EXPLORING THE RELATIONSHIP BETWEEN LIGAMENT MICROSTRUCTURE AND MECHANICS IN OA-AFFECTED HUMAN KNEES

Abstract

INTRODUCTION

Knee ligaments play a critical role in providing joint stability and limiting excessive motion. Altered joint loading due to OA can affect various knee tissues, including the ligaments. Previous studies using post-traumatic OA animal models have reported changes in affected knee’s ligament viscoelasticity. This study investigates the microstructural characteristics of OA-affected human knee ligaments and examines whether these features are related to their viscoelastic properties.

OBJECTIVE

This study examines whether clustered bundle thickness and the proportion of non-collagenous volume to total ligament volume differs among ligaments in OA-affected knees. It also examines if the mechanical properties of the ligaments are dependent on the bundle thickness and proportion of non-collagenous volume.

METHODS

Anterior (ACL; n = 6) and posterior (PCL; n = 7) cruciate ligaments, and medial (MCL; n = 8) and lateral (LCL; n = 7) collateral ligaments were collected from eight fresh-frozen cadaveric knees (five female; age: 65 ± 8 years). All knees had histology-confirmed osteoarthritis (average OARSI grade of tibial cartilage samples: >2). Following preconditioning, samples underwent a mechanical testing protocol that included a two-step stress relaxation (to 4% and 8% strain, 30 min each) and cyclic loading up to 5.0 Hz with ±0.5% strain amplitude and 20 cycles per frequency. Equilibrium modulus was derived from the stress-relaxation data, while dynamic modulus and phase difference were calculated from cyclic loading. After mechanical testing, samples were stored in formalin and underwent gradual dehydration in ethanol and critical point drying. Subsequently, they were imaged by an Xradia 610 Versa X-ray microscopy (XRM, with 4x objective, 40kV voltage, 2s exposure, 10µm voxel size, and binning of 4). The reconstructed XRM images of ligaments were visualized in CTVox, and collagen bundle thickness and non-collagenous volume (by open and close porosity analyses) were calculated in CTAn software. Pearson correlation analysis between collagen bundle thickness and non-collagenous volume and mechanical properties was performed in R 4.2.2.

RESULTS

Among the four ligaments, the LCL exhibited the highest bundle thickness, while the PCL showed the highest non-collagenous volume ratio; however, these differences were not statistically significant. Equilibrium modulus was negatively correlated with bundle thickness across all ligaments, and with non-collagenous volume in all but the MCL. The phase difference at 5 Hz in the PCL showed a strong positive correlation with bundle thickness (r = 0.82). The ACL displayed a strong negative correlation between dynamic modulus at 5 Hz and non-collagenous volume (r = –0.82).

CONCLUSION

Although collagen is the primary load-bearing component of ligaments, increased bundle thickness did not correlate with either equilibrium modulus or dynamic modulus at 5 Hz. In contrast, greater bundle thickness was associated with a higher phase difference, indicating increased viscosity, possibly due to increased number of fibers and interaction surface. Additionally, higher non-collagenous volume in the ACL was associated with reduced stiffness.