2 Research recommendations
- 2.1 Treatments for osteoarthritis in very old people
- 2.2 Combinations of treatments for osteoarthritis
- 2.3 Treating common presentations of osteoarthritis for which there is little evidence
- 2.4 Biomechanical interventions in the management of osteoarthritis
- 2.5 Treatments that modify joint structure in people with osteoarthritis
The Guideline Development Group has made the following recommendations for research, based on its review of evidence, to improve NICE guidance and patient care in the future. The Guideline Development Group's full set of research recommendations is detailed in the full guideline.
What are the short-term and long-term benefits of non-pharmacological and pharmacological treatments for osteoarthritis in very old people (for example, aged 80 years and older)?
Very little data exist on the use of pharmacological and non-pharmacological treatments for osteoarthritis in very old people. This is highly relevant, not only because of the ageing population but also because of the high incidence of comorbidities in this population – osteoarthritis may be one of many health problems affecting function, and this may influence the appropriateness of management options. The acceptability, nature and setting for exercise strategies for this population is one area suggested for further study. Any non-pharmacological intervention for which a reduction in the need for drug treatment can be demonstrated is desirable. NSAIDs are frequently contraindicated in older people with comorbidities (such as renal failure, cardiovascular or gastrointestinal intolerance), and effective pharmacological options for this group warrant further study. Outcome and intervention studies are also needed for very old people in whom joint replacement surgery is not recommended because of risks associated with comorbidities.
What are the benefits of combinations of treatments for osteoarthritis, and how can these be included in clinically useful, cost-effective algorithms for long-term care?
Most people with osteoarthritis have symptoms for many years, and over this time they will receive several treatments, sometimes in combination. This may involve a combination of non-pharmacological and pharmacological treatments, such as using a walking stick and taking analgesics at the same time. Perhaps more commonly, a person may take different analgesics at the same time (for example, NSAIDs and opioids). However, most of the osteoarthritis trial evidence only evaluates single treatments, and often such trials are of short duration (for example, 6 weeks). We need to understand the benefits of combination treatments relevant to particular anatomical sites of osteoarthritis (for example, hand compared with knee) and whether particular combinations provide synergistic benefit in terms of symptom relief. Also needed is an understanding of how combinations of treatments can be included in algorithms (for example, dose escalation or substitution designs) for use in clinical practice. Trials to address this area may need to utilise complex intervention methodologies with health economic evaluations, and will need to stratify for comorbidities that affect the use of a particular intervention.
What are effective treatments for people with osteoarthritis who have common but poorly researched problems, such as pain in more than one joint or foot osteoarthritis?
Although people with osteoarthritis typically have symptoms that affect one joint at any particular time, there are still many people, especially older people, who have more than one painful joint. For example, it is common for osteoarthritis to affect both knees, or for a person to have pain in one knee and in one or more small joints such as the base of the thumb or the big toe. The mechanisms that cause pain may differ in people with one affected joint compared with those who have pain in several joints. For example, altered use because of pain in one joint often leads to increased mechanical stress and pain at other sites, and having chronic pain at one site can influence the experience of pain elsewhere in the body. However, almost all trials of treatments for osteoarthritis focus on a single joint, and if a participant has bilateral symptoms or additional symptoms at a different joint site only one 'index' joint (the most painful) is assessed. Whether systemic treatments for osteoarthritis work less well if a person has more than one painful site, and whether local treatment of one joint (for example, injection of corticosteroid into a knee) can lead to benefits at other sites (for example, the foot) remains unknown. A further caveat to current research evidence is that most trials focus on treatment of knee osteoarthritis, and to a lesser extent hip or hand osteoarthritis, but there are very few trials that examine other prevalent sites of osteoarthritis such as the first metatarsophalangeal (bunion) joint, the mid-foot joints, the ankle or the shoulder. Trials should be undertaken to determine the efficacy of available treatments, both local and systemic, at such sites. New outcome instruments to measure pain, stiffness and function specific to osteoarthritis at each site may need to be developed and validated for use in such trials.
Which biomechanical interventions (such as footwear, insoles, braces and splints) are most beneficial in the management of osteoarthritis, and in which subgroups of people with osteoarthritis do they have the greatest benefit?
In many people, osteoarthritis is made worse by weight-bearing or biomechanical forces through an affected joint. For example, base of thumb pain may be worse with grabbing and lifting items. Local support for the joint, in this case via a thumb splint, may improve pain and function. A large range of devices are available to help people with osteoarthritis in different joints, but there are very few trials to demonstrate their efficacy, and in particular little data to guide healthcare professionals on which people would benefit most from these aids. For example, there are many knee braces available, but few well designed randomised controlled trials of their efficacy, and few suggestions for clinicians on which patient subgroups might benefit from their use. Trials in the device area require careful attention to design issues such as the selection of control or sham interventions, blinded assessments and the choice of validated outcome measures that reflect the specific joint or functional ability being targeted.
In people with osteoarthritis, are there treatments that can modify joint structure, resulting in delayed structural progression and improved outcomes?
There is evidence from observational studies that factors affecting structural joint components, biomechanics and inflammation in and around the joint influence the progression of osteoarthritis. Symptoms appear to be more closely linked to structure than was once thought, so preventing progression of the structural deterioration of a joint is expected to deliver symptomatic benefits for people with osteoarthritis, as well as delaying joint replacement in some. There have been published randomised controlled trials with interventions targeting structural components of cartilage (glucosamine sulphate) and bone (strontium ranelate). However, several limitations have been identified with the glucosamine sulphate studies, and it is unclear whether cardiovascular concerns will prevent approval of strontium ranelate for treating osteoarthritis. Randomised, placebo-controlled trials of adequate power and duration (related to the structural end point under consideration) should be undertaken to determine the benefits and side effects of agents with disease-modifying osteoarthritis drug potential for treating both hip and knee osteoarthritis (separately). Appropriate structural end points may include progression of radiographic joint space narrowing or MRI features of osteoarthritis. Associated clinical end points could include measures of pain, function and health-related quality of life. Studies should also include rates of subsequent joint replacement (preferably maintaining original blinding, even if extensions are open label). Later phase trials should include a health economic evaluation.