Total knee arthroplasty (TKA) has been shown to improve knee joint function during gait post-operatively. However, there is considerable patient to patient variability, with most gait mechanics metrics not reaching asymptomatic levels. To understand how to target functional improvements with TKA, it is important to identify an optimal set of functional metrics that remain deficient post-TKA. The purpose of this study was to identify which combination of knee joint kinematics and kinetics during gait best discriminate pre-operative gait from postoperative gait, as well as post-operative from asymptomatic.

Seventy-three patients scheduled to receive a TKA for severe knee osteoarthritis underwent 3D gait analysis 1 week before and 1 year after surgery. Sixty asymptomatic individuals also underwent analysis. Eleven discrete gait parameters were extracted from the gait kinematic and kinetic waveforms, as previously defined (Astephen et al., J Orthop Res., 2008). Stepwise linear discriminant analyses were used to determine the sets of parameters that optimally separated pre-operative from post-operative gait, and post-operative from asymptomatic gait. Cross-validation was used to quantify group classification error.

Knee flexion angle range, knee adduction moment first peak, and gait velocity were included in the optimal discriminant function between the pre- and post-operative groups (P<0.05), with relatively equal standardised canonical coefficients (0.567, −0.501, 0.565 respectively), and a total classification rate of 74%. A number of metrics were included in the discriminant function to optimally separate post-operative and asymptomatic gait function, including the knee flexion angle range, peak stance knee flexion angle, minimum late stance knee extension moment, minimum mid-stance knee adduction moment, and peak knee internal rotation moment (P<0.05). The mid-stance knee adduction moment had the largest standardised canonical coefficients in the function, and 89.5% of cases were correctly classified.

Separation of pre and post-operative gait patterns included only three parameters, suggesting that current standard of care TKA significantly improves only walking velocity, knee flexion angle range, and the peak value of the knee adduction moment. A number of gait metrics, which were included in the discriminant function between post-operative and asymptomatic gait, could benefit from further improvement either through rehabilitation or design. With almost 90% classification, separation of post-operative gait function from asymptomatic levels is significant. The consolidation of knee joint function during gait into single, discrete discriminant scores allows for an efficient summary representation of patient-specific (or implant-specific) improvement in gait function from TKA surgery.

Purpose: The goal of this study was to investigate if musculoskeletal activation patterns measured with electromyography (EMG) are predictive of migration of total knee replacements (TKR) measured with radiostereometric analysis (RSA).

Method: 37 TKR patients who were part of a larger randomized controlled RSA trial were recruited to this study. Study participants had been randomized to receive the Nexgen LPS Trabecular Metal tibial monoblock component (n = 19), or the cemented NexGen Option Stemmed tibial component (n = 18) (Zimmer, Warsaw IN). Ethics approval was received from the institutional review board. In the week prior to their surgery, the patients went to the dynamics of human motion laboratory and underwent EMG data collection. Surface electrodes were placed over the vastus lateralis, vastus medialis, rectus femoris, the lateral and medial hamstrings, and the lateral and medial gastrocnemius using standardized placements (Hubley-Kozey et al., 2006). The variability in subject EMG patterns was captured with a set of discrete scores that represented weightings on objectively-extracted features of the gait waveform data using principal component analysis (PCA). Within four days of surgery and at six months post-operatively, patients had bi-planar knee x-rays taken. RSA analysis was performed with MB-RSA (MEDIS, Leiden). RSA results were reported as maximum total point motion (MTPM), and six degrees of freedom translations and rotations at six months.

Results: A correlation was found between the third principle component of the lateral gastrocnemius muscle (representing high gastrocnemius activation in late stance) and the anterior migration of the component (R2=0.247 P=0.002). A correlation was found between the vastus medialis principle component three (representing low vastus medialis activation in late stance) and the anterior migration of the component (R2= 0.338, P=0.000). A stepwise regression model was developed for anterior migration of the tibial component. To reduce the number of terms in the model only the two EMG variables that were correlated with anterior migration, implant type and BMI were entered leaving four possible terms. The stepwise regression eliminated all variables but the lateral gastrocnemius and the vastus medialis. The regression equation was Anterior-Posterior Migration = 0.01 +0.12*Vastus Medialis PC3 + 0.074*Lateral Gastrocnemius PC3 (R2=0.487, R2 Adj=0.457, P< 0.0001)

Conclusion: It has previously been shown that anterior shear on the tibial component of TKR is temporally localized to the last third of stance phase of gait. Both the gastrocnemius and vastus muscle groups have the ability to produce large anterior posterior shear on an the knee during late stance. This result shows that variables which capture the temporal activation patterns of these muscles preoperatively are related to the migration of the tibial component of TKR postoperatively. This may have implications for rehabilitation of these patients.

Purpose: Adults with knee osteoarthritis (OA) show biomechanical changes in gait which may be linked to the quadriceps weakness often associated with knee OA. The purpose of this study was to mimic the effect of quadriceps weakness by inducing fatigue to determine if this produced gait characteristics similar to those present in knee OA.

Method: Sixteen healthy, sedentary female subjects between the ages of 19 and 35 years participated. Subjects were randomly assigned to perform a quadriceps-fatiguing protocol using a CybexTM isokinetic dynamometer (n=9) or a resting control group (n=7). Gait was evaluated before and after the rest or the fatiguing protocol. Infrared and virtual markers were used to record the locations of 16 anatomical landmarks. Marker position was recorded using an Optotrak motion capture system. An AMTI force plate collected ground reaction forces. Joint kinematics and kinetics were calculated using standard techniques. Maximum, minimum and time to peak were calculated for knee flexion angle, and the flexion, adduction and rotation moments during stance. A 2-factor (group, pre-post) mixed model ANOVA was used to test main effects and interactions (alpha = 0.05). Post hoc Bonferroni testing (alphaadj=0.0125) was used to determine pair wise differences.

Results: The two groups were statistically similar at baseline (p> 0.05) in terms of their age, mass, height, strength, and self-selected walking velocity. The fatigue protocol resulted in a 49 ± 12% decrease in peak knee extensor torque. The statistical results showed that knee flexion and external rotation moments decreased, the time to peak knee flexion angle increased, and the minimum knee adduction moment increased from pre-test to post-test in the fatigued group (p< 0.05).

Conclusion: Induced quadriceps fatigue alters kinematic and kinetic gait parameters. The changes are consistent with the gait characteristics observed in patients with knee OA and imply a change in joint motion and loading. These results provide preliminary evidence of a direct link between quadriceps weakness and the mechanical environment of the knee joint. This may be useful in developing more specific management programs for knee OA.

Purpose: To test for a decrease in knee musculature co-activation at one- and two-year post total knee replacement (TKR) compared to pre-TKR values.

Method Thirty men and 35 women with knee osteoarthritis (OA), after providing informed consent, participated in this study. Surface electromyograms (EMG) recorded the activation of seven muscle sites (rectus femoris, vastus lateralis and medialis, lateral and medial hamstrings, lateral and medial gastrocnemius) while subjects ambulated, at a self-selected velocity, along a six-meter walkway; one week prior to TKR surgery, and one- and two-years post-TKR. Linear enveloped EMG waveforms, amplitude normalized to 100%MVIC and time normalized to 100% of the gait cycle, were entered into a principal component analysis model [1]. A two-factor (pre-post, muscle) repeated measures ANOVA was applied to test statistically significant main effects (pre-post, muscle) and interactions (alpha = 0.05).

Results: The mean age, mass and height pre-TKR were 63.4 years, 91.4 kg and 1.69 m, respectively. Forty seven and 25 subjects completed the one- and two-year follow up, respectively. Walking velocity increased from 0.9 m/s pre-TKR to 1.1 m/s for both one- and two-year post TKR. Four principal components explained 89% of the variance in the waveform data. PC1 (60% of the variance), associated with co-activation throughout the entire stance phase, was found to be statistically significant (p< 0.05). The post hoc analysis revealed no significant differences between one- and two-year post-TKR PC1 scores, but the two vasti muscles and lateral hamstring had significantly lower PC1 scores post-TKR compared to pre-TKR values. There were significant differences among muscles with the two vasti muscles higher pre-TKR compared to all other muscles, and lateral hamstring higher pre-TKR than the remaining four muscles.

Conclusion: PC1 captured a co-activation pattern illustrating muscle activation during the majority of the stance phase of gait for the vasti and lateral hamstring muscles. This pattern, previously shown in severe OA gait [1], is considered an adapted response to the pain and joint instability associated with latter stages of OA progression. These findings support that the neuromuscular control strategies are altered within one year with no additional change for the two year post-TKR measures. These results support a more efficient neuromuscular control strategy post-TKR and perhaps an associated decrease in metabolic cost and an improvement in function.

Purpose: The purpose of this study was to determine what differences exist in the knee flexion, rotation and adduction moments and periarticular knee muscle activation patterns between subjects with medial compartment knee osteoarthritis (OA) and those with lateral compartment knee OA.

Method: Forty eight individuals with knee OA were studied. The group was divided into those with predominantly medial compartment involvement (38 subjects, age 63 ± 8 years) and those with lateral involvement (10 subjects, age 63 ± 9 years). Three-dimensional motion (Optotrak) and ground reaction force (AMTI) data were collected while the subjects walked at a self-selected velocity. The knee flexion, rotation and adduction moments, time normalized to the percentage of one gait cycle, were calculated using an inverse dynamics approach. Electromyograms (EMG) were also collected from the rectus femoris, vastus lateralis, vastus medialis, medial and lateral hamstrings, and medial and lateral gastrocnemius and normalized to maximum voluntary isometric contractions. Knee moments and waveforms for each muscle for one complete gait cycle were analyzed for group differences using principal component analysis (PCA) followed by Student’s t-tests (alpha-adj = 0.017) for the PCA scores.

Results: The two groups were statistical similar in terms of age, height, weight, and walking velocity (p> 0.05). PCA analysis revealed statistically significant differences (p< 0.017) in patterns for the knee adduction moment, medial gastrocnemius, and lateral hamstrings between the two groups.

Conclusion: As expected, there was a difference in the knee adduction moment between the two groups. What is novel is that the muscle activation patterns from the lateral site group are consistent with an attempt to unload that compartment. The results of this study provide evidence that biomechanical and neuromuscular differences do exist, depending on the OA site. This could have implications for developing site-specific conservative management approaches

The purpose of this study was to characterise the neuromuscular patterns associated with different severities of knee osteoarthritis (OA).

Forty-five patients with moderate OA, thirty-seven with severe OA and thirty-eight asymptomatic controls underwent a complete gait analysis with only the electromyographic (EMG) findings presented in this abstract. Severity levels were established through the Kellgren-Lawrence radiographic grading system, functional ability, and those classified with severe OA were tested within one-week of total knee replacement surgery. All OA patients had medial joint involvement. Subjects walked along a five-meter walkway a total of five times at a self- selected walking speed. Muscle activation patterns of the vastus medialis and lateralis, medial and lateral hamstring and medial and lateral gastrocnemius were recorded and normalised to maximum voluntary isometric contractions. All EMG waveforms were analyzed for group differences using PCA [1] followed by an ANOVA (group by muscle) for the PCA scores for each muscle group. These scores reflect both magnitude and shape changes.

The control group was significantly younger (53.3 ±9.5 yrs) and lighter (77.5 ±14.5 Kg) than the patient groups (Moderate =59.8 ±8.0 years and 94.2 ±19.2 Kg and Severe = 63.1 ±7.9 yrs and 95.8 ±14.6Kg). The severe OA group walked significantly slower (0.9 ±0.2 m/s) than the asymptomatic (1.3 ±0.1) m/s) and the moderate OA (1.2 ±0.2 m/s) groups. The PCA analysis of the EMG waveforms revealed statistically significant differences (P< 0.05) in patterns among the three groups and between muscles within the three muscle groups tested.

The neuromuscular differences found among groups during gait demonstrate that the role of the musculature surrounding the knee is altered slightly in those with moderate OA and altered drastically in those with end-stage OA compared to asymptomatic subjects, reflecting a progression. The differences are consistent with the severe group adopting a co-activation strategy of agonist and antagonists, more lateral activation and a reduction in plantar flexion during push off. These are consistent with strategies to increase dynamic stability and reduce medial joint loading. The moderate OA group illustrates a trend toward adopting this pattern but with only very subtle differences from asymptomatic subjects as has been previously reported. These neuromuscular alterations have implications with respect to muscle function and may assist in defining severity.

The purpose of this investigation was to determine the changes in frontal plane kinetics (loading) and neuromuscular responses pre and post unilateral total knee replacement surgery (TKR) during walking.

Thirty-four patients with severe knee osteoarthritis (within one week prior to TKR surgery) underwent a gait analysis. 3D kinematics, kinetics and electromyographic (EMG) recruitment patterns from seven lower limb muscles (vastus medialis and lateralis, medial and lateral hamstrings, medial and lateral gastrocnemius and rectus femoris) were recorded while walking at their self-selected walking speed. This was repeated one-year post-TKR surgery. EMG data were normalised to maximum voluntary isometric contractions and the knee adduction moment was normalised to body mass. All waveforms were normalised in time to 100% of the gait cycle. Principal component analysis was applied to the pre-and post-TKR waveforms. T-tests and ANOVA models tested pre-post TKR differences and differences between muscles.

At pre-TKR, the average age of the subjects was 66 ± 6.6 years and there were no statistically significant differences between pre and post TKR measures of mass (90Kg). The walking velocity significantly (p< 0.05) increased from the pre-TKR (.9 ±.23 m/s) to the post-TRK (1.07 ±.21 m/s). There were statistically significantly (p< 0.05) magnitude and shape differences between the pre-and-post-TKR waveforms for the knee adduction moment and the EMG waveforms. In general there were reduced adduction moments and EMG amplitudes for quadriceps and hamstrings post-TKR.

The results show improved function with the increased walking velocity, but more important are the differences with respect to joint loading and muscle function. The decreased knee adduction moment post-TKR reflects reduced loading on the medial compartment of the prosthesis. The alterations in the quadriceps and hamstrings illustrate that post-TKR the muscles no longer co-activate at high percentage of their maximum during the majority of the gait cycle as was shown in the pre-TKR waveforms. Finally the high lateral hamstring activity found pre-operatively was reduced resulting in a more balanced activation between the medial and lateral sites post operatively. These post-TKR changes have implications for improved joint loading, reduced risk of muscle fatigue and decreased metabolic costs associated with walking.

To compare strength and recruitment of periarticular knee muscles in subjects with severe osteoarthritis (OA) one week before and one year after a total knee replacement (TKR).

Twenty-eight subjects, mean age = 64.5 years, with severe knee OA performed maximum voluntary isometric contractions for six exercises designed to test knee flexor and extensor and plantarflexor muscle strength. Torque and surface electromyograms (EMG) from the lateral and medial gastrocnemius, lateral and medial hamstring, vastus lateralis and medialis and rectus femoris muscles were recorded. Exercises included knee extension and flexion at mid range (45°) and closed-pack (15°) positions and plantarflexion with knee extended. Subjects completed WOMAC questionnaires to assess function. Custom software written in Matlab version 7.0.4 was used to calculate muscle torque and process EMG data. Paired Student t-tests (alpha = 0.05) were used to detect significant differences between pre-test and post-test data. Statistical analyses were performed in Minitab.

Post-TKR torque increases ranged from 1.6% to 19.7%, but only knee extension with the subject’s knee at 45° showed a statistically significant (p< 0.05) increase (74.3 ± 29.5 Nm to 86.1 ± 28.5 Nm). EMG amplitudes increased for the quadriceps and hamstring muscles (p< 0.05) post TKR, but the relative contributions of each muscle did not change, excepting rectus femoris. Within each exercise, some subjects increased their torque, but almost as many decreased their post-TKR torque. WOMAC scores for pain, stiffness, and function improved significantly (p< 0.05) by one year after TKR.

TKR surgery is becoming more common as a treatment for OA, but few studies have examined muscle strength before and after, which impacts patient function and the lifespan of the implant. By one year post-TKR subjects reported significant decreases in pain and stiffness, and significant improvements in function. This is consistent with the literature. Half of the subjects decreased in muscle strength to levels lower than pre-surgery. The results provide evidence that post-TKR management must address muscular strength deficits in addition to subjective assessments of improved symptoms to measure success.

The knee adduction moment is indicative of the degree of medial compartmental loading at the knee joint and has been related to the presence and progression of knee osteoarthritis (OA). Studies have reported differences between OA and asymptomatic groups when measuring the adduction moment at the knee; however, there have been various biomechanical models used to describe this moment. In addition, non-invasive interventions have been shown to decrease the adduction moment but only at certain portions of the gait cycle. The objective of the study was to determine if changing the biomechanical model would affect the ability to detect differences between OA and asymptomatic gait and whether these differences depended on which portion of the gait cycle was analysed.

The gait of forty-four asymptomatic and forty-four moderate OA subjects was measured. The adduction moment was calculated using three different biomechanical models commonly used in the literature:

a 2D representation of the lower limb,

The choice of biomechanical model changed the overall magnitude and shape of the adduction moment waveform. These changes affected the ability to detect group differences using commonly reported parameters of the adduction moment. However, group separation was achieved (regardless of model) when analyzing the overall magnitude of the adduction moment across stance phase and the mid-stance portion of the gait cycle.

These results demonstrate that the OA subjects are not unloading the medial compartment of the knee at full weight acceptance as well as the healthy controls. Furthermore, the OA subjects are experiencing a higher medial compartment load that is being sustained for the duration of the stance phase of the gait cycle. Group differences that are not model dependent may be important in understanding the pathomechanics of OA and evaluating interventions. These findings support the need for a better understanding of the anatomical mechanisms associated with the adduction moment.

Determine the association between net external knee adduction moment (KAM) characteristics and foot progression angle (FPA) in asymptomatic individuals and those with moderate and severe osteoarthritis through discrete variable and principal component analysis (PCA).

Fifty-nine asymptomatic (age 52 ± 10 years), fifty-five with moderate knee OA (age 60 ± 9 years) and sixty-one individuals with severe knee OA (age 67 ± 8 years, tested within one week of total knee replacement surgery) participated. Three-dimensional (3D) motion (Optotrak) and ground reaction force (AMTI) data were recorded during gait. Subjects walked at a self-selected velocity. The KAM, calculated using inverse dynamics was time normalised to one complete gait cycle. FPA was calculated using stance phase kinematic gait variables. The discrete variable, peak KAM, was extracted for the interval (30–60%) of the gait cycle. PCA was used to extract the predominant waveform features (Principal Components (PC)) of which PC-Scores were computed for each original waveform. Pearson Product Moment Correlations were calculated for the FPA and both the PC-scores and peak KAM. Alpha of 0.05 used to test significance.

No significant correlations were noted for the groups between peak KAM and the FPA, or for the first PC-Scores (PC1) of which captured the original KAM waveforms overall magnitude and shape. The second PC (PC2) captured the shape and magnitude during the second interval of stance (30–60%) with respect to the first. Correlations of FPA to PC2 were significant for the asymptomatic group(r=−0.40, p=0.002) and the moderate OA group (r=−0.32, p=0.017) but not for the severe group(r=−0.13, p=0.316).

No relationship between FPA and peak KAM was found across the groups using discrete variable analysis despite reports of associations in asymptomatic subjects. The PCA results suggest a toe out FPA was moderately correlated to a decreased KAM during 30–60% of the gait cycle for asymptomatic and moderate OA individuals only. These individuals respond to a toe out progression angle, altering the KAM which directly affects medial knee compartment loading, where those with severe OA do not.

Modern gait analysis offers a unique means to measure the biomechanical response to diseases of the musculoskeletal system during activities of daily living. The objective of this on-going study is to quantify the biomechanical environment of the knee joint in subjects with moderate knee osteoarthritis (OA). We collected 3-D motion, ground reaction force, and electromyographic data from seven normal subjects and five subjects with moderate knee OA. There were no differences in stride characteristics or joint motion patterns between the two groups. In contrast, we found differences in knee joint kinetics between the moderate OA subjects and the normal control subjects.

The objective of this on-going study is to quantify the biomechanical environment of the knee joint in subjects with moderate knee osteoarthritis (OA). Our goal is to identify biomechanical characteristics related to treatment interventions.

The moderate knee OA patients walked with a visibly normal gait as measured by stride characteristics and joint angles. Differences were detected in the joint loading (ie adduction and flexion moments).

The biomechanical differences between normal and osteoarthritic knees will provide the basis upon which to design and evaluate non-invasive treatments for knee OA.

Subjects performed, in random order, five trials of their normal selected speed, and a fast walk (150% of the normal speed). Three-dimensional motion and force data were used to calculate three dimensional joint angles, moments and forces.

There were no differences in stride characteristics (walking speeds, stride lengths, or stride times) between the two groups. The moderate OA patients walked with normal knee joint motion patterns. In contrast, we found differences in knee joint kinetics between the moderate OA subjects and the normal control subjects. The magnitude of the adduction moment during stance was larger for the moderate OA patients at both walking speeds (p< 0.05). We also identified differences in the pattern of the flexion moment, but only at the higher walking speed (p< 0.05).

Gait analysis can provide insight into the mechanical factors of knee osteoarthritis by quantifying the dynamic loading and alignment of the knee during activities of daily living

Spinal stabilization through appropriate neuromuscular responses to external perturbations is important in the prevention and rehabilitation of low back pain (LBP). Muscle synergism, coordination and imbalances are terms used to describe the neuromuscular strategies considered important to actively maintain spinal stability. We recorded surface electromyographic (EMG) recordings from healthy controls (CON) and those with chronic, mechanical low back pain during performance of an exercise model that dynamically challenged lumbar-pelvic stability. Those with LBP showed greater variation in amplitude in response to the perturbations imposed by the exercise model, and demonstrated a lack of synergistic and antagonistic coactivation compared to the CON group.

The purpose of this study was to compare the neuromuscular control strategies used by those with LBP and those without to complete a standardized task aimed at dynamically challenging stability of the lumbar spine and pelvis.

Those with LBP activated their muscles in a more asynchronous manner than normal controls, illustrating an alteration in neuromuscular control that should be a focus of therapeutic intervention strategies aimed at prevention and rehabilitation of LBP.

These data illustrate a need for neuromuscular retraining, focusing on muscle coactivation in response to dynamic perturbations rather than a single perturbation.

Surface EMG recordings from two trunk extensor and five abdominal muscle sites were recorded from twenty-four men without LBP and fourteen men with chronic LBP while they performed a task that dynamically challenged lumbar spine and pelvis stability. The EMG amplitudes recorded from the upper and lower rectus abdominus sites were significantly (p< 0.05) lower for the LBP compared to the CON group. The temporal EMG profiles were compared using a statistical pattern recognition technique. This analysis showed that the LBP group used different patterns of synergistic muscle coactivity compared to the highly coordinated manner in which all seven muscles were recruited for the CON.

These results quantify the neuromuscular control differences between the two groups providing a foundation for developing an objective classifier of neuromuscular control impairments associated with LBP. In future this approach could assist in directing therapeutic interventions in particular those aimed at muscle reeducation.

The objective of this study was to determine if abnormal neuromuscular patterns exist in individuals with knee Osteoarthritis compared to those with healthy knees. We collected surface electromyographic signals during preferred speed and fast walk conditions from seven muscles crossing the knee joint. We found differences between the two groups that could lead to differences in joint loading, with the OA group having higher coactivity between hamstrings and quadriceps during initial loading. Further investigating these differences is warranted in particular given the trend for lower external extensor moments for the OA group at the fast walking speed.

The purpose of this study was to compare neuromuscular control of knee joint motion during walking between those with moderate Osteoarthritis (OA) and those with healthy knees (CON).

Moderate OA neuromuscular control patterns differed from those with healthy knees.

Detecting neuromuscular alteration associated with mild to moderate knee OA is important to direct therapeutic strategies aimed to slow down or possibly reverse disease progression.

Surface electromyographic (EMG) recordings were collected from seven muscles crossing the knee joint of CON (n=7) and those with moderate OA (n=4) during preferred speed and a fast-paced walks. A pattern recognition technique was applied to the EMG profiles. No differences (> 0.05) were reported between the two groups for spatial and temporal gait parameters or knee joint kinematics. Statistical differences were found (p< 0.05) in muscle activation patterns between the two groups and the differences were more prominent at the faster walking speed. The two vasti muscles had double peaks during stance and higher amplitudes at heel strike for the OA group. There was higher activity in the two hamstring muscles at heel contact and a burst of activity during late stance for the OA group.

The disproportionately higher knee flexor coactivity at heel strike may reflect a guarded response to pain, whereas the burst during weight transfer may reflect a stabilizing response as the knee moment changes from a flexor to an extensor moment. At normal walking speeds the neuromuscular control patterns were similar between groups, but differences were exaggerated when the system was stressed at higher speed.