Background
Osteoarthritis (OA) is a chronic, localised joint disease affecting approximately one-third of adults, with the disease prevalence increasing with advancing age [
1]. The economic impact of knee OA is also a large and growing problem for health care systems. In Australia, the estimated financial costs of OA and other arthritic diseases in 2000 totalled almost 9 billion dollars [
2]. Demographic predictions indicate that people aged over 65 years will comprise more than 20% of the population by 2040 [
3], thus knee OA will only become more prevalent.
The knee is the most common lower limb site for OA, with the disease affecting the tibiofemoral and patellofemoral joints either in isolation or combination. The medial tibiofemoral compartment is the most commonly affected (medial 67% versus lateral 16% [
4]). Patients with knee OA frequently report symptoms of knee pain and difficulty with activities of daily living, such as walking, stair-climbing and housekeeping [
5]. Ultimately, pain and disability associated with the disease lead to a loss of functional independence and a profound reduction in quality-of-life.
Management strategies for knee OA may be regarded as
primary prevention (reduction of risk factors to reduce disease incidence);
secondary prevention (interventions to slow/prevent progression to serious disease) or;
tertiary prevention (treatment of pain and disability) [
6]. To date, most knee OA research has focussed on tertiary management strategies, primarily drug therapies. Although effective, drug therapies have side effects and are expensive [
7]. Accordingly, recent knee OA clinical guidelines reinforce the importance of non-pharmacological strategies in the management of the condition [
8,
9] yet there is an absence of high quality evidence to support the use of such therapies [
8]. Thus there is a clear need for future clinical trials to evaluate specific non-pharmacological therapies in order to better guide clinical decision-making.
Given that there is currently no cure for knee OA and the only established treatment for end-stage OA is costly joint replacement, slowing of structural disease progression is essential to help reduce the personal and societal burden of knee OA. Traditionally, disease progression has been assessed by measuring loss of joint space over time from serial x-rays. There is now increasing use of magnetic resonance imaging to measure knee cartilage volume as it has proved to be a valid and reproducible technique that is more sensitive to change than x-rays [
10]. Given the absence of a sufficiently sensitive measure of structural change until recently, this has meant that few interventions have been tested as to their effect on disease progression.
Increased load across the joint is important in the pathogenesis of knee OA. Interventions that alter knee load may reduce symptoms and slow disease progression in patients with knee OA. Direct measurement of knee joint loads is not feasible because of the invasive nature of this in vivo method. However, gait analysis can calculate external joint-loading moments that are directly related to internal joint loads. The external knee adduction moment determines load distribution across the medial and lateral tibial plateaus, with force across the medial compartment almost 2.5 times that of the lateral [
11]. This may explain the much higher prevalence of medial compared with lateral tibiofemoral joint OA.
The magnitude of the adduction moment is partly determined by the mechanical alignment of the knee. In medial knee OA, mechanical alignment becomes varus as the medial joint space narrows. Varus malalignment causes the ground reaction force vector to pass more medially to the knee joint centre, resulting in a higher knee adduction moment. Cross-sectional studies demonstrate that patients with knee OA have a higher knee adduction moment during walking when compared to healthy age-matched controls [
12,
13].
Recent research has found that a higher adduction moment is associated with more severe knee pain [
14] and greater radiographic disease severity [
15]. Severity of knee malalignment is also significantly associated with knee pain severity and physical function [
16]. Longitudinal studies have demonstrated that as little as a one-unit increase in the adduction moment is associated with up to a 6.5-fold increase in the risk of disease progression [
14,
17]. Similarly, knee joint varus malalignment is also correlated with disease progression [
16,
18,
19]. Given the importance of the knee adduction moment and joint alignment with regard to both symptom severity and disease progression in knee OA, conservative strategies to alter these biomechanical factors constitute a logical rehabilitative approach.
In 1987, Sasaki and Yasuda [
20,
21] first reported the potential of a laterally wedged insole in the shoe to treat medial knee OA. They demonstrated that the insole statically aligned the knee in a more upright position by shifting the calcaneus into a valgus position relative to the tibia. The authors concluded that such an alteration helped reduce excessive loading of the medial joint surface, leading to mitigation of knee pain. Biomechanical studies have since evaluated the effects of laterally wedged insoles on knee alignment and medial compartment loading. Ogata et al [
22] demonstrated that lateral thrust of the knee (which forces the knee into varus alignment) was reduced with lateral wedges in people with normal and osteoarthritic knees. Similar results have been reported by more recent studies [
23‐
25], with Kerrigan et al [
24] demonstrating that a 5° laterally wedged insole significantly reduces the knee adduction moment by 6% in medial knee OA. Finally, Giffin et al [
26] observed a significant varus to valgus shift in knee alignment on static radiographs with a lateral heel wedge although this has not been a consistent finding [
27]. Thus there is evidence that lateral wedges can reduce varus malalignment and the adduction moment, two key biomechanical features that are associated with knee OA symptoms and disease progression.
However, despite their biomechanical effects, few randomised controlled trials have evaluated their clinical efficacy [
28‐
30]. In 156 patients with medial knee OA, no significant effect of laterally wedged insoles on symptoms over 2 years was demonstrated, as compared to control insoles [
28]. However, a significant reduction in non-steroidal anti-inflammatory drug intake and greater compliance was observed with laterally wedged insoles, leading the authors to conclude that the results favoured a beneficial effect of the intervention. A recent double-blind, randomized, crossover trial in 90 patients found no effect of 6 weeks of lateral wedge use on pain or [
30]. The samples selected for these studies may partly explain the non-significant effect of laterally wedged insoles. Whilst patients demonstrated medial knee OA, selection criteria did not consider knee joint alignment. Significant effects are more likely in participants with evidence of varus malalignment, rather than in a diverse cohort where alignment is likely to range from valgus (in which lateral wedges could be detrimental) to neutral to varus.
As previously stated, it is also important to evaluate the effect of treatment on disease progression. Only one clinical trial has investigated the effect of lateral wedges on knee OA progression [
29]. Using annual x-rays for two years, this study failed to show that lateral wedges slowed joint space narrowing over time compared to control insoles. This non-significant finding may relate to the use of relatively insensitive x-rays as a measure of disease progression, as well as to an inappropriate patient group and the use of a short rather than full length wedge. Given the paucity of data attesting to the structural effects of lateral wedges over time, further research is necessary using more sensitive methods of measuring disease progression such as magnetic resonance imaging. Furthermore, no study to date has evaluated the cost-effectiveness of lateral wedges.
The aim of this trial is to determine whether wearing lateral wedge insoles for 12 months improves pain, physical function and health-related quality of life, slows structural disease progression and is more cost-effective than control flat insoles in individuals with medial knee joint OA and varus malalignment.
Discussion
This study uses a double-blind randomised controlled trial design to investigate whether lateral wedge insoles have greater effects on symptoms and disease progression and are more cost-effective than control flat insoles in people with mild to moderately severe medial knee OA and varus malalignment.
Recent research has highlighted the existence of subgroups of patients with knee OA particularly with regards to local factors such as malalignment. It is known that these subgroups show different risks of progression and may respond differently to interventions [
62]. Thus, our inclusion/exclusion criteria aim to recruit a subgroup of people with medial knee OA that would most likely respond to lateral wedge insoles. We excluded those with severe disease because there is little scope for structural progression, one of our primary outcomes and recent studies have also suggested that benefits with wedges might be confined to those with mild to moderate disease [
63].
Lateral wedge insoles may vary in their angle of wedge and in their length. Whilst a higher lateral wedge may produce greater biomechanical effects, it is difficult to fit comfortably inside the shoe. We are testing a full length wedge rather than a heel wedge because we found that in 13 people with medial knee OA, a full length wedge reduced the knee adduction moment by 12% compared to no insoles whereas moments obtained with rearfoot wedges were not significantly different to those obtained without insoles (Hinman et al unpublished data).
Our choice of outcome measures are those recommended for use in clinical trials of OA by international rheumatology bodies [
35,
36]. These include measures of pain, function, quality of life and global response to treatment. Magnetic resonance imaging represents an advance in the measurement of structural disease progression. Traditionally, serial x-rays have been used to measure loss of joint space over time. However, measuring joint space width using conventional radiography provides only an approximation of articular cartilage, while MRI directly visualizes joint cartilage. MRI is recognised as a valid, accurate and reproducible tool to measure articular cartilage volume [
38,
64]. It is sensitive to change in both normal subjects [
65] and those with OA [
52] and has been shown to relate to clinically relevant end points including worsening of knee symptoms [
66] and the risk of knee replacement [
53].
It is anticipated that all participants will be recruited by the end of 2007 with data acquisition completed a year later. The results from this trial will contribute to evidence based recommendations for the usefulness of lateral wedged insoles in the management of medial knee OA.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
KB and RH conceived and designed the trial protocol. KB, RH, FC, CP and AH procured the project funding. CP was responsible for the podiatric screening assessment and the insole design and acquisition. FC designed the MRI data acquisition and analysis. AH designed the economic analysis. RO designed the statistical analysis. K-AB is the research assistant on the project. KB drafted the manuscript and RH, CP, FC, K-AB, AH and RO contributed to the manuscript. All authors read and approved the final manuscript.