This site is intended for health professionals only
In July 2018, the National Institute for Health and Care Excellence (NICE) published revised guidelines for the management of rheumatoid arthritis (RA) disease in adults.1 Some clinicians will find it challenging to adhere to these, but they reflect best practice. Management of RA depends on a multidisciplinary approach and shared care between secondary and primary care. The guideline is relevant to non-specialist health professionals who are involved in the initial assessment of RA symptoms and ongoing care of people diagnosed with RA. What are the implications of these guidelines for commissioners and providers of services for people with RA?
RA is a chronic, disabling autoimmune disease characterised by synovitis of small and large joints causing swelling, stiffness, pain, and progressive joint destruction. Approximately 1% of the UK population have RA, and as many as 15% of these people may have severe active disease at any point in time. It affects roughly three times as many women as men. People tend to develop RA between 40 and 60 years of age, although it can occur at any age. The early signs of RA of joint pain and swelling usually present in primary care. Fast and accurate referral to rheumatology services is important to achieve early remission and prevent or reduce disability.2
The management of RA has evolved in the nine years since the previous NICE guideline on RA was published, with greater emphasis on a treat-to-target strategy rather than specific drug regimens,3 and debate about the merit of initiating treatment with combination drug therapy.4 Technologies such as ultrasound have been increasingly used for diagnosis and monitoring of synovitis where it is unclear from clinical examination.5 These aspects of management were investigated by the Guideline Committee, and recommendations have been updated using
new evidence, leading to changes to the recommendations for treatment with conventional synthetic disease modifying anti-rheumatic drugs (csDMARDs), glucocorticoids for bridging treatment, and choice of treatment for symptom control. Several aspects of the guideline have remained unchanged since its publication in 2009.
NICE publishes evidence-based recommendations for health and care in England (not Wales or Scotland, although they can also be used there). The express aim of the Institute is to prevent ill health, to promote and protect good health, to improve the quality of care and services and to adapt and provide health and social care services. The guidelines are widely used to define ‘minimum standards of care’ in the UK, so that patients and carers using the National Health Service (NHS) know what they are entitled to receive from healthcare providers. Commissioners and Trusts are expected to adhere to NICE guidelines and to assure the process through regular audit. If this does not happen, then providers would be open to censure, for example by the Health Service Ombudsman in the event of a complaint, and may lose their eligibility to bid for provision of specialised services.
NICE also publishes quality standards in the form of statements that are designed for commissioners and providers to identify gaps in service provision and areas for improvement, to facilitate measurement of quality of care and demonstration of high quality care, with the aim to facilitate commissioning of high quality services. The quality standards for RA were last published in 2013 but are currently being revised. Clinicians would normally be expected to undertake regular audit against these standards, and commissioners might be expected to receive assurance that this is undertaken.
The new recommendations are:
The guideline emphasises the importance of rapid referral to a rheumatologist for any adult with suspected persistent synovitis of undetermined cause independent of investigations including blood tests for acute phase response or rheumatoid factor. The guideline recommends referral in any patient when:
Referral should be guided by clinical examination and should not be delayed by waiting for results of any investigations as they may be normal especially in early disease. When positive, anti-cyclic citrullinated peptide (CCP) antibodies and/or radiographic erosions at diagnosis in combination with a raised C-reactive protein (CRP) are indicators of a poor prognosis. It is recommended that CCP, CRP and X-rays are arranged at initial diagnosis in secondary care if they were not undertaken before referral.
The guideline recommends that the rheumatologist should inform those with risk factors of a poor prognosis that they have an increased risk of radiological progression. This is in order to emphasise the importance of the patient monitoring their condition and seeking rapid access to specialist care if disease worsens or they have a flare.
Disease activity can be measured by various tools, such as the DAS28, which is based on a composite score from clinical assessment of the number of tender joints, swollen joints, global pain, and a biomarker for inflammation (either erythrocyte sedimentation rate (ESR) or CRP). Definitions of remission or low disease activity vary according to the measure used. For example, with the DAS28, remission is a score of <2.6 and low disease activity is ≤3.2. The previous guideline did not recommend a specific target other than agreeing a target with the patient. The revised guideline now recommends that patients with active RA are treated with the aim of achieving a target of remission or low disease activity if remission cannot be achieved (treat-to-target). This is more challenging than the recommendation in 2009 and could have resource implications as patients might have more treatment and follow up appointments. The guideline also recommended that clinicians should consider making the target remission rather than low disease activity for people with an increased risk of radiological progression (that is, those with positive anti-CCP antibodies or erosions on X-ray at baseline assessment).
Initial pharmacological management is led by specialists. Conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDs) are differentiated from targeted synthetic (ts)DMARDs (such as Janus kinase inhibitors) and biologic (b)DMARDs (including inhibitors of cytokines such as tumour necrosis factor and interleukin-6), which were not within the scope of this guideline. The previous guideline recommended initial treatment with a combination of two or more csDMARDs including methotrexate. However, the evidence for this approach was re-evaluated and a meta-analysis did not find superiority for any individual drug. This is surprising to those who consider methotrexate to be superior although this is not supported by current data. Extensive literature review also did not find superiority for initial combination compared with a step up strategy. In contrast to the previous recommendation, the current guideline therefore recommends initiation with a single csDMARD (either sulfasalazine, methotrexate, or leflunomide) and sequentially adding further drugs in a step-up approach if the target is not met. Thereafter, if the patient remains with severe active disease (DAS>5.1) they would be eligible for a bDMARD. NICE at present does not have a recommendation for those who have not met the target of at least low disease activity and do not have severe active disease; this is currently under review because of the reduction in the cost of some drugs following the introduction of tsDMARDs and biosimilar bDMARDs.
Once a patient has achieved and maintained their treatment target of remission or low disease activity for at least a year without glucocorticoids, the guideline recommends the rheumatologist should consider cautiously reducing drug doses or stopping drugs in a step-down strategy but to return promptly to the previous DMARD regimen if the treatment target is no longer met.
What has proved challenging for some rheumatologists was the recommendation in the 2009 guideline to follow patients monthly until their target was met. However, although this may have resource implications the recommendation remains that all patients should be reviewed monthly in their rheumatology unit until they are in remission or low disease state. Thereafter it is recommended that patients should have a review appointment after six months to ensure that the target has been maintained and if stable to be reviewed at least on an annual basis. The annual review should be a comprehensive evaluation and include an assessment of disease activity and damage and any need for surgery, a measure of functional ability (using, for example, the Health Assessment Questionnaire) and impact on life, a check for the development of comorbidities, such as hypertension, ischaemic heart disease, osteoporosis and depression, an assessment of symptoms that suggest complications such as vasculitis and disease of the cervical spine, lung or eyes, and appropriate cross-referral within the multidisciplinary team. An annual review was also included in the previous guideline but many rheumatologists have found a comprehensive review to be difficult to deliver. The availability of specialist nurses is often instrumental in supporting these recommendations and service planning should consider the resources required to deliver both monthly monitoring and annual review. Although labour intensive, this approach may prove cost effective by reducing the number of patients who need to be prescribed a bDMARD.
One problem with csDMARDS is that they have a gradual onset of action over weeks to months. In order to provide a rapid reduction in symptoms, most rheumatologists recommend short term bridging treatment with glucocorticoids (oral, intramuscular, or intra-articular). The guideline committee were unable to strengthen the recommendation and advise all patients to receive bridging therapy because of the lack of research evidence. However, rheumatologists should be encouraged to prescribe short term steroids when initiating or changing a DMARD and also for disease flares. The guideline recommendation is to “Consider short term bridging treatment with glucocorticoids (oral, intramuscular, or intra-articular) when starting a new conventional synthetic DMARD.”
Although control of synovitis with csDMARD and corticosteroids improves symptoms, some patients require additional analgesia. The committee found very limited evidence for paracetamol, opioids, and tricyclic antidepressants for symptom control in rheumatoid arthritis, so the recommendation for “other analgesics” was removed from the update of this guideline and replaced with a recommendation for NSAIDs alone – to consider oral non-steroidal anti-inflammatory drugs (NSAIDs), including traditional NSAIDs and COX II selective inhibitors, when control of pain or stiffness is inadequate taking account of potential gastrointestinal, liver, and cardio-renal toxicity, and the person’s risk factors, including age and pregnancy. The lowest effective dose for the shortest possible time of NSAIDs was recommended with co-prescription of a proton pump inhibitor and regular review of risk factors for adverse events.
When patients with RA have met their target, monitoring patients on DMARDs should be shared between primary and secondary care. However, flares of disease are characteristic of many patients with RA and there should be rapid access to specialist care for flares and this is emphasised in the guideline. In addition, although the use of ultrasound has expanded in rheumatology as well as other specialties, the role of ultrasound in the management of RA is unclear.5 Following an extensive literature review the conclusion in the guideline was not to recommend ultrasonography for routine monitoring of disease activity in adults with RA.
Some rheumatologists who have not adopted a treat-to-target strategy may need a change in practice. This will require revision of local protocols in order that step-up protocols may be implemented rather than initial combination therapy. The recommendation to especially target patients with poor prognostic markers will need to be included in new protocols. In addition, there may be challenges to health professionals in primary and secondary care when explaining risk factors for progression to some patients. Ultrasound scanning of joints is increasing, and the recommendation not to use ultrasound routinely may need to be reflected in the revision of local protocols.
Although current evidence suggests that all people with RA should be offered the same management strategy, it is possible that those identified with a risk of poor prognosis should be treated differently. A high priority research recommendation has been included to answer this question. Research is also needed to identify the best use of corticosteroids in RA, and whether ultrasound can improve management. In addition, In view of the considerable difference in cost between subcutaneous and oral methotrexate, further research needs to be undertaken to determine whether there is greater efficacy of subcutaneous methotrexate compared with oral therapy.
Patients with a DAS28 between 3.2 and 5.1 are often referred to as having moderate disease and at present NICE do not have guidance for this group of patients if they have failed csDMARDs; they are not currently eligible for a bDMARD or tsDAMRD unless they have a DAS28 >5.1. This has proved difficult in managing these patients but, with the reduction in costs of bDMARDs, it is hoped that revised health economic analyses will find that it will be cost effective for those with moderate disease to be treated with biosimilar bDMARDs.