Advances in Insoles and Shoes for Knee Osteoarthritis

Rana S Hinman; Kim L Bennell


Curr Opin Rheumatol. 2009;21(2):164-170. 

In This Article

Effect of Footwear on Knee Loading and Osteoarthritis Symptoms

Wearing appropriate footwear is a management strategy recommended in recent knee osteoarthritis clinical guidelines.[16] This recommendation is based solely on expert opinion. With the exception of the work commenced by Kerrigan et al.[35,36,37,38] 10 years ago, most research into the potential role of shoes for knee osteoarthritis has been conducted in the past 2 years only. A proportion of this work is available only in abstract form. Other limitations of the current research include the use of younger healthy individuals without osteoarthritis and primarily biomechanical evaluations addressing effects of shoes on loading rather than symptoms.

Effect of Shoes on Knee Joint Loading Compared with That of Bare Feet

Evidence from two studies in knee osteoarthritis has shown that wearing shoes significantly increases medial knee load compared with walking in barefeet. A recent study[39•] evaluated the effects of wearing the participant's own everyday shoes on the knee adduction moment in 40 people with medial compartment osteoarthritis. Compared with walking barefoot, wearing shoes resulted in a 7.4% increase in the adduction moment. These findings confirmed those of an earlier study.[40] Changes in joint loading when wearing shoes are not explained by alterations in gait pattern,[40] suggesting that the design of everyday off-the-shelf shoes may predispose knee osteoarthritis patients to excessive joint loading. This hypothesis is supported by data demonstrating that the effect of shoes on knee loads is not systematic. Although most participants in the study by Kemp et al.[39•] demonstrated an increased adduction moment with shoes, considerable individual variation was observed, with 15% of individuals showing reduced knee loading with shoes and some showing increased loading by 20-30% (compared with barefoot), far in excess of the average 7.4% increase demonstrated by the entire cohort. As participants wore their own shoes in this study, it seems likely that individual shoe characteristics may mediate effects of footwear on knee loading. On the basis of research showing that a one-unit increase in the adduction moment increases the risk of osteoarthritis progression 6.5 times,[41] the more conservative findings of Kemp et al.[39•] suggest that wearing shoes may increase the risk of disease progression by a factor of 2.8 on average. As walking barefoot is impractical and potentially dangerous, research is focused on determining the optimal off-the-shelf shoe type for knee osteoarthritis, as well as exploring whether shoe modifications or custom-designed shoes have a role to play in knee osteoarthritis.

Off-the-shelf Shoes and Their Effects on Knee Load

Some off-the-shelf shoes increase knee loads more than others. Studies in healthy women without knee osteoarthritis revealed that walking in high-heeled shoes (2.5-2.8 inches) resulted in greater loads across the medial knee compartment and the patellofemoral joint compared with bare-foot walking,[38] irrespective of the heel width.[37] Although these early studies did not include a flat-heeled comparison shoe, a later study[35] demonstrated that moderate-heeled (1.5 inches) shoes resulted in 9-14% increases in late stance varus knee torque (adduction moment) in younger and older healthy women compared with flat-heeled control shoes. In contrast, men's dress shoes and sneakers, with an average 0.5-inch heel, did not significantly affect joint torques (other than that explained by changes in walking speed) compared with barefoot walking in healthy men.[36] These studies clearly show that shoes with a heel of more than 1.5 inch detrimentally influence medial knee loads and should be avoided or worn minimally by patients with medial knee osteoarthritis. Long-term effects of wearing high-heeled shoes on disease incidence and progression remains unknown, as is their effect on knee symptoms.

A recent abstract compared the effects of four common shoe types (clogs, stability athletic shoes, flexible walking shoes and flip-flops) to barefoot walking in a small cohort with knee osteoarthritis.[42••] Both clogs and stability athletic shoes significantly increased the adduction moment compared with barefoot walking, whereas the flexible walking shoes and flip-flops had no effect. It appears that flexible footwear better approximates barefoot walking and minimizes knee loading whereas supportive stability footwear may have detrimental biomechanical effects. Further research is required to determine which off-the-shelf shoes are optimal for knee osteoarthritis and to evaluate their long-term effects on symptoms and disease progression.

Footwear Modifications and Their Potential Applications in Knee Osteoarthritis

Recent research has focused on the development and evaluation of shoes modified to reduce the knee adduction moment. The 'mobility' shoe is a flexible light-weight shoe designed to mimic essential features of natural foot motion via specialized grooves placed at the major foot flexion points.[43••] Data from participants with knee osteoarthritis show that the adduction moment recorded with the mobility shoe approximates that of barefoot walking. Research demonstrates it can significantly reduce the adduction moment by 12% compared with a control shoe and by 8% compared with conventional self-selected walking shoes. Although immediate benefits are evident, it is unknown whether reductions in knee load can be maintained with prolonged use of the mobility shoe.

'Unstable' shoes have been designed to strengthen lower extremity muscles that contribute to static and dynamic stability.[44,45] The shoe has a multilayered sole that changes flat hard surfaces into uneven ground, demanding increased muscle activity to maintain stability. A small study[45] in young healthy people showed that 'unstable' shoes produced changes and trends in kinematic, kinetic and electromyographic characteristics deemed by the authors to be advantageous. A 12-week randomized controlled trial[44] conducted in 123 people with knee osteoarthritis compared unstable shoes to off-the-shelf walking shoes. Both groups demonstrated reductions in knee pain with generally no differences between the two shoe types tested. Although there were trends towards improved balance performance in the group wearing unstable shoes, differences were not significant between groups. No differences between groups regarding joint motion or strength were observed. At this stage, a specially designed unstable shoe appears to offer little clinical benefit to patients with knee osteoarthritis.

On the basis of the promising results of studies evaluating inserted lateral wedge insoles, some studies have evaluated the effects of shoes modified to include a laterally angled sole. In young healthy adults, Fisher et al.[46••] demonstrated that shoes with either a 4 or 8° valgus shoe sole angle significantly reduced the knee adduction moment compared with a control shoe. Participants with higher knee adduction moments in the control shoe prior to intervention demonstrated greater reductions in moments with the wedged sole shoes. In contrast, another study that attached a 14° wedge to the shoe did not reduce the varus knee moment unless combined with a rigid ankle-stabilizing orthosis.[32] It is not clear why results of these studies differ, but differences in the type of shoe used or the length of the sole wedging may be responsible.

Modifications to shoe sole stiffness have also been tested for their potential application to knee osteoarthritis. A variable-stiffness sole that is denser (rigid) laterally than medially can significantly reduce the adduction moment in healthy young adults compared with control shoes of uniform sole stiffness.[46••] Similarly to the wedged sole shoe, participants with higher knee adduction moments in the control shoe prior to intervention demonstrated greater reductions in moments with the variable-stiffness sole shoes. Data from older individuals with medial knee osteoarthritis show similar results. Erhart et al.[47••] tested a variable-stiffness shoe in 79 people with medial osteoarthritis and found average reductions in the knee adduction moment ranging from 2.4 to 6.2%, depending on walking speed and compared with a constant-stiffness control shoe. Importantly, reductions in the knee adduction moment did not occur at the expense of overloading other lower joints. Furthermore, an abstract[48••] reporting findings from a small randomized controlled trial showed improvements in pain and function at 12 months with variable-stiffness shoes compared with control shoes. Variable-stiffness sole shoes offer great potential for medial knee osteoarthritis and should be subjected to further larger clinical trials that include measures of structural disease progression.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: