Non-Invasive Characterization of Cartilage Properties Using MR Imaging

Osteoarthritis (OA) is a degenerative disease affecting articular cartilage, leading to loss of its structure and function. Early stage OA is characterized by changes in the extracellular matrix (ECM), including a reduction in proteoglycans (PG) concentration, increased water content within the tissue, and increased synthesis and degradation of matrix molecules with disorganization of collagen network [1, 2]. The ability to noninvasively quantify PG changes in cartilage would therefore be useful for early OA diagnosis, monitoring cartilage response to therapies, and assessing efficacy of cartilage repair procedures [3]. T1rho and T2 weighted magnetic resonance (MR) imaging techniques have been shown to have potential in tracking early biochemical compositional changes within cartilage associated with degeneration [3, 4]. Additionally, this method has the potential to be a powerful tool to better understand how cartilage responds to different loading environments both acutely and over time. The main objective of this work is to validate T1rho and T2 relaxation times as non-invasive measures for the assessment of biochemical and biomechanical properties of cartilage.

The first two studies presented in this work focus on the validation of this T1rho and T2 imaging for non-invasive cartilage assessment. The first study examines both normal and osteoarthritic cartilage containing both OA defect regions and healthy appearing areas. This study aims to comprehensively assess the relationship between OA cartilage composition, biochemical, and biomechanical properties with T1rho and T2 relaxation times in order to validate this technique as an in vivo diagnostic method for early stage OA. The second study utilizes targeted enzymatic depletion of both glycosaminoglycan (GAG) and collagen to determine the specific effect of each ECM component on T1rho and T2 relaxation times. A repeated measures design examines the effect of targeted enzymatic cartilage degradation (to isolate changes in cartilage biochemical composition, mechanical properties, and histology) on the T1rho and T2 relaxation times. These studies utilize confined compression for biomechanical analysis of cartilage, biochemical assays for the determination of S-GAG and collagen content, and histology for visualization of cartilage structure and composition. These measures are compared to the associated T1rho and T2 relaxation times. The results of these studies indicate that increases in T1rho relaxation times are correlated with S-GAG depletion, increased percent extractable collagen, decreases in mechanical strength of cartilage, and areas of OA defects (within which the previously mentioned biomechanical and biochemical conditions exist). Together, the results of these two studies validate T1rho and T2 quantitative imaging techniques for the in vivo diagnosis of early OA and the non-invasive assessment of cartilage biomechanical and biochemical properties.

Altered patterns of mechanical loading can result in morphological and compositional changes to cartilage that lead to cartilage degeneration. Quantitative MR imaging is a unique tool with the potential to provide insight into the relationship between biomechanics and the biophysical environment of cartilage, which is vital to better understanding the development of OA and degeneration of cartilage. The third study presented utilizes T1rho as a method for assessing localized changes to cartilage with dynamic activity. Sagittal MR images were obtained before and immediately after subjects completed a single legged hopping activity to dynamically load cartilage. A system of equally spaced grid points were registered to 3D surface mesh models of the tibial and femoral cartilage surfaces constructed from the MR images. T1rho relaxation times were then determined at each grid point to examine site-specific changes before and after exercise. A significant decrease in relaxation times was found after exercise in both the tibial plateau and the femoral condyle, with a greater decrease observed in the lateral femoral cartilage than in the medial femoral cartilage. No significant correlation between location and exercise was found. At each grid point, T1rho cartilage maps were also divided into superficial and deep regions of cartilage to determine where the greatest changes occurred. Ongoing analysis of the layer specific results will provide insight into where in the cartilage thickness these changes are most localized. The decrease in relaxation times after loading is likely due to the relative increase in PG content that results from the exudation of water from the cartilage ECM due to loading. This study demonstrates how T1rho may be used to non-invasively provide insight into the biophysical environment of cartilage with loading.

T1rho and T2 imaging represent a very powerful tool for the non-invasive assessment of articular cartilage. This work is significant in that it validates this method for the assessment of cartilage biomechanical and biochemical properties. Additionally, these methods can be used in future work to better understand how various risk factors contribute to OA development and to give valuable insight into the connection between biomechanical factors, biochemical composition, and the development of cartilage degeneration.