The purpose of this proposed study is to investigate both the acute and chronic response of frontal plane knee moment after gait retraining and to assess the effects on the biomechanics of the contralateral side. Fifty-one patients diagnosed with tibiofemoral joint osteoarthritis (TFJ OA) will be recruited to participate in the proposed study. Participants will complete baseline trials to assess gait kinematic and kinetic parameters. Following baseline, each participant will perform six conditions of the foot progression gait or three conditions of the trunk lean gait modifications to determine which strategy is most effective in reducing frontal plane knee moment. Participants will then be randomized to either the control or experimental group based on their identified preferred strategy.

Participants will complete eight gait retraining sessions using patient-specific gait modifications (tailored foot progression and tailored lateral trunk lean), or normal gait (control) during the training period. A fading feedback design will be implemented. Real-time haptic biofeedback will be provided on every step during the first two weeks and reduced by 25% every subsequent two weeks. No feedback will be provided during baseline and at testing sessions. Measures of pain and function will also be collected at all testing sessions. Variables of interest include ankle, knee, hip sagittal and frontal plane moments. In addition, sagittal and frontal plane impulse will also be assessed.

Descriptive statistics will be calculated for foot progression angle, trunk lean angle, frontal and sagittal plane hip, knee, and ankle angles as well as moments. Descriptive statistics for frontal plane knee impulse will also be calculated. A multivariate analysis of variance (MANOVA) will be conducted to compare frontal plane knee moment, frontal plane knee impulse and the absolute sagittal plane moment will be compared across three groups at four different time points. Repeated measures analysis of variance (ANOVA) will be conducted to compare both sagittal and frontal plane joint biomechanics for the contralateral limb. The p-value will be set at .05.

A randomized controlled study design with a waitlist will be used to investigate the effects of a 10-week gait retraining intervention on the knee adduction moment. Participants will be randomly assigned to either the intervention group or the control group. The intervention group will be further split into two groups. One group will consist of participants who reduced knee adduction moment (KAM) most with altered foot progression gait and the other will consist of participants who reduce KAM most effectively with trunk lean gait during pre-assessment. Participants will then perform 8 gait retraining sessions over 8 weeks (one session per week) using either their specific gait modification strategy (intervention) or with their normal gait (control). As part of the waitlist design, at the end of 10 weeks, patients assigned to the control group will be reassigned to their previously determined patient specific gait modification intervention. The goal is to minimize attrition and to increase the effective sample size.

On arrival to the laboratory, participants will be required to read and sign an informed consent form. Participant height and mass will be recorded. Pain and function will be assessed using the Western Ontario and McMaster Universities Arthritis (WOMAC) questionnaire and a numeric rating scale (NRS) from zero to 10, where zero represents no pain and 10 the highest level of pain. The experimental limb for the purpose of this study is defined as the leg diagnosed with symptomatic TFJ OA or the most symptomatic limb in the case of bilateral OA. Participants will be equipped with 4 surface electromyogram (sEMG) placed on the rectus femoris, vastus medialis, bicep femoris and semitendinosus of the experimental limb. EMG sensors will be used to record muscle activity during the baseline test. Participants will then have 53 retroreflective markers attached to their trunk and lower extremities. A static calibration trial of the VICON motion capture system will be acquired by having the participants stand on the foremost force plate with both feet aligned with the anterior-posterior axis of the laboratory. Participants will also perform a dynamic calibration by completing three clockwise rotations of the hip which will be used to estimate the hip joint centers. Participants will then complete 5 baseline walking trials along with a 6-meter walkway at their preferred speed. Participants will be required to walk for 12 minutes on a treadmill, and additional data will be recorded. Participants will then be required perform 9 conditions with altered foot progression gait or trunk lean gait to determine which strategy is most effective in reducing KAM. Participants will first receive standardized verbal instructions on how to achieve the instructed gait modification. Participants will then be provided haptic real-time biofeedback to ensure that they successfully achieve the required magnitudes of gait modification. Kinematic data collected in Vicon will be streamed to Matlab (Mathworks, Natick, MA) for real-time computation of joint angles. Participants will receive feedback from tactile sensors attached either to the lateral-proximal aspect of the fibula (foot progression) or the center of the scapula (lateral trunk lean) of the experimental side with hypoallergenic double-sided tape. One vibration will indicate a required decrease in the target gait parameter, while two vibrations will indicate a required increase. Feedback will be provided on each step and no vibration will indicate that no correction is needed. A trial will only be considered valid if the participant fully contacted the force plates twice with the foot of the experimental limb and the modified parameter was in the prescribed target range. Additionally, participants will be required to maintain an average gait speed ±5% relative to baseline for trials to be considered successful. The modification to be completed in the baseline session will be randomized for each participant. Once these trials are completed, participants will return to the lab within the same week to complete 5 additional baseline trials and three trials of whichever gait modification they did not perform the first day. Each baseline data collection session will take approximately one hour.

Participants will then be randomly assigned to either the intervention group or the control group. The intervention group will be further split into two groups. One group will consist of participants who reduced KAM most with altered foot progression gait and the other will consist of participants who reduce KAM most effectively with trunk lean gait. Participants will then perform 8 gait retraining sessions over 8 weeks (one session per week) using either their specific gait modification strategy (intervention) or with their normal gait (control). During gait-retraining sessions, participants will walk on a Woodway Desmo treadmill (Woodway, Waukesha, WI) placed in the center of a calibrated volume area (approximately 0.5 x 1.5 meters). A three-camera high-speed motion analysis system (Vicon, Oxford, England) sampling at 200 Hz will be used to record gait kinematics. The indicated anatomical landmarks (either C7 and T10 or posterior calcaneus and 2nd metatarsal phalangeal joint) will be marked with an ultraviolet pen, which will allow for visibility for a week, and will be reapplied at subsequent visits. This will improve marker placement repeatability during the gait retraining phase. A five-minute dynamic warm-up will be provided prior to the commencement of each gait retraining session; participants will then walk with their individualized gait modification strategy for 20 minutes. Participants will be provided with haptic feedback in the same manner outlined during the individualization phase or will continue to walk without feedback. A fading feedback design will be implemented across sessions to gradually integrate task acquisition and transfer and to help facilitate the internalization of the learned skill. During the first 2 weeks, real-time biofeedback (RTB) will be delivered on every step. For the third and fourth week, RTB will be provided on the first 3 foot strikes by the experimental leg and withheld on the fourth indicating a 25% reduction. During the fifth and sixth week, feedback will be provided on alternating foot strikes reducing RTB frequency delivery to 50%. For the final two weeks of gait-retraining, no feedback will be provided on the first three steps but delivered on the fourth reducing RTB frequency to 25%.

Between gait retraining sessions, subjects will be instructed to practice their acquired gait strategy on their own outside of the laboratory session, which will occur in the absence of feedback. They will be instructed to practice at least 10 min per day and will be provided weekly activity logs to record time of day and amount practiced each day during the eight weeks of gait retraining. Practice logs will be submitted weekly. Over-ground gait analysis and 12 minutes of treadmill walking will be performed at week 4 (first post-test) and 9 (2nd post-test) over the course of the intervention to track skill acquisition. This testing will be the same as the baseline trials, however, participants will be instructed to walk using only their specific gait modification strategy. No feedback will be provided during any of the skill acquisition or retention tests. Follow-up testing will occur at one, three, and six months as well as one-year post-intervention to measure retention of prescribed gait modifications.

Inclusion Criteria:

• a clinical diagnosis of knee osteoarthritis by a qualified health professional such as an orthopedic surgeon or physical therapist
• between the ages of 18 and 80
• able to walk unaided for a minimum of 20 minutes

Exclusion Criteria:

• body mass index greater than 35
• history of lower back, hip or, knee surgery within the last 2 years
• knee arthroscopy or pharmacological injection in the previous six months
• neurological, or musculoskeletal conditions affecting ambulation
• cognitive impairment that would inhibit motor learning
• use of gait aid, orthotic shoe inserts, or hinged knee brace