Background Platelet-rich plasma (PRP) has a still conflicting efficacy for knee osteoarthritis (KOA) and might be a minimally invasive and safe treatment alternative. The potential benefit of only plasma (non-enriched) has never been investigated. Our aim was to evaluate the efficacy of intra-articular platelet-rich plasma (PRP) and plasma to improve pain and function in participants with KOA over 24 weeks. Methods Randomized, double-blind, placebo-controlled trial with 3 groups (n = 62): PRP (n = 20), plasma (n = 21) and saline (n = 21). Two ultrasound-guided knee injections were performed with a 2-week interval. The primary outcome was visual analog scale 0-10 cm (VAS) for overall pain at week 24, with intermediate assessments at weeks 6 and 12. Main secondary outcomes were: KOOS, OMERACT-OARSI criteria and TUGT. Results At baseline, 92% of participants were female, with a mean age of 65 years, mean BMI of 28.0 Kg/m2and mean VAS pain of 6.2 cm. Change in pain from baseline at week 24 were -2.9 (SD 2.5), -2.4 (SD 2.5) and -3.5 cm (SD 3.3) for PRP, plasma and saline, respectively (p intergroup = 0.499). There were no differences between the three groups at weeks 6 and 12. Similarly, there were no differences between groups regarding secondary outcomes. The PRP group showed higher frequency of adverse events (65% versus 24% and 33% for plasma and saline, respectively, p = 0.02), mostly mild transitory increase in pain. Conclusions PRP and plasma were not superior to placebo for pain and function improvement in KOA over 24 weeks. The PRP group had a higher frequency of mild transitory increase in pain. Trial registration ClinicalTrials.gov, NCT03138317, 03/05/2017.

Background: Platelet-rich plasma (PRP) has a still conflicting efficacy for knee osteoarthritis (KOA) and might be a minimally invasive and safe treatment alternative. The potential benefit of only plasma (non-enriched) has never been investigated. Our aim was to evaluate the efficacy of intra-articular platelet-rich plasma (PRP) and plasma to improve pain and function in participants with KOA over 24 weeks. Methods: Randomized, double-blind, placebo-controlled trial with 3 groups (n = 62): PRP (n = 20), plasma (n = 21) and saline (n = 21). Two ultrasound-guided knee injections were performed with a 2-week interval. The primary outcome was visual analog scale 0-10 cm (VAS) for overall pain at week 24, with intermediate assessments at weeks 6 and 12. Main secondary outcomes were: KOOS, OMERACT-OARSI criteria and TUGT. Results: At baseline, 92% of participants were female, with a mean age of 65 years, mean BMI of 28.0 Kg/m2and mean VAS pain of 6.2 cm. Change in pain from baseline at week 24 were -2.9 (SD 2.5), -2.4 (SD 2.5) and -3.5 cm (SD 3.3) for PRP, plasma and saline, respectively (p intergroup = 0.499). There were no differences between the three groups at weeks 6 and 12. Similarly, there were no differences between groups regarding secondary outcomes. The PRP group showed higher frequency of adverse events (65% versus 24% and 33% for plasma and saline, respectively, p = 0.02), mostly mild transitory increase in pain.Conclusions: PRP and plasma were not superior to placebo for pain and function improvement in KOA over 24 weeks. The PRP group had a higher frequency of mild transitory increase in pain.Trial registration: ClinicalTrials.gov, NCT03138317, 03/05/2017.

onset and progression (ACLR-OA develops more rapidly than ACLT-OA). Both ACL injury models induced specific meniscal molecular changes with a greater increase in the medial meniscus compared to the lateral. These changes may be predictive of long term meniscal pathology and have the potential to contribute to the development of medial tibiofemoral OA though reduced mechanical function and increased production of degradative enzymes. 692SYNOVIAL FLUID MICRORHEOLOGY TO IDENTIFY POTENTIAL CANDIDATES FOR VISCOSUPPLEMENTATION R. Yin, J. Su, M. Colville, M. Paszek, H. Reesink. Cornell Univ., Ithaca, NY, USAPurpose: Hyaluronic acid (HA) is a large, hydrophilic glycosaminoglycan that functions as the primary viscous lubricant in synovial fluid. Studies suggest that HA can be decreased, increased, or unchanged in concentration and/or molecular weight (MW) distribution in osteoarthritis (OA). Although HA viscosupplementation can be used to treat OA-related pain, it is difficult to predict which patients will respond best to HA supplementation. One potential explanation for the variable clinical response to HA is the presence of distinct OA phenotypes, some of which may benefit from HA supplementation whereas others may not. Here, we first asked whether HA concentration and MW distribution were altered in naturally occurring equine carpal osteoarthritis as a model for naturally occurring OA in humans. Multiple particle-tracking microrheology (MPTM) was performed to determine whether viscosity correlated with HA concentration and/or molecular weight. Methods: All protocols were approved by the university IACUC. Synovial fluid sampling: Synovial fluid samples were obtained from healthy research horses and clinical cases with carpal OA (n¼87; 25 healthy and 62 OA joints). HA ELISA: Synovial fluid HA concentrations were quantified for all samples using both a sandwich ELISA (HA DuoSet ELISA, R&D Systems) and a competitive ELISA (HA ELISA, Echelon Biosciences). HA agarose gel electrophoresis: PBS-diluted and proteinase K-digested SF samples were run on 0.5% agarose gels, stained with 0.005% Stains-All and de-stained with 10% ethanol. Densitometric scans of stained gels were used to detect relative HA molecular weight distribution into four categories: >6.1MDa, 6.1-3.1MDa, 3.1-1.5MDa and <1.5MDa. Microrheology: Synovial fluid viscosity was measured using MPTM (n¼24; 12 healthy and 12 OA joints). 15ml of synovial fluid was loaded with 0.5 nm yellow-green fluorescent beads and imaged on a custom epifluorescence microscope with a high-speed EMCCD detector. Statistical analysis: Data was log-transformation to achieve normality, and t-tests were used to compare healthy and OA groups. Pearson productmoment correlation coefficients were calculated. Results: Full cohort (n¼87): Synovial fluid HA concentrations varied substantially between individual animals (range: 0.04 to 0.9 mg/mL, R&D Systems; 0.14 to 2.74 mg/mL, Echelon Biosciences), but did not differ between healthy and OA joints (median: 0.35 and 0.29 mg/mL, R&D Systems; 0.70 and ...