Purpose: Patients with mild to moderate osteoarthritis (OA) have been treated with hyaluronic acid (HA) viscosupplementation for decades as it functions to increase the viscosity of diseased synovial fluid and improve cartilage lubrication. Viscosupplementation has shown mixed results with reports of no effect and reports of reduced pain up to 6 months after injections. As a result, it is unknown if viscosupplementation may benefit specific OA disease states. Measuring shear strains is a powerful metric for assessing lubrication therapies because increased strains in articular cartilage have been linked to negative cellular responses including apoptosis. The objective of this study was to investigate the effect of viscosupplementation on friction behavior and shear strains in articular cartilage between inflammatory and non-inflammatory arthritic phenotypes of human synovial fluid, and identify if friction parameters or synovial fluid composition predicted tissue strains. Methods: Synovial fluid (SF) samples were obtained with consent from the knee joints of osteoarthritic patients (n=12). Samples with a WBC count greater than 2000 cells/mm3 and a PMN composition over 25% were classified as inflammatory (n=6), and remaining samples were classified as non-inflammatory (n=6). Lubricin concentration was quantified using a peanut agglutinin sandwich ELISA with mAb 9G3, and purified lubricin from bovine SF as the standard. Cartilage explants from the femoral condyles of neonatal bovids were bisected, fluorescently stained, and mounted on a test frame on an inverted confocal microscope. While bathed in SF, explants were axially compressed 15% then sheared against glass at a sliding speed of 1 mm/s. Depth-dependent shear strains were calculated from the displacements of photobleached lines perpendicular to the cartilage surface (Fig 1A). Following testing, the SF was mixed in a 1:1 ratio with Hymovis, a modified HA viscosupplement, and the same test was performed on a new cartilage sample. Friction coefficients were calculated as the ratio of shear force divided by axial force. The ratio of time in static friction over one sliding cycle describes the amount of time the tissue is in contact with the glass plate before sliding, and was calculated by dividing the duration the tissue spent in static friction (tμ static, Fig 1B ) by the period (tcycle). Lubricin concentration, tμ static/ tcycle, and friction coefficients were compared between SF groups using a t-test. Shear strains were compared between SF groups +/- HA using a two-way repeated measures ANOVA. Results: For all SF samples +/- HA, shear strains were highest at the tissue surface and decreased with depth (Fig 2A). The addition of HA to the inflammatory group significantly reduced shear strains in the first 125μm (p<0.01), but did not decrease tissue strains in the non-inflammatory group at any depth (p>0.41 for all comparisons). There was no difference in lubricin concentration between inflammatory and non-inflammatory groups (p=0.93). However, regardless of arthritic phenotype, increased lubricin concentration correlated with decreased shear strains at the tissue surface (Fig 2B, fit to dose-response curve). Average friction coefficients were greater in inflammatory SFs compared to non-inflammatory, but were not significant (p=0.09). Ratios of time in static friction (tμ static/ tcycle) were higher in inflammatory SF compared to non-inflammatory (p<0.05). We find that the coefficient of friction is a weak predictor of strain, while the relative time in static friction is a strong predictor (Fig 2C,D). Conclusions: Our results describe distinct tribological phenotypes of inflammatory and non-inflammatory SF where strains are higher with inflammatory SF as a lubricant. Additionally, viscosupplementation has greater efficacy when combined with inflammatory SF compared to non-inflammatory. While there was no difference in lubricin levels between the arthritic SF phenotypes, lower levels of lubricin correlated with higher strains at the tissue surface for both SF groups. Higher tissue strains are linked to damaging mechanobiological consequences, and predicting whether viscosupplementation would decrease tissue strains could identify patients that would benefit the most from this treatment.Figure 2A) HA addition significantly decreased strains at surface for inflammatory group (p<0.01), but had no effect on non-inflammatory samples (p>0.41). Inset shows surface strains. B) Increased lubricin concentration correlates with decreased surface shear strains (dose-response fit). C) Relative time in static friction had a strong positive correlation with surface shear strains, while D) coefficient of friction had a weak correlation.View Large Image Figure ViewerDownload Hi-res image Download (PPT)