Rationale and impact
- Information and support
- Therapeutic exercise
- Weight management
- Manual therapy
- Topical, oral and transdermal medicines
- Intra-articular injections
- Follow-up and review
- Imaging for management of osteoarthritis
- Referral for joint replacement
- Arthroscopic procedures
These sections briefly explain why the committee made the recommendations and how they might affect practice.
There was no evidence showing that imaging is beneficial for diagnosing osteoarthritis. The committee agreed that imaging adds little value and that osteoarthritis can be diagnosed by taking a thorough history and doing an examination. The committee agreed that imaging can be useful if atypical features are present that could suggest an alternative or additional diagnosis, such as other inflammatory forms of arthritis (for example, rheumatoid arthritis) and malignancy.
Imaging is frequently used when diagnosing osteoarthritis, despite uncertainties about its benefit, the resource implications and potential for delays in starting management. The recommendations should reduce unnecessary resource use and be cost saving.
Evidence showed that generally people with osteoarthritis wanted more information about their condition. This included information about the causes, what their diagnosis means for the future and where to find more information on self‑management. The committee based their recommendations on the evidence and their experience. They agreed that it is important to tell people that diagnosis is made clinically without imaging, that imaging rarely provides any extra information helpful for diagnosing or planning non-surgical treatment for osteoarthritis, and that it would only be used if there were suspicion of an alternative diagnosis or other complications. This would help reassure and dispel any belief that X-rays or other forms of imaging are needed to diagnose osteoarthritis.
The committee noted the importance of information that offers hope for the future and supports self-management strategies (for example, information that emphasises symptom-reducing behaviours, like therapeutic exercise). They agreed that explaining the core treatments for osteoarthritis would help people understand that pharmacological treatments are not a long-term solution. They also agreed that information about recognising flares and how to manage changes in pain would help the person better understand how their condition may vary over time and what they can do about it. The committee noted more evidence was needed on information about managing flares and information for different populations of people with osteoarthritis, and so made recommendations for research on what information people with osteoarthritis need.
The committee agreed that each person's experience of osteoarthritis differs and therefore tailoring the information to their needs, as described in NICE's guideline on patient experience in adult NHS services, is important. They also agreed that osteoarthritis is more common in older people who are likely to have other conditions. Therefore, the recommendations on delivering an approach to care that takes account of multimorbidity in NICE's guideline on multimorbidity are particularly relevant to people with osteoarthritis.
The recommendations generally reflect current practice because information is likely to be already provided, but they advise on areas in which practice can be improved.
The evidence showed that exercise has a clinically important benefit for people with osteoarthritis, as well as general health benefits and a superior safety profile compared with other common treatments, such as analgesia. In particular, the committee highlighted the importance of therapeutic exercise to help manage and reduce symptoms and improve or maintain physical functioning over the long term. For most benefit, they recommended this be tailored to the needs of the person, such as joint-site-specific exercises.
Limited evidence showed that supervised exercise had some benefits compared with unsupervised exercise. The committee's expert consensus was that supervised exercise is likely to be of greater benefit than unsupervised exercise for people with osteoarthritis. This is because supervised exercise may enable tailored exercise and social support, which may increase adherence and lead to people with osteoarthritis forming a regular exercise habit.
The committee also agreed that shared decision making is important when deciding the form of exercise delivery and type of exercise, as well as considering personal preference and service availability. The committee, acknowledging the importance of exercise, made further recommendations to support people to continue therapeutic exercise by emphasising its benefits while acknowledging that exercise may initially be difficult. They wanted to reassure people with osteoarthritis and healthcare professionals that exercise is not harmful to osteoarthritic joints, and that doing regular and consistent exercise over a long period of time can reduce pain and increase functioning and quality of life.
The committee noted that further evidence may be needed and made a recommendation for research to compare the clinical and cost effectiveness of supervised group and individual exercise with unsupervised exercise.
Evidence showed that treatment packages had a clinically important benefit on physical function compared with education or behaviour change interventions alone. They also had consistent beneficial changes in quality of life, pain and physical function compared with standard care. However, they showed no superiority to individual therapies (such as exercise, manual therapy and electrotherapy). The committee agreed that a person-centred approach is important. Additional education or behavioural change approaches may help some people achieve their goals, but others may not need this. Therefore, the committee recommended combining therapeutic exercise as part of a structured treatment package because this may be more suitable for some people and motivate them to continue with therapeutic exercise.
Current practice around therapeutic exercise varies. These recommendations may lead to a change in practice by recommending tailored exercise (including supervised exercise) and using treatment packages.
The committee acknowledged that evidence on the effects of weight loss for people with osteoarthritis had limitations. However, for people with knee osteoarthritis, the evidence generally showed that as more weight was lost, the benefits for quality of life, pain and physical function increased. The committee acknowledged the challenges people can have with losing weight and maintaining this weight loss, and recommended that they are supported.
The committee acknowledged that, for people who are overweight, losing more than 10% of their body weight may be the most beneficial, but this may not be achievable for all. They wanted to emphasise that losing any amount of weight would be beneficial, but that losing more would have more benefits. They agreed that also explaining that losing 10% of their body weight is likely to be better than 5% might help provide an incentive and encourage weight loss. They also agreed that advising on the amount of weight to lose can help people with osteoarthritis by providing a target for them to work towards.
The committee determined that, although evidence was from people with knee osteoarthritis, this could be applied to people with other osteoarthritis-affected joints. This is because of the potential additional benefits of weight loss seen in other populations, such as reducing inflammatory factors, that may be beneficial for all joint sites and general wellbeing. The committee also agreed that osteoarthritis is a multi-joint disease and people presenting with the condition in 1 joint may end up getting it in another. Weight loss may help reduce this risk.
This recommendation somewhat reflects current practice but may lead to a change in how people with osteoarthritis and healthcare professionals discuss weight loss. This is unlikely to have a significant resource impact.
The committee acknowledged recent evidence that showed some clinical benefits of manual therapy for hip and knee osteoarthritis, with no evidence being identified for other joint sites. However, the benefits were stronger if manual therapy was combined with exercise. Clinical and economic evidence showed that exercise alone was more effective than both manual therapy alone and the combination of manual therapy and exercise. So, the committee concluded that manual therapy should only be considered alongside therapeutic exercise. Most studies provided therapy for less than 3 months and on average for 7 weeks. The committee agreed that the duration of manual therapy would be similar, but would vary according to the person's needs. They agreed that further research was needed, in particular evidence from well‑powered, high-quality studies with adequate blinding and on other osteoarthritis‑affected joints. They made a recommendation for research on manual therapy for people with osteoarthritis used alone and in combination with therapeutic exercise.
Current practice around manual therapy varies. Adding manual therapy to exercise would be a slight change in practice, but it would not have a substantial resource impact because it is not needed long term.
The available evidence was predominantly for knee osteoarthritis. This showed a lack of benefits of acupuncture and some evidence of harm. Economic evidence also showed that using acupuncture for osteoarthritis is not cost effective, so the committee did not recommend using acupuncture or dry needling. There was some evidence of clinical benefit and cost effectiveness for electroacupuncture but this was of very low quality because of small study sizes and inconsistency between studies. The evidence for electroacupuncture suggested it showed a benefit compared with sham acupuncture but not compared with acupuncture or no treatment. The committee considered that the inconsistent evidence could be the result of some people responding more to electroacupuncture than others. Because there is uncertainty about who might benefit from electroacupuncture, the committee made a recommendation for research on electroacupuncture for osteoarthritis.
The recommendation reflects current practice so the committee agreed there should be no change in practice or resource impact to the NHS.
Although there were many studies on electrotherapy, the findings were inconsistent and mostly showed little benefit. The committee acknowledged that most studies were small, with fewer than 100 participants, and that evidence from direct comparisons of electrotherapy with other interventions was uncertain. The committee agreed there is not enough evidence to recommend electrotherapy for people with osteoarthritis.
Extracorporeal shockwave therapy showed some evidence of benefit compared with a sham intervention. However, this evidence was uncertain because of the small trial sizes and challenges in using appropriate sham techniques. The committee agreed that further research using an appropriate sham with more than 50 participants in each study arm is needed and made a recommendation for research on extracorporeal shockwave therapy.
The committee noted that although the use of some electrotherapy modalities in the NHS has decreased, other modalities with unclear evidence of benefit continue to be used. Also, people with osteoarthritis may self-prescribe electrotherapy devices. This recommendation may reduce the prescription and use of TENS.
Evidence from a small study on walking aids showed that they benefit quality of life and reduce pain compared with no device. The committee agreed that walking aids have the advantage of reducing the pressure in the leg joints, which helps stability and movement to encourage physical activity and independence. This is particularly the case while waiting for joint replacement or if surgery cannot be undertaken, because the aid helps support exercise and confidence with walking. Overall, they agreed that the evidence, supported by their expert opinion, was enough to recommend walking aids for people with lower limb osteoarthritis.
The committee concluded that there was not enough evidence to support the routine use of insoles, braces, tape, splints or supports. They also noted that there is a potential risk that some of these devices could cause significant adverse events, such as blistering and other pressure damage. The committee acknowledged that research on devices has challenges, such as appropriate sham devices for comparisons.
The committee agreed that further research is needed on devices, through studies that have a larger number of participants, sufficient blinding and allocation concealment. Because most of the evidence was for knee osteoarthritis, they made a recommendation for research on devices for painful foot and ankle osteoarthritis. The committee also noted that evidence on footwear had limitations and made a recommendation for research on footwear for managing lower limb osteoarthritis.
Use of devices for osteoarthritis is varied in current practice. Currently, insoles, braces, tape or supports may be used by some people with osteoarthritis because they were previously recommended by NICE. The recommendations may change practice with using devices. But the practice of using walking aids is unlikely to change. There is a potential for some cost savings to the NHS.
The committee agreed that pharmacological treatments may be useful for reducing symptoms and supporting people to start other more effective treatments, such as therapeutic exercise. However, they noted that the risks of pharmacological treatments should be understood and that treatments should not be overused or used when they are not needed. The committee agreed that it was difficult to define treatment strengths and durations that would be generalisable to everyone. This is because people with osteoarthritis can have a variety of comorbidities and factors that might influence treatment. Therefore, the committee emphasised that treatments should use the lowest effective dose for the shortest possible time.
Topical non-steroidal anti-inflammatory drugs (NSAIDs) were clinically effective in reducing pain for people with knee osteoarthritis and generally the most cost‑effective medicine for osteoarthritis. They were also associated with minimal adverse events. Evidence on topical NSAIDs came from studies including people with knee osteoarthritis and 1 study including people with hand osteoarthritis. The evidence showed no clinically important difference for hand osteoarthritis, but the committee noted this was uncertain because it was based on 1 study. The committee noted that although evidence only showed benefit for knee osteoarthritis, other people with osteoarthritis-affected joints may also benefit from topical NSAIDs. There was some evidence showing that topical capsaicin reduces pain in knee osteoarthritis, but not hand osteoarthritis, and has minimal adverse events. However, capsaicin is more expensive and topical NSAIDs were considered a better option. The committee made a recommendation for research on topical medicines for osteoarthritis-affected joints other than the knee.
Oral NSAIDs were found to be cost effective and evidence showed they slightly reduced pain and increased physical function. The committee acknowledged the Medicines and Healthcare products Regulatory Agency (MHRA) safety warnings on NSAIDs for cardiovascular safety, renal safety and gastrointestinal risk. They agreed that NSAIDs, as well as other pharmacological treatments for osteoarthritis, should be used for as short a time as possible and that the potential harms for gastrointestinal, cardiovascular, liver and kidney adverse events should be carefully considered when prescribing.
Evidence showed that adding gastroprotection can reduce gastrointestinal bleeding or perforation. However, this was associated with an increase in cardiovascular adverse events compared with oral NSAIDs alone. The committee agreed that this may be unrelated to the addition of gastroprotection and that randomised controlled trial evidence alone may not be the best source for safety evidence, because the population size and length of follow-up are usually limited. Therefore, they also used their clinical experience and guidance from other organisations, including the MHRA. Based on this, the committee agreed that use of gastroprotection should be offered with NSAIDs.
Evidence showed that opioids also have the potential for harm, including gastrointestinal and central nervous system adverse events. The committee acknowledged further potential harms such as physical dependence, opioid-induced hyperalgesia and tolerance. Cost-effectiveness evidence showed that buprenorphine, a transdermal opioid, was generally more cost effective than oral strong opioids (such as morphine, oxycodone and tramadol). This evidence was from people having buprenorphine who had not had opioids before, but this was generally not the case for people having oral strong opioids. All people had already tried a type of analgesia such as NSAIDs or paracetamol. However, the committee acknowledged the MHRA safety warning on opioids and recommendations in NICE's guideline on medicines associated with dependence or withdrawal symptoms, which advises against the use of modified-release opioids. Therefore, the committee recommended against the use of strong opioids. Evidence from 1 small study of weak opioids showed a clinically important benefit in reducing pain. The committee agreed that there was not enough evidence of benefit and on potential risks.
Although paracetamol has a low potential to cause adverse events, evidence showed that it has no additional benefit in reducing osteoarthritis pain and improving quality of life and physical function compared with placebo. However, the committee discussed that some people cannot use NSAIDS. Therefore, they recommended against the routine use of paracetamol but noted some circumstances where it may be used.
Evidence on glucosamine was inconsistent and the largest benefits were shown by smaller studies that were of lower quality. Because glucosamine is not used in current practice and there is no strong evidence of benefit the committee recommended against its use for people with osteoarthritis.
The committee determined that there was not enough evidence to make recommendations on antiepileptics, antidepressants, rubefacients and topical local anaesthetics. The committee made recommendations for research on antiepileptics, antidepressants, weak oral opioids and topical local anaesthetics for osteoarthritis. Duloxetine was not included in the recommendation for research because many studies have already investigated its use, but there is less evidence for other antidepressants that may be used more regularly in the NHS to manage pain (such as tricyclic antidepressants). The committee did not make a recommendation for research on rubefacients because they did not think that these would benefit people with osteoarthritis.
The committee agreed that pharmacological treatments should be periodically reviewed with the person. They recommended this should be done according to clinical need.
Current practice in pharmacological treatment for osteoarthritis varies in the types of treatments used and how people access treatment (such as buying medicines over the counter instead of accessing them through healthcare services). These recommendations may cause changes in current practice towards using medicines for a shorter time, increasing use of topical NSAIDs, and reducing use of paracetamol and opioids.
There was no evidence showing that hyaluronan injections improved quality of life or physical function, or reduced pain, in people with knee or hip osteoarthritis. Evidence showed a potential harm for hip osteoarthritis. Limited evidence for other osteoarthritis-affected joints showed inconsistent benefits and some potential harms. Based on their expert opinion, the committee agreed that these results were generalisable to other forms of osteoarthritis and that hyaluronan injections should not be offered.
Evidence showed that corticosteroid injections had inconsistent benefits on improving quality of life and physical function for people with hip osteoarthritis, and reducing pain for people with knee osteoarthritis. There was no evidence showing long-term benefit beyond 3 months. Given the potential benefits and committee expert opinion, they agreed that intra-articular corticosteroids could be considered for people with osteoarthritis if other treatments have not worked, provided the person was made aware that the injection would only provide short-term relief (2 to 10 weeks). The committee agreed that when used, corticosteroid injections should only be used to supplement and support people to participate in therapeutic exercise where possible. Based on their expert opinion, the committee agreed that this evidence was generalisable to other osteoarthritis-affected joints.
The committee acknowledged that there was a lack of consistent evidence on corticosteroids (especially for non-knee joint sites), so they made a recommendation for research on intra-articular corticosteroids.
There was some evidence from very small studies that showed a potential benefit of stem cell injections. The committee noted that this is an experimental treatment and agreed that it should not be used outside research. They made a recommendation for research on intra-articular stem cell injections.
The recommendation for intra-articular hyaluronan injections reflects current practice so the committee agreed there should be no change in practice or resource impact to the NHS. Corticosteroid injections are used in current practice, but recommending them only for the short-term relief of symptoms may lead to a reduction in their use and a cost saving to the NHS.
There was no evidence on follow-up for people with osteoarthritis. Therefore, the committee based their recommendations on their expert opinion. In current practice, follow-up is mainly symptom-led or people with osteoarthritis raise the condition as a concern during follow-up consultations for other conditions. The committee agreed that symptom-led follow-up is likely to be appropriate for most people with osteoarthritis. This is because they may be able to self-manage their condition effectively after getting information and guidance on management strategies. However, the committee also acknowledged that follow-up should focus on the person's needs, so there are some situations in which planned follow-up may be necessary. The committee noted that agreeing a specific time for people to seek additional help if the management is not improving their symptoms is important. They also agreed that it was important to manage osteoarthritis and other conditions the person may have holistically. Because there was no evidence in this area the committee also made a recommendation for research on the effectiveness of patient-initiated compared with routine follow-up.
These recommendations generally reflect current practice. Because they include self-management and pre-existing appointments for other conditions, they are unlikely to cause a substantial increase in costs.
There was no evidence on using imaging to manage osteoarthritis. Therefore, the committee used their expertise to inform the recommendation. They acknowledged that imaging was important for confirming the severity of structural joint changes when planning or considering surgery. But it was unclear who should do this imaging because some surgeons may only accept a referral for surgery if they are provided with imaging results, whereas others may prefer to do their own imaging after referral. However, in most cases, imaging should not be needed for managing osteoarthritis because it does not guide how the condition will respond to treatment. The committee made recommendations for research on using imaging to inform non-surgical and pre-surgical management of osteoarthritis.
The recommendation reflects current practice, so the cost impact is likely to be minimal. It may be cost saving by reducing the use of imaging for people with osteoarthritis.
Evidence on referral criteria for joint replacement was limited. This evidence suggested that non-response to analgesics may be associated with a need for joint replacement. Longer duration of symptoms did not appear to be associated with the need for joint replacement, which may show that the symptom duration is less relevant than non-response to treatments. Evidence for the Oxford Hip and Knee scores and the Knee injury and Osteoarthritis Outcome score (KOOS) and Hip disability and Osteoarthritis Outcome Score (HOOS) summary score showed that these numerical scales alone were unlikely to determine whether someone should have surgery, so they were not recommended for use. The committee agreed that the decision to refer someone for joint replacement should be based on clinical assessment after trying all appropriate treatments for that person. These should have been tried for a sufficient length of time to ensure they are not effective at reducing symptoms before referral happens. Given the absence of evidence, the committee made a recommendation for research on indicators for joint replacement in people with osteoarthritis.
Evidence on weight loss before surgery showed that, after hip or knee replacement, there was no difference in outcomes for people in different body mass index (BMI) categories. People who were overweight or obese based on BMI did not have an increased mortality rate after surgery and had improved health-related quality of life and patient-reported outcome measures. For people who were underweight based on BMI, evidence showed an increased mortality rate. However, the committee considered that this may be due to comorbidities and that the effect may be exaggerated by the smaller number of underweight participants in studies. Some studies combined the healthy weight group with the underweight group, which made interpreting the evidence more difficult. The committee acknowledged that BMI can give a false impression of the risks and that other factors need to be considered, such as comorbidities. The committee concluded that BMI, and other measurements of whether someone is overweight or obese, should not be a barrier to joint replacement.
Similarly, the committee agreed that everyone should be treated equally, and people should not be excluded from referral for joint replacement based on their age, sex or gender, if they smoke or any comorbidities. They agreed that there are few contraindications to surgery and the surgeon would be best placed to assess and discuss suitability of joint replacement on a case-by-case basis. The committee also recommended that the varying risks of surgery in relation to a person's specific circumstances should be explained.
Current practice is inconsistent. If all centres adopt these recommendations, then it may lead to an increase in the number of referrals for surgery and subsequently more joint replacements done overall.
There was no evidence showing that arthroscopic procedures reduce pain and improve physical function. Evidence also showed possible harms with arthroscopic procedures compared with sham procedures. Cost-effectiveness evidence showed that arthroscopic procedures were more costly than standard care.
The committee agreed that arthroscopic procedures were not commonly used in clinical practice for osteoarthritis.
The recommendation reflects current practice, so there should be no change in practice or resource impact.