Lateral wedge insoles for medial knee osteoarthritis: Effects on lower limb frontal plane biomechanics

Clinical Biomechanics 27 (2012) 27–33

Contents lists available at ScienceDirect

Clinical Biomechanics j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n b i o m e c h

Lateral wedge insoles for medial knee osteoarthritis: Effects on lower limb frontal plane biomechanics Rana S. Hinman ⁎, Kelly Ann Bowles, Ben B. Metcalf, Tim V. Wrigley, Kim L. Bennell Centre for Health, Exercise and Sports Medicine, Physiotherapy, School of Health Sciences, The University of Melbourne, Victoria, Australia

a r t i c l e

i n f o

Article history: Received 1 November 2010 Accepted 19 July 2011 Keywords: Osteoarthritis Knee Orthotics Biomechanics Load Adduction moment

a b s t r a c t Background: Lateral wedges reduce the peak knee adduction moment and are advocated for knee osteoarthritis. However some patients demonstrate adverse biomechanical effects with treatment. Clinical management is hampered by lack of knowledge about their mechanism of effect. We evaluated effects of lateral wedges on frontal plane biomechanics, in order to elucidate mechanisms of effect. Methods: Seventy three participants with knee osteoarthritis underwent gait analysis with and without 5° lateral wedges. Frontal plane parameters at the foot, knee and hip were evaluated, including peak knee adduction moment, knee adduction angular impulse, center of pressure displacement, ground reaction force, and knee-ground reaction force lever arm. Findings: Lateral wedges reduced peak knee adduction moment and knee adduction angular impulse (− 5.8% and − 6.3% respectively, both P b 0.001). Although reductions in peak moment were correlated with more lateral center of pressure (r = 0.25, P b 0.05), less varus malalignment (r values 0.25–0.38, P b 0.05), reduced knee-ground reaction force lever arm (r = 0.69, P b 0.01), less hip adduction (r = 0.24, P b 0.05) and a more vertical frontal plane ground reaction force vector (r = 0.67, P b 0.001), only reduction in knee-ground reaction force lever arm was significantly predictive in regression analyses (B = 0.056, adjusted R 2 = 0.461, P b 0.001). Interpretation: Lateral wedges significantly reduce peak knee adduction moment and knee adduction angular impulse. It seems a reduced knee-ground reaction force lever arm with lateral wedges is the central mechanism explaining their load-reducing effects. In order to understand why some patients do not respond to treatment, future evaluation of patient characteristics that mediate wedge effects on this lever arm is required. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Knee osteoarthritis (OA) is one of the most common musculoskeletal disorders in the world, and a leading cause of knee pain and disability in the elderly population. Knee OA typically affects the medial tibiofemoral joint compartment (Ledingham et al., 1993), probably because of the greater load applied to this compartment (relative to the lateral) during walking and other weightbearing activities (Andriacchi and Mundermann, 2006; Schipplein and Andriacchi, 1991). As there is no cure for knee OA, and arthroplasty is typically reserved for end-stage disease, conservative treatment is the mainstay for managing this condition. While relief of symptoms is a major aim, an increasingly important priority is to minimise the risk of disease progression over time. As excessive knee load is a significant risk factor for increased risk of structural progression (Miyazaki et al., 2002), treatment strategies to reduce load constitute ⁎ Corresponding author at: Centre for Health Exercise and Sports Medicine, School of Physiotherapy, The University of Melbourne, Parkville, Victoria, 3010, Australia. E-mail address: [email protected] (R.S. Hinman). 0268-0033/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2011.07.010

a logical option. Lateral wedge shoe insoles are one such treatment option frequently recommended in clinical guidelines (Zhang et al., 2008) for the management of medial knee OA. Lateral wedges are orthotic devices placed within the shoes. During gait, the ground reaction force (GRF) vector passes medially to the knee joint center, creating an external knee adduction moment throughout stance. In people with knee OA, lateral wedges have been shown to reduce the peak knee adduction moment, an indirect biomechanical surrogate for medial compartment knee load, by approximately 4–12% on average (Butler et al., 2007; Hinman et al., 2008a, 2008b; Kerrigan et al., 2002; Kakihana et al., 2005; Kuroyanagi et al., 2007; Shimada et al., 2006). Limited literature is available regarding the effects of lateral wedges directly on medial compartment load, however mathematical modelling has shown that they can reduce medial compartment contact force by 3% in healthy individuals (Crenshaw et al., 2000). Unfortunately, clinical trials have failed to demonstrate a significant impact of lateral wedges on symptoms of disease (Baker et al., 2007; Barrios et al., 2009; Bennell et al., 2011; Maillefert et al., 2001; Pham et al., 2003). Our data and that of others suggest that approximately 13–18% of people with knee OA actually

28

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33

have an adverse biomechanical response to lateral wedges (Hinman et al., 2008b; Kakihana et al., 2005, 2007), which may explain the disappointing findings of clinical trials. These data suggest there may be a mechanical limitation of lateral wedges in some patients with knee OA. Before research can be undertaken to identify and classify this sub-group of non-responders, a greater understanding of the mechanical mechanism of effect underpinning the reduced knee adduction moment with lateral wedges in knee OA is required. Little is known about effects of lateral wedges at joints other than the knee (Butler et al, 2009). Initial propositions for laterally wedged insoles for medial knee OA suggested that wedges improved mechanical alignment of the lower limb (Sasaki and Yasuda, 1987; Yasuda and Sasaki, 1987). Although subsequent studies have failed to demonstrate a significant change in static lower limb alignment with lateral wedges (Hinman et al., 2008b; Maly et al., 2002), it is possible that a change in dynamic knee alignment may occur during gait. This has not adequately been investigated in people with knee OA. Research also suggests that lateral wedges may laterally shift the center of pressure of the GRF, reducing its adduction lever arm at the knee (and hence the external knee adduction moment), however findings across studies are inconsistent at present and are limited by generally small sample sizes (Kakihana et al., 2004, 2005, 2007; Maly et al., 2002). In fact, no study has evaluated the effect of lateral wedges on the knee lever am in the frontal plane. It is also possible that lateral wedge insoles may affect the magnitude/orientation of the resultant frontal plane GRF vector at times of peak medial compartment loading (which would also alter the knee adduction moment). Lateral wedge insoles could have this effect by influencing the pattern of vertical and medio-lateral center of mass acceleration during stance. Understanding mechanical mechanisms of effect with lateral wedges is important as this information will help guide future research to determine which patient subgroups may best benefit from treatment (or conversely, be unlikely to benefit), as well as highlight potential adverse effects at the foot and/or hip. The aim of this study was to evaluate the immediate effects of lateral wedge insoles on lower limb frontal plane biomechanics in a large cohort with medial knee OA, in order to elucidate mechanisms underpinning their knee unloading effects. It was hypothesised that lateral wedge insoles would result in changes in both lower limb alignment and in the orientation and magnitude of the GRF relative to the knee at the time of peak knee adduction moment, and that these changes would correlate with reductions in the peak knee adduction moment.

2. Methods 2.1. Participants Participants in this study comprised a subset of volunteers enrolled in a 12 month randomised controlled trial of lateral wedges, recruited from the community by advertisements (Bennell et al., 2007). Seventy three people aged over 50 years with medial compartment knee OA participated in this study. Selection criteria were necessarily based on those used for the larger randomised controlled trial. Participants were included if they reported knee pain on most days of the previous month and demonstrated medial tibiofemoral osteophytes on X-ray (Altman et al., 1986). Other inclusion criteria were an average knee pain N3 on an 11-point Likert scale when walking. Exclusion criteria included questionable or severe radiographic disease (Kellgren & Lawrence Grade 1 or 4); valgus knee alignment N185° on a standardised standing knee X-ray (Kraus et al., 2005); use of a gait aid; lateral tibiofemoral compartment joint space narrowing greater than medial; body mass index ≥ 36 kg/m 2; hip or knee joint replacement; knee surgery or injection (past 6 months); use of insoles or foot orthotics (past

6 months), foot or ankle problem precluding use of insoles and; usual footwear incompatible with insoles. In this study, only the symptomatic knee was tested. In the case of participants with bilateral eligible knees, the more symptomatic was deemed the test limb. Testing for this study was performed as part of the baseline assessment for randomised controlled trial. The University of Melbourne Human Research Ethics Committee approved the study and all participants provided written informed consent. 2.2. Lateral wedge insoles Participants underwent 3D gait analysis, both with and without, a new pair of lateral wedge insoles. Testing occurred in randomised order. Standardised non-customised lateral wedge insoles made of high-density ethyl vinyl acetate were used, and sized according to shoe size. They were wedged approximately 5° (as greater wedging has been associated with foot discomfort (Kerrigan et al., 2002)) and were worn bilaterally inside the participant's own shoes. The insoles were wedged along the lateral edge of the entire length of the foot, as we have previously shown full-length wedges to be more effective at reducing the knee adduction moment than heel wedges (Hinman et al., 2008a). A Shore Durometer type A reading was taken on 20 samples of the wedge material from different batches of material used for the larger clinical trial and the mean hardness was found to be 57.5 (+/−2.5) units. 2.3. Gait analysis A Vicon motion analysis system with six-eight M2/MX cameras operating at 120 Hz (Vicon, Oxford, UK) was used to measure lower limb frontal plane kinematics and kinetics. The standard Plug-in-Gait marker set was used (anterior superior iliac spine, posterior superior iliac spine, mid-lateral thigh, lateral knee joint, lateral shank, lateral malleolus, on the shoe over the second metatarsal head and over the posterior calcaneus). Additional medial knee and ankle markers were used during the single static standing trial to assist in determining the knee and ankle joint flexion-extension axes, halfway along which the respective joint centers were placed. GRFs were measured by two 0R6-6-2000 force plates (Advanced Mechanical Technology Inc., Watertown, MA) embedded in the floor at the midpoint of a 10 m walkway at 1080 Hz, in synchrony with the cameras. Participants walked at their usual comfortable pace and data were collected from 5 successful trials for each test condition. Participants were not informed about the embedded force plates to prevent them “targeting” the plates and thus altering their gait pattern. Several practice trials ensured that participants walked naturally and landed the whole foot of the test limb on the force plate. Walking speed was monitored by two photoelectric beams and verbal feedback ensured that speed during each trials varied not more than 10% from the average speed of the first. A successful trial was that in which the participant walked naturally, landed the whole foot of the test limb on the force plate and where speed did not vary by more than 10% of the first. Net external joint moments were calculated via inverse dynamics (Vicon Plug-In-Gait v1.9). Joint moments were normalised for body weight and height and reported in Nm/BW*HT%. A custom-written Body Builder program (Vicon, Oxford, UK) was used to calculate additional variables related to the frontal plane biomechanics of the foot, knee and hip, in particular the GRF-to-knee lever arm, position of the center of pressure in relation to the foot, and certain knee angles. The variables of interest for this study are defined in Table 1. All discrete variables were averaged over the 5 trials for each test condition. Test–retest reliability for measuring the peak knee adduction moment in our laboratory is excellent (intra-class correlation coefficients (ICC(3,5)) of 0.92–0.97, in eleven elderly patients with knee pain tested one week apart).

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33 Table 1 Biomechanical variables of interest. Variable

Definition

Peak external knee adduction moment in first half of stance Positive area under the knee adduction moment-time graph. This measure incorporates both the mean magnitude of the (positive) moment and the time for which it is imposed on the knee a Center of pressure offset Distance of the centre of pressure from the line of the (mm) foot (calcaneus to 2nd metatarsal), where positive values indicate medial offset Foot progression angle (°) Angle between foot long axis (ankle joint center to 2nd metatarsal) and laboratory forward axis, where negative values indicate toe out (averaged from 2550% stance) Base of support (mm) Perpendicular distance between the ankle joint center of the test leg in stance and the straight line between the ankle joint center of the non-test leg in the previous and subsequent steps. Knee varus-valgusa (°) Varus–valgus angle calculated as first Euler-Cardan angle for shank with respect to thigh (equivalent to shank varus–valgus angle projected on thigh coordinate system), where positive values indicate varus Hip-knee-ankle anglea (°) Angle formed from hip-knee-ankle centers, in laboratory frontal plane, where positive values indicate varus a Knee-GRF lever arm (mm) Perpendicular distance between GRF and knee joint center in laboratory frontal plane Femur anglea (°) Angle of hip-knee center vector to vertical in laboratory frontal plane, where positive values indicate varus. Tibia anglea (°) Angle of knee-ankle center vector in laboratory frontal plane, where positive values indicate varus. Peak hip adduction (°) Peak hip adduction angle during stance phase Total hip adduction Absolute range of hip adduction during stance phase. excursion (°) Peak hip adduction Peak external hip adduction moment during stance moment (Nm/BW*HT%) phase GRF anglea (°) Angle of GRF vector in laboratory frontal plane GRF magnitudea (N) Resultant magnitude of GRF in laboratory frontal plane Peak knee adduction moment (Nm/BW*HT%) Knee adduction angular impulse (Nm.s/BW*HT%)

a

At time of first peak knee adduction moment; GRF = ground reaction force.

2.4. Other measures Radiographic disease severity was assessed using the Kellgren and Lawrence system (Kellgren et al., 1963). The Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index was used to assess knee pain (score range 0–20, higher scores indicating worse pain) and impairment in physical function (score range 0–68, higher scores indicating worse function) (Bellamy et al., 1988).

29

3. Results Characteristics of the study cohort are described in Table 2. There were more women than men involved in the study, and participants demonstrated mild-moderate radiographic and symptomatic disease severity, as described by the Kellgren and Lawrence system and the WOMAC. The effect of lateral wedges on the selected frontal plane biomechanical parameters is summarised in Table 3. A significant reduction in both the peak knee adduction moment and the knee adduction angular impulse were observed when lateral wedges were inserted into the shoes (− 5.8% and − 6.3% respectively). At the foot, insertion of lateral wedge insoles resulted in a significant lateral shift in the center of pressure, an increase in toeout angle and a slightly wider base of support. At the knee, insertion of lateral wedge insoles resulted in a less varus knee angle, a shorter knee-GRF lever arm, as well as a more vertical femur in the frontal plane. There was no significant change in the angle of the tibia in the frontal plane. At the hip, insertion of lateral wedge insoles resulted in a significant increase in peak adduction. There was no significant change in either the total hip adduction excursion or in the peak hip adduction moment. Regarding the frontal plane GRF, insertion of lateral wedge insoles resulted in a more vertically orientated GRF vector but its actual magnitude in the frontal plane remained unchanged. Gait speed did not differ across test conditions, with the mean (SD) velocity in the natural state recorded as 1.26 m/s (0.19) compared to 1.25 m/s (0.18) when walking with wedges (P = 0.43). Correlations between changes in biomechanical variables with lateral wedges are reported in Table 4. Although most relationships were significant, stepwise regression analysis demonstrated that only change in knee-GRF lever arm significantly predicted change in the peak knee adduction moment (B = 0.056, SE = 0.007, P b 0.001), explaining 46.1% of the variance (adjusted R 2 = 0.461). All other variables were excluded from the model. Fig. 1 graphically depicts the individual relationship between the % change in knee-GRF lever arm and the % change in the peak knee adduction moment across individuals in the cohort. This individual data demonstrates not only the relatively consistent relationship between a decrease (or increase) in the knee-GRF lever arm and a decrease (or increase) in the peak knee adduction moment, but also the relatively variable magnitude of response in these parameters across individuals. 4. Discussion This study evaluated the immediate effects of lateral wedge insoles on lower limb frontal plane biomechanics in a large cohort with medial knee OA. Findings from our study confirmed those of Table 2 Presenting characteristics of the cohort. Values are reported as mean (SD) unless otherwise indicated.

2.5. Statistical analysis

Age (y) Gender, n (%)

Statistical analyses were performed using PASW Statistics 18 (SPSS Inc. Chicago, Illinois, USA) and an alpha level of 0.05. Data were checked for normality prior to analysis and as most data were normally distributed, parametric tests were used. The effects of lateral wedge insoles on biomechanical variables were evaluated using paired t tests. For those variables that were significantly different with lateral wedges, relationships between their mean change and the mean change in the peak knee adduction moment were evaluated using Pearson r correlation coefficients. Those demonstrating a significant correlation were then subjected to stepwise regression (probability of entry = 0.05 and probability of removal = 0.10), with change in peak knee adduction moment as the dependent variable.

Height (m) Body mass (kg) Body mass index (kg/m2) WOMACa Pain subscale Physical function subscale Osteoarthritis severityb, n (%) Knee alignmentc (°) a

63.3 (8.4) 45 (62) female 28 (38) male 1.67 (0.09) 77.2 (14.5) 27.7 (3.6) 7 (3) 25 (13) 41 (56) Grade 2 32 (44) Grade 3 180.9 (2.6)

Western Ontario and MacMaster Universities Osteoarthritis Index, where higher scores indicate worse symptoms. b Radiographic severity according to Kellgren and Lawrence grading system, where higher grades indicate more severe disease. c Measured from a standing knee-X-ray, where a score of 184.0° corresponds to neutral alignment and higher scores indicate valgus and lower scores indicate varus.

30

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33

Table 3 Change in biomechanical parameters with insertion of lateral wedges.

Peak knee adduction moment (Nm/BW*HT%) Knee adduction angular impulse (Nm.s/BW*HT%) Center of pressure offseta (mm) Foot progression angle (°) Base of support (mm) Knee varus–valgusa (°) Hip-knee-ankle anglea (°) Knee-GRF lever arma (mm) Femur anglea (°) Tibia anglea (°) Peak hip adduction (°) Total hip adduction excursion (°) Peak hip adduction moment (Nm/BW*HT%) GRF anglea (°) GRF magnitudea (N) a b c

No wedges

Wedges

Mean (SD)

Mean (SD)

3.82 (0.78) 1.26 (0.37) − 5.6 (4.3) − 6.06 (5.56) 89.0 (24.8) 2.11 (4.94) 1.91 (4.59) 53.14 (11.89) − 3.54 (3.35) 5.45 (2.19) 6.37 (4.35) 30.17 (11.12) 6.02 (1.25) 6.19 (1.45) 840.5 (174.5)

3.60 (0.75) 1.18 (0.38) − 9.1 (4.6) − 7.78 (5.78) 93.5 (26.3) 1.63 (4.94) 1.47 (4.58) 50.20 (12.03) − 3.97 (3.41) 5.44 (2.16) 6.81 (4.16) 32.27 (13.67) 6.07 (1.16) 5.79 (1.47) 837.0 (174.7)

Mean difference (95% CI) b

Changec

P valuea

− 0.22 (− 0.16, −0.29) − 0.08 (− 0.06, −0.10) − 3.4 (− 2.8, −4.1) − 1.72 (− 1.34, −2.10) 4.5 (8.4, 0.6) − 0.48 (− 0.27, −0.68) − 0.45 (− 0.30, −0.59) − 2.94 (− 2.11, −3.77) − 0.43 (− 0.23, −0.62) − 0.02 (− 0.19, 0.16) 0.44 (0.68, 0.19) 2.10 (− 0.47, 4.67) 0.05 (− 0.04, 0.14) − 0.40 (− 0.30, −0.50) − 3.5 (− 10.6, 3.7)

− 5.8% − 6.3% − 60.7% − 28.4% 5.1% − 22.7% − 23.6% − 5.5% − 12.1% − 0.4% 6.9% 7.0% 0.8% − 6.5% − 0.4%

b0.001 b0.001 b0.001 b0.001 0.024 b0.001 b0.001 b0.001 b0.001 0.85 0.001 0.11 0.27 b0.001 0.34

At time of first peak knee adduction moment; GRF = ground reaction force. Calculated as no wedges condition minus wedges condition. Calculated as mean difference divided by mean score with no wedges, multiplied by 100.

numerous others that lateral wedge insoles significantly reduce the peak knee adduction moment in people with knee OA (Butler et al., 2007; Hinman et al., 2008a, 2008b; Kerrigan et al., 2002; Kakihana et al., 2005; Kuroyanagi et al., 2007; Shimada et al., 2006). Importantly, in contrast to most previous research, our study also sought to elucidate mechanisms underpinning the unloading effects of wedges. We have demonstrated logical and consistent biomechanical effects of lateral wedges at the foot and knee (and to a lesser extent at the hip), whereby the primary mechanism of action is related to a reduced frontal plane knee-GRF lever arm. The consistency of this relationship across individuals in our large cohort supports the notion that lateral wedges are a credible mechanical intervention for reducing medial compartment knee load. This is of particular relevance in this patient population where reduction of load is an important priority in order to reduce risk of structural disease progression over time (Miyazaki et al., 2002), and potentially delay total joint replacement. Our observed mean reduction in the peak knee adduction moment by 5.8% is consistent with the findings of others who have also evaluated a 5° lateral wedge in people with knee OA (Butler et al., 2007; Hinman et al., 2008a, 2008b; Kerrigan et al., 2002; Kakihana et al., 2005; Kuroyanagi et al., 2007; Shimada et al., 2006). No other research group has reported the effect of lateral wedges on the knee adduction angular impulse, precluding comparison of findings. Our finding that this parameter reduced on average by 6.3% is consistent

Table 4 Univariate correlations between change in peak knee adduction moment and change in other measured biomechanical variables. Change in…

Change in peak knee adduction moment (Nm/BW*HT%)

Knee adduction angular impulse (Nm.s/BW*HT%) Center of pressure offseta (mm) Foot progression angle (°) Base of support (mm) Knee varus–valgusa (°) Hip-knee-ankle anglea (°) Knee-GRF lever arma (mm) Femur anglea (°) Peak hip adduction (°) GRF anglea (°)

0.42** 0.25* − 0.08 0.01 0.38** 0.25* 0.69** − 0.02 0.24* 0.67**

*P b 0.05. **P b 0.01. a At time of first peak knee adduction moment; GRF = ground reaction force.

with the magnitude of change in the peak moment, a finding that is not surprising given that its calculation effectively incorporates the mean magnitude of the moment, and that our data demonstrated a significant correlation between these measures of medial knee joint load (Table 4). Few studies have evaluated the effects of lateral wedged insoles on joints other than knee. In separate cohorts of 51 and 13 people with knee OA, one research group (Kakihana et al., 2005, 2007) have evaluated a 6° lateral wedge taped to the sole of the barefoot on the frontal plane angles and moments at the knee and subtalar joints, and on center of pressure. Compared to no wedge, a significant increase in the subtalar joint valgus moment was observed. Similar to our findings, the authors observed a significant shift laterally in the center of pressure with lateral wedges, which resulted in an increased subtalar joint moment arm. No significant changes in knee alignment were observed. Whilst significant concomitant reductions in the peak knee adduction moment were also noted (5.6–6.0% change), the authors did not statistically analyse the relationship between changes at the subtalar joint and in the knee adduction moment with wedges. Similar to our present findings the authors observed diverse, sometimes reversed, effects with the lateral wedges among their OA cohort, suggesting that individual patient characteristics may mediate insole effectiveness. This particular variability in response to lateral wedges appears unique to people with knee OA. Testing in healthy cohorts (Kakihana et al., 2004, 2005, 2007) has shown that all healthy people demonstrate both a lateral shift in center of pressure and a reduction in knee adduction moment with lateral wedges. It is possible that the variability in response in knee OA is related to foot posture, as recent research has highlighted that people with knee OA have a more pronated foot type compared to those without OA (Levinger et al., 2010). Severity of knee malalignment may also influence response. Characteristics of shoes worn by individual participants may also mediate effect, such as shoes containing a medial arch support (Nakajima et al., 2009) or shoes that are worn down on one side more than the other. More recently, a study involving 30 people with medial knee OA examined the effect of lateral wedge insoles that were customised to the individual (wedging ranged from 5 to 15°) on rearfoot and hip frontal plane mechanics (Butler et al., 2009). In contrast to the present study, participants wore the insoles in a standardised pair of walking shoes and progressively increased use of the insoles over a week prior to undergoing 3D gait analysis with the insoles. Peak eversion, eversion excursion and peak eversion moment were increased with

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33

31

Fig. 1. Individual % change (y-axis) in the knee-GRF lever arm (circles) and in the peak knee adduction moment (columns) with lateral wedge insoles across the 73 individual participants (x-axis).

lateral wedges, whilst a reduction in the peak knee adduction moment (10%) was observed. Interestingly, no significant relationship between the changes in these variables could be demonstrated. This may reflect an insufficient sample size given that the relationship between change in rearfoot moment and change in the knee moment approached significance (r = −0.35, P = 0.06). In contrast to this study, we observed a significant increase in peak hip adduction with lateral wedges. This is probably because of our larger sample size. In contrast to our findings, a small study in only 9 people with medial knee OA failed to find a relationship between the knee adduction moment and center of pressure excursion in the mediolateral plane (Maly et al., 2002). It is unclear whether this finding is

related to the small sample size, or the nature of the wedge tested. The insole was edged only at the heel and was actually ineffective at reducing the knee adduction moment. We have previously shown that such an insole is sub-optimal for reducing knee load in comparison to an insole wedged along the full length of the foot (Hinman et al., 2008a). This is the first study to evaluate a range of frontal plane biomechanical parameters at the foot, knee and hip in order to elucidate the mechanism of effect of lateral wedges in people with knee OA. Our data, in combination with the limited findings of others (Butler et al., 2009; Kakihana et al., 2004, 2005, 2007), provide considerable insight into the likely mechanism of effect of lateral

More adducted hip Femur more vertical

Reduced external knee adduction moment (KAM)

Reduced varus angle

Reduced kneeGRF lever arm

Knee-GRF lever arm Ground reaction force (GRF)

Lateral wedge Laterally shifted center of pressure (COP) Fig. 2. Mechanism of action of lateral wedged insoles.

Laterally shifted, more vertical GRF

32

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33

wedges in people with knee OA (Fig. 2). It appears that lateral wedges result in a lateral shift in the center of pressure at the foot (which in turn would likely lead to an increased subtalar eversion moment). This shifts the frontal plane GRF vector towards the knee joint center. Together with the more vertically oriented GRF that we observed, both serve to reduce the knee-GRF lever arm. Accordingly, a reduction in the peak knee adduction moment is observed (despite no change in the frontal plane GRF magnitude). Although a reduction in the dynamic varus malalignment of the knee was also seen, the unchanged tibial alignment suggests that this must result from a proximal adaptation, which was observed in the more vertically aligned femur. This in turn is consistent with the mean increased peak hip adduction also noted. Such a straightening of the femur in the frontal plane would require a lateral displacement of the pelvis (and center of mass), which could explain the more vertically oriented frontal plane GRF that we measured. Thus it seems that the reduced knee lever arm observed with lateral wedges results primarily from changes in the frontal plane orientation and location of the GRF vector relative to the knee center, rather than changes in the location of the knee center relative to the GRF. It must be recognised that many of our observed biomechanical changes were small, and the clinical relevance of such small changes is unknown at present. Future studies are needed to determine the magnitude of change required in biomechanical parameters to influence symptoms and/or disease progression. An understanding of the mechanical mechanism of effect of lateral wedges is important as this information has implications for future research evaluating patient sub-groups most likely (and unlikely) to respond to treatment with wedges. For most patients, a significant reduction in the knee-GRF lever arm is required for lateral wedges to be effective in reducing medial knee load. For this to happen, a lateral shift in the center of pressure at the foot is typically needed (by approximately 3.4 mm). The relationship between shift in center of pressure and change in knee-GRF lever arm does not appear to be absolute however, as of the 8/73 participants in our study who had a medial shift in center of pressure with wedges, 5 still managed to reduce their knee lever arm. This highlights the fact that change in knee-GRF lever arm with wedges explains only 46% of the variability in change in the peak adduction moment and suggests that factors other than those evaluated in this study also play a role. It remains unclear why a small but considerable proportion of people demonstrate reverse effects of lateral wedges. In our study, 23% of participants (17/73) increased their knee adduction moment with lateral wedges, which is consistent with previous research (Hinman et al., 2008b; Kakihana et al., 2005, 2007). However, variability in response to orthotic intervention is not unique to people with knee OA and has in fact also been observed in healthy individuals (Nigg et al., 2003; Stacoff et al., 2000), which suggests that anatomical differences or variations in neuromotor activity may mediate response to orthotics across individuals. Future research is required in knee OA to identify and characterise non-responders to lateral wedges. Given that factors such as radiographic disease severity and knee alignment do not appear to mediate biomechanical effect (Hinman et al., 2008b; Kakihana et al., 2007), characteristics of static and dynamic foot posture and changes in muscle activity with wedges may warrant future investigation. It is also possible that individual changes in gait pattern (such as increased toe out) may lead to variation in response to lateral wedges. In conjunction with the findings of others (Butler et al., 2009; Kakihana et al., 2005), our data also highlight potential sources of adverse effects of lateral wedges. Although subtalar joint moments were not evaluated in our study, the significant lateral shift in the center of pressure observed with lateral wedges likely translates to an increase in the eversion moment at this joint. It is thus possible that lateral wedges could adversely affect patients with foot and/or ankle pathology. Careful screening for foot and/or ankle pathology is

required prior to prescribing lateral wedges and patients using lateral wedges should be regularly monitored against the development of foot and/or ankle pain. Although lateral wedges cause an increase in peak hip adduction, the small nature of the change (mean 0.4°), taken in conjunction with the unchanged hip adduction moment, suggests that this is probably not of clinical importance. The major strength of our study is its large cohort of people with painful medial knee OA (n = 73), far larger than any previous biomechanical study of lateral wedges. This enabled us to evaluate a range of frontal plane biomechanical parameters at the foot, knee and hip in order to explore relationships between change in variables with lateral wedges. In order to maximise generalisability, we permitted participants to wear their own usual footwear when using the lateral wedges. Although a strength, this too could also be considered a limitation as the variability in footwear across our cohort may have contributed somewhat to the variability in response to wedges. Further research is needed to determine how different shoe types influence the effects of lateral wedges. A limitation of our study is the lack of frontal plane ankle/subtalar kinetic and kinematic data. It was not possible to measure these parameters because of the limitations of the current Plug-In-Gait model at the ankle. In conclusion, this study aimed to evaluate the immediate effects of lateral wedge insoles on lower limb frontal plane biomechanics in a large cohort with medial knee OA, in order to elucidate mechanisms underpinning their knee unloading effects. Findings demonstrated a significant reduction in peak knee adduction moment and knee adduction angular impulse with lateral wedges. Although lateral wedges significantly changed many other frontal plane parameters, it would seem that a reduction in knee-GRF lever arm with lateral wedges is the central mechanism explaining the load-reducing effect of lateral wedges. Acknowledgements This study was supported by funding from the National Health & Medical Research Council of Australia (Project Grant #61788). The study sponsors had no role in the study design, in the collection, analysis and interpretation of data: in the writing of the manuscript; nor in the decision to submit the manuscript for publication. References Altman, R., Asch, E., Bloch, D., Bole, G., Borenstein, D., Brandt, K., et al., 1986. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Arthritis Rheum 29, 1039–1049. Andriacchi, T.P., Mundermann, A., 2006. The role of ambulatory mechanics in the initiation and progression of knee osteoarthritis. Curr. Opin. Rheumatol. 18, 514–518. Baker, K., Goggins, J., Xie, H., Szumowski, K., Lavalley, M., Hunter, D., et al., 2007. A randomized crossover trial of a wedged insole for treatment of knee osteoarthritis. Arthritis Rheum. 56, 1198–1203. Barrios, J.A., Crenshaw, J.R., Royer, T.D., Davis, I.S., 2009. Walking shoes and laterally wedged orthoses in the clinical management of medial tibiofemoral osteoarthritis: a one-year prospective controlled trial. Knee 16, 136–142. Bellamy, N., Buchanan, W.W., Goldsmith, C.H., Campbell, J., Stitt, L.W., 1988. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J. Rheumatol. 15, 1833–1840. Bennell, K.L., Bowles, K., Payne, C., Cicuttini, F., Osborne, R., Harris, A., et al., 2007. Effects of laterally wedged insoles on symptoms and disease progression in medial knee osteoarthritis: a protocol for a randomised, double-blind, placebo controlled trial. BMC Musculoskel Dis. 8, 96. Bennell, K.L., Bowles, K., Payne, C., Cicuttini, F., Williamson, E., Forbes, A., et al., 2011. Lateral wedge insoles for medial knee osteoarthritis: 12 month randomised controlled trial. BMJ 342 (Published 18 May 2011). doi:10.1136/bmj.d2912. Butler, R.J., Marchesi, S., Royer, T., Davis, I.S., 2007. The effect of a subject-specific amount of lateral wedge on knee mechanics in patients with medial knee osteoarthritis. J. Orthop. Res. 25, 1121–1127. Butler, R.J., Barrios, J.A., Royer, T., Davis, I.S., 2009. Effect of laterally wedged foot orthoses on rearfoot and hip mechanics in patients with medial knee osteoarthritis. Prosthet. Orthot. Int. 33, 107–116. Crenshaw, S.J., Pollo, F.E., Calton, E.F., 2000. Effects of lateral-wedged insoles on kinetics at the knee. Clin. Orthopaedics Rel. Res. 375, 185–192. Hinman, R.S., Bowles, K., Payne, C., Bennell, K.L., 2008a. Effect of length on laterally wedged insoles in knee osteoarthritis. Arthritis Care Res. 59, 144–147.

R.S. Hinman et al. / Clinical Biomechanics 27 (2012) 27–33 Hinman, R.S., Payne, C., Metcalf, B.R., Wrigley, T.V., Bennell, K.L., 2008b. Lateral wedges in knee osteoarthritis: what are their immediate clinical and biomechanical effects and can these predict a three-month clinical outcome? Arthritis Care Res. 59, 408–415. Kakihana, W., Akai, M., Yamasaki, N., Takashima, T., Nakazawa, K., 2004. Changes of joint moments in the gait of normal subjects wearing laterally wedged insoles. Am. J. Phys. Med. Rehabil. 83, 273–278. Kakihana, W., Akai, M., Nakazawa, K., Takashima, T., Naito, K., Torii, S., 2005. Effects of laterally wedged insoles on knee and subtalar joint moments. Arch. Phys. Med. Rehabil. 86, 1465–1471. Kakihana, W., Akai, M., Nakazawa, K., Naito, K., Torii, S., 2007. Inconsistent knee varus moment reduction caused by a lateral wedge in knee osteoarthritis. Am. J. Phys. Med. Rehabil. 86, 446–454. Kellgren, J.H., Jeffrey, M.R., Ball, J., 1963. The epidemiology of chronic rheumatism: atlas of standard radiographs. Blackwell Scientific, Oxford. Kerrigan, D.C., Lelas, J.L., Goggins, J., Merriman, G.J., Kaplan, R.J., Felson, D.T., 2002. Effectiveness of a lateral-wedge insole on knee varus torque in patients with knee osteoarthritis. Arch. Phys. Med. Rehabil. 83, 889–893. Kraus, V.B., Parker Vail, T., Worrell, T., Mcdaniel, G., 2005. A comparative assessment of alignment angle of the knee by radiographic and physical examination methods. Arthritis Rheum. 52, 1730–1735. Kuroyanagi, Y., Nagura, T., Matsumoto, H., Otani, T., Suda, Y., Nakamura, T., et al., 2007. The lateral wedged insole with subtalar strapping significantly reducs dynamic knee load in the medial compartment- gait analysis on patients with medial knee osteoarthritis. Osteoar Cart. 15, 932–936. Ledingham, J., Regan, M., Jones, A., Doherty, M., 1993. Radiographic patterns and associations of osteoarthritis of the knee in patients referred to hospital. Ann. Rheum. Dis. 52, 520–526. Levinger, P., Menz, H.B., Fotoohabadi, M.R., Feller, J.A., Bartlett, J.R., Bergman, N.R., 2010. Foot posture in people with medial compartment knee osteoarthritis. J. Foot Ankle Res. 3, 29. Maillefert, J.F., Hudry, C., Baron, G., Kieffert, P., Bourgeois, P., Lechevalier, D., et al., 2001. Laterally elevated wedged insoles in the treatment of medial knee osteoarthritis: a prospective randomised controlled study. Osteoar Cart. 9, 738–745.

33

Maly, M.R., Culham, E.G., Costigan, P.A., 2002. Static and dynamic biomechanics of foot orthoses in people with medial compartment knee osteoarthritis. Clin. Biomech. 17, 603–610. Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, 617–622. Nakajima, K., Kakihana, W., Nakagawa, T., Mitomi, H., Hikita, A., Suzuki, R., et al., 2009. Addition of an arch support improves the biomechanical effect of a laterally wedged insole. Gait Posture. 29, 208–213. Nigg, B.M., Stergiou, P., Cole, G., Stefanyshyn, D., Mundermann, A., Hunble, N., 2003. Effect of shoe inserts on kinematics, center of pressure, and leg joint moments during running. Med. Sci. Sports Exercise 35, 314–319. Pham, T., Maillefert, J.F., Hudry, C., Kieffert, P., Bourgeois, P., Lechevalier, D., et al., 2003. Laterally elevated wedged insoles in the treatment of medial knee osteoarthritis. A two year prospective randomized controlled study. Osteoar Cart 12, 46–55. Sasaki, T., Yasuda, K., 1987. Clinical evaluation of the treatment of osteoarthritic knees using a newly designed wedged insole. Clin. Orthop. 221, 181–187. Schipplein, O.D., Andriacchi, T.P., 1991. Interaction between active and passive knee stabilizers during level walking. J. Orthop. Res. 9, 113–119. Shimada, S., Kobayashi, S., Wada, M., Uchida, K., Sasaki, S., Kawahara, H., et al., 2006. Effects of disease severity on response to lateral wedged shoe insole for medial compartment knee osteoarthritis. Arch. Phys. Med. Rehabil. 87, 1436–1441. Stacoff, A., Reinschmidt, C., Nigg, B.M., van den Bogert, A.J., Lundberg, A., Denoth, J., et al., 2000. Effects of foot orthoses on skeletal motion during running. Clin. Biomech. 15, 54–64. Yasuda, K., Sasaki, T., 1987. The mechanics of treatment of the osteoarthritic knee with a wedged insole. Clin. Orthop. 215, 162–172. Zhang, W., Moskowitz, R.W., Nuki, G., Abramson, S., Altman, R.D., Arden, N., et al., 2008. OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoar Cart. 16, 137–162.