Association Between Knee Joint Muscle Activation and Knee Joint Moment Patterns During Walking in Moderate Medial Compartment Knee Osteoarthritis: Implications for Secondary Prevention.

OBJECTIVE: To determine associations between knee moment features linked to osteoarthritis (OA) progression, gait muscle activation patterns, and strength. DESIGN: Cross-sectional secondary analysis. SETTING: Gait laboratory. PARTICIPANTS: Convenience sample of 54 patients with moderate, medial knee OA (N=54). INTERVENTIONS: None. MAIN OUTCOME MEASURES: Knee moments and quadriceps and hamstrings activation were examined during walking. Knee extensor and flexor strength were measured. Waveform patterns were extracted using principal component analysis. Each measured waveform was scored against principal components (PCs) that captured overall magnitude (PC1) and early to midstance difference (PC2) features, with higher PC2 scores interpreted as greater moment differential and more prolonged muscle activity. Correlations were calculated between moment PC scores and muscle PC and strength scores. Regression analyses determined moment PC score variance explained by muscle PC scores and strength. RESULTS: All correlations for knee adduction moment difference feature (KAMPC2) and prolonged muscle activity (PC2) were significant (r=-0.40 to -0.54). Knee flexion moment difference feature (KFMPC2) was significantly correlated with all quadriceps and medial hamstrings PC2 scores (r=-0.47 to -0.61) and medial hamstrings magnitude feature (PC1) (r=-0.52). KAMPC2 was significantly correlated with knee flexor strength (r=0.43), and KFMPC2 was significantly correlated with knee extensor (r=0.60) and flexor (r=0.55) strength. Regression models including muscle PC2 scores and knee flexor strength explained 46% of KAMPC2 variance, whereas muscle PC2 scores and knee extensor strength explained 59% of KFMPC2 variance. CONCLUSIONS: Muscle activation patterns and strength explained significant variance in moment difference features, highest for the knee flexion moment. This supports that exercises such as neuromuscular training, focused on appropriate muscle activation patterns, and strengthening have the potential to alter dynamic loading gait patterns associated with knee OA clinical progression.

Quantifying Achievable Levels of Improvement in Knee Joint Biomechanics During Gait After Total Knee Arthroplasty Relative to Osteoarthritis Severity.

Total knee arthroplasty (TKA) surgery improves knee joint kinematics and kinetics during gait for most patients, but a lack of evidence exists for the level and incidence of improvement that is achieved. The objective of this study was to quantify patient-specific improvements in knee biomechanics relative to osteoarthritis (OA) severity levels. Seventy-two patients underwent 3-dimensional (3D) gait analysis before and 1 year after TKA surgery, as well as 72 asymptomatic adults and 72 with moderate knee OA. A combination of principal component analysis and discriminant analyses were used to categorize knee joint biomechanics for patients before and after surgery relative to asymptomatic, moderate, and severe OA. Post-TKA, 63% were categorized with knee biomechanics consistent with moderate OA, 29% with severe OA, and 8% asymptomatic. The magnitude and pattern of the knee adduction moment and angle (frontal plane features) were the most significant contributors in discriminating between pre-TKA and post-TKA knee biomechanics. Standard of care TKA improves knee biomechanics during gait to levels most consistent with moderate knee OA and predominately targets frontal plane features. These results provide evidence for the level of improvement in knee biomechanics that can be expected following surgery and highlight the biomechanics most targeted by surgery.

Differences in Baseline Joint Moments and Muscle Activation Patterns Associated With Knee Osteoarthritis Progression When Defined Using a Clinical Versus a Structural Outcome.

Both structural and clinical changes can signify knee osteoarthritis progression; however, these changes are not always concurrent. A better understanding of mechanical factors associated with progression and whether they differ for structural versus clinical outcomes could lead to improved conservative management. This study examined baseline gait differences between progression and no progression groups defined at an average of 7-year follow-up using 2 different outcomes indicative of knee osteoarthritis progression: radiographic medial joint space narrowing and total knee arthroplasty. Of 49 individuals with knee osteoarthritis who underwent baseline gait analysis, 32 progressed and 17 did not progress using the radiographic outcome, while 13 progressed and 36 did not progress using the arthroplasty outcome. Key knee moment and electromyography waveform features were extracted using principal component analysis, and confidence intervals were used to examine between-group differences in these metrics. Those who progressed using the arthroplasty outcome had prolonged rectus femoris and lateral hamstrings muscle activation compared with the no arthroplasty group. Those with radiographic progression had greater mid-stance internal knee rotation moments compared with the no radiographic progression group. These results provide preliminary evidence for the role of prolonged muscle activation in total knee arthroplasty, while radiographic changes may be related to loading magnitude.

Torque and Muscle Activation Impairment Along With Insulin Resistance Are Associated With Falls in Women With Fibromyalgia.

Góes, SM, Stefanello, JMF, Homann, D, Lodovico, A, Hubley-Kozey, CL, and Rodacki, ALF. Torque and muscle activation impairment along with insulin resistance are associated with falls in women with fibromyalgia. J Strength Cond Res 30(11): 3155-3164, 2016-Fibromyalgia (FM) is a chronic pain condition associated with reduced muscle strength, which can lead to functional incapacity and higher risk of falls. The purpose of the study was to compare maximal ankle joint torque, muscle activation, and metabolic changes between women with and without FM. In addition, the relationship between those aspects and retrospectively reported falls in women with FM was determined. Twenty-nine middle-aged women with FM and 30 controls were recruited. Fall history, pain intensity, and pain threshold were assessed. Plasma glucose levels and insulin resistance (IR) were determined. Peak torque and rate of torque development (RTD) were calculated, and muscle activation was assessed from maximum isometric voluntary ankle dorsiflexion and plantar flexion contractions. In addition, voluntary muscle activation failure of the anterior tibialis muscle during maximal dorsiflexion was calculated. When compared to controls, women with FM reported higher number of retrospectively reported falls, exhibited higher IR, showed reduced plantar flexion and dorsiflexion RTD, had lower plantar flexion peak torque, and demonstrated more antagonist coactivation and higher muscle activation failure (p ≤ 0.05). Higher muscle activation failure was explained by glucose level and pain intensity (adj R = 0.28; p ≤ 0.05). Reduced plantar flexion and dorsiflexion peak torque explained 80% of retrospectively reported falls variance; also, high antagonist coactivation (odds ratio [OR] = 1.6; p ≤ 0.05) and high IR (OR = 1.8; p ≤ 0.05) increased the chance of falls in the FM group. A combination of metabolic factors and muscle function increased the odds of retrospectively reporting a fall in FM. Both aspects may be considered in interventions designed for reducing falls in this population.

Knee joint biomechanics and neuromuscular control during gait before and after total knee arthroplasty are sex-specific.

The future of total knee arthroplasty (TKA) surgery will involve planning that incorporates more patient-specific characteristics. Despite known biological, morphological, and functional differences between men and women, there has been little investigation into knee joint biomechanical and neuromuscular differences between men and women with osteoarthritis, and none that have examined sex-specific biomechanical and neuromuscular responses to TKA surgery. The objective of this study was to examine sex-associated differences in knee kinematics, kinetics and neuromuscular patterns during gait before and after TKA. Fifty-two patients with end-stage knee OA (28 women, 24 men) underwent gait and neuromuscular analysis within the week prior to and one year after surgery. A number of sex-specific differences were identified which suggest a different manifestation of end-stage knee OA between the sexes.

Temporal patterns of the trunk muscles remain altered in a low back-injured population despite subjective reports of recovery.

OBJECTIVE: To compare temporal activation patterns from 24 abdominal and lumbar muscles between healthy subjects and those who reported recovery from recent low back injury (LBI). DESIGN: Cross-sectional comparative study. SETTING: University neuromuscular function laboratory. PARTICIPANTS: Healthy adult volunteers (N=81; 30 LBI, 51 asymptomatic subjects). INTERVENTIONS: Trunk muscle electromyographic activity was collected during 2 difficulty levels of a supine trunk stability test aimed at challenging lumbopelvic control. MAIN OUTCOME MEASURES: Principal component (PC) analysis was applied to determine differences in temporal and/or amplitude electromyographic patterns between groups. Mixed-model analyses of variance were performed on PC scores that explained more than 89% of the variance (α=.05). RESULTS: Four PCs explained 89% and 96% of the variance for the abdominal and back muscles, respectively, with both muscle groups having similar shapes in the first 3 PCs. Significant interactions or group main effects were found for all PC scores except PC4 for the back extensors. Overall activation amplitudes for both the abdominal and back muscles (PC1 scores) were significantly (P<.05) higher for the LBI group, with both abdominal and back muscles of the LBI group demonstrating an increased response to the leg-loading phase (PC2 scores) compared with the asymptomatic group. Differences were also found between groups in their preparatory activity (PC3 scores), with the LBI group having a higher early relative amplitude of abdominal and back extensor activity. CONCLUSIONS: Despite perceived readiness to return to work and low pain scores, muscle activation patterns remained altered in this LBI group, including reduced synergistic coactivation and increased overall amplitudes as well as greater relative amplitude differences during specific phases of the movement. Electromyographic measures provide objective information to help guide therapy and may assist with determining the level of healing and return-to-work readiness after an LBI.

Body mass index affects knee joint mechanics during gait differently with and without moderate knee osteoarthritis.

OBJECTIVE: Obesity is a highly cited risk factor for knee osteoarthritis (OA), but its role in knee OA pathogenesis and progression is not as clear. Excess weight may contribute to an increased mechanical burden and altered dynamic movement and loading patterns at the knee. The objective of this study was to examine the interacting role of moderate knee OA disease presence and obesity on knee joint mechanics during gait. METHODS: Gait analysis was performed on 104 asymptomatic and 140 individuals with moderate knee OA. Each subject group was divided into three body mass categories based on body mass index (BMI): healthy weight (BMI<25), overweight (25≤BMI≤30), and obese (BMI>30). Three-dimensional knee joint angles and net external knee joint moments were calculated and waveform principal component analysis (PCA) was applied to extract major patterns of variability from each. PC scores for major patterns were compared between groups using a two-factor ANOVA. RESULTS: Significant BMI main effects were found in the pattern of the knee adduction moment, the knee flexion moment, and the knee rotation moment during gait. Two interaction effects between moderate OA disease presence and BMI were also found that described different changes in the knee flexion moment and the knee flexion angle with increased BMI with and without knee OA. CONCLUSION: Our results suggest that increased BMI is associated with different changes in biomechanical patterns of the knee joint during gait depending on the presence of moderate knee OA.

Is there a dose response for valgus unloader brace usage on knee pain, function, and muscle strength?

OBJECTIVE: To examine whether there was a dose response for valgus unloader brace wear on knee pain, function, and muscle strength in participants with medial compartment knee osteoarthritis. DESIGN: In this single-group study, participants with medial compartment knee osteoarthritis were followed for approximately 6 months. SETTING: Recruitment was conducted in the general community, and testing was performed at a university laboratory. PARTICIPANTS: A convenience sample of patients (N=32) who were prescribed a valgus unloader brace agreed to participate, met the inclusion criteria, and completed the baseline data collection. Twenty-four participants (20 men, 4 women) completed baseline and follow-up collections. INTERVENTION: Participants wore their valgus unloader brace as needed. MAIN OUTCOME MEASURES: Knee extensor, flexor, and plantar flexor strength was tested at baseline and follow-up. Participants filled out Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and Medical Outcomes Study 36-Item Short-Form Health Survey questionnaires to assess pain and function. Self-selected walking velocity and stride length were objective measures of function. Brace usage (dose) and activity (step count) were recorded at least 4 days/week for the study duration. RESULTS: Positive relationships existed between brace wear usage and percent change in step count (r=.59, P=.006) and percent change in hamstrings strength (r=.37, P=.072). At follow-up, there was significant improvement in hamstrings strength (P=.013), and trends toward improvements in WOMAC pain (P=.059) and WOMAC function (P=.089). CONCLUSIONS: Our results indicate that greater brace use may positively affect physical activity level, but there was minimal effect of brace wear dosage on lower-limb muscle strength. Only knee flexion showed a positive relationship. Our finding of no decreased muscle strength indicates that increased brace use over a 6-month period does not result in muscle impairment.

Effect of a high intensity quadriceps fatigue protocol on knee joint mechanics and muscle activation during gait in young adults.

The purpose of this study was to determine the effect of impaired quadriceps function on knee joint biomechanics and neuromuscular function during gait. Surface electromyograms, three-dimensional motion and ground reaction forces were collected during gait before and after 20 healthy adults completed a high intensity quadriceps fatigue protocol. Pattern recognition techniques were utilized to examine changes in amplitude and temporal characteristics of all gait variables. The fatigue protocol resulted in decreased knee extensor torque generation and quadriceps median power frequencies for 18 of 20 participants (p < 0.05). The gait data from these 18 participants was analyzed. The knee external rotation angle increased (p < 0.05), the net external flexion and external rotation moments decreased (p < 0.05), and the net external adduction moment increased (p < 0.05). Post-fatigue changes in periarticular muscle activation patterns were consistent with the biomechanical changes, but were not significantly altered. Even for this low demand task of walking the knee motion and loading characteristics were altered following a high intensity fatigue protocol in a manner that may place the knee joint at greater risk for joint pathology and injury.

The effect of total knee arthroplasty on knee joint kinematics and kinetics during gait.

This study determined how total knee arthroplasty (TKA) altered knee motion and loading during gait. Three-dimensional kinematic and kinetic gait patterns of 42 patients with severe knee osteoarthritis were collected 1 week prior and 1-year post-TKA. Principal component analysis extracted major patterns of variability in the gait waveforms. Overall and midstance knee adduction moment magnitude decreased. Overall knee flexion angle magnitude increased due to an increase during swing. Increases in the early stance knee flexion moment and late stance knee extension moment were found, indicating improved impact attenuation and function. A decrease in the early stance knee external rotation moment indicated alteration in the typical rotation mechanism. Most changes moved toward an asymptomatic pattern and would be considered improvements in motion, function, and loading.

Association of Low Physical Activity Levels With Gait Patterns Considered at Risk for Clinical Knee Osteoarthritis Progression.

OBJECTIVE: Although gait analysis provides an estimate of joint loading magnitude and patterns during a typical step, accelerometry provides information about loading frequency. Understanding the relationships between these components of loading and knee osteoarthritis (OA) progression may improve conservative management, as gait interventions may need to account for physical activity levels or vice versa. The primary objective was to examine relationships between gait patterns that have previously been associated with OA progression and accelerometer-derived metrics of loading frequency. The secondary objective examined the association of accelerometer-derived metrics and total knee arthroplasty (TKA) at a mean follow-up of 3.5 years. METHODS: Fifty-seven individuals with knee OA underwent gait analysis and 1 week of accelerometer wear. Spearman correlations were calculated between accelerometer-derived metrics and gait patterns. Differences across quartiles of step count were examined with Jonckheere-Terpstra tests. In a subsample, baseline differences between TKA and no TKA groups were examined with Mann-Whitney U-tests. RESULTS: Gait variables previously related to progression were correlated to both step count and moderate- to vigorous-intensity, but not lower-intensity, physical activity. Individuals in the lowest quartile (~4000 steps/day) exhibited gait patterns previously related to progression. There were no differences in any baseline accelerometer-derived metrics between those that did and did not undergo TKA at follow-up. CONCLUSION: Complex relationships exist between gait, physical activity, and OA progression. Accelerometer-derived metrics may contribute unique information about overall loading for individuals above a certain activity threshold, but for those with lower activity levels, gait may be sufficient to predict clinical progression risk, at least over the short term.

Baseline Gait Muscle Activation Patterns Differ for Osteoarthritis Patients Who Undergo Total Knee Arthroplasty Five to Eight Years Later From Those Who Do Not.

OBJECTIVE: To determine if baseline quadriceps and hamstrings muscle activity patterns differed between those with medial-compartment knee osteoarthritis (OA) who advanced to total knee arthroplasty (TKA) and those who did not advance to TKA, and to examine associations between features extracted from principal component analysis (PCA) and discrete measures. METHODS: Surface electromyograms of the vastus lateralis and medialis, rectus femoris, and lateral and medial hamstrings during walking were collected from 54 individuals with knee OA. Amplitude and temporal characteristics from PCA, co-contraction indices (CCI) for lateral and medial muscle pairs, and root mean square (RMS) amplitudes for early, mid, late, and overall stance were calculated from electromyographic waveforms. At follow-up 5 to 8 years later, 26 participants reported having undergone TKA. Analysis of variance models tested for differences in principal component (PC) scores and discrete measures between TKA and no-TKA groups (α = 0.05). Pearson's product moment correlation coefficients were calculated between PC scores and discrete variables. RESULTS: The TKA group had higher hamstrings activity magnitudes (PC1), prolonged activity in mid stance (PC2) for all muscles, and greater lateral CCI. TKA had higher RMS hamstrings activity for all stance phases, and higher RMS mid- and late-stance quadriceps activity. PC1 was highly correlated with RMS amplitude (highest overall and early stance). PC2 was correlated with mid- and late-stance RMS. CCIs were correlated with PC1 and PC2, with greater variance explained for PC1. CONCLUSION: Those who advanced to TKA had higher magnitudes and more prolonged agonist and antagonist activity, consistent with less joint unloading. These gait muscle activation patterns indicate a potential conservative intervention target.

Directed attention alters the temporal activation patterns of back extensors during trunk flexion-extension in individuals with chronic low back pain.

In chronic low back pain patients (CLBP), neuromuscular and pain intensity have been identified as contributing factors in the disability of the individual. However, it is unclear whether pain intensity influences neuromuscular activation and if directed attention mediates this relationship. Thus, the purpose of this study was to determine the effect of directed attention in individuals with different pain intensities on back extensor activation profiles. Fifty-four CLBP patients were separated into either high- or low-pain groups. Surface electromyograms were recorded from back muscles while the subjects performed a trunk flexion motion for four different attention conditions. Pattern recognition and repeated measures ANOVAs were used to examine the effect of sex, attention and pain intensity on temporal muscle activation patterns. The results showed that there was a significant sex x attention x pain interaction. The largest changes in muscle timing were observed in the low-pain group when their attention was focused on their pain, but the pattern of muscle activation differed between sexes. For males, a rapid decline in activation at mid-extension occurred, whereas females showed delayed activation at the beginning of extension. Overall, this study demonstrated that directed attention on pain had an effect on trunk muscle temporal recruitment, and that this relationship differed between sexes and pain groups. This suggests that sex-specific mechanisms may alter the neuromuscular control of the spine in CLBP patients for different pain levels.

Temporal coactivation of abdominal muscles during dynamic stability exercises.

The purpose of this study was to determine abdominal muscle temporal responses to a leg-loading exercise protocol and if differences exist between those able and unable to minimize lumbar-pelvic motion during this protocol. The focus was a supine bilateral leg-loading task that incorporated a slide (level 4) or no slide (level 5). Thirty-three healthy subjects (mean age 24 years) completed the task while surface electromyograms (EMG) from 5 abdominal muscle sites were recorded. Subjects were assigned to stable or unstable groups based on their ability to minimize lumbar-pelvic motion. After time and amplitude normalization, electromyography waveforms were entered into a pattern recognition procedure and scores for each principal pattern were calculated. Four principal patterns explained 90% of variance in the waveform data, with these principal patterns capturing the mean pattern, the relative amplitude change during the leg-extension phase, and subtle changes in shape throughout the exercise. Significant interactions (p < 0.05) were found for principal patterns; 1, 2, and 4 scores; and significant main (p < 0.05) effects for principal pattern 3 scores. These results illustrate temporal synchrony among the abdominal wall muscle activation during the bilateral leg-loading tasks; however, there was less variability in the activation patterns during the leg-lift and leg extension-phases for those who were able to minimize lumbar-pelvic motion compared to those who were unable to perform the task correctly. These results illustrate the need to focus on coordinated recruiting of the abdominal wall muscles in an organized manner and not simply increasing the intensity of activation for stabilization training.

Characterisation of trunk muscle activation amplitude patterns during a simulated checkstand operation with continuously changing flexor and lateral moment demands.

While the typical physical exposure to modern-day workers has changed from heavy to low level repetitive demands, there is limited research that examines light occupations. This study examined trunk muscle recruitment strategies in response to a simulated checkout operation. Surface electromyography and kinematic variables were recorded from 29 healthy subjects. Four principal patterns accounted for 95.3% of the variation. Significant differences in scores captured different strategies in response to reach conditions and external moment directions. Synergistic co-activation of ipsilateral back sites and differential activation among external oblique and erector spinae sites suggests that the central nervous system may control different regions of the trunk musculature to optimally account for asymmetrical demands. The strategy between the internal oblique and back extensor sites suggests that a specific co-activation strategy may be needed during lighter work. During low-load occupational tasks, several recruitment strategies were required to maintain spinal stability and account for changing external moments. STATEMENT OF RELEVANCE: Different recruitment strategies found in response to changing external moments offer new insights into neuromuscular control for lighter work. Specifically, multiple trunk muscle sites interact in a complex manner, taking into account task specificity and individual variation that are valuable in workstation design, evaluating injury risk and estimating spinal loads.

Activation amplitude patterns do not change for back muscles but are altered for abdominal muscles between dominant and non-dominant hands during one-handed lifts.

It is assumed when lifting with the dominant hand that the relationship between contralateral and ipsilateral trunk muscle responses are similar to when lifting with the non-dominant hand. The purpose of this study was to quantify trunk muscle activation amplitude patterns during right- and left-handed lifts. Surface electromyography (EMG) and kinematic variables were recorded from 29 healthy subjects. Minimal trunk and pelvis motion was observed. Three principal patterns accounted for 95% of the variation in the EMG data indicating minimal variation in the pattern. Significant differences in scores captured different recruitment strategies for reach and hand. Selective and differential recruitment of back sites characterized lifts at greater distances from the body, whereas co-activation between internal oblique and back sites characterized lifts closer to the body. While the results showed no handedness effect for back muscles, the external oblique responded differently between right- and left-handed lifts. Specific recruitment strategies were used to account for subtle changes in reach and asymmetrical demands.

Do Older Adults and Those Recovered from Low Back Injury Share Common Muscle Activation Adaptations?

Theoretical models suggest trunk muscle activation compensates for spinal systems impairments. The purpose of this study was to determine if two populations (older adults and those recovered from a lower back injury (rLBI)) with spinal system impairments have similar muscle activation patterns to each other, but differ from controls. Trunk electromyograms collected from 12 older adults, 16 rLBI, and 19 controls during two dynamic tasks showed that older adults and rLBI had higher activation amplitudes, sustained temporal and more synergistic activation relative to controls. However, differences found between older adults and rLBI suggest that spinal system impairments differed between groups or that recent pain (rLBI) uniquely influenced muscle activation. This sheds light on our understanding of the relationship between spinal system impairments and muscle activation.

Trunk Muscle Activation Patterns Differ Between Those With Low and High Back Extensor Strength During a Controlled Dynamic Task.

It is proposed that reduced function in one of the spinal systems (active, passive, and neural) outlined by Panjabi could increase the risk of experiencing a low back injury (LBI). Also proposed is that reduced function in any one system can be compensated for by adjusting the time-varying recruitment of trunk muscles. This study addressed whether those with reduced active system function (WEAK), measured as back extensor strength, would have different trunk muscle activation patterns than those with higher function (STRONG), and secondly whether this relationship would be modified following recovery from a LBI. Sixty men participated, 30 recently recovered from LBI (rLBI, 4-12 weeks post injury) and 30 who had not had a LBI in the last year (ASYM). ASYM and rLBI participants were separated into STRONG and WEAK sub-groups if their isometric back extensor strength was above or below their group median, respectively. Trunk electromyograms from 24 muscle sites were recorded during a highly controlled horizontal transfer task. Principal component analysis captured key muscle activation patterns (amplitude and temporal); then analysis of variance models tested for strength or group*strength effects on these patterns consistent with the two main objectives. Significant strength, or group by strength effects were found for 3/4 electromyographic comparisons. In general, the WEAK group required higher activation amplitudes of abdominal and back extensor muscles, and greater temporal responsiveness of back extensor muscles only to the changing external moments than those who were STRONG. Group by strength interactions found that participants in the rLBI group had greater differences between WEAK and STRONG participants for overall muscle activation amplitudes in both abdominal and back extensor muscles. This increase in muscle activation was interpreted as compensation for lower maximum force properties whereas the increased temporal responsiveness captured a greater need to modify the agonist back extensors muscle activation patterns only in response to changes in the dynamic moments. Interactions captured that the recent experience of pain (rLBI) modified the magnitude of adjustment in muscle activation patterns potentially adapting to an increased risk of instability (painful flare) events associated with a deficit (lower strength) of the active system.

Longitudinal evidence links joint level mechanics and muscle activation patterns to 3-year medial joint space narrowing.

BACKGROUND: It is currently not known if there are different mechanical factors involved in accelerated rates of knee osteoarthritis structural progression. Data regarding the role of the transverse plane moment along with the contributions to joint loading from muscle activity, a primary contributor to the joint loading environment, is not well represented in the current literature on knee OA radiographic progression. The objective of this study was to understand if a 3-year end point corroborates what has been shown for longer term radiographic progression or provides more insight into factors that may be implicated in more accelerated radiographic progression than those shown previously. METHODS: 52 participants visited the Dynamics of Human Motion laboratory at baseline for three-dimensional, self-selected speed over ground walking gait analysis. Differences in magnitude and patterns of 3D knee moments and electromyography waveforms between participants who progressed radiographically from those that did not were compared using t-tests (P < 0.05). FINDINGS: Features of the frontal and transverse plane knee moments along with muscle activation patterns for the lateral gastrocnemius and lateral hamstrings differentiated the progression group from the non-progression group at baseline. INTERPRETATION: In general, the walking gait biomechanics of the progression group in this 3-year radiographic study aligned well with previously reported characteristics of diagnosed or symptomatic osteoarthritis. The higher rotation moment range during stance found with the progression group is a novel finding that points to a need to better understand torsional joint loading and its implications for loading of the knee joint tissues.