Influence of sex on tibiofemoral cartilage response to running in young healthy men and women

Abstract

Purpose: Women are more likely to experience greater degeneration and more severe symptoms related to osteoarthritis (OA) than men. It is unknown whether sex influences the acute response of cartilage in the knee to loading. Identifying sex-differences, if present, may improve our understanding of the differences in OA experienced between men and women at a tissue level. The purpose of this study was to determine if there is an influence of sex on loading-induced changes in regional tibiofemoral cartilage thickness as measured by magnetic resonance imaging (MRI) following running in young men and women. Methods: Young healthy men (N=15, 25.6±4.1 y, 23.7±2.7 kg/m2) and women (N=15, 26.1±3.4 y, 23.1±3.1 kg/m2) were recruited. Participants were excluded if they were unfit to participate in physical activity as captured by the Get Active! Questionnaire, scored below 74 on the Lower-Extremity Functional Scale, self-reported lower-limb injuries in the last 3 months, had orthopaedic surgery in the limb of interest (right), or had contraindications to MRI. Participants were asked to attend one MRI imaging session in the morning. Participants then rested supine for 30 minutes to mitigate effects of morning loading activities (i.e., walking) on the cartilage. Pre-activity scans were acquired with a 3.0T MR750 Discovery MRI scanner. Specifically, sagittal 3D fat-saturated fast-spoiled gradient recalled (FSPGR) scans were obtained to clearly delineate articular cartilage from surrounding tissues (FOV: 160 mm, Matrix: 512 x 512, In-Plane Resolution: 0.3125 x 0.3125 mm, Slice Spacing: 1 mm, TR: 17.388 ms, TE: 5.832 ms, Frequency: 127.78 MHz, Flip Angle: 18 degrees). Following pre-activity scans, participants ran on a treadmill adjacent to the MRI scanning room for 15 minutes. Then, participants immediately completed post-activity imaging (FSPGR). Tibiofemoral cartilage were segmented from pre- and post-activity FSPGR scans using a convolutional neural network (NeuralSeg). Medial and lateral tibial, and medial and lateral weight-bearing femoral cartilage were extracted and analyzed separately. Thickness maps were created by finding the minimum Euclidean distance from each bone-cartilage interface voxel to the articular surface. These maps were then flattened and warped to the same dimensions for further analyses. Pixelwise statistical parametric mapping using F-statistics was completed to examine the influence of sex on cartilage response to exercise. First, maps were smoothed using a Gaussian kernel filter with a full-width half maximum of 4 mm. Next, pixelwise F-statistic maps were generated for the factors of Sex, Exercise, and the interaction of Sex and Exercise (Sex#Exercise). F-maps were thresholded at a corresponding alpha level of p=0.01. Finally, the significance of thresholded clusters was determined using Gaussian random field (GRF) theory at an alpha level of p=0.001. Results: A minimum cluster size of 20 pixels was determined as a significant change. For the factor of Sex, significant clusters were identified for all four tissues, indicating that female sex was associated with overall thinner cartilage. These clusters of thinner cartilage in women in the medial and lateral tibia were located centrally, aligning approximately with the location of articulation of the femur with the tibia (Fig 1). Similarly, significant medial and lateral weight-bearing femoral clusters were located at the approximate location of the articulation with the tibia. The effect of exercise caused significant clusters for the lateral tibia only, suggesting cartilage thinning following exercise. This cluster was located centrally. The interaction Sex#Exercise did not contain significant clusters. Conclusions: This study compared cartilage thickness changes in young healthy men and women. Consistent with previous literature, women had thinner cartilage; the clustering analysis emphasizes that this thinner cartilage appears in the weight-bearing regions of the knee among women. Running produced systematic deformation patterns in cartilage of the lateral tibia. Finally, no interaction of sex and exercise was found, suggesting that sex does not cause distinct cartilage deformation patterns in response to running. Future work should aim to include other factors that may affect cartilage response to loading, such as age or disease.