Effectiveness of insoles in treating medial osteoarthritis of the knee

In the linked randomised controlled trial (doi:10.1136/bmj.d2912) Bennell and colleagues assessed the effect of lateral wedge insoles on the symptoms and progression of medial knee osteoarthritis.1 They found that such insoles worn for one year provide no symptomatic or structural benefit compared with a flat control insole.

The medial tibiofemoral compartment is most commonly affected in osteoarthritis of the knee, probably because an external adduction moment (the tendency of a force to twist or rotate an object) in the knee during walking causes compression of the medial compartment.2 Varus knee alignment (bowleg) has been reported as one of the best predictors of a high knee adduction moment during walking, and there is strong evidence that varus alignment or a high adduction moment predicts faster progression of knee osteoarthritis.3 Lateral wedged insoles and valgus knee braces have therefore been developed as a conservative approach to altering mechanical loading and reducing the symptoms and progression of knee osteoarthritis. Several studies show that lateral wedged insoles reduce the knee adduction moment; this effect seems more pronounced in mild osteoarthritis (early to mild radiographic stage) than in more advanced stages (moderate to severe).2

A systematic review of braces and orthoses for knee osteoarthritis included three randomised controlled trials on insoles.4 One found a significant decrease in the use of non-steroidal anti-inflammatory drugs in people prescribed wedged insoles compared with neutral insoles after six months but found no differences in symptom scores.5 The second showed that the femorotibial angle was significantly improved in people using elastically strapped wedged insoles rather than traditional wedged insoles at six months and two years; symptom scores were also significantly better in people using strapped insoles at six months.6 The third trial showed that symptoms were borderline significantly better after six weeks of treatment with a strapped wedged insole compared with a sock-type wedged insole.7

Since the systematic review, four other trials on the effect of lateral wedged insoles have been published. A crossover trial compared lateral wedged insoles with neutral insoles for six weeks in patients with mainly severe medial knee osteoarthritis and found that wedged insoles had no significant effect on symptoms.8 The second trial randomised patients with mild to severe medial osteoarthritis of the knee and varus alignment into five groups treated for 12 weeks with different kinds of neutral and lateral wedged insoles; strapped wedged insoles had the best effect on symptoms, but the authors did not report between group differences.9 A recent randomised controlled trial showed no difference in scores for pain, function, or stiffness after one month to one year of a lateral wedged insole compared with a neutral insole in people with mild to severe medial knee osteoarthritis.10 The fourth and most recent randomised controlled trial compared lateral wedged insoles with a valgus brace in patients with mild to moderate medial knee osteoarthritis, and it showed no difference in symptom scores after six months of treatment between the intervention groups.11

Bennell and colleagues’ trial is the largest trial (n=200) to look at wedged insoles in osteoarthritis of the knee.1 Inclusion criteria were medial knee pain during walking in combination with mild to moderate osteoarthritis of the medial knee and a varus knee alignment. In addition to assessing the symptoms at the end of the one year trial, the annual structural change was assessed by magnetic resonance imaging. The daily use of the insoles was within the recommended range; the loss to follow-up was acceptable; and randomisation, blinding, and analyses procedures were adequate. This trial therefore adds important evidence for the lack of clinical effectiveness of a wedged insole in patients with mild to moderate medial knee osteoarthritis. Moreover, it also found no structural benefit. The pragmatic design is one of the strengths of the study because it aims to reflect how the intervention would be used in clinical practice. However, such a design may also result in non-optimal use of the intervention; for example, the participants in this trial were allowed to wear the wedged insole in their own (sometimes high heeled) shoes. In addition, almost 80% of the participants showed joint space narrowing of the lateral knee compartment too, and the increased load at the lateral compartment may have affected this compartment negatively.

In this trial, as in many others, half of the patients already had radiographic evidence of moderate osteoarthritis of the knee, and wedged insoles have the greatest effect on the adduction moment in the knee in early to mild osteoarthritis.2 Future studies should therefore focus on the effect of wedged insoles, especially elastically strapped ones, in early stage osteoarthritis.

Evidence so far indicates that the use of traditional lateral wedged insoles will on average not benefit patients with mild to moderate medial knee osteoarthritis and cannot be recommended.