S88 Poster Presentations/Osteoarthritis and Cartilage 19S1 (2011) S53–S236 analysis was used. Results are reported as odds ratios (ORs) and 95% confidence intervals (CIs). Results: The 250 participants had a mean age of 64.8 (SD 10.2), BMI 28.6 (5.6), and 189 (76%) were women. 90 persons (36%) reported low knee confidence. The Table shows the relationship between each frontal plane motion measure (independent variable) and low knee confidence (dependent variable) for each regression model. Greater peak knee varus angle during the entire stance phase and during the mid and terminal stance subphases were each associated with low knee confidence (significant results are bolded). Conclusions: Higher peak knee varus angle during the stance phase of gait was associated with low knee confidence in persons with knee OA. These results should be evaluated further in longitudinal studies. Minimizing excessive knee varus motion during ambulation through neuromuscular stabilization or gait training may be beneficial in potentially improving knee confidence and, thereby, poor function outcome. Table: Relationship between knee frontal plane motion and low knee confidence, logistic regression models Measure of frontal plane motion during gait Mean (SD) Age and gender-adjusted OR per 1° (95% CI) Peak knee varus angle during stance (°) 2.20 (1.29) 1.26 (1.02, 1.57) Peak knee varus angular velocity (°/sec) 0.35 (0.18) 0.55 (0.12, 2.40) Total knee varus-valgus range of motion (°) 10.28 (3.28) 0.96 (0.89, 1.04) Peak knee varus angle during early stance (°) 1.03 (1.04) 1.26 (0.98, 1.63) Peak knee varus angle during mid stance (°) 1.76 (1.44) 1.23 (1.01, 1.49) Peak knee varus angle during terminal stance (°) 1.81 (1.16) 1.29 (1.01, 1.65) Peak knee varus angle during pre swing (°) 0.41 (1.47) 1.11 (0.92, 1.33) 176 DOES INCREASED LOADING OCCUR ON THE CONTRALATERAL SIDE IN MEDIAL KNEE OSTEOARTHRITIS AND WHAT IMPACT DO LATERAL WEDGE INSOLES HAVE ON THIS? R.K. Jones 1 , G.J. Chapman 1 , A.H. Findlow 1 , M. Parkes 2 , L. Forsythe 2 , D.T. Felson 2 . 1 Univ. of Salford, Salford, United Kingdom; 2 Univ. of Manchester, Manchester, United Kingdom Purpose: A common conservative treatment in medial knee osteoarthritis (OA) are lateral wedge insoles which can be worn long- term and have been shown to reduce the medial knee loading. In persons with medial knee OA, the contralateral knee is at a high risk of concurrent or later OA, usually also medial OA, so a reduction in medial loading might prevent disease there. Given the lack of available structure modifying treatments, strategies to prevent development or progression of OA are needed. The majority of work on lateral wedge insoles has focussed on the reduction of the external knee adduction moment (EKAM) on the affected side. To date, there have been no reports on what happens to the contralateral side. The study hypothesised in addition to a decrease on the affected side, that lateral wedge insoles would reduce medial loading on the contralateral side. The purpose of this study was to determine what effects on joint loading lateral wedge insoles have on the affected and contralateral sides during walking. Methods: As part of a randomised trial of different footwear/insoles for medial knee OA, we tested a control shoe and two lateral wedge insoles both of which have been shown to reduce the EKAM modestly in patients with medial knee OA. The affected side was defined as the more painful knee. Subjects underwent a 3-dimensional kinematic (Qualisys OQUS, Gothenburg, Sweden) and kinetic (AMTI, USA) analysis whilst walking in a control shoe and lateral wedge insoles. Walking conditions were randomised and lateral wedge insoles were worn bilaterally so the effects of lateral wedges could be assessed on both limbs. A minimum of three trials were assessed in each condition and the 1 st peak in the EKAM and the knee adduction angular impulse (KAAI) were computed for each trial for each person. The data was normally distributed and the data is presented as the mean with changes from control shoe to lateral wedge insole in the same limb were tested using a paired t test. Results: We studied 36 patients with medial knee OA (mean age 59.3 years, 47% Female, mean BMI 32.4, 54% KL grade 2 and 46% KL grade 3). Fifty five percent had a higher EKAM on their contralateral limb in comparison with the affected limb (mean difference in EKAM 3% higher and KAAI 7% higher). When using the lateral wedge insoles, there was a significant reduction in EKAM on the affected side compared with the control (mean difference in EKAM = −4.86%, t = 2.16, p = 0.038) and a significant reduction on the contralateral side compared to the control (mean difference in EKAM = 10.10%, t = 4.34, p < 0.001). The KAAI showed similar trends, with significant reductions on the affected side (mean difference in EKAM = −4.57%, t = 2.58, p = 0.01), and on the contralateral side (mean change in KAAI = −10.67%, t = 4.58, p < 0.001). Conclusions: We, like others, found modest reductions in medial knee loading in patients with medial knee OA when treated with lateral wedge insoles. We also demonstrate that medial loading can be reduced substantially on the contralateral side perhaps even more than on the affected side. Conservative management options such as lateral wedge insoles might assist in preventing knee OA on the contralateral side especially in the very early stages of the disease. This has implications for individuals with severe OA and/or post-surgery when loading on the unaffected side is likely to be higher. Future studies should identify the longer term effects of lateral wedge insoles on the contralateral side to determine their effectiveness following surgery. 177 GAIT COMPENSATIONS IN A RAT MEDIAL MENISCUS TRANSECTION MODEL OF OSTEOARTHRITIS K.D. Allen, B.A. Mata, M.A. Gabr, J.L. Huebner, V.B. Kraus, L.A. Setton. Duke Univ., Durham, NC, USA Purpose: Knee instability due to ligament or meniscus injury is associated with significant risk of developing osteoarthritis (OA). Cartilage lesions resembling those found in OA can be modeled in the rat through surgical transection of the medial collateral ligament (MCL) and medial meniscus (MM). This pre-clinical OA model has documented evidence of cartilage loss, osteophyte formation, and limb hypersensitivity. The objective of this study was to evaluate changes in gait, synovial fluid and serum biomarkers, and histopathology associated with MCL and MM transection in the rat knee. Methods: The right MCL (MCL sham) or the right MCL and MM (MMT) were transected in 12 male Lewis rats (200–250g, n = 6 per group); 4 rats that did not receive a surgery served as naïve controls. Weight bearing (incapacitance meter) and gait were evaluated on post-operative day 9, 16, and 23. Spatiotemporal characteristics of gait were evaluated through digitization of five high-speed videos of rats walking in an open gait arena (200 frames per sec). Gait dynamics were characterized through a Hall-effect force transducer (+4.9 N vertical, ±2.45 braking/propulsion, 200 Hz) connected to an isolated section of floor (5 x 10 cm). At sacrifice, serum and synovial fluid were collected via 100 microliter lavage, and cytokine activity was analyzed via a 10-plex assay (Bio-rad Luminex). Knees were then processed for histological scoring using the OARSI histopathology assessment scheme. Results: Static weight distribution was not affected by either MCL sham or MMT surgery; however during ambulation, the gait of MMT rats became increasingly asymmetric. The spatiotemporal variable gait symmetry measures the consistency of foot-strike events for a limb pair in time; when symmetry equals 0.5, this indicates that left and right foot-strike events are equally spaced in time. With the gait of MMT rats becoming progressively asymmetric (being significantly different from a symmetric gait pattern by day 23, Fig 1, left-hand panel), our data indicate that gait symmetry may reflect the progression of knee damage due to instability in rats. MMT rats also had a percentage stance time imbalance in their hind limbs. Percentage stance time imbalance is a comparison of the left and right limb stance times normalized to stride time; in balanced gait, percentage stance time imbalance should be near- zero, indicating equal time is being spent on both limbs of a limb pair. A percentage stance time of more than zero in MMT rats indicates that MMT rats tend to spend more time on their contralateral limb than the operated limb while walking (Fig. 1, right-hand panel). In general, the gait of MCL sham animals did not significantly vary from naïve controls. Ground reaction forces analysis was also able to describe operated limb compensation in rodent gait following knee instability. MMT rats had lower peak vertical force in the operated limb relative to the contralateral limb (Fig. 2). Moreover, peak vertical force in the MMT group tended to be lower than MCL shams (p = 0.062, not shown) and naïve controls (p = 0.060, not shown). In addition, MMT rats had lower peak propulsive forces relative to MCL shams (p = 0.022, not shown) and naïve controls (p = 0.014, not shown). Cytokine concentrations in synovial fluid lavage were not above the detection limit, likely due to dilution from the 100 microliter lavage collection method. Serum differences were also non-significant amongst groups.