| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dartmouth Medical School Hanover, NH 03755
Mosher et al. [1] have published an excellent study revealing the difficulty of predicting the T2 of articular cartilage [1]. Their work provides important additional insight into the influence of T2 decay on the MR appearance of articular cartilage. Their study found that the T2 values of cartilage in the posterior aspect of a single image through the lateral femoral condyle differed substantially from predicted values. Their work suggests that the effect of orientation on cartilage T2 is less than that predicted from prior ex vivo studies. It therefore calls into question widely held assumptions regarding the interrelationship between cartilage structure, orientation of the static magnetic field (B0), and T2.
We believe that the results of this study [1] actually provide additional support for the argument that the structure of cartilage has a strong influence on T2 decay and that variations in orientation relative to B0 still provide a satisfactory explanation for the observed T2 heterogeneity. The results, however, illuminate two important considerations that must be accounted for when imaging articular cartilage. First, joint surfaces are irregular and curved in three dimensions. Second, the structure of hyaline cartilage is not uniform. In reviewing the results of this study, it is helpful to consider the shape of the femoral condyle surface and the complexity of cartilage structure.
The predicted T2 values are based on an assumption that the structural elements of cartilage that impart a strong influence on T2 radiate perpendicularly from the subchondral bone. It is further assumed that the angle of this radiating structure can be described by drawing a line perpendicular to the surface of a tomographic sagittal plane image of the femoral condyle. The surface of the femoral condyle, however, is a curved three-dimensional structure, and a sagittal plane image will not account for the curvature of the surface out of the image plane. Therefore, a line drawn perpendicular to the surface of the image is not necessarily perpendicular to the bone surface: this is evident if one examines a three-dimensional model of a femur (Fig. 1A,1B).
|
|
It is probably even more important to recognize that the extracellular matrix of cartilage has a fibrous structure. We have shown in an earlier study that the shape of this structure, not the orientation of collagen fibrils, has a strong influence on the MR appearance of cartilage—an influence that is consistent with the magic angle effect [2]. This complex three-dimensional structure is not uniform. The shape of cartilage at the periphery of a joint differs considerably from the more central portion of the joint [3] (Fig. 2A,2B). Because of this variability of form, predicting the orientation of cartilage relative to B0 at a given location is extremely difficult. We have found, however, that the architecture or structure of cartilage tends to be characteristic for different joint surfaces, suggesting that variations in cartilage T2 can be predicted if the structure is understood. MR imaging is uniquely capable of providing additional insights into this structure.
|
|
Given the complexity of cartilage composition, its frequently oblique orientation at the bone surface, the curved surface of the bone, and the relatively large voxel dimensions of the T2 maps, the results of the study by Mosher et al. [1] are not surprising. In fact, higher and more variable T2 values in the posterior femoral condyle cartilage should be expected. The study is important because it underscores the difficulty in quantifying the T2 of cartilage while reinforcing the importance of measuring T2 in clinical studies. The results show that MR imaging and quantification of articular cartilage must account for the considerable complexity of joint surfaces and cartilage structure. As Kneeland [4] has written in his commentary, the relationship between cartilage structure and its MR appearance is clearly in need of more investigative work.
References
M.S. Hershey Medical Center Pennsylvania University College of
Medicine Hershey, PA 17033
Children's Hospital Medical Center Cincinnati, OH 45229
We appreciate the interest of Goodwin and Dunn regarding our recent publication evaluating orientation dependence of T2 in femoral cartilage [1]. We agree with their comments that complex joint geometry and tissue heterogeneity are important factors in evaluating the MR imaging appearance of articular cartilage.
Studies addressing the dependence of T2 on orientation of collagen fibrils in articular cartilage have historically used the theoretical model illustrated in figure 4 of our article [1], which is based on the arcade model originally proposed by Benninghoff [2]. In this model, collagen fibrils in the radial zone are assumed to be oriented perpendicular to the cortical surface, to have a more random orientation in the transitional zone, and to be parallel to the cortical surface in the superficial lamina splendens. This model correlates well with observed orientation dependence of T2 in small excised cartilage samples measured at high magnetic field strength [3]. Goodwin and Dunn describe features of femoral cartilage that limit application of this model to interpretation of our in vivo results. First, because the posterior femoral condyle has a complex curved surface, a sagittal slice is not orthogonal to the cortical bone. Thus, collagen fibers in the radial zone are imaged tangentially and their true orientation is indeterminate. Second, results from freeze fracture experiments suggest the fibrous architecture of cartilage has a complex planar or leaf organization, and raises questions regarding the validity of using the Benninghoff model to describe orientation behavior of T2.
Although in-plane curvature is a potential artifact, the experimental protocol was designed to minimize this effect. In the axial plane, the posterior femoral condyle has a round contour with a surface curvature that can be approximated with a 3- to 4-cm-diameter circle. A 2-mm-thick sagittal slice centered in the lateral femoral condyle would transect a surface with approximately 2° of in-plane curvature. Our data analysis clustered profiles in 10° intervals for comparison. Therefore, it is unlikely that in-plane curvature produced significant additional error in determination of orientation effect. Also, because the major axis of curvature of the condyle is in the axial rather than coronal plane, in-plane curvature will primarily rotate the orientation of the collagen fibers around the z-axis and not have an important effect on relative angle with B0. Finally, the spatial resolution of T2 maps in our experiment is substantially higher than that reported in prior in vivo studies describing an orientation dependence of cartilage signal intensity [4]. It is unlikely that the effect of inplane curvature in our experiment would be greater than that observed in standard clinical imaging; in fact, it is likely to be substantially less. Therefore, although curvature of the femoral condyle does limit precision in determining relative orientation of the radial zone with B0, it is unlikely to be a significant source of error.
Goodwin and Dunn question the assumption that the "structural elements of cartilage that impart a strong influence on T2 radiate perpendicularly from the subchondral bone." We agree. Results of our study [1] do not support the proposed Benninghoff model of collagen fiber orientation frequently used as a basis for the magic angle effect. Alternative models of the collagen matrix architecture may be more appropriate for interpretation of in vivo cartilage T2 data.
We were interested to note the similarity of the MR imaging appearance of cartilage in figure 2A of their letter and in figure 1 of our article [1]. In both are seen loss of radial striations and increased signal intensity in regions of cartilage in which orientation of the radial zone of cartilage approaches the magic angle. However, in our study, this qualitative change in signal intensity corresponded to only a small quantitative difference in T2, with the greatest increase in T2 observed near the articular surface. The Benninghoff model [2] would predict the greatest increase in T2 in the radial zone, with little orientation effect in the transitional zone as a result of random arrangement of collagen fibrils.
Goodwin and Dunn propose an intriguing explanation for this observation. Images presented in figure 2 of their letter and prior publication [5] suggest that a complex three-dimensional planar arrangement of the fibrous architecture, rather than individual fiber orientation, correlates with the MR imaging appearance of cartilage. This three-dimensional model proposed by Jeffery et al. [6] describes a fine network of collagen fibers arranged in curved layers or leaves extending from the cortical to the articular surface according to the split-line pattern. In the transitional zone of the posterior femoral condyle, it is possible that preferential orientation of the collagen layers, rather than discrete collagen fibrils, approaches 55°. Likewise, in the radial zone of the posterior femoral condyle, preferential orientation of the layers may be far from the magic angle. Because this three-dimensional model describes an anisotropic arrangement of the collagen matrix in the transitional zone, it provides a theoretical basis for our observed variation in T2 between regions of the femoral condyle, with greater orientation effect observed in the more superficial cartilage.
As a cautionary note, the inherent heterogeneity of articular cartilage in a joint limits extrapolation of results obtained on small excised tissue specimens to the in vivo setting. As knowledge accumulates, it is becoming apparent that articular cartilage is a complex tissue having regional differences in structure and composition within a joint—and between joints. Because cartilage T2 appears sensitive to these structural features, it will likely be an important tool for unraveling the complexity of articular cartilage. However, much work needs to be done.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. M. White, M. S. Sussman, M. Hurtig, L. Probyn, G. Tomlinson, and R. Kandel Cartilage T2 Assessment: Differentiation of Normal Hyaline Cartilage and Reparative Tissue after Arthroscopic Cartilage Repair in Equine Subjects. Radiology, November 1, 2006; 241(2): 407 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Recht, D. W. Goodwin, C. S. Winalski, and L. M. White MRI of Articular Cartilage: Revisiting Current Status and Future Directions Am. J. Roentgenol., October 1, 2005; 185(4): 899 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Van Breuseghem, H. T. C. Bosmans, L. V. Elst, F. Maes, S. D. Pans, P. P. M. A. Brys, E. A. Geusens, and G. J. Marchal T2 Mapping of Human Femorotibial Cartilage with Turbo Mixed MR Imaging at 1.5 T: Feasibility Radiology, November 1, 2004; 233(2): 609 - 614. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
