretically, this marker could be used almost exactly the same way that cartilage T2 is used. However, even less is known about its diagnostic accuracy, responsiveness to disease and therapy, dynamic range, and measurement precision. Accordingly, further characterization is needed.

As mentioned above, methods for evaluating the integrity of the proteoglycan matrix by probing regional variations in fixed negative charged density in articular cartilage have recently been developed. The histological and biochemical validity of this approach has been well demonstrated by a number of groups.71,72,73 Using cartilage-nulling inversion recovery sequences at high spatial resolutions and high field strength, Bashir et al.71 demonstrated high histological correlation of the distribution of anionic Gd-DTPA2- with perichondrocytic GAG depletion following incubation of cartilage explants with IL-1 (interleukin-1). Subsequent studies have shown a linear correlation between T1 associated with Gd-DTPA2- and cartilage GAG ranging from 10 mg/mL to 70 mg/mL as measured directly biochemically.89 In a study by Trattnig et al.,73 areas of abnormal Gd-DTPA2- uptake in cartilage specimens harvested at total knee replacement surgery all corresponded to sites of collagen loss based on azan staining at histology. Unfortunately, this study did not report the correlation with areas of abnormal T2, if any were present. The study also reported marked interindividual variation in the pattern of Gd-DTPA2- uptake in eight normal volunteers that were examined, as well as marked differences in the diffusion times observed for cartilages of different thickness. Accordingly, while Gd-DTPA2- uptake appears to be a valid method for quantifying GAG concentration and its distribution in articular cartilage, with good dynamic range properties relative to GAG concentration, the relationship of this marker to cartilage T2 has yet to be examined. Does abnormal Gd-DTPA2- uptake precede abnormal T2 temporally? What is the relative performance of these two markers in terms of sensitivity, specificity, responsiveness to disease and therapy, dynamic range, predictive power for subsequent cartilage loss, other structural changes associated with OA, and clinical outcomes of OA? Finally, what is the optimal in vivo acquisition technique for cartilage Gd-DTPA2- uptake as a marker?

Imaging Other Articular Components in Osteoarthritis

In addition to evaluating the articular cartilage, MRI is uniquely capable of imaging all of the other structures that make up the joint, including the synovium and joint fluid, articular bones, intra-articular menisci, labra and discs, cruciate ligaments, collateral and other capsular ligaments, and periarticular tendons and muscles. Moreover, using the same voxel-counting technique employed for quantifying articular cartilage in 3D reconstructed images,14,90 it is possible to determine the volume of each of these components within the same joint.

Some degree of synovial thickening can be found in a majority of osteoarthritic joints.91 Whether this synovitis contributes directly to articular cartilage loss in OA, or simply arises in reaction to the breakdown of cartilage by other causes remains a controversy.92 However, synovitis may be important to the symptoms and disability of OA, and may pose different treatment requirements than those directed only toward chondroprotection . MRI is capable of imaging thickened or inflamed synovium, but usually this requires the use of special techniques, such as magnetization-transfer subtraction,12 fat-suppressed, T1-weighted imaging,12

P.185


or intravenous injection of Gd-containing contrast material12,93,94,95 (Fig. 9-24). By monitoring the rate of synovial enhancement with Gd-containing contrast over time using rapid, sequential MRI, it is furthermore possible to grade the severity of the synovitis in these patients. The majority of work in this area has, however, focused on rheumatoid arthritis.

Figure 9-24 Synovial imaging with MRI. Transverse images of the suprapatellar recess of the knee of a patient with rheumatoid arthritis using magnetization-transfer subtraction A and fat-suppressed T1-weighted gradient-echo B both delineate the thickened synovial tissue with high contrast. (From

Peterfy C, Majumdar S, Lang P, van Dijke CF, Sack K, Genant HK: MR imaging of the arthritic knee: improved discrimination of cartilage, synovium and effusion with pulsed saturation transfer and fat-suppressed, T1-weighted sequences. Radiology 191:413 419, 1994.

)

Osseous changes in OA are superbly depicted by MRI. Both cortical and trabecular bone can be visualized with MRI, and because of the tomographic nature of this modality, MRI is better at delineating structures, such as osteophytes (Fig. 9-25) and subchondral cysts, that are often obscured by overlying structures on conventional radiographs. Using high-resolution MRI techniques,96,97 it may be possible to monitor trabecular changes in the subchondral bone (Fig. 9-26) in order to determine their importance in the development and progression of OA.

Figure 9-25 Delineating osteophytes with MRI. Sagittal A, and coronal B, images of a knee of a patient with OA clearly delineate marginal and central osteophytes. (Courtesy of Synarc, Inc.)

In addition to delineating the calcified components of a bone, MRI is uniquely capable of imaging the marrow. Subchondral marrow edema is occasionally associated not only with acute trauma but with progressive OA.98,99 Focal bone marrow edema in OA may be due to subchondral injuries caused by shifting articular contact points at sites of biomechanically failing cartilage (Fig. 9-27), or pulsion

P.186


of synovial fluid into uncovered subchondral bone. However, osteonecrosis, infection, and infiltrating neoplasms could theoretically produce a similar MRI appearance. Conventional radiographs are usually unremarkable in areas of bone marrow edema; however, bone scintigraphy may show increased uptake in these areas.

Figure 9-26 High resolution MRI of cortical and trabecular bone. Axial high-resolution (~150 m in-plane, 500- m slice thickness) fast 3D GRE image of the distal ulna and radius delineates both cortical and trabecular bone with high detail. (Courtesy of Synarc, Inc.)

The menisci in the knee (Fig. 9-28) and glenoid labrum in the shoulder are important to the stability and functional integrity of these joints. Equally important are the cruciate (Fig. 9-29) and collateral ligaments and the glenohumeral ligaments. The utility of MRI for evaluating these articular structures is already well established.100

Figure 9-27 Subchondral bone edema in OA. Sagittal fat-suppressed intermediate-weighted FSE image of an osteoarthritic knee showing local bone marrow edema in the antero-lateral femur (asterisk). Note the focus of increased signal (arrow) in the articular cartilage overlying this region. Similar findings are also present in the patella of this image. (Courtesy Synarc, Inc.)

A whole-organ MRI scoring method (WORMS), has been developed for clinical research in the knee.101 This scoring method examines 5 articular surface features (articular cartilage, subarticular marrow edema, subarticular cysts, subarticular bone attrition, and marginal osteophytes) in 15 regions of the knee, along with 8 other features (medial and lateral menisci, medial and lateral collateral ligaments, anterior and posterior cruciate ligaments, synovium and synovial effusion, and perarticular bursae and cysts). The test-retest reproducibility of these scores are high when done by trained, experienced radiologists. WORMS is currently being used in numerous longitudinal clinical trials and epidemiological studies.

Figure 9-28 MRI of the meniscus. Sagittal, fat-suppressed proton-density image shows a minimally displaced tear (arrow) of the posterior horn of the medial meniscus. This is associated with partial-thickness thinning (arrowhead) of the femoral articular cartilage immediately adjacent to the torn meniscus. (Courtesy Synarc, Inc.)

Challenges in Imaging Specific Joints

The Knee

Each joint poses different challenges to proper imaging with MRI. Most work thus far has focused on the knee, because not only is the knee frequently affected by OA and because loss of knee function can be severely disabling, but because the knee is a comparatively easy joint to image. Reasons for this include the large size of this joint, which lowers demands on spatial resolution, and the relatively cylindrical shape of the knee, which minimizes perturbation of the static magnetic field; field homogeneity is critical to the performance of frequency-selective fat suppression or

P.187


water excitation techniques, and important in quantitative studies based on signal intensity measurements. The cylindrical shape also allows the use of circumferential imaging coils, which show greater homogeneity than surface coils.15 Additionally, because the knee is a relatively incongruent joint, contact areas between the hyaline cartilage plates in all but the most severely degenerated joints are small. Articular surfaces are therefore easy to separate from each other on MR images. Delineating the articular surfaces is facilitated by the relative abundance of synovial fluid in the knee, which provides high contrast at this interface on T2-weighted images and fat-suppressed, T1-weighted images. Because the articular surfaces are only gently curved, partial-volume averaging is not a major problem. Because of these forgiving imaging features and the availability of surgical and arthroscopic therapies for many internal derangements of the knee, MRI experience with the knee is greater than for any other joint in the body.

These advantages, however, are offset to some extent by a number of disadvantages. The knee is a highly complex joint composed of three articular compartments, one of which involves a sesamoid bone the patella. The hyaline cartilage covering each of the articular surfaces accordingly shows somewhat different biomechanical properties and vulnerabilities. The joint contains two intra-articular ligaments, an intra-articular tendon, two menisci, intracapsular-extrasynovial fat pads, complex capsular ligaments (particularly laterally), and variable ontological remnants (plicas). Joint failure in the knee involves an equally complex interplay among these numerous articular constituents. Because the knee is a large joint, full coverage of the synovial cavity, including the suprapatellar recess, requires a relatively large FOV (12 cm to 18 cm). Because loose bodies tend to collect in the eddy pools within synovial recesses, incomplete coverage can result in important oversights. This can be particularly problematic in cases with large popliteal cysts dissecting down the calf. Larger fields of view, however, necessitate proportionately larger imaging matrices in order to maintain spatial resolution, and thereby increase the imaging time.1 A more thorough description of MRI techniques for whole-organ evaluation of the knee joint is provided in a review by Peterfy et al.102

Figure 9-29 MRI of the anterior cruciate ligament. A, Sagittal, fat-suppressed proton density-weighted image shows an intact anterior cruciate ligament (arrowheads). B, Similar image of a different knee shows a torn anterior cruciate ligament. (Courtesy of Synarc, Inc.)

The Hip

Next to the knee, the hip is the most important joint affected by OA from a disability standpoint. Despite this, however, the hip has received only scant attention in MRI evaluation for OA. This is at least in part because the hip poses significant challenges to proper imaging with MRI. It is a highly congruent joint, which makes separating the articular surfaces difficult. Delineation of the surfaces is further hampered by the relative lack of joint fluid in the tight synovial cavity of the hip. Moreover, the articular surfaces are highly curved, giving rise to severe partial-volume effects in all planes unless extremely high spatial resolution is employed. Accordingly, cartilage thickness measurements in the hip using MRI have been somewhat disappointing.103 Achieving high spatial resolution in the hip is, itself, not an entirely straightforward matter. Since the hip is a relatively deep joint, signal drop off with small (<5 cm) surface coils is usually prohibitive. Larger surface coils could be employed, but these offer lower resolution and do not provide homogeneous signal for quantitative measurements. The anatomy of the hip prevents the use of small circumferential coils, which could provide homogeneous images with high resolution. A large circumferential coil, such as the body coil, could be used in this way, but does not provide sufficient SNR to support the high spatial resolution needed. Multiple coils configured in a phased array about the hip offer high SNR along with high spatial resolution (Fig. 9-30) and are probably the best alternative for this purpose.

P.188


Figure 9-30 MRI of the hip using phased array technique. Coronal T2-weighted FSE image of a normal hip acquired using multiple coils arranged in a flexible phased array shows high S/N despite the relatively high-resolution employed. F = femoral head, A = superior acetabulum. (Courtesy of Synarc, Inc.)

Figure 9-31 MRI of OA shoulder. Oblique coronal (in plane with the long axis of the supraspinatus tendon), T1-weighted spin-echo image of an osteoarthritic shoulder shows exuberant osteophyte formation along the inferior margin of the humeral head. (Courtesy of Synarc, Inc.).

Figure 9-32 MRI of the proximal interphalangeal (PIP) joints using sagittal GRE water-selective excitation imaging. A, The articular cartilage (arrows) in the proximal interphalangeal (PIP) joint and the normal extensor tendon (arrowhead) of a normal subject. B, The PIP joint of a patient with severe chronic osteoarthritis, demonstrating complete loss of the articular cartilage. Note the thickening and high signal change in the extensor tendon close to its insertion (arrow). Note also the high signal in the bone marrow representing edema at the tendon enthesis site (arrowhead) and the large dorsal osteophyte (*). C, A commonly seen pattern of cartilage loss predominantly affecting the volar aspect articular surfaces, with more dorsal cartilage preservation. Severe soft tissue swelling around the dorsum of the joint is also present along with prominent dorsal osteophytes (arrowheads). V = volar aspect of joint. (From

Tan AL, Grainger AJ, Tanner SF, Shelley DM, Pease C, Emery P, McGonagle D. High-resolution magnetic resonance imaging for the assessment of hand osteoarthritis. Arthritis Rheum. 52:2355 2365, 2005.

)

P.189


The Shoulder

Like the hip, the shoulder is a congruent, ball-in-socket joint with closely opposing articular surfaces104 (Fig. 9-31). Because of the angular shape of the shoulder, magnetic field heterogeneities tend to develop laterally near the greater tuberosity.105 Although the field appears relatively undisturbed at the glenohumeral joint, lateral heterogeneities can limit the performance of fat suppression and complicate evaluation of the rotator cuff. Accurate assessment of the tendons of the rotator cuff is important because the shoulder relies heavily on these structures for stability, and rotator cuff tear is an important risk factor for the development of OA in this joint.106 Shoulder stability is also dependent on the integrity of the glenoid labrum and the glenohumeral ligaments. However, reliable imaging of these labrocapsular structures can be extremely difficult, particularly in the absence of joint distention by significant synovial effusion. This can be improved by intra-articular injection of saline107 or Gd-containing MRI contrast material (MR arthrography).108,109

Hand and Finger Joints

The joint most commonly affected by OA is the distal interphalangeal joint of the finger. The major challenge to imaging this small joint is the demand on spatial resolution. For this reason, small-bore, high-field magnets and small circumferential imaging coils are usually necessary13,110 (Fig. 9-32). The metacarpophalangeal joints are less frequently affected by OA, but are larger joints, and have been successfully imaged using conventional clinical MRI systems.90

Advances in MRI Technology and the Impact on Magnetic Resonance Imaging of Osteoarthritis

Over the last few years, MRI technology has undergone a significant upgrade, with both higher field scanners (for example 3.0T) and dedicated lower field extremity scanners (0.2T to 1.0T) becoming available. This is a result of scientific advances in the field of superconductivity, digital signal processing, amplifier and networking electronics, and image visualization. The advances in technology make it possible to image joints with higher spatial resolution, lower time, and improved patient comfort.

First generation 3T scanners had a number of limitations that minimized their utility for joint imaging in OA. First of all, the image resolution and characteristics at 1.5T are generally considered satisfactory for clinical imaging of joints and the increased field strength considered unnecessary. Second, the need for additional space and shielding, combined with the poor magnet homogeneity and shortage of FDA approved coils, as well as the increased cost, limited the use of 3T scanners, notwithstanding the ability to visualize various tissue compartments with higher spatial resolution.

Second generation 3T scanners, however, have overcome many of these limitations. Field homogeneity, and thereby fat saturation, is no longer a problem, and a number of coils are available. Enhancements in superconductivity have allowed the bore to become shorter, and the space requirements are now quite similar to 1.5T scanners. Workhorse sequences commonly used on 1.5T such as T2 FSE and 3D-GRE are also available on the 3T scanner, at a higher SNR. One can take advantage of this improved SNR to either improve the spatial resolution, or to decrease the exam time. In particular, the exam time can be significantly shortened with the use of phased array coils and parallel imaging. Phased array coil technology was originally developed to improve the intensity uniformity of MR images obtained using surface coils, while preserving their inherent gain of SNR. Recently, new methods for encoding the MRI signal are being adopted that fall under the generic name of parallel imaging. Parallel imaging methods use the unique spatial perspective of the signal that comes from individual coils, along with the known sensitivity profiles of the surface coil elements within the array. This strategy allows overall improvement in coverage, which can be traded off for a reduction in the amount of time required to obtain the MR image up to a factor related to the number of independent coil channels within the array. Multiple RF channels are required to process these data independently, and in principle, an eight-channel coil would be able to image eight times as fast. However, practical considerations limit image acceleration to values well below the maximum allowed by theory. The driver for these high field scanners has been neuroimaging and cardiovascular imaging, and OA imaging stands to benefit from the increased presence of 3T scanners. Improved spatial resolution allows for more sensitive assessment of change in slowly progressing markers such as cartilage volume. Alternatively, parallel imaging allows for reductions of 20% to 40% in imaging time, thereby improving patient comfort.

Low-Field Imaging

Another way to improve patient comfort and reduce overall cost burden to the health care system is to use low-field MRI units.111 Conventional whole-body MRI is still relatively expensive and inconvenient, and although it is free of ionizing radiation, it is contraindicated in patients with pacemakers, aneurism clips, and other metal objects. Additionally, some patients find the experience unpleasant, and about 5% are unable to complete the examination because of claustrophobia. Low field-strength extremity MRI systems were introduced a decade ago as lower-cost alternatives in such circumstances. Because these systems operate at lower magnetic field strength, typically 0.2T to 1.0T, they can be made much smaller and operated less expensively. Also, whereas conventional 1.5T magnets require placing the entire body into the bore, imaging with extremity MRI systems requires patients to insert only their limb into the magnet while sitting or lying next to the unit. This eliminates claustrophobia, and reduces risks associated with metal in the body or in the examination room. Because of the small fringe-field, low weight, and small footprint of these systems, they can operate in environments that were previously inaccessible to MRI,

P.190


such as medical offices. The smallest extremity MRI system currently available commercially is described by Shellock et al.112 This system can operate in as little as 4 square meters of space, and is actually portable. Recent work has shown additional utility in one of these scanners111 in providing a range of magnetic field strengths, which can be used to obtain T1-dispersion contrast for protein imaging.113 The main disadvantage of extremity MRI systems is that their low magnetic field strength cannot support as much image resolution or as many image contrast mechanisms as conventional whole-body 1.5T systems. 24 Additionally, the small size of these systems precludes imaging other body parts, such as the shoulders, hips, spine, chest, abdomen, and pelvis, which is a capability that most radiology services require. Because of these limitations, extremity MRI systems were not initially felt by mainstream radiologists to provide sufficient performance for their needs. Higher field strength (1.0T) extremity systems that can support higher spatial resolution and broader contrast mechanisms, as well as larger low-field systems that can accommodate additional anatomical sites, such as the shoulders, have become available, but at the expense of larger space requirements and greater cost, and even these systems still offer some performance deficit in the eyes of many radiologists. It is important to keep in mind, however, that the needs of radiology are not the same as those of orthopedists and rheumatologists. The circumstances and therefore the technical performance requirements for MRI in these disciplines are very different. Orthopedists and rheumatologists do not have as much need to image multiple body parts at least not in patients with OA. Imaging the knees, and perhaps the hands, is usually sufficient. Fueled by increasing utilization in both OA and rheumatoid arthritis, extremity MRI systems can in turn be expected to continuously improve their technical performance in order to keep pace.

Conclusion

MRI is clearly a tool of unprecedented capabilities for evaluating joint disease and its potential treatments. MRI's unparalleled tissue contrast allows it to directly examine all components of a joint simultaneously and thus evaluate the joint as a whole organ and OA as a disorder of organ failure, in which dysfunction may result from any one of a number of different causes. Especially intriguing is the unique potential of this technology for identifying very early changes associated with cartilage degeneration, and its ability to quantify subtle morphological and compositional variations in different articular tissues over time. Employing these techniques, MRI may provide more objective measures of disease progression and treatment response than are currently attainable by other methods. This will facilitate both the assessment of new therapies for OA and investigations of the pathophysiology in this disorder. However, with this growing armamentarium comes a greater need for technical sophistication on the part of the clinician and growing pressures to contain costs. These demands necessitate a deeper understanding of the tradeoffs associated with choosing different diagnostic approaches. There is a particular need for clinicians to become sophisticated in applications of MRI in this disease, not only to better understand the growing number of studies that utilize this modality, but to assist in directing its development to better serve the needs of clinicians and their patients.

References

1. Abragam A. The Principles of Nuclear Magnetism. 1983, London: Oxford University Press.

2. Budinger T, Lauterbur P. Nuclear magnetic resonance technology for medical studies. Science 226:288 298, 1984.

3. Haacke E, Tkach J. Fast MR imaging: techniques and clinical applications. AJR 155:951 964, 1990.

4. Pykett I. NMR imaging in medicine. Sci Am 246:78 88, 1982.

5. Young S. Magnetic Resonance Imaging: Basic Principles. 1988, New York: Raven Press.

6. Xia Y, Farquhar T, Burton-Wurster N, et al. Dffiusion and relaxation mapping of cartilage-bone plugs and excised disks using microscopic magnetic resonance imaging. Magn Reson Med 31:273 282, 1994.

7. Burstein D, Gray ML, Hartman AL, et al. Diffusion of small solutes in cartilage as measured by nuclear magnetic resonance (NMR) spectroscopy and imaging. J Orthop Res 11:465 478, 1993.

8. Hargreaves BA, Vasanawala SS, Pauly JM, et al. Characterization and reduction of the transient response in steady-state MR imaging. Magn Reson Med. 46(1):149 58, 2001.

9. Scheffler K, Heid O, Hennig J. Magnetization preparation during the steady state: fat-saturated 3D TrueFISP. Magn Reson Med 45(6):1075 80, 2001.

10. Hennig J, Speck O, Scheffler K. Optimization of signal behavior in the transition to driven equilibrium in steady-state free precession sequences. Magn Reson Med 48(5):801 9, 2002.

11. Woolf SD, Chesnick S, Frank JA, et al. Magnetization transfer contrast: MR imaging of the knee. Radiology 179: 623-628, 1991.

12. Peterfy CG, Majumdar S, Lang P, et al. MR imaging of the arthritic knee: improved discrimination of cartilage, synovium and effusion with pulsed saturation transfer and fat-suppressed T1-weighted sequences. Radiology 191:413 419, 1994.

13. Hall LD, Tyler JA. Can quantitative magnetic resonance imaging detect and monitor the progression of early osteoarthritis?, in Osteoarthritic Disorders, K.E. Kuetner and V.M. Goldberg, Editors. 1995, American Accadamy of Orthopaedic Surgeons: Rosemont, IL. pp 67-84.

14. Peterfy CG, van Dijke CF, Janzen DL, et al. Quantification of articular cartilage in the knee by pulsed saturation transfer and fat-suppressed MRI: optimization and validation. Radiology 192:485 491, 1994.

15. Kneeland JB, Hyde JS. High-resolution MR imaging with local coils. Radiology 171:1 7, 1989.

16. Mosher T, Dardzinski B, Smith M. Human articular cartilage: influence of aging and early symptomatic degeneration on the spatial variation of T2 preliminary findings at 3 T. Radiology 241(1):259 266, 2000.

17. Chandnani VP, Ho C, Chu P, et al. Knee hyaline cartilage evaluated with MR imaging: a cadaveric study involving multiple imaging sequences and intraarticular injection of gadolinium and saline solution. Radiology 178:557 561, 1991.

18. Dardizinski B, Mosher T, Li S, et al. Spatial variation of T2 in human articular cartilage. Radiology 205:546 550, 1997.

19. Kim DJ, Suh JS, Jeong EK, et al. Correlation of laminated MR appearance of articular cartilage with histology, ascertained by artificial landmarks on the cartilage. J Magn Reson Imaging 10(1):57 64, 1999.

20. Xia Y, Farquhar T, Burton-Wurster N, et al. Origin of cartilage laminae in MRI. JMRI 7:887 894, 1997.

21. Rubenstein JD, Kim JK, Morava-Protzner I, et al. Effects of collagen orientation on MR imaging characteristics of bovine cartilage. Radiology 188:219 226, 1993.

P.191


22. Erickson SJ, Prost RW, Timins ME. The magic angle effect: background physics and clinical relevance. Radiology 188:23 25, 1993.

23. Erickson SJ, Waldschmidt JG, Caervionke LF, et al. Hyaline cartilage: truncation artifact as a cause of trilaminar appearance with fat suppressed three dimensional spoiled gradient recalled sequences. Radiology 201:260 264, 1996.

24. Frank LR, Brossmann J, Buxton RB, et al. MR imaging truncation artifacts can create a false laminar appearance in cartilage. AJR Am J Roentgenol 168(2): 547-54, 1997.

25. Yao L, Gentili A, Thomas A. Incidental magnetization transfer contrast in fast spin-echo imaging of cartilage. JMRI 6:180 184, 1996.

26. Miyazaki M, Takai H, Kojima F, et al. Control of magnetization transfer effects in fast SE imaging. in Radiological Society of North America. 1994. Chicago, IL.

27. Santyr GE. Magnetization transfer effects in multislice MR imaging. Magn Reson Imaging 11:521 522, 1993.

28. Gold GE, Hargreaves BA, Reeder SB, et al. Controversies in protocol selection in the imaging of articular cartilage. Semin Musculoskelet Radiol 9(2):161 72, 2005.

29. Recht MP, Kramer J, Marcelis S, et al. Abnormalities of articular cartilage in the knee: analysis of available MR techniques. Radiology 187:473 478, 1993.

30. Recht MP, Pirraino DW, Paletta GA, et al. Accuracy of fat-suppressed three-dimensionl spoiled gradient-echo FLASH MR imaging in the detection of patellofemoral articular cartilage abnormalities. Radiology 198:209 212, 1996.

31. Disler DG, McCauley TR, Kelman CG, et al. Fat-suppressed three-dimensional spoiled gradient-echo MR imaging of hyaline cartilage defects in the knee: comparison with standard MR imaging and arthroscopy. AJR 167:127 132, 1996.

32. Disler DG, McCauley TR, Wirth CR, et al. Detection of knee hyaline articular cartilage defects using fat-suppressed three-dimensional spoiled gradient-echo MR imaging: comparison with standard MR imaging and correlation with arthroscopy. AJR 165:377 382, 1995.

33. Carr H.Y. Steady-state free precession in nuclear magnetic resonance. Phys. Rev 112:1693 1701, 1958.

34. Heid O. True FISP cardiac fluoroscopy. In: Proceedings of the 5th International Society for Magnetic Resonance in Medicine, Vancouver, 1997.

35. Duerk JL, Lewin JS, Wendt M, et al. Remember true FISP? A high SNR, near 1-second imaging method for T2-like contrast in interventional MRI at .2 T. J Magn Reson Imaging 8(1):203 8, 1998.

36. Hargreaves BA, Gold GE, Beaulieu CF, et al. Comparison of new sequences for high-resolution cartilage imaging. Magn Reson Med 49(4):700 9, 2003.

37. Mosher TJ, Pruett SW. Magnetic resonance imaging of superficial cartilage lesions: role of contrast in lesion detection. J Magn Reson Imaging 10(2):178 82, 1999.

38. Gold G, Thedens D, Pauly J, et al. MR imaging of articular cartilage of the knee: new methods using ultrashort TEs. AJR 170:1223 1226, 1998.

39. Kijowski R, Lu A, Block WF, et al. Evaluation of the articular cartilage of the knee joint with vastly undersampledisotropic projection reconstruction steady-state free precession imaging. J Magn Reson Imaging. 24(1):168 175, 2006.

40. Vasanawala SS, Pauly JM, Nishimura DG. Fluctuating equilibrium MRI. Magn Reson Med 42(5):876 883, 1999.

41. Absil J, Denolin V, Metens T. Fat attenuation using a dual steady-state balanced-SSFP sequence with periodically variable flip angles. Magn Reson Med 55(2):343 351, 2006.

42. Leupold J, Hennig J, Scheffler K. Alternating repetition time balanced steady state free precession. Magn Reson Med 55(3): 557-565, 2006.

43. Vasanawala SS, Pauly JM, Nishimura DG. Linear combination steady-state free precession MRI. Magn Reson Med 43(1):82 90, 2000.

44. Hargreaves BA, Vasanawala SS, Nayak KS, et al. Fat-suppressed steady-state free precession imaging using phase detection. Magn Reson Med 50(1): 210-213, 2003.

45. Hargreaves BA, Bangerter NK, Shimakawa A, et al. Dual-acquisition phase-sensitive fat-water separation using balanced steady-state free precession. Magn Reson Imaging 24(2):113 122, 2006.

46. Vasanawala SS, Hargreaves BA, Pauly JM, et al. Rapid musculoskeletal MRI with phase-sensitive steady-state free precession: comparison with routine knee MRI. AJR Am J Roentgenol 184(5):1450 1455, 2005.

47. Reeder SB, Pelc NJ, Alley MT, et al. Rapid MR imaging of articular cartilage with steady-state free precession and multipoint fat-water separation. AJR Am J Roentgenol 180(2):357 362, 2003.

48. Reeder SB, Pineda AR, Wen Z, et al. Iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL): application with fast spin-echo imaging. Magn Reson Med 54(3): 636-644, 2005.

49. Gold G, Reeder S, Yu H, et al. Cartilage Morphology at 1.5T: Comparison of 3D FS-SPGR and IDEAL SPGR Imaging. International Society for Magnetic Resonance in Medicine, Miami, 2005.

50. Kornaat PR, Reeder SB, Koo S, et al. MR imaging of articular cartilage at 1.5T and 3.0T: comparison of SPGR and SSFP sequences. Osteoarthritis Cartilage 13(4):338 344, 2005.

51. Broderick L, Turner D, Renfrew D, et al. Severity of articular cartilage abnormality in patients with osteoarthritis: evaluation with fast spin-echo MR vs arthroscopy. Am J Roentgenol 162:99 103, 1994.

52. Rose PM, Demlow TA, Szumowski J, et al. Chondromalacia patellae: fat-suppressed MR imaging. Radiology 193:437 440, 1994.

53. Yulish BS, Montanez J, Goodfellow DB, et al. Chondromalacia patellae: assessment with MR imaging. Radiology 164:763 766, 1987.

54. Quinn SF, Rose PM, Brown TR, et al. MR imaging of the patellofemoral compartment. MRI Clin N Am 2:425 439, 1994.

55. Zaim S, Lynch JA, Li J, et al. MRI of early cartilage degeneration following meniscal surgery: a three-year longitudinal study. International Society for Magnetic Resonance in Medicine, Glasgow, Scotland, 2001. In proceedings.

56. Gold GE, Butts K, Fechner KP, et al. In vivo diffusion-weighted imaging of cartilage. 6th Annual Meeting of the International Society of Magnetic Resonance in Medicine, Sydney, Australia, 1998.

57. Frank LR, Wong EC, Luh W, et al. Articular cartilage in the knee: mapping of the physiologic parameters at MR imaging with a local gradient coil - preliminary results. Radiology 210:241 246, 1999.

58. Foy BD, LLM, Gray ML, et al. NMR parameters of intersitial sodium in cartilage [abstract]. in Society of Magnetic Resonance in Medicine, Amsterdam, 1989.

59. Jelicks LA, PPK, O'Byrne EM, et al. Hydrogen-1, sodium-23, and ccarbon-13 MR spectroscopy of cartilage degradation in vitro. J Magn Reson Imaging 3:565 568, 1993.

60. Paul PK, O'Byrne EM, Gupta RK, et al. Detection of cartilage degradation with sodium NMR [letter]. Br J Rheumatol 30(4):318, 1991.

61. Bashir A, Gray M, Burstein D. Sodium T1 and T2 in control and defraded cartilage: implications for determination of tissue proteoglycan content. In: Proceedings of the 14th annual meeting of the Society of Magnetic Resonance in Medicine, Nice, 1995.

62. Insko E, Reddy R, Kaufman J, et al. Sodium spectroscopic evaluation of early articular cartilage degradation. In: Proceedings of the 4th annual International Society of Magnetic Resonance in Medicine, New York, 1996.

63. Callaghan P. Principles of Nuclear Magnetic Resonance Microscopy. 1991, New York, Oxford University Press.

64. Reddy R, Insko EK, Noyszewski EA, et al. Sodium MRI of human articular cartilage in vivo. Magn Reson Med 39(5):697 701, 1998.

65. Granot J. Sodium imaging of human body organs and extremities in vivo. Radiology 167(2):547 550, 1988.

66. Ra JB, Hilal SK, Oh CH, et al. In vivo magnetic resonance imaging of sodium in the human body. Magn Reson Med 7(1):11 22, 1988.

67. Insko EK, Kaufman JH, Leigh JS, et al. Sodium NMR evaluation of articular cartilage degradation. Magn Reson Med 41(1):30 34, 1999.

68. Shapiro E, Saha P, Kaufman J, et al. In vivo evaluation of human cartilage compression and recovery using 1H and 23Na MRI. In: Proceedings of the 7th Annual Meeting of the International Society of Magnetic Resonance in Medicine, Philadelphia, 1999.

P.192


69. Regatte RR, Kaufman JH, Noyszewski EA, et al. Sodium and proton MR properties of cartilage during compression. J Magn Reson Imaging 10(6):961 967, 1999.

70. Wu E, Gao E, Cham E, et al. Application of HTS RF Coil for sodium imaging on a high field system. In: Proceedings of the 7th Annual Meeting of the International Society of Magnetic Resonance in Medicine. Philadelphia, 1999.

71. Bashir A, Gray ML, Burstein D. Gd-DTPA as a measure of cartilage degradation. Magn Reson Med 36:665 673, 1996.

72. Bashir A, Gray ML, Hartke J, et al. Nondistructive imaging of human cartilage glycosaminoglycan concentration by MRI. Magn Reson Med 41:857 865, 1999.

73. Trattnig S, Mlynarkck V, Breilenseher M, et al. MR visualization of proteoglycan depletion in articular cartilage via intravenous injection of Gd-DTPA. Magn Reson Imaging 17:577 583, 1999.

74. Kusaka Y, Grunder W, Rumpel H, et al. MR microimaging of articular cartilage and contrast enhancement by manganese ions. Magn Reson Med 24:137 148, 1992.

75. Fujioka M, Kusaka Y, Morita Y, et al. Contrast-enhanced MR imaging of articular cartilage: a new sensitive method for diagnosis of cartilage degeneration. In: 40th Annual Meeting, Orthopaedic Research Society, New Orleans, 1994.

76. Lesperance LM, Gray ML, Burstein D. Determination of fixed charge density in cartilage using nuclear magnetic resonance. J Orthop Res 10:1 13, 1992.

77. Borthakur A, Hulvershorn J, Gualtieri E, et al. A pulse sequence for rapid in vivo spin-locked MRI. J Magn Reson Imaging 23(4):591 596, 2006.

78. Li X, Han ET, Ma CB, et al. In vivo 3T spiral imaging based multi-slice T(1rho) mapping of knee cartilage in osteoarthritis. Magn Reson Med 54(4):929 936, 2005.

79. Menezes NM, Gray ML, Hartke JR, et al. T2 and T1rho MRI in articular cartilage systems. Magn Reson Med 51(3):503 509, 2004.

80. Peterfy CG, Guermazi A, Zaim S, et al. Whole-organ evaluation of the knee in osteoarthritis using MRI. in XIV European League Against Rheumatism Congress. Glasgow, Scotland, 1999.

81. Wildy K, Zaim S, Peterfy C, et al. Reliability of the Whole-Organ review MRI scorning (WORMS) method for knee osteoarthritis (OA) in a multicenter study. In: 65th Annual Scientific Meeting of the American College of Rheumatology, San Francisco, Nov. 11-15, 2001.

82. Eckstein F, Sitteck H, Gavazzenia A, et al. Assessment of articular cartilage volume and thickness with magnetic resonance imaging (MRI). Trans Orthop Res Soc 20:194, 1995.

83. Burgkart R, Glaser C, Hyhlik-D rr A, et al. Magnetic resonance imaging-based assessment of cartilage loss in severe osteoarthritis: Accuracy, precision and diagnostic value. Arthritis Rheum 44(9):2072 2077, 2001.

84. Peterfy C, White D, Zhao J, et al. Longitudinal measurement of knee articular cartilage volume in osteoarthritis. Annual Meeting of Am Coll Rheum. American College of Rheumatology, San Diego, 1998.

85. Eckstein F, Cicuttini F, Raynauld JP, et al. Magnetic resonance imaging (MRI) of articular cartilage in knee osteoarthritis (OA): morphological assessment. Osteoarthritis Cartilage 2006,

86. Pilch L, Stewart C, Gordon D, et al. Assessment of cartilage volume in the femorotibial joint with magnetic resonance imaging and 3D computer reconstruction. J Rheum 21:2307 2321, 1994.

87. Cohen ZA, McCarthy DM, Ateshian GA, et al. In vivo and in vitro knee joint cartilage topography, thickness, and contact areas from MRI. Annual Meeting of Orthopaedic Research Society, San Francisco, February 1997,

88. Eckstein F, Gavazzeni A, Sittek H, et al. Determination of knee joint cartilage thickness using three-dimensional magnetic resonance chondro-Crassometry (3D MR-CCM). Magn Reson Med 36:256 265, 1996.

89. Bashir A, Gray ML, Hartke J, et al. Validation of gadolinium-enhanced MRI for GAG measurement in human cartilage. International Society of Magnetic Resonance in Medicine, Philadelphia, 1998.

90. Peterfy CG, van Dijke CF, Lu Y, et al. Quantification of articular cartilage in the metacarpophalangeal joints of the hand: accuracy and precision of 3D MR imaging. AJR 165:371 375, 1995.

91. Fernandez-Madrid F, Karvonen RL, Teitge RA, et al. Synovial thickening detected by MR imaging in osteoarthritis of the knee confirmed by biopsy as synovitis. Magn Reson Imaging 13:177 183, 1995.

92. Brandt KD. Insights into the natural history of osteoarthritis and the potential for pharmacologic modification of the disease afforded by study of the cruciate-deficient dog, in Osteoarthritic Disorders, K.E. Keutner and V.M. Goldberg, Editors. American Acadamy of Orthopaedic Surgeons: Rosemont, IL 1995, pp 419-426.

93. Palmer WE, Rosenthal DI, Shoenberg OI, et al. Quantification of inflammation in the wrist with gadolinium-enhanced MR imaging and PET with 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology 196:645 655, 1995.

94. K nig H, Sieper J, Sorensen M, et al. Contrast-enhanced dynamic MR imaging in rheumatoid arthritis of the knee joint: follow-up study after cortisol drug therapy. In: 77th Scientific Assembly and Annual Meeting of the Radiological Society of North America, Chicago, IL, 1991.

95. Yamato M, Tamai K, Yamaguchi T, et al. MRI of the knee in rheumatoid arthritis: Gd-DTPA perfusion dynamics. J Comput Assist Tomogr 17: 781-785, 1993.

96. Weinstein RS, Majumdar S. Fractal geometry and vertebral compression fractures. J Bone Min Res 9:1797 1802, 1994.

97. Majumdar S, Genant HK, Grampp S, et al. Analysis of trabecular structure in the distal radius using high resolution magnetic resonance images. Euro Radiol 4:517 524, 1994.

98. Vellet AD, Marks P, Fowler P, et al. Occult posttraumatic lesions of the knee, prevalence, classification, and short-term sequelae evaluated with MR imaging. Radiology 178:271 276, 1991.

99. Felson DT, Chaisson CE, Hill CL, et al. The association of bone marrow lesions with pain in knee osteoarthritis. Ann Intern Med 134(7):541 549, 2001.

100. Resnick D. Internal derangements of joints, in Diagnosis of Bone and Joint Disorders, D. Resnick, Ed. W.B. Saunders, Philadelphia, 1995, pp 3063-3069.

101. Peterfy CG, Guermazi A, Zaim S, et al. Whole-Organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage 12(3):177 190, 2004.

102. Peterfy CG, Gold G, Eckstein F, et al. MRI protocols for whole-organ assessment of the knee in osteoarthritis. Osteoarthritis Cartilage, 14 Suppl A:A95-111, 2006.

103. Hodler J, Trudell D, Pathria MN, et al. Width of the articular cartilage of the hip: quantification by using fat-suppression spin-echo MR imaging in cadavers. AJR 159:351 355, 1992.

104. Hodler J, Loredo R, Longo C, et al. Assessment of articular cartilage thickness of the humeral head: MR-anatomic correlation in cadavers. AJR 165:615 620, 1995.

105. Peterfy C. Technical considerations, in Shoulder Magnetic Resonance Imaging, L. Steinbach, et al. Eds. Lippincott-Raven, Philadelphia, 1998, pp 37-63.

106. Peterfy C, Genant H, Mow V, et al. Evaluating arthritic changes in the shoulder with MRI, in Shoulder Magnetic Resonance Imaging, L. Steinbach, et al., Eds. Lippincott-Raven, Philadelphia, 1998, pp 221-237.

107. Tirman PFJ, Stauffer AE, Crues JV, et al. Saline magnetic resonance arthrography in the evaluation of glenohumeral instability. Arthroscopy 9:550 559, 1993.

108. Palmer WE, Brown JH, Rosenthal DN. Labral-ligamentous complex of the shoulder: evaluation with MR arthrography. Radiology 190:645 651, 1994.

109. Tirman PFJ, Bost FW, Garvin GJ, et al. Posterosuperior glenoid impingement: MRI and MR arthrographic findings with arthroscopic correlation. Radiology 193:431 436, 1994.

110. Tan AL, Grainger AJ, Tanner SF, et al. High-resolution magnetic resonance imaging for the assessment of hand osteoarthritis. Arthritis Rheum 52(8): 2355-2365, 2005.

111. Macovski A, Conolly S. Novel approaches to low-cost MRI. Magn Reson Med 30(2):221 230, 1993.

112. Shellock FG, Stone KR, Crues JV. Development and clinical application of kinematic MRI of the patellofemoral joint using an extremity MR system. Med Sci Sports Exerc 31(6):788 791, 1999.

113. Ungersma SE, Matter NI, Hardy JW, et al. Magnetic resonance imaging with T(1) dispersion contrast. Magn Reson Med 55(6):1362 1371, 2006.



Osteoarthritis. Diagnosis and Medical. Surgical Management
Osteoarthritis: Diagnosis and Medical/Surgical Management
ISBN: 0781767075
EAN: 2147483647
Year: 2007
Pages: 19

Similar book on Amazon

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net