Non-pharmacological approaches for the treatment of osteoarthritis

For the most part, non-pharmacological approaches are recommended for osteoarthritis treatment. This recommendation is based mainly on biomechanical observations leading to a modulation of the symptomatic loading joint. Approaches include orthoses, insoles, exercise, diet and patient education. The approach used for each osteoarthritis site must be adapted for the individual patient. Here, we use an evidence-based approach, including the European League Against Rheumatism (EULAR) and Osteoarthritis Research Society International (OARSI) recommendations, to summarise the non-pharmacological treatments available for knee, hip and hand osteoarthritis and to help the physician in daily clinical practice.

Treatment of osteoarthritis (OA) consists of a combination of non-pharmacological and pharmacological approaches. Non-pharmacological approaches include orthoses, insoles, exercise, diet and patient education. For each OA anatomical site, the non-pharmacological approach must be adapted to the individual patient. Pharmacological treatments, however, are usually the same, whatever the anatomical site.

Orthoses or braces and insoles are mainly prescribed to modulate mechanical stress on the symptomatic joint compartment. Besides this mechanical effect, they may have effects on muscle co-contraction and proprioception . Orthoses are external devices used for knee and hand OA. For knee OA, orthoses are rest orthoses, knee sleeves and unloading knee braces. Orthoses have no use in hip OA. Insoles are commonly prescribed for knee and hip OA. Foot pronation and cushioning insoles are largely prescribed for lower-limb OA. Information on the use of orthoses and insoles is elaborated by the physiotherapist, occupational therapist, podiatrist, prosthetist and orthotist, depending on the typical use and any professional legislation in each country.

Exercise therapy is widely used for lower-limb and hand OA to improve joint range of motion, muscle strength, tendon lengthening, aerobic performance and proprioception. The modalities of exercise are numerous and should be adapted to the joint affected and the health of the patient. Exercise therapy is delivered mainly by physical and occupational therapists. Even if exercise therapy is a key treatment modality in OA, the optimal content of this treatment has yet to be precisely described, so physicians face difficulties in prescribing such treatment for particular OA sites and for particular patients.

Obesity is a major risk factor for the onset and progression of knee OA, and reducing weight can alleviate impairment and disability . In hip OA, the influence of obesity is less important . Surprisingly, obesity is associated with osteoarthritis of the hand, even in non-weight-bearing joints . This observation suggests the existence of systemic pathogenic mechanisms linked to obesity, besides mechanical factors . Because of these mechanical and metabolic apsects, overweight OA patients should aim to lose weight.

Non-pharmacological treatments must be explained to patients to improve adherence to treatment. This main goal of patient education, the final step of non-pharmacological approaches to OA treatment, involves various forms of vehicle information.

Pharmacological treatment of OA involves a combination of acetaminophen, non-steroidal anti-inflammatory drugs (NSAIDs) and symptomatic slow-acting drugs in OA (SYSADOAs). Depending on the country, some SYSADOAs are licensed for use as medications and/or nutritional supplements (i.e., glucosamine, chondroitin sulphate and avocado/soybean unsaponifiable). We consider SYSADOAs and nutritional supplements as part of the pharmacological approach, so we do not describe them here.

Recommendations for treating OA are numerous. They are mainly based on literature analysis to calculate the effect size of a treatment modality, along with expert opinion. The most important information for the physician in daily practice is that effect sizes of non-pharmacological treatments are in the same range (0–1) as those of pharmacological modalities, which favours non-pharmacological approaches in treating OA. The main international recommendations are from the Osteoarthritis Research Society International (OARSI) and the European League Against Rheumatism (EULAR). Besides these international guidelines, national recommendations are adapted to the local health-care system. As examples, the American College of Rheumatology, the United Kingdom National Institute of Health and Clinical Excellence and the French Society of Physical Medicine and Rehabilitation have formulated national clinical guidelines for OA. Recently, Misso et al. listed 18 clinical practice guidelines devoted to knee OA .

Practice points

  • The effect sizes of non-pharmacological treatments for OA are similar to those of pharmacological treatments, which suggest the importance of the use of non-pharmacological approaches in treating OA.

  • Non-pharmacological approaches for OA include orthoses, insoles, exercise, diet and patient education.

  • Some non-pharmacological modalities can be used whatever the joint affected (aerobic exercise, diet and patient education), whereas other modalities are specific to the joint affected (orthoses, insoles and specific exercise).

  • The non-pharmacological approach must be adapted to the OA site and clearly explained to the patient.

Research agenda

  • Specific methodology to evaluate non-pharmacological approaches to OA must be used in future randomised clinical trials.

  • The absence of randomised clinical trial for a treatment modality does not mean that the modality is ineffective.

Research agenda

  • Specific methodology to evaluate non-pharmacological approaches to OA must be used in future randomised clinical trials.

  • The absence of randomised clinical trial for a treatment modality does not mean that the modality is ineffective.

Non-pharmacological approaches for the treatment of knee OA


For knee OA, orthosis consists of rest orthoses, knee sleeves and unloading knee braces. Rest orthoses are used for joint immobilisation, which excludes any dynamic effect. Rest orthoses are created by a stiff composite, by casting or a line. The effectiveness of rest orthoses for lower-limb OA has not been studied in clinical trials. To date, they cannot be recommended . Knee sleeves are elastic non-adhesive orthoses associated with various devices aimed at patellar alignment or frontal femoro-tibial stabilisation. Unloading knee braces are, like knee sleeves, functional devices. They are composed of external stems, hinges and straps. Their purpose is to decrease compressive loads transmitted to the joint surfaces, either in the medial or lateral femoro-tibial compartment, depending on the valgus or varus position of the device.

The EULAR recommendation #3 is devoted to the non-pharmacological approach in the treatment of knee OA . Orthotic devices are recommended mainly on the basis of expert opinion, not evidence. Only one randomised clinical trial, comparing one group with a knee sleeve, one group with a valgus brace and a control group without a knee sleeve or a brace, has been described . This study showed a significant improvement in pain and function for the two intervention groups. In addition, valgus brace treatment is more effective than the simple knee sleeve. Effect size could not be calculated, but the strength of the recommendation was moderate (B), which points out the importance of expert opinion in this recommendation. Cochrane reviews and the OARSI recommendations include one more study, but the device used was an unloading knee brace, not a knee sleeve . The OARSI recommends a knee brace for patients with knee OA and mild or moderate varus or valgus instability to reduce pain, improve stability and diminish the risk of falling . For the unloading knee braces, Beaudreuil et al., in a recent review, found 16 studies . Unfortunately, these studies were of poor quality, and the recommendation for using an unloading knee brace was again based on the Kirkley study . Finally, Beaudreuil et al. pointed out the potential adverse effects of unloading knee braces, such as venous thrombo-embolic events . This last point has to be considered in daily medical practice.

Practice points

  • Knee sleeves can be used for knee OA to decrease pain and disability.

  • Unloading valgus knee braces can be used for medial knee OA to decrease pain and disability.

  • Unloading valgus knee braces may be more effective than knee sleeves but have more adverse effects.

Research agenda

  • Randomised clinical trials are needed to determine whether bracing can be effective to decrease structural damage in knee OA.

  • Instability, malalignment and the main symptomatic knee compartment (medial, lateral and patellar) must be evaluated for their importance in the response to bracing treatment.

  • Basic research is needed to evaluate the biomechanical and neurological (proprioception) mechanisms of the bracing effect.

Insoles and footwear

Insoles can be neutral, lateral-wedged or cushioning. The biomechanical concept for the use of insoles for knee OA has not yet been clearly validated and has been described only for lateral-wedged insoles. Two different biomechanical theories have been elaborated: adduction moment theory and kinetic chain theory . The adduction moment theory estimates the loading on the medial knee compartment by the product of the mechanical axis force and the distance between this axis and the knee joint centre. With a lateral-wedged insole, the mechanical axis might reach a more upright position and reduce the distance between the axis and the joint centre, thus decreasing the knee medial loading ( Fig. 1 ). For the kinetic chain theory, a lateral-wedged insole creates a valgus hind foot and then a valgus knee joint, as well as a decrease in medial compartment load. Whatever the theory, lateral-wedged insoles could decrease the load on the knee joint and probably more on the medial compartment.

Fig. 1

Biomechanical effect of lateral-wedge insole in medial femoro-tibial osteoarthritis: the adduction moment theory. A: in the absence of a lateral-wedge insole. B: in the presence of a lateral-wedge insole. Adapted from Gelis et al. .

The EULAR recommendation #9 describes insoles, without specification, with a strength of recommendation of B . The calculation of the effect size was not feasible. For the OARSI recommendations, the authors added one more study ; the effect size not being calculated for the same reason. Finally, the Cochrane review described a better pain effect with a strapped insole as compared with an inserted insole but with a poor long-term adherence . However, strapped insoles are not disseminated in many countries. In light of the literature and recommendations, lateral-wedged insoles ( Fig. 2 ) could be of interest to decrease pain and NSAID consumption in patients with medial knee OA. To date, no evidence suggests a structural or functional impact of insoles . No results from a randomised controlled trial exist to comment on the other sites (lateral and patellar).

Fig. 2

Back view of lateral-wedge insole.

No randomised controlled trial has evaluated footwear and a cushioning insole. The OARSI recommendations point out an interest to advise patients about appropriate footwear. The optimal shoes could have a flat or low heel, be flexible and have lateral-wedged soles .

Practice points

  • A lateral-wedged insole could reduce symptoms in medial knee OA.

  • A strapped insole could have a good symptomatic effect.

  • A neutral and cushioning insole could be of interest in knee OA, whatever the compartment affected.

Research agenda

  • Medial-wedged insoles should be evaluated for lateral knee OA.

  • Randomised controlled trials evaluating footwear are needed.

  • The optimal thickness and duration of wear of insoles should be evaluated.


Exercise therapy for knee OA should improve joint range of motion, muscle and tendon lengthening, strength and endurance and decrease pain and loading on the symptomatic compartment. Functional improvements are expected in walking ability and daily activities, even sport. The possible modalities of exercise treatments are numerous and depend on the rhythm, duration and type or technique, conducted individually or in groups and supervised or not by a physiotherapist. Exercise therapy can be divided into two modalities. The first is aerobic. Aerobic exercise is, by definition, non-specific and aims to improve general physical performance. The second type is analytic exercise, focusses on the symptomatic joint and aims to improve joint range of motion, to increase muscle strength and to decrease loading of the symptomatic joint compartment. Analytic exercise is based on a precise evaluation of the joint and muscle impairment.

Exercise therapy is always recommended in clinical guidelines for managing knee OA. The EULAR recommendation #3 points out the importance of exercise, although not specified. The range of the effect size is between 0.57 and 1.00 . The OARSI recommendations are more precise and include “regular aerobic, muscle strengthening, and range of motion exercises.” Pooled effect sizes for pain relief are 0.52 for aerobic exercise and 0.32 for muscle strengthening. A recent Cochrane review of 32 studies concluded that exercise for knee OA has at least a short-term benefit in terms of reduced knee pain and improved physical function ; even if the magnitude of the treatment effect is considered small, it is comparable to that reported with NSAIDs . Unfortunately, the Cochrane review did not define ‘exercise therapy’ The MOVE consensus, involving a literature search and a Delphi expert process, gives more precision regarding the content of exercise therapy. The consensus mainly recommended aerobic and strengthening exercises to decrease pain and to improve health status . Group and home exercises are equally effective , and patients may benefit from referral to a physiotherapist . Finally, proprioceptive exercise could have the same efficacy as strength training .

Which muscle to strengthen? To date, responding to this question in terms of the international recommendations is difficult. The response can only come from expert opinion. Strengthening the quadriceps may improve joint stability as may hamstring muscle strengthening. To decrease loading in the medial compartment, strengthening the lateral muscle knee chain would radically open the medial compartment. By contrast, for lateral knee OA, medial muscle chain strengthening could decrease the load on the lateral compartment. These last recommendations are based only on the clinical experience of the authors and need the involvment of a physiotherapist.

Finally, knee flessum is probably important to detect and treat with postural exercise in extension. This last aspect of the treatment is never defined in international recommendations but may improve the natural course of knee OA and have better results after total knee replacement.

Practice points

  • Aerobic, strengthening, range-of-motion and proprioceptive exercise is recommended to decrease pain and improve function and quality of life in knee OA.

  • Inexpensive aerobic exercise can include sport, walking, swimming, cycling and any physical activity the patient particularly enjoys (authors’ recommendations).

  • Quadriceps and hamstring muscle strengthening may improve joint stability.

  • For medial knee OA, strengthening the lateral muscle knee chain may decrease the load on the medial compartment (authors’ recommendations).

  • For lateral knee OA, strengthening the medial muscle knee chain may decrease the load on the medial compartment (authors’ recommendations).

  • Specific postural exercises in extension may guard against flessum (authors’ recommendations).

Only gold members can continue reading. Log In or Register to continue

Nov 11, 2017 | Posted by in RHEUMATOLOGY | Comments Off on Non-pharmacological approaches for the treatment of osteoarthritis
Premium Wordpress Themes by UFO Themes