Predicting Knee Joint Contact Forces During Normal Walking Using Kinematic Inputs With a Long-Short Term Neural Network.

Knee joint contact forces are commonly estimated via surrogate measures (i.e., external knee adduction moments or musculoskeletal modeling). Despite its capabilities, modeling is not optimal for clinicians or persons with limited experience. The purpose of this study was to design a novel prediction method for knee joint contact forces that is simplistic in terms of required inputs. This study included marker trajectories and instrumented knee forces during normal walking from the "Grand Challenge" (n = 6) and "CAMS" (n = 2) datasets. Inverse kinematics were used to derive stance phase hip (sagittal, frontal, transverse), knee (sagittal, frontal), ankle (sagittal), and trunk (frontal) kinematics. A long-short term memory network (LSTM) was created using matlab to predict medial and lateral knee force waveforms using combinations of the kinematics. The Grand Challenge and CAMS datasets trained and tested the network, respectively. Musculoskeletal modeling forces were derived using static optimization and joint reaction tools in OpenSim. Waveform accuracy was determined as the proportion of variance and root-mean-square error between network predictions and in vivo data. The LSTM network was highly accurate for medial forces (R2 = 0.77, RMSE = 0.27 BW) and required only frontal hip and knee and sagittal hip and ankle kinematics. Modeled medial force predictions were excellent (R2 = 0.77, RMSE = 0.33 BW). Lateral force predictions were poor for both methods (LSTM R2 = 0.18, RMSE = 0.08 BW; modeling R2 = 0.21, RMSE = 0.54 BW). The designed LSTM network outperformed most reports of musculoskeletal modeling, including those reached in this study, revealing knee joint forces can accurately be predicted by using only kinematic input variables.

Influencing factors analysis of asymmetry in knee adduction moment among patients with unilateral knee osteoarthritis.

BACKGROUND: The knee adduction moment(KAM) of both lower limbs in patients with unilateral knee osteoarthritis(KOA) exhibits asymmetry during walking, but the factors influencing this asymmetry remain unclear. This study aimed to explore the influencing factors of KAM asymmetry in patients with unilateral KOA. METHODS: A total of 148 patients with unilateral medial compartment KOA were selected for this retrospective study, and general data such as gender, age, and duration of disease were collected. The hip-knee-ankle (HKA) angle, degree of pain, and knee-extension muscle strength on the affected side were assessed through radiographic outcomes, the visual analog scale(VAS), and the Biodex isokinetic system. The peak KAM of both lower limbs was analyzed using a BTS motion-capture system and force platform. The asymmetry index(ASI) of KAM was calculated, and the patients were further categorized into the KAM symmetry group(ASI value ≤ 10%) and the KAM asymmetry group(ASI value>10%).Binary logistic regression analysis was employed to analyze the factors influencing the asymmetry of KAM. RESULTS: 90 patients were categorized into the KAM asymmetry group, representing 60.8% of the cohort. A significant difference in the ASI value of KAM was observed between the two groups. Correlation analysis identified nine factors, including sex, age, and BMI, that were positively correlated with the ASI value of KAM. In contrast, knee-extension muscle strength and per-capita monthly household income were negatively correlated with the ASI value of KAM. Regression analysis revealed that being female(OR = 1.752), older age(OR = 2.472), increased BMI(OR = 1.535), larger varus angle(OR = 3.965), higher VAS score(OR = 2.617), Kellgren-Lawrence(K-L) grade IV(OR = 4.474), history of knee joint trauma(OR = 5.684), and living in a rural location(OR = 1.554) increased the risk of KAM asymmetry. Conversely, increased knee-extension muscle strength(OR = 0.758) and a per-capita monthly household income of 3000 ~ 6000 yuan(OR = 0.814) decreased the risk of KAM asymmetry. CONCLUSION: Female gender, older age, increased BMI, larger varus angle, higher VAS score, K-L grade IV, history of knee joint trauma, and living in a rural location are identified as risk factors for KAM asymmetry. Conversely, increased knee-extension muscle strength and a per-capita monthly household income of 3000 ~ 6000 yuan serve as protective factors against this asymmetry.

The Estimation of Knee Medial Force with Substitution Parameters during Walking and Turning.

PURPOSE: Knee adduction, flexion moment, and adduction angle are often used as surrogate parameters of knee medial force. To verify whether these parameters are suitable as surrogates under different walking states, we investigated the correlation between knee medial loading with the surrogates during walking and turning. METHODS: Sixteen healthy subjects were recruited to complete straight walk (SW), step turn (ST), and crossover turn (CT). Knee joint moments were obtained using inverse dynamics, and knee medial force was computed using a previously validated musculoskeletal model, Freebody. Linear regression was used to predict the peak of knee medial force with the peaks of the surrogate parameters and walking speed. RESULTS: There was no significant difference in walking speed among these three tasks. The peak knee adduction moment (pKAM) was a significant predictor of the peak knee medial force (pKMF) for SW, ST, and CT (p < 0.001), while the peak knee flexion moment (pKFM) was only a significant predictor of the pKMF for SW (p = 0.034). The statistical analysis showed that the pKMF increased, while the pKFM and the peak knee adduction angle (pKAA) decreased significantly during CT compared to those of SW and ST (p < 0.001). The correlation analysis indicated that the knee parameters during SW and ST were quite similar. CONCLUSIONS: This study investigated the relationship between knee medial force and some surrogate parameters during walking and turning. KAM was still the best surrogate parameter for SW, ST, and CT. It is necessary to consider the type of movement when comparing the surrogate predictors of knee medial force, as the prediction equations differ significantly among movement types.

Comparing Knee Kinetics and Kinematics in Healthy Individuals and Those With Knee Osteoarthritis, With and Without Flat Feet.

Individuals with knee osteoarthritis (KOA) and flat feet are more likely to experience increased pain and cartilage damage. This study aimed to investigate the knee kinetics, kinematics, pain, and physical function in individuals with moderate symptomatic KOA, in comparison to asymptomatic control participants. Thirty volunteers with moderate KOA (with flat feet n = 15, with normal feet n = 15) and 30 asymptomatic people (with flat feet n = 15, with normal feet n = 15) were evaluated. The knee adduction angular impulse, knee flexion moment, knee flexion angular impulse, and knee flexion angle were measured during level walking. The pain was assessed in patients with KOA. The study found that individuals with KOA had a significant increase in the knee adduction angular impulse compared with the asymptomatic people (P < .05). The KOA with flat feet group had significantly lower knee flexion moment, knee flexion angular impulse, and knee flexion angle values than the KOA with normal feet group (P < .05). Furthermore, the KOA with flat feet group had a higher pain score than the KOA with normal feet group. Individuals with osteoarthritis and flat feet had lower knee flexion moments which may indicate reduced knee force exerted through compensatory mechanisms. Despite this reduction, they reported significantly higher levels of pain compared with those without flat feet, a finding that warrants further investigation in future studies.

Restoration of the native tibial joint line obliquity in total knee arthroplasty with inverse kinematic alignment does not increase knee adduction moments.

PURPOSE: Patient-specific alignment in total knee arthroplasty (TKA) has shown promising patient-reported outcome measures; however, the clinical and biomechanical effects of restoring the native knee anatomy remain debated. The purpose of this study was to compare the gait pattern between a mechanically aligned TKA cohort (adjusted mechanical alignment-aMA) and a patient-specific alignment TKA cohort (inverse kinematic alignment-iKA). METHODS: At two years postoperatively, the aMA and iKA groups, each with 15 patients, were analyzed in a retrospective case-control study. All patients underwent TKA with robotic assistance (Mako, Stryker) through an identical perioperative protocol. The patients' demographics were identical. The control group comprised 15 healthy participants matched for age and gender. Gait analysis was performed with a 3D motion capture system (VICON). Data collection was conducted by a blinded investigator. The primary outcomes were knee flexion during walking, knee adduction moment during walking and spatiotemporal parameters (STPs). The secondary outcomes were the Oxford Knee Score (OKS) and Forgotten Joint Score (FJS). RESULTS: During walking, the maximum knee flexion did not differ between the iKA group (53.0°) and the control group (55.1°), whereas the aMA group showed lower amplitudes of sagittal motion (47.4°). In addition, the native limb alignment in the iKA group was better restored, and although more in varus, the knee adduction moments in the iKA group were not increased (225 N mm/kg) compared to aMA group (276 N mm/kg). No significant differences in STPs were observed between patients receiving iKA and healthy controls. Six of 7 STPs differed significantly between patients receiving aMA and healthy controls. The OKS was significantly better in patients receiving iKA than aMA: 45.4 vs. 40.9; p = 0.05. The FJS was significantly better in patients receiving iKA than aMA: 84.8 vs. 55.5; p = 0.002. CONCLUSION: At two years postoperatively, the gait pattern showed greater resemblance to that in healthy controls in patients receiving iKA rather than aMA. The restoration of the native coronal limb alignment does not lead to increased knee adduction moments due to the restoration of the native tibial joint line obliquity. LEVEL OF EVIDENCE: Level III.

The medial inclination of the proximal tibia is associated with the external knee adduction moment in advanced varus knee osteoarthritis.

PURPOSE: Whether the inclined articular surface on the medial proximal tibia and the external knee adduction moment (KAM) correlate remains unclear. The hypothesis was that a steeper inclined articular surface correlated with a larger KAM in advanced knee osteoarthritis (OA). METHODS: A total of 44 females (non-OA, 9 knees; early OA, 14 knees; advanced OA, 21 knees; mean age, 58 ± 16 years) were examined. Three-dimensional (3D) assessment was used on biplanar long-leg radiographs and 3D bone models using a 3D to 2D image registration technique. The approximation plane in the proximal tibia was determined using the least-square method. The joint moments were mathematically calculated in a gait analysis, applying a motion capture system and force plates. The main evaluation parameters were the femorotibial angle (FTA), the coronal inclination of the approximation plane in the medial proximal tibia (coronal inclination), and internal knee joint moments. The KAM means the external moments balanced with the internal knee abduction moments. RESULTS: The advanced OA showed a larger internal abduction moment (p = 0.017) at the loading response than the other groups. The larger FTA and steeper coronal inclination correlated with the larger internal abduction moment (FTA, p < 0.001; coronal inclination, p = 0.003) at the loading response. CONCLUSIONS: As the clinical relevance, the association among the coronal inclination of the medial proximal tibia, lower extremity alignment, and KAM is one of the key factors to help better understand the etiology of knee OA. LEVEL OF EVIDENCE: III.

Variable Stiffness Shoes for Knee Osteoarthritis: An Evaluation of 3-Dimensional Gait Mechanics and Medial Joint Contact Forces.

The study aim was to quantify the impact of a commercially available variable stiffness shoe (VSS) on 3-dimensional ankle, knee, and hip mechanics and estimated knee contact forces compared with a control shoe. Fourteen participants (10 females) with knee osteoarthritis completed gait analysis after providing informed consent. Shoe conditions tested were control shoe (New Balance MW411v2) and VSS (Abeo SMART3400). An OpenSim musculoskeletal model with static optimization was used to estimate knee contact forces. There were no differences in joint kinematics or in the knee adduction or flexion moments (P = .06; P = .2). There were increases in the knee internal and external rotation (P = .02; P = .03) and hip adduction and internal rotation moments for VSS versus control (P = .03; P = .02). The estimated contact forces were not different between shoes (total P = .3, medial P = .1, and lateral P = .8), but contact force changes were correlated with changes in the knee adduction moment (medial r2 = .61; P < .007). High variability in knee flexion moment changes and increases in the internal rotation moment combined with small decreases in the knee adduction moment did not lead to decreases in estimated contact forces. These results suggest that evaluation of VSS using only the knee adduction moment may not adequately capture its impact on osteoarthritis.

Reducing Knee Joint Load during a Golf Swing: The Effects of Ball Position Modification at Address.

As the modern golf swing has changed, the incidence of knee pain in professional golfers is increasing. For those with previous knee injuries, developing a golf-swing modification that reduces knee loading may be necessary to recover performance after injury. The purpose of this study was to test whether ball position modification reduces knee joint load in a golf swing. Thirteen male professional golfers participated in the study. Golf swings were captured using a three-dimensional motion capture system and two force platforms, with conditions for self-selected ball position and eight additional ball positions. Knee internal rotation and adduction moments were calculated. The length of one golf ball (4.27 cm) backward ball position (closer to the golfer) significantly reduced the peak internal rotation moment of the lead knee (- 13.8%) (p < 0.001) and the length of one golf ball (4.27 cm) away from the target ball position significantly reduced the peak adduction moment of the lead knee (- 11.5%) (p < 0.001) compared with that of the self-selected ball position. Based on these observations, we conclude that the backward ball position modification might be suggested for golfers with anterior cruciate ligament injuries, and the away from the target modification might be suggested for golfers with medial compartment knee osteoarthritis.

Frontal plane knee alignment mediates the effect of frontal plane rearfoot motion on knee joint load distribution during walking in people with medial knee osteoarthritis.

OBJECTIVE: To examine the nature of differences in the relationship between frontal plane rearfoot kinematics and knee adduction moment (KAM) magnitudes. DESIGN: Cross-sectional study resulting from a combination of overground walking biomechanics data obtained from participants with medial tibiofemoral osteoarthritis at two separate sites. Statistical models were created to examine the relationship between minimum frontal plane rearfoot angle (negative values = eversion) and different measures of the KAM, including examination of confounding, mediation, and effect modification from knee pain, radiographic disease severity, static rearfoot alignment, and frontal plane knee angle. RESULTS: Bivariable relationships between minimum frontal plane rearfoot angle and the KAM showed consistent negative correlations (r = -0.411 to -0.447), indicating higher KAM magnitudes associated with the rearfoot in a more everted position during stance. However, the nature of this relationship appears to be mainly influenced by frontal plane knee kinematics. Specifically, frontal plane knee angle during gait was found to completely mediate the relationship between minimum frontal plane rearfoot angle and the KAM, and was also an effect modifier in this relationship. No other variable significantly altered the relationship. CONCLUSIONS: While there does appear to be a moderate relationship between frontal plane rearfoot angle and the KAM, any differences in the magnitude of this relationship can likely be explained through an examination of frontal plane knee angle during walking. This finding suggests that interventions derived distal to the knee should account for the effect of frontal plane knee angle to have the desired effect on the KAM.

A neural network to predict the knee adduction moment in patients with osteoarthritis using anatomical landmarks obtainable from 2D video analysis.

OBJECTIVE: The knee adduction moment (KAM) can inform treatment of medial knee osteoarthritis; however, measuring the KAM requires an expensive gait analysis laboratory. We evaluated the feasibility of predicting the peak KAM during natural and modified walking patterns using the positions of anatomical landmarks that could be identified from video analysis. METHOD: Using inverse dynamics, we calculated the KAM for 86 individuals (64 with knee osteoarthritis, 22 without) walking naturally and with foot progression angle modifications. We trained a neural network to predict the peak KAM using the 3-dimensional positions of 13 anatomical landmarks measured with motion capture (3D neural network). We also trained models to predict the peak KAM using 2-dimensional subsets of the dataset to simulate 2-dimensional video analysis (frontal and sagittal plane neural networks). Model performance was evaluated on a held-out, 8-person test set that included steps from all trials. RESULTS: The 3D neural network predicted the peak KAM for all test steps with r2( Murray et al., 2012) 2 = 0.78. This model predicted individuals' average peak KAM during natural walking with r2( Murray et al., 2012) 2 = 0.86 and classified which 15° foot progression angle modifications reduced the peak KAM with accuracy = 0.85. The frontal plane neural network predicted peak KAM with similar accuracy (r2( Murray et al., 2012) 2 = 0.85) to the 3D neural network, but the sagittal plane neural network did not (r2( Murray et al., 2012) 2 = 0.14). CONCLUSION: Using the positions of anatomical landmarks from motion capture, a neural network accurately predicted the peak KAM during natural and modified walking. This study demonstrates the feasibility of measuring the peak KAM using positions obtainable from 2D video analysis.

Exploring the relationship between pain intensity and knee moments in participants with medial knee osteoarthritis: a cross-sectional study.

BACKGROUND: High biomechanical loading is believed to be a risk factor to pain in people with knee osteoarthritis (OA), but controversial findings have been reported on the relationship between external knee adduction moment (KAM) and pain. A more comprehensive analysis considering other factor such as external knee flexion moment (KFM) could help better reveal this relationship. This study explored the relationship between external knee adduction moment and pain intensity in participants with knee osteoarthritis (OA) using an integrated path analysis model. METHODS: This was a cross-sectional study based on laboratory setting. Forty-seven participants with clinical and radiographic medial knee OA were analyzed for their external knee adduction moment (KAM) and knee flexion moment (KFM) during walking using a motion analysis system. Pain intensity was measured by visual analogue scale (VAS) and the pain subscale of the Knee Injury and Osteoarthritis Outcome Score. Varus/valgus alignment was captured and quantified using a bi-planar X-ray system. Using a path analysis model, the relationships between pain intensity, KAM, KFM, OA radiographic severity, knee varus angle and walking speed were examined. RESULTS: The proposed path model met the goodness-of-fit criteria. Based on this model, KAM had a negative effect on VAS pain indirectly through the mediation of KFM. The model indicated KAM and KFM were negatively related to one another; and KFM was positively related to VAS. The KAM index, defined as (KAM/ (KAM + KFM)), was negatively related to VAS. CONCLUSIONS: Path analysis enabled the construction of a more integrated pathokinematic framework for people with knee OA. The KAM index which reflected the load sharing on the frontal and sagittal planes also revealed its relationship with pain. Re-distribution of mechanical loading from frontal to sagittal plane might be a strategy for pain avoidance associated with mechanical irritation.

Estimation of the External Knee Adduction Moment during Gait Using an Inertial Measurement Unit in Patients with Knee Osteoarthritis.

Although the external knee adduction moment (KAM) during gait was shown to be a quantitative parameter of medial knee osteoarthritis (OA), it requires expensive equipment and a dedicated large space to measure. Therefore, it becomes a major reason to limit KAM measurement in a clinical environment. The purpose of this study was to estimate KAM using a single inertial measurement unit (IMU) during gait in patients with knee OA. A total of 22 medial knee OA patients (44 knee joints) performed conventional gait analysis using three-dimensional (3D) motion capture system. At the same time, we attached commercial IMUs to six body segments (sternum, pelvis, both thighs, and both shanks), and IMU signals during gait were recorded synchronized with the motion capture system. The peak-to-peak difference of acceleration in the lateral/medial axis immediately after heel contact was defined as the thrust acceleration (TA). We hypothesized that TA would represent the lateral thrust of the knee during the stance phase and correlate with the first peak of KAM. The relationship between the peak KAM and TA of pelvis (R = 0.52, p < 0.001), shanks (R = 0.57, p < 0.001) and thighs (R = 0.49, p = 0.001) showed a significant correlation. The root mean square error (RMSE) of linear regression models of pelvis, shanks, and thighs to estimate KAM were 0.082, 0.079, and 0.084 Nm/(kg·m), respectively. Our newly established parameter TA showed a moderate correlation with conventional KAM. The current study confirmed our hypothesis that a single IMU would predict conventional KAM during gait. Since KAM is known as an indicator for prognosis and severity of knee OA, this new parameter has the potential to become an accessible predictor for medial knee OA instead of KAM.

Effects of walking with a "draw-in maneuver" on the knee adduction moment and hip muscle activity.

[Purpose] To investigate the effect of performing a draw-in maneuver (DI) on knee adduction moment (KAM) and hip and trunk muscle activities while walking. [Participants and Methods] We included 30 healthy young adults (21.5 ± 0.6 years, 16 males and 14 females) in this study. We measured the KAM and lever arm while participants walked with either a normal gait or a DI gait. We also performed surface electromyography (EMG) of the hip and trunk muscles (i.e., internal oblique abdominal muscle [IO], external oblique abdominal muscle [EO], multifidus muscle [MF], and gluteus medius muscle [GM]). [Results] The 1st peak of the KAM was significantly lower when walking with a DI gait compared to when walking with a normal gait. The integrated EMG activity of the IO, EO, and GM during the 1st half of the stance phase, and of the IO and EO during the 2nd half of the stance phase was significantly higher during the DI than during normal gait. [Conclusion] Compared with a normal gait, a DI gait leads to a decrease in the 1st peak of the KAM as a result of the shorter lever arm, and an increase in the muscular activity of the GM, IO, and EO.

Proprioceptive neuromuscular facilitation improves pain and descending mechanics among elderly with knee osteoarthritis.

PURPOSE: Knee osteoarthritis (KOA) is a common disease that causes pain and limits functionality in the elderly during daily activities, especially during stair descent. Proprioceptive neuromuscular facilitation (PNF) practices promote multiple-plane joint movements, which relieve pain and increase joint range of motion (ROM). This study aims to examine the effects of a 12-week PNF intervention on pain relief, passive and active joint ROM, external knee adduction moment (KAM), and hip adduction moment (HAM) in the elderly with KOA during stair descent. MATERIALS AND METHODS: Seventy-six elderly who were diagnosed with KOA were assessed for eligibility and, 36 of them met the inclusive criteria, were randomly divided into two groups: the twelve-week PNF intervention group and the control group. Pain score was measured by the Western Ontario and McMaster Universities Arthritis Index (WOMAC). Passive joint ROM was measured using a goniometer. Active joint ROM, KAM, and HAM during stair descent were measured using a motion analysis system with a force platform. All the data were recorded at weeks 0, 6, and 12. RESULTS: Compared to the control group, the PNF group showed a decreased pain score; increased passive hip, knee, and ankle ROM; a decreased minimum knee flexion angle, and increased HAM during stair descent. PERSPECTIVE: Proprioceptive neuromuscular facilitation intervention is a successful method to relieve symptoms of KOA. It relieves pain without increasing KAM, enhances passive ROM, increases active knee flexion ROM, and increases HAM during stair descent in the elderly with KOA.

Modifying Stride Length in Isolation and in Combination With Foot Progression Angle and Step Width Can Improve Knee Kinetics Related to Osteoarthritis; A Preliminary Study in Healthy Subjects.

The purpose of this study was to determine the effects of modifying stride length (SL) on knee adduction and flexion moments, two markers of knee loading associated with medial-compartment knee osteoarthritis (OA) progression. This study also tested if SL modifications, in addition to foot progression angle (FP) and step width (SW) modifications, provide solutions in more subjects for reducing knee adduction moment (KAM) without increasing knee flexion moment (KFM), potentially protecting the joint. Fourteen healthy subjects (six female) were enrolled in this preliminary study. Walking trials were collected first without instructions, and then following foot placement instructions for 50 combinations of SL, FP, and SW modifications. Repeated measures analysis of variance was used to detect group-average effects of footprint modifications on maximum KAM and KFM and on KAM impulse. Subject-specific dose-responses between footprint modifications and kinetics changes were modeled with linear regressions, and the models were used to identify modification solutions, per subject, for various kinetics change conditions. Shorter SL significantly decreased the three kinetics measures (p < 0.01). Potential solutions for 10% reductions in maximum KAM and KAM impulse without increasing maximum KFM were identified for five subjects with FP and SW modifications. A significantly higher proportion of subjects had solutions when adding SL modifications (11 subjects, p = 0.04). In conclusion, SL is a valuable parameter to modify, especially in combination with FP and SW modifications, to reduce markers of medial knee loading. Future work is needed to extend these findings to osteoarthritic knees.

Comparing Anterior Cruciate Ligament Injury Risk Variables Between Unanticipated Cutting and Decelerating Tasks.

To examine the relationship between anterior cruciate ligament injury risk factors in unanticipated cutting and decelerating. Three-dimensional kinematics and ground reaction forces were collected on 11 females (22 [2] y, 1.67 [0.08] m, and 68.5 [9.8] kg) during 2 unanticipated tasks. Paired samples t tests were performed to compare dependent variables between tasks. Spearman rank correlation coefficients were calculated to analyze the relationship between peak internal knee adduction moment and peak anterior tibial shear force (ASF) during 2 unanticipated tasks. Significantly greater knee abduction angles, peak knee adduction moments, and peak ASFs were observed during cutting (P ≤ .05). A strong positive correlation existed between decelerating ASF and cutting ASF (ρ = .67), while correlations between decelerating knee adduction moment and cutting knee adduction moment and decelerating ASF and cutting knee adduction moment were not significant. In situations where time management is a necessity and only one task can be evaluated, it may be more appropriate to utilize an unanticipated cutting task rather than an unanticipated deceleration task because of the increased knee adduction moment and ASF. These data can help future clinicians in better designing more effective anterior cruciate ligament injury risk screening methods.

Clinical and biomechanical changes following a 4-month toe-out gait modification program for people with medial knee osteoarthritis: a randomized controlled trial.

OBJECTIVE: To compare changes in knee pain, function, and loading following a 4-month progressive walking program with or without toe-out gait modification in people with medial tibiofemoral knee osteoarthritis. DESIGN: Individuals with medial knee osteoarthritis were randomized to a 4-month program to increase walking activity with (toe-out) or without (progressive walking) concomitant toe-out gait modification. The walking program was similar between the two groups, except that the gait modification group was trained to walk with 15° more toe-out. Primary outcomes included: knee joint pain (WOMAC), foot progression angles and knee joint loading during gait (knee adduction moment (KAM)). Secondary outcomes included WOMAC function, timed stair climb, and knee flexion moments during gait. RESULTS: Seventy-nine participants (40 in toe-out group, 39 in progressive walking group) were recruited. Intention-to-treat analysis showed no between-group differences in knee pain, function, or timed stair climb. However, the toe-out group exhibited significantly greater changes in foot progression angle (mean difference = -9.04° (indicating more toe-out), 95% CI: -11.22°, -6.86°; P < 0.001), late stance KAM (mean difference = -0.26 %BW*ht, 95% CI: -0.39 %BW*ht, -0.12 %BW*ht, P < 0.001) and KAM impulse (-0.06 %BW*ht*s, 95% CI: -0.11 %BW*ht*s, -0.01 %BW*ht*s; P = 0.031) compared to the progressive walking group at follow-up. The only between-group difference that remained at a 1-month retention assessment was foot progression angle, with greater changes in the toe-out group (mean difference = -6.78°, 95% CI: -8.82°, -4.75°; P < 0.001). CONCLUSIONS: Though both groups experienced improvements in self-reported pain and function, only the toe-out group experienced biomechanical improvements. TRIALS REGISTRY NUMBER: NCT02019108.

Moderators and mediators of effects of unloading shoes on knee pain in people with knee osteoarthritis: an exploratory analysis of the SHARK randomised controlled trial.

OBJECTIVE: To investigate moderators and biomechanical mediators of effects of unloading shoes on knee pain in people with knee osteoarthritis (OA). METHODS: Exploratory analysis from 164 participants in a clinical trial comparing unloading (ASICS GEL-Melbourne OA) to conventional walking shoes. The primary outcome was 6-month change in knee pain (11-point numerical rating scale (NRS)). Moderators included baseline peak knee adduction moment (KAM), radiographic severity (Kellgren & Lawrence (KL) scale), body mass, foot posture, neuropathic pain and diffuse knee pain. Mediators included change in peak KAM and KAM impulse. RESULTS: Radiographic severity was the only moderator to interact with footwear group (P = 0.02). Participants with KL = 2 experienced greater pain reductions with conventional compared to unloading shoes (mean difference in change in pain -1.64 units, 95% CI -3.07, -0.21), while unloading shoes tended to result in greater pain reductions than conventional shoes in KL = 3 (0.98, 95% CI -0.44, 2.39) and KL = 4 (0.64, 95% CI -0.64, 1.93). No variable showed any significant mediating effect in the entire cohort. However, there was some evidence that unloading shoes may reduce pain through reductions in peak KAM (indirect effect -0.31, 95% CIs -0.65, 0.03; P = 0.07) in people with KL ≥ 3, compared to conventional shoes. CONCLUSION: Unloading shoes conferred additional symptomatic benefits over conventional shoes in people with moderate to severe knee OA. There was some evidence effects may be mediated by a reduction in peak KAM. However, we were underpowered for subgroup analyses. These patients may represent a subgroup to which biomechanical interventions designed to reduce the KAM may be more effectively targeted.

Relationship between knee joint contact forces and external knee joint moments in patients with medial knee osteoarthritis: effects of gait modifications.

OBJECTIVE: To evaluate 1) the relationship between the knee contact force (KCF) and knee adduction and flexion moments (KAM and KFM) during normal gait in people with medial knee osteoarthritis (KOA), 2) the effects on the KCF of walking with a modified gait pattern and 3) the relationship between changes in the KCF and changes in the knee moments. METHOD: We modeled the gait biomechanics of thirty-five patients with medial KOA using the AnyBody Modeling System during normal gait and two modified gait patterns. We calculated the internal KCF and evaluated the external joint moments (KAM and KFM) against it using linear regression analyses. RESULTS: First peak medial KCF was associated with first peak KAM (R2 = 0.60) and with KAM and KFM (R2 = 0.73). Walking with both modified gait patterns reduced KAM (P = 0.002) and the medial to total KCF ratio (P < 0.001) at the first peak. Changes in KAM during modified gait were moderately associated with changes in the medial KCF at the first peak (R2 = 0.54 and 0.53). CONCLUSIONS: At the first peak, KAM is a reasonable substitute for the medial contact force, but not at the second peak. First peak KFM is also a significant contributor to the medial KCF. At the first peak, walking with a modified gait reduced the ratio of the medial to total KCF but not the medial KCF itself. To determine the effects of gait modifications on cartilage loading and disease progression, longitudinal studies and individualized modeling, accounting for motion control, would be required.

Kinematically aligned total knee arthroplasty reduces knee adduction moment more than mechanically aligned total knee arthroplasty.

PURPOSE: Knee adduction moment (KAM) has been recognized as a good clinical surrogate for medial tibiofemoral joint loading and is associated with implant durability after total knee arthroplasty (TKA). This study aimed to examine the effects of joint line obliquity in kinematically aligned TKA (KA-TKA) on KAM during gait. METHODS: The study enrolled 21 knees from 18 patients who underwent cylindrical axis reference KA-TKA and a matched group of 21 knees from 18 patients who underwent mechanically aligned (MA)-TKA as controls. Gait analyses were performed the day before TKA and at an overall mean of 2.6 years postoperatively. First peak KAM and variables associated with frontal knee kinetics were determined and compared between groups. RESULTS: In KA-TKA, the proximal tibia was resected with 3.4° ± 1.5° of varus in relation to the mechanical axis, and the final femorotibial shaft axis was 176.7° ± 3.8° with KA-TKA and 174.4° ± 3.0° with MA-TKA. KAM was significantly smaller with KA-TKA than with MA-TKA (p < 0.032). Regarding variables affecting KAM, significant differences were evident between the two TKAs for knee adduction angle (p = 0.0021), lever arm (p = 0.028), and Δlever arm (p = 0.0001). CONCLUSIONS: In KA-TKA, joint line obliquity reduced peak KAM during gait, despite slight varus limb alignment, and this reduced KAM in KA-TKA can tolerate constitutional varus alignment. In clinical settings, KA-TKA thus represents a promising technical option for patients with large coronal bowing of the shaft carrying a risk of increased KAM after TKA. LEVEL OF EVIDENCE: III.