KNEE OSTEOARTHRITIS (OA) is a common chronic

Transcription

1 822 ORIGINAL ARTICLE Gait Differs Between Unilateral and Bilateral Knee Osteoarthritis Mark W. Creaby, PhD, Kim L. Bennell, PhD, Michael A. Hunt, PhD ABSTRACT. Creaby MW, Bennell KL, Hunt MA. Gait differs between unilateral and bilateral knee osteoarthritis. Arch Phys Med Rehabil 2012;93: Objectives: To compare walking biomechanics in the most painful leg, and symmetry in biomechanics between legs, in individuals with (1) unilateral pain and structural osteoarthritis (OA), (2) unilateral pain, but bilateral structural OA, and (3) bilateral pain and structural OA and in (4) an asymptomatic control group. Design: Cohort study. Setting: Laboratory based. Participants: Participants with symptomatic and/or radiographic medial tibiofemoral OA in one or both knees (n 91), and asymptomatic control participants (n 31). Interventions: Not applicable. Main Outcome Measure: The peak knee adduction moment, peak knee flexion moment, knee varus-valgus angle, peak knee flexion angle, toe-out, and trunk lean were computed from 3-dimensional analysis of walking at a self-selected speed. Results: After controlling for walking speed, greater trunk lean toward the more painful knee and reduced flexion in the more painful knee were observed in all OA groups compared with the control group. Between-knee asymmetries indicating greater varus angle and a lower external flexion moment in the painful knee were present in those with unilateral pain and either unilateral or bilateral structural OA. Knee biomechanics were symmetrical in those with bilateral pain and structural OA and in the pain free control group. Conclusions: The presence of pain unilaterally appears to be associated with asymmetries in knee biomechanics. Contrary to this, bilateral pain is associated with symmetry. This suggests that the symptomatic status of both knees should be considered when contemplating unilateral or bilateral biomechanical interventions for medial knee OA. Key Words: Biomechanics; Knee; Osteoarthritis; Pain; Rehabilitation by the American Congress of Rehabilitation Medicine From the Centre of Physical Activity Across the Lifespan, School of Exercise Science, Australian Catholic University, Brisbane, Queensland, Australia (Creaby); Centre for Health, Exercise and Sports Medicine, University of Melbourne, Melbourne, Victoria, Australia (Creaby, Bennell, Hunt); and Department of Physical Therapy, University of British Columbia, Vancouver, British Columbia, Canada (Hunt). Supported by the National Health and Medical Research Council, Australia (project grant no ); and an Australian Research Council Future Fellowship. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Correspondence to Mark W. Creaby, PhD, Centre of Physical Activity Across the Lifespan, School of Exercise Science, Australian Catholic University, QLD, 4014 Australia, mark.creaby@acu.edu.au. Reprints are not available from the author. In-press corrected proof published online on Mar 5, 2012, at /12/ $36.00/0 doi: /j.apmr KNEE OSTEOARTHRITIS (OA) is a common chronic joint disease that mainly affects the medial tibiofemoral compartment. 1 While the precise pathophysiology of the disease is not fully understood, it is typically conceptualized as mechanically driven. Indeed, the external knee adduction moment (KAM) during walking, a surrogate measure of the load on the medial tibiofemoral compartment during walking, 2,3 has been associated with radiographic disease severity 4,5 and the risk of structural deterioration associated with medial knee OA progression. 6 Further evidence of the involvement of mechanical loading in medial knee OA comes from various cross-sectional reports of a higher KAM in the affected knee compared with the knees of healthy individuals. 4,7-9 Given that global factors, such as body weight, are known to influence gait mechanics in both knees, an alternative approach to elucidate differences in gait mechanics with knee OA, independent of such global factors, is to compare the affected knee with the contralateral knee. Using this approach, a higher peak KAM in the operative versus nonoperative knee of patients scheduled for high tibial osteotomy has been observed. 10 Briem and Snyder-Mackler, 11 however, reported no between-limb difference in peak KAM, but altered kinematics at the knee and hip in a smaller group of patients scheduled for the same surgery. In both of these studies, 10,11 the symptomatic status and radiographic status of the contralateral limb were not clearly defined. This leaves open the possibility that pain and/or structural disease may be present in the contralateral limb, influencing ipsilateral joint loading and the adoption of compensatory gait strategies. A reduction in self-selected walking velocity is consistently reported in knee OA in comparison with asymptomatic agematched individuals. 8,12-14 It is argued that this may be a deliberate strategy to reduce knee moments and thus joint load in the painful knee(s). 15 Other kinematic strategies to reduce joint load, such as increased toe-out, greater trunk lean toward the affected limb, and reduced knee flexion, 21 have also been reported. Conceivably, individuals with bilateral pain would adopt compensations bilaterally, but this has not previously been investigated. Similarly, it is not known if individuals with unilateral pain adopt strategies in only the affected limb, or also in the contralateral limb. A further consideration is the influence on gait mechanics of structural disease in the contralateral knee. While the influence on gait of disease in the ipsilateral knee is well established, the effect of structural OA in the contralateral knee which may or may not be painful is not known. In this respect, true delineation of distinct subgroups of those CI KAM K-L OA List of Abbreviations confidence interval knee adduction moment Kellgren-Lawrence osteoarthritis

2 GAIT IN UNILATERAL AND BILATERAL KNEE OSTEOARTHRITIS, Creaby 823 with knee OA can be based on the systematic identification of the presence or absence of pain and radiographic changes, resulting in 4 possibilities: (1) unilateral pain, unilateral structural disease; (2) unilateral pain, bilateral structural disease; (3) bilateral pain, unilateral structural disease (this presentation is rare); and (4) bilateral pain, bilateral structural disease (the most common presentation 22 ). The purpose of this study, therefore, was to compare walking biomechanics in the most painful leg, and symmetry in biomechanics between legs, in individuals with (1) unilateral pain and structural OA, (2) unilateral pain, but bilateral structural OA, and (3) bilateral pain and structural OA and in (4) an asymptomatic control group. First, we examined betweengroup differences in gait mechanics of the most symptomatic limb (or study limb, in the case of the control group). Second, we assessed the interlimb symmetry in these gait mechanics for each of the 4 groups. We expected a higher KAM, higher knee flexion moment, greater varus angle, more toe-out, more trunk lean, and less knee flexion in the OA groups compared with controls. Furthermore, we expected asymmetries in the groups with unilateral pain compared with those with no pain or with bilateral pain. METHODS Participants One hundred twenty-two participants, including 91 with medial compartment knee OA and 31 asymptomatic, healthy controls, were recruited from the local community. Participants in the OA groups were originally recruited for a randomized controlled trial of hip strengthening, 23 and the measurements included in this study were taken at baseline prior to intervention. Inclusion criteria for the OA participants were as follows: age over 50 years, knee pain on most days of the previous month (average level 3cm on a 10-cm visual analog scale), knee alignment 182 on a standardized semiflexed standing posteroanterior knee radiograph (corresponding to a mechanical axis angle of 180 on a full leg radiograph, indicating varus alignment), 24 a predominance of pain or tenderness over the medial region of the knee, osteoarthritic changes in the medial compartment of the tibiofemoral joint (osteophytes and/or joint space narrowing), and at least grade 2 knee OA on the Kellgren-Lawrence (K-L) scale. 25 The exclusion criteria were as follows: lateral tibiofemoral joint space width less than medial, symptoms originating predominantly from the patellofemoral joint as determined by clinical examination, knee surgery or intraarticular corticosteroid injection within the past 6 months, current or past (within 4wk) oral corticosteroid use, systemic arthritic conditions, a history of tibiofemoral or patellofemoral joint replacement or tibial osteotomy, any other muscular, joint, or neurologic condition affecting lower limb function, and the inability to walk without a gait aid. Control participants were aged over 50 years and in general good health. They were excluded if they (1) reported a history of knee pain, injury, or pathology; (2) reported past lower limb surgery that may alter gait patterns; (3) reported any other muscular, joint, or neurologic condition affecting lower limb function; (4) were unable to walk without a gait aid; or (5) reported any condition that affected mobility for more than 1 week in the past year. Ethical approval was obtained from the University of Melbourne Human Research Ethics Committee and from the Department of Human Services Victoria, Radiation Safety Committee. All participants provided written informed consent. Instrumentation and Procedures The study limb was defined as the more painful knee in the OA participants and was randomly assigned for the pain-free control group; the contralateral knee was defined as the nonstudy knee. Standardized semiflexed posteroanterior knee radiographs for each knee were obtained in barefoot standing for participants with knee OA. Radiographic severity of tibiofemoral OA was assessed using the K-L scale by an experienced musculoskeletal researcher and physiotherapist (M.A.H.) whose intrarater reliability was.84 (linearly incremental weighted kappa statistic). According to this system, disease severity is rated on an ordinal scale from 0 to 4 (0 normal; 1 possible osteophytes; 2 minimal osteophytes and possible joint space narrowing [mild OA]; 3 moderate osteophytes, some narrowing, and possible sclerosis [moderate OA]; and 4 large osteophytes, definite joint space narrowing, and severe sclerosis [severe OA]). 25 Anatomic knee alignment was measured using the posteroanterior radiographs. Alignment measured in this manner is strongly correlated with the mechanical axis obtained from a long leg radiograph (r.75; P.001) 24 and avoids the additional cost and radiation associated with a long leg radiograph. A prediction equation was used for conversion. 24 The average pain felt over the last week in the study knee of each participant was assessed with an 11-point Likert scale numbered from 0 to 10 (a higher score indicating worse pain). All participants performed barefoot walking trials at their self-selected, normative walking speed. Speed was monitored using timing gates to ensure that intertrial variation was not 10%. Force plate data were collected from 2 plates at 1080Hz. a Synchronized 3-dimensional kinematic data were collected at 120Hz using a Vicon motion analysis system with eight M2 CMOS cameras. b Reflective markers were placed on the thorax, pelvis, and lower limbs to define the segments of the trunk, pelvis, thigh, shank and foot, and the joint centers of the hip, knee, and ankle. Each participant performed 5 successful walking trials for each limb (trials where the foot landed within the borders of the force plate), and mean data were used for analyses. External joint moments were calculated using inverse dynamic techniques (Vicon Plug-In-Gait v2 b ) and the peak KAM and knee flexion moment values during stance were identified. A joint coordinate system approach 26 was used to determine knee flexion/extension and knee varus/valgus. Trunk lean and toe-out were defined relative to the laboratory coordinate system using the same convention. 26 While peak values for knee flexion during stance and trunk lean toward the ipsilateral limb were selected for analysis, the average knee varus-valgus and toe-out angles from 25% to 50% stance were selected for analysis, as the magnitudes of these variables do not change considerably during this portion of stance. Asymmetry values in these gait characteristics were determined by subtracting the nonstudy knee from the study knee. 27,28 Statistical Analysis Experimental groups were defined based on the absence or presence of knee pain and definite radiographic knee OA (K-L 1): controls (no knee pain); Uni-pain/Uni-xray (pain in 1 knee and radiographic OA in that knee); Uni-pain/Bi-xray (pain in 1 knee, but radiographic OA in both knees); Bi-pain/ Uni-xray (pain in both knees, but radiographic OA in 1 knee); Bi-pain/Bi-xray (pain in both knees and radiographic OA in both knees). Analyses were performed using SPSS for Windows, Version 15. c Data were checked for normality prior to analyses. One-

3 824 GAIT IN UNILATERAL AND BILATERAL KNEE OSTEOARTHRITIS, Creaby way analysis of variance and chi-square analysis were used to evaluate differences in demographics and disease characteristics between groups. To examine between-group differences in discrete values of the study knee, one-way analysis of covariance was used with walking speed included as a covariate. An a priori alpha level of.05 was set for these analyses. Ninetyfive percent confidence intervals (CIs) were used to evaluate between-limb symmetry in each of the experimental groups; 95% CIs not overlapping zero were considered indicative of asymmetry. RESULTS Participant characteristics, stratified by group, are shown in table 1. Of the 91 participants with knee OA, 11 were classified as Uni-pain/Uni-xray, 22 were classified as Unipain/Bi-xray, and 56 as Bi-pain/Bi-xray. Two knee OA participants had bilateral pain but unilateral evidence of knee OA on radiograph and were not considered in our analysis. Groups were similar in age, height, and sex distribution. The Uni-pain/Uni-xray group had a significantly higher body mass index than the other 3 groups; the Bi-pain/ Bi-xray group had a significantly higher body mass index than the control group. Radiographic disease severity for the OA groups is presented in table 2. In the groups with bilateral evidence of OA, 30% and 41% of individuals in the Bi-pain/Bi-xray and Uni-pain/Bi-xray groups, respectively, had less severe radiographic OA in their less painful knee, and around 50% of individuals in these groups had the same radiographic severity in both knees. In the remaining 9% to 18%, the K-L grade was more severe in the less painful knee. Comparison of gait characteristics in the more painful (study) limb between groups did not reveal any difference in the peak KAM after controlling for walking speed (P.05) (table 3). Knee flexion moments, however, were higher in the study limb of the control group compared with the study limb of each of the OA groups (P.001). Trunk lean was greater in each of the OA groups compared with the control group (P.001). Furthermore, the Uni-pain/Bi-xray and Bi-pain/Bixray groups exhibited less peak knee flexion, less toe-out, and a more varus knee angle than the control group (all P.05). In all experimental groups, an overlap with zero of the 95% CIs for peak KAM indicates a similar peak moment in both knees (fig 1). Contrary to this, varus-valgus angle asymmetry in the Uni-pain/Uni-xray and Uni-pain/Bi-xray groups were 5.01 and 2.54, respectively, and 95% CIs did not overlap with zero. This indicates that, on average, the varus angle in the painful knees was greater than in the pain-free knees. The control and Bi-pain/Bi-xray groups, however, demonstrated symmetry in the knee varus-valgus angle. Peak knee flexion angle asymmetry was observed in the Uni-pain/Bi-xray group, indicating less flexion in the painful knee. However, this asymmetry was not observed in any of the other groups. Regarding the peak knee flexion moment, negative asymmetry (indicating higher flexion moments in the painfree limb) was observed in the Uni-pain/Uni-xray group and Uni-pain/Bi-xray group. Positive trunk lean asymmetry was observed in the Uni-pain/Uni-xray and Bi-pain/Bi-xray groups, indicating greater lean toward the more painful knee. No asymmetries were apparent in peak toe-out angle. DISCUSSION Our data illustrate that between-limb asymmetries in gait were principally limited to individuals with unilateral knee pain, and either unilateral or bilateral structural disease, whereas relative symmetry in gait is apparent in those with bilateral pain and in asymptomatic individuals. These findings may have important implications regarding the tailoring of gait intervention strategies on the basis of the symptomatic and structural status of both knees in individuals presenting with medial knee OA. Numerous differences in walking biomechanics have been documented in individuals with symptomatic medial compartment knee OA compared with asymptomatic individuals. 4,7-9 Moreover, evidence of differences in gait between the more and less affected limbs in individuals with OA indicate that asymmetrical gait may be a feature of the disease, 10,11,27 whereas gait appears to be largely symmetrical in asymptomatic individuals. 29,30 Our data provide further support for the presence of symmetrical gait patterns in asymptomatic individuals. This is consistent with our hypotheses, that individuals with no pain, history of pain, or injury in either leg would demonstrate similar biomechanics on both sides. Symmetry was also evident in knee and foot biomechanics for individuals with bilateral pain resulting from bilateral structural knee OA, yet biomechanics of the most affected leg in this group differed substantially from asymptomatic individuals. Thus, it appears that bilateral pain and structural OA are associated with abnormal gait mechanics that equally affect both legs. Therefore, in the prescription of interventions to address the biomechanics of medial knee OA in those with bilateral pain, it would seem prudent to treat both the more and less painful limbs. Interpretation of differences in gait with unilateral knee pain but bilateral evidence of structural knee OA may assist us in further understanding the influence of structural disease, inde- Table 1: Participant Characteristics of the 4 Experimental Groups Characteristics Controls (n 31) Uni-pain/Uni-xray (n 11) Uni-pain/Bi-xray (n 22) Bi-pain/Bi-xray (n 56) Age (y) Height (m) Mass (kg) BMI (kg/m 2 ) * *, Sex, n (%) Men 11 (35) 6 (55) 10 (45) 29 (55) Women 20 (65) 5 (45) 12 (55) 24 (45) Walking pain (/10) * * * Walking speed (m/s) * *, * NOTE. Values are mean SD or as otherwise indicated. Abbreviation: BMI, body mass index. *Different from controls (P.05). Different from Uni-pain/Uni-xray (P.05).

4 GAIT IN UNILATERAL AND BILATERAL KNEE OSTEOARTHRITIS, Creaby 825 Table 2: K-L Disease Severity in the Knees of OA Participants Characteristics Uni-pain/ Uni-xray (n 11) Uni-pain/ Bi-xray (n 22) Bi-pain/ Bi-xray (n 56) Disease severity,* n (%) Grade 1 0 (0) 0 (0) 0 (0) Grade 2 2 (18) 10 (45) 17 (30) Grade 3 3 (27) 9 (41) 17 (30) Grade 4 6 (55) 3 (14) 22 (40) Severity in less symptomatic versus more symptomatic limb, n (%) Less severe 11 (100) 9 (41) 17 (30) Same 0 (0) 11 (50) 29 (52) More severe 0 (0) 2 (9) 10 (18) *K-L disease severity of the more painful limb. pendent of the effect pain may have on OA gait. In the painful limb of this group, the peak knee flexion angle and moment were lower, and knee varus angle was greater compared with controls. Moreover, this group exhibited asymmetrical gait in these same variables, indicating that gait dissimilar to that in the asymptomatic group was only present in the painful limb. This could be interpreted as an indication that a difference in gait from controls is associated with pain rather than with the presence of structural OA, supporting the findings of earlier studies of knee pain manipulation. For example, we observed a lower knee flexion moment in the painful knee compared with controls, and an asymmetry toward a lower moment in the painful knee. Consistent with this, others have demonstrated that the removal of knee OA pain leads to an increase in knee flexion moments, 31 whereas the introduction of experimental knee pain in otherwise healthy individuals leads to a reduction in flexion moments. 32 This suggests that gait with lower flexion moments may be employed as a deliberate pain avoidance strategy in the painful knee, but is not reflected in the pain free knee. The presence of higher joint moments in the pain free knee may accelerate the onset of symptomatic OA in the pain free knee. Thus, in this group, strategies to encourage reduced knee flexion moments in the pain free knee are likely to be desirable. With the exception of trunk lean and the peak knee flexion moment, gait in the affected limb of individuals with unilateral pain and unilateral structural disease was remarkably similar to our healthy control group. This group demonstrated a tendency toward more severe structural disease, which is typically associated with a higher KAM. 4,5 In our population, however, after statistically correcting for differences in walking speed, the KAM did not differ from the control or other OA groups. Other recent studies have also reported no difference in the KAM with OA after controlling for walking speed. 14,15,33 This is still somewhat surprising given that the greater knee varus angle in the OA groups compared with the controls (not significant in the Uni-pain/Uni-xray group) suggests a longer KAM lever arm, which would contribute to a higher peak KAM. We propose that the apparent adoption of a trunk lean gait strategy in the more affected knee of the OA groups may allow the KAM to be normalized in the presence of dynamic knee varus. It has previously been demonstrated that increased lateral trunk lean toward the ipsilateral knee and increased toe-out are related to reductions in peak KAM 19,20,34 and arguably are employed as strategies to minimize pain in those with medial knee OA. 4,15,21 While this appears true for trunk lean (as discussed in the preceding paragraph), our data indicate less, not more, toe-out in participants with bilateral structural OA compared with controls. Theoretically, this would contribute to higher joint loading and thus accelerated structural disease progression in these individuals. Indeed, less toe-out has been associated with greater joint space narrowing in individuals with medial knee OA. 35 The small degree of toe-out in this symptomatic population highlights it as a potentially important target for intervention. Study Limitations The cross-sectional design of this study means it was not possible to determine the causal relationship between gait biomechanics and the uni- or bilateral development of knee OA. Further longitudinal work is required to establish the role of gait asymmetries in disease development and progression. Another limitation associated with this study was the absence of radiographs in the healthy control group, leaving open the possibility that some structural deterioration associated with knee OA may have been present in this group. However, our main finding, that gait asymmetries are associated with the presence of unilateral pain, is dependent on symptoms, not structural disease status, and thus still holds in the absence of radiographs in our control group. CONCLUSIONS Our data indicate that individuals with unilateral knee pain and either unilateral or bilateral structural medial knee OA demonstrate asymmetries in knee biomechanics during walking in the frontal (knee varus angle) and sagittal (peak flexion moment) planes. Contrary to this, individuals with bilateral Gait Characteristics Table 3: Gait Mechanics of the More Painful (Study) Limb of Each Group Controls (n 31) Uni-pain/Uni-xray (n 11) Uni-pain/Bi-xray (n 22) Bi-pain/Bi-xray (n 56) P F Kinematics Average knee varus-valgus ( ) * * Peak knee flexion ( ) * * Average toe-out ( ) * * Average trunk lean ( ) * * * Kinetics Peak KAM (%Bw.Ht.) Peak knee flexion (%Bw.Ht.) * * * NOTE. Data are presented as the estimated marginal mean SE or as otherwise indicated. Abbreviations: %Bw.Ht., percentage body weight body height. *Different from controls (P.05). Difference identified in gait characteristic using one-way analysis of covariance (P.05).

5 826 GAIT IN UNILATERAL AND BILATERAL KNEE OSTEOARTHRITIS, Creaby Peak KAM limb diff (%Bw.Ht.) Varus-Valgus angle limb diff (deg.) Peak Knee Flexion moment limb diff (%Bw.Ht.) Peak Knee Flexion limb diff (deg.) Toe-out limb diff (deg.) Trunk lean limb diff (deg.) Control Uni-pain/Uni-xray Uni-pain/Bi-xray Bi-pain/Bi-xray Fig 1. Asymmetry values for the gait characteristics of each group, presented as the mean difference with 95% CI error bars. Positive values indicate the variable was greater in the more painful (study) limb; negative values indicate the variable was greater in the less painful (study) limb. Abbreviations: diff, difference; %Bw.Ht., percentage body weight body height. pain and structural disease exhibited symmetrical biomechanics at the knee during walking. Thus, interventions addressing the underlying biomechanics of knee OA may be best applied to both knees in patients with bilateral pain. Moreover, our cohort of OA participants did not use an increased toe-out gait strategy. Because increased toe-out is suggested

6 GAIT IN UNILATERAL AND BILATERAL KNEE OSTEOARTHRITIS, Creaby 827 to reduce knee load and pain, this may represent an important target for intervention. Acknowledgments: We thank Fiona McManus, BPhysio(Hons), for assisting with participant recruitment and data collection. References 1. Ledingham J, Regan M, Jones A, Doherty M. Radiographic patterns and associations of osteoarthritis of the knee in patients referred to hospital. Ann Rheum Dis 1993;52: Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991; 9: Zhao D, Banks SA, Mitchell KH, D Lima DD, Colwell CW Jr, Fregly BJ. Correlation between the knee adduction torque and medial contact force for a variety of gait patterns. J Orthop Res 2007;25: Hurwitz DE, Ryals AB, Case JP, Block JA, Andriacchi TP. The knee adduction moment during gait in subjects with knee osteoarthritis is more closely correlated with static alignment than radiographic disease severity, toe out angle and pain. J Orthop Res 2002;20: Sharma L, Hurwitz DE, Thonar EJ, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum 1998;41: Miyazaki T, Wada M, Kawahara H, Sato M, Baba H, Shimada S. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann Rheum Dis 2002;61: Baliunas AJ, Hurwitz DE, Ryals AB, et al. Increased knee joint loads during walking are present in subjects with knee osteoarthritis. Osteoarth Cart 2002;10: Astephen JL, Deluzio KJ, Caldwell GE, Dunbar MJ. Biomechanical changes at the hip, knee, and ankle joints during gait are associated with knee osteoarthritis severity. J Orthop Res 2008; 26: Gok H, Ergin S, Yavuzer G. Kinetic and kinematic characteristics of gait in patients with medial knee arthrosis. Acta Orthop Scand 2002;73: Hunt MA, Birmingham TB, Giffin JR, Jenkyn TR. Associations among knee adduction moment, frontal plane ground reaction force, and lever arm during walking in patients with knee osteoarthritis. J Biomech 2006;39: Briem K, Snyder-Mackler L. Proximal gait adaptations in medial knee OA. J Orthop Res 2009;27: Al-Zahrani KS, Bakheit AM. A study of the gait characteristics of patients with chronic osteoarthritis of the knee. Disabil Rehabil 2002;24: Weidow J, Tranberg R, Saari T, Kärrholm J. Hip and knee joint rotations differ between patients with medial and lateral knee osteoarthritis: gait analysis of 30 patients and 15 controls. J Orthop Res 2006;24: Zeni JA Jr, Higginson JS. Differences in gait parameters between healthy subjects and persons with moderate and severe knee osteoarthritis: a result of altered walking speed? Clin Biomech (Bristol, Avon) 2009;24: Mundermann A, Dyrby CO, Hurwitz DE, Sharma L, Andriacchi TP. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity: reduced walking speed. Arthritis Rheum 2004;50: Andrews M, Noyes FR, Hewett TE, Andriacchi TP. Lower limb alignment and foot angle are related to stance phase knee adduction in normal subjects: a critical analysis of the reliability of gait analysis data. J Orthop Res 1996;14: Jenkyn TR, Hunt MA, Jones IC, Giffin JR, Birmingham TB. Toe-out gait in patients with knee osteoarthritis partially transforms external knee adduction moment into flexion moment during early stance phase of gait: a tri-planar kinetic mechanism. J Biomech 2008;41: Wang J, Kuo K, Andriacchi T, Galante J. The influence of walking mechanics and time on the results of proximal tibial osteotomy. J Bone Joint Surg Am 1990;72: Hunt MA, Birmingham TB, Bryant D, et al. Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarth Cart 2008;16: Mundermann A, Asay JL, Mundermann L, Andriacchi TP. Implications of increased medio-lateral trunk sway for ambulatory mechanics. J Biomech 2008;41: Mundermann A, Dyrby CO, Andriacchi TP. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum 2005;52: Gunther KP, Sturmer T, Sauerland S, et al. Prevalence of generalised osteoarthritis in patients with advanced hip and knee osteoarthritis: the Ulm Osteoarthritis Study. Ann Rheum Dis 1998;57: Bennell KL, Hunt MA, Wrigley TV, et al. Hip strengthening reduces symptoms but not knee load in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Osteoarth Cart 2010;18: Kraus VB, Vail TP, Worrell T, McDaniel G. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum 2005;52: Kellgren JH, Lawrence JS. Radiological assessment of osteoarthrosis. Ann Rheum Dis 1957;16: Grood ES, Suntay WJ. A joint coordinate system for the clinical description of three-dimensional motions: application to the knee. J Biomech Eng 1983;105: Bejek Z, Paroczai R, Illyes A, Kiss RM. The influence of walking speed on gait parameters in healthy people and in patients with osteoarthritis. Knee Surg Sports Traumatol Arthrosc 2006;14: Mockel G, Perka C, Labs K, Duda G. The influence of walking speed on kinetic and kinematic parameters in patients with osteoarthritis of the hip using a force-instrumented treadmill and standardised gait speeds. Arch Orthop Trauma Surg 2003;123: Lythgo N, Wilson C, Galea M. Basic gait and symmetry measures for primary school-aged children and young adults. II: walking at slow, free and fast speed. Gait Posture 2011;33: Teichtahl AJ, Wluka AE, Morris ME, Davis SR, Cicuttini FM. The associations between the dominant and nondominant peak external knee adductor moments during gait in healthy subjects: evidence for symmetry. Arch Phys Med Rehabil 2009;90: Hurwitz DE, Ryals AR, Block JA, Sharma L, Schnitzer TJ, Andriacchi TP. Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 2000;18: Henriksen M, Graven-Nielsen T, Aaboe J, Andriacchi TP, Bliddal H. Gait changes in patients with knee osteoarthritis are replicated by experimental knee pain. Arthritis Care Res 2010;62: Heiden TL, Lloyd DG, Ackland TR. Knee joint kinematics, kinetics and muscle co-contraction in knee osteoarthritis patient gait. Clin Biomech (Bristol, Avon) 2009;24: Hunt MA, Simic M, Hinman RS, Bennell KL, Wrigley TV. Feasibility of a gait retraining strategy for reducing knee joint loading: increased trunk lean guided by real-time biofeedback. J Biomech 2011;44: Chang A, Hurwitz D, Dunlop D, et al. The relationship between toe-out angle during gait and progression of medial tibiofemoral osteoarthritis. Ann Rheum Dis 2007;66: Suppliers a. Advanced Mechanical Technology Inc, 176 Waltham St, Watertown, MA b. Vicon, 14 Minns Business Park, West Way, Oxford, OX2 0JB, UK. c. IBM Corp, New Orchard Rd, Armonk, NY