Quantitative magnetic resonance imaging (qMRI) parameters have generated much interest as potential biomarkers for detecting pre-arthritic alterations in populations at-risk for developing osteoarthritis (OA) (Amano et al., 2016, Pedoia et al., 2017, Theologis et al., 2014, Williams et al., 2019). Of particular interest, qMRI parameters T1ρ and T2* are associated with the matrix composition of articular cartilage (Wheaton et al., 2005, Williams et al., 2010). Advancements in studying the composition of cartilage may provide valuable information in the development of novel therapeutic interventions that prevent OA onset and progression before irreversible morphological changes occur (Matzat et al., 2013).

While mechanical loading is thought to play a key role in the progression of OA, traditional qMRI of knee articular cartilage is done without applied load. The challenge of studying qMRI of the articular cartilage in a loaded state is the need for an MRI-compatible loading device. While studies of articular cartilage before and after exercise provide insight into physiological loading, prior work in knee articular cartilage after squatting has shown relaxation on the order of 15 min (Van Ginckel and Witvrouw, 2013), and typical qMRI scan times take the majority of that timeframe. MRI-compatible loading devices permit consistent loading during the scan, as well as participant-specific load application that does not vary with the type of exercise nor the patient-specific manner by which the exercise is completed. MRI-compatible loading devices are difficult to construct due to constraints on the material and size, and the requirement that the subject’s knee remains motionless during scan acquisition (Jerban et al., 2020). As a result, our understanding of how the matrix composition of human articular cartilage changes with the application of load is limited.

Currently, the studies that investigated the effects of acute loading on T1ρ relaxation of human knee articular cartilage in-vivo examined weightbearing region of interests (ROIs) that combine the tibial and femoral cartilage in contact with each other (Souza et al., 2010, Subburaj et al., 2012b). To our knowledge, no study assessed the effects of acute loading on T1ρ when analyzing the human femoral and tibial articular cartilage separately. Separate analyses for the femoral and tibial articular cartilage may provide bone-specific insights into articular cartilage degeneration—in this study and in future studies of patients at-risk for OA. Furthermore, no study assessed the effects of acute loading on T2*, which may be of critical importance as prior work has found that T2* is sensitive to changes in deep cartilaginous tissue in a patient population at high risk for post-traumatic OA (Williams et al., 2019).

Understanding the effects of loading on qMRI has been limited to the regions of cartilage directly transmitting the load. Analysis of non-contact cartilage regions could explain if changes attributable to the application of load are concentrated to cartilage-on-cartilage contacting regions, or if load affects the entire cartilage volume. Additionally, a layer-wise analysis of the response to compressive load may be insightful due to articular cartilage’s microstructure varying from superficial to deep. Overall, a better understanding of how healthy tibial and femoral cartilage changes in response to load may aid in the identification of an abnormal response produced by injury or disease.

Therefore, the primary aim of this study was to quantify the effect of acute compressive loading on T1ρ and T2* relaxation in articular cartilage of the femur and tibia separately. Secondary aims were to i) determine if the effects of load in the femoral regions that are in contact with the tibial regions are consistent with those in the femoral non-contact regions located posteriorly, ii) determine if the effects of load are different between the deep and superficial layers, and iii) determine if the effects of load differ across multiple visits. We hypothesize that there will be an effect of load in the femoral and tibial regions, that the effect of load on the femoral cartilage that is in contact with the tibia will be significantly larger than that of femoral cartilage in non-contacting regions, that the effect of load will depend on the layer analyzed, and that the effect of load will not differ across visits.