Sensory nervous system impact on subchondral bone pathology in a murine OA model

Abstract

Purpose: There is strong evidence to suggest altered knee biomechanics and aberrant joint tissue biology are associated with the development and progression of knee osteoarthritis (OA), however our understanding of the link between mechanics and biology in humans with knee pathology is lacking. Human biomechanics studies have shown a link of medial knee OA to knee varus malalignment and increased medial knee compartment loading. Furthermore, changes in bone structure and inflammation are thought to be responsible for long term joint deterioration. We hypothesized that joint inflammation and altered bone remodeling in the degenerative knee are a consequence of medial knee overloading. Therefore, this study aimed to determine if indicators of dynamic frontal plane knee loading or knee varus alignment are associated with synovial fluid biomarkers of joint inflammation, altered bone turnover or bone mechanobiology in subjects with medial knee degeneration. Methods: Subjects with medial knee OA (KL grade II-IV) undergoing high tibial osteotomy surgery or isolated medial knee ICRS grade III focal cartilage defects undergoing microfracture surgery had synovial fluid (SF) aspirated from the affected knee and underwent 3D motion capture analysis within a 4-week period. Kinematic marker data and ground reaction forces from level barefoot walking at a self-selected pace were measured using a 12 infra-red camera (Qualisys) and 6 force plate (Bertec) set up, with a modified Helen Hayes marker-set. A six-degrees-of-freedom musculoskeletal model with 8 segments was built (Visual 3D) to calculate early-and late-stance peak knee adduction moments (1 st peak and 2 nd peak KAMs, respectively) representative of peak medial knee compartmental loading, knee adduction angles during KAM peaks (1 st peak and 2 nd peak KAAs) representative of dynamic joint alignment during peak loads, and the knee adduction angular impulse (KAAI) representative of cumulative medial knee loading over stance-phase. SF aspirated from the affected knee was centrifuged at 5000G for 15m to remove cells and stored at -80 C. Multiplex chemiluminescence (Mesoscale Discovery) or enzyme-linked immunosorbent assays quantified the concentrations of 10 SF molecules relating to inflammation (proinflammatory: TNF-a, IL-6, IL-8; anti-inflammatory: IL-10), bone remodeling (formation: ALP, OPG; resorption: CTX-I, RANKL) and bone mechanobiology (glutamate, sclerostin). The RANKL:OPG ratio was also calculated as an indication of osteoclast activity. Normality of biomarker data was tested using Shapiro Wilks Test. Non-normal biomarker data was log-transformed prior to regression to satisfy assumptions. Multiple regression was applied to predict biomarker levels from biomechanical parameters, whilst controlling for age and BMI as covariates.Results: Biomechanical and biological data was collected from 10 medial knee pathology subjects (all male; mean (SD) age ¼ 50.3 (6.2) years, BMI ¼ 28.7 (3.9) kg/m 2 ). For the effect of knee mechanics on inflammatory markers, 2 nd peak KAMs ...