Definition

Rheumatoid arthritis (RA) is a common systemic inflammatory disease (affecting approximately 1% of the population and women more commonly than men) characterized by the presence of a destructive inflammatory polyarthritis. The inflammatory arthritis has a predisposition for the hands, particularly the proximal interphalangeal joints, the metacarpophalangeal joints and the wrists, but can affect any synovial joint. Although synovitis is the characteristic feature, RA is a multi-system disease with potential involvement of the lungs, heart, eyes, vascular tree, haematopoietic system and nervous system (known as extra-articular manifestations). Some 80% of adults affected by RA have a positive rheumatoid factor (RF – immunoglobulin (Ig) G or IgM antibodies recognizing IgG) at some point during the disease (although up to 60% of patients may be rheumatoid factor-negative at presentation and up to 10% of the normal population may have a positive rheumatoid factor). More recently, antibodies to cyclic citrullinated peptides (ACPA) have been shown to have greater specificity and similar sensitivity to RF. Importantly ACPA titres often rise several months or years prior to the onset of symptoms, making them more helpful in diagnosing and classifying early disease.

Classification criteria for RA were defined by the American Rheumatology Association (ARA) in 1987 and have been used extensively in research studies, showing high sensitivity and specificity for the diagnosis of RA. These criteria (Box 16.1), particularly the four joint-associated clinical components (i.e. prolonged early morning stiffness with polyarticular symmetrical hand joint involvement), still give a useful guide to the most common clinical features of RA. However, these criteria have now been superseded by the 2010 EULAR (European League Against Rheumatism)/ACR (American College of Rheumatology) Classification Criteria. The 2010 criteria (Box 16.2) have been shown to have greater sensitivity and specificity in early RA.

Box 16.2

2010 EULAR/ACR classification criteria for rheumatoid arthritis

Add the scores of categories A–D

A score ≥6 indicates DEFINITE RHEUMATOID ARTHRITIS

Large joints: shoulders, elbows, hips, knees and ankles

Small joints: excludes DIPJs, 1st MCPJs and 1st MTPJs

RA, as defined by either of these classifications, is probably a heterogeneous syndrome rather than a single disorder, and the variability of the disease (including the pattern of presentation, presence or absence of particular antibodies, prognosis, response to treatment and extra-articular manifestations) probably reflects this.

Aetiology and pathogenesis

The cause of RA is unknown. There is evidence of a genetic contribution; individuals with certain HLA types are at increased risk of disease, and having an identical twin with the disease markedly increases the risk. As is discussed in Chapter 3 (Epidemiology and genetics of rheumatic diseases), several allelic variations of the HLA-DRB1 gene (such as HLA-DRB1 0401 and 0404) have been shown to be associated with disease development in different populations; these alleles share a common amino acid sequence in the DR-β chain (residues 70–74 – the ‘shared epitope’), although how this is involved in the disease aetiology remains an active research question. Although it is a risk factor, carrying an HLA genotype associated with RA or having an identical twin affected does not inevitably mean that disease will develop. Additional environmental processes must also be important. Smoking is certainly one of the contributory environmental processes associated with a significantly increased risk, particularly in patients carrying the shared epitope and who are seropositive for either RF or ACPA. Perhaps the most likely hypothesis, given the association with HLA-DR alleles, is that an aberrant T-lymphocyte response to an antigen results in activation of the immune system (including B-lymphocyte production of rheumatoid factor and ACPA) and an autoimmune driven inflammatory response focused on the synovium. The specific antigen involved in RA has not been clearly identified, although much interest has centred recently on the role of citrullinated proteins, following the identification of ACPA as a disease marker in a subgroup of patients with RA. Citrullination is the post-translation modification of the amino acid arginine (positively charged at physiological pH) to citrulline (uncharged) in a protein. Citrullination changes both the peptide sequence and the charge, and could therefore allow a novel peptide sequence to occur which could escape conventional tolerance mechanisms. Bacterial proteins can also be citrullinated – Porphyromonas gingivalis (a bacterium responsible for periodontitis, frequently found in smokers) unusually has an enzyme that can convert arginine to citrulline. Therefore, bacterial citrullination of self-proteins at sites of inflammation could also break tolerance.

Whatever triggers the initial immune activation, there is clear evidence of an active downstream innate and acquired immune response in patients with RA. It is this chronic inflammatory response within the synovium that accounts for the musculoskeletal clinical features. It also causes the destructive changes in the cartilage and bone that are characteristic of the disease. Treatment with disease-modifying agents aims to control this process. In the majority of patients with RA, the inflammatory process is associated with a pro-inflammatory cytokine milieu dominated by cytokines such as tumour necrosis factor (TNF) and interleukin (IL)-6, produced by macrophage-like synoviocytes. This cytokine production may be driven by either autoreactive T cells, immune complexes containing RF and other autoantibodies, or activation of innate immunity. One of the most dramatic recent developments in RA has been the use of antibodies and fusion proteins that specifically block the action of these cytokines. These therapies show marked efficacy, not only in reducing symptoms but also in slowing disease progression.

Clinical features

History

As highlighted in Chapter 1 (Clinical history and examination), the characteristic clinical history feature of any inflammatory arthritis is early morning stiffness. In inflammatory arthritis this stiffness is present for at least 30 minutes, and usually for several hours on waking. It is often better in the afternoon but recurs with immobility and towards the evening (diurnal variation). Any patient who spontaneously offers a history of early morning stiffness needs thorough investigation to exclude an inflammatory arthropathy.

In RA, the joint stiffness is characteristically accompanied by pain and swelling, most commonly initially of the wrists, the small joints of the hands and the feet in a symmetrical distribution. However, the arthritis can affect any synovial joint and, although patients frequently present with hand symptoms, presentation with other joint involvement does not exclude the diagnosis. An additional useful indicator that symptoms reflect an underlying inflammatory arthritis is the responsiveness of symptoms to non-steroidal anti-inflammatory drugs (NSAIDs), and this should be explored in the history. (It is of note that over-the-counter doses of ibuprofen are not anti-inflammatory.)

RA can cause significant disability through functional impairment, restricted participation and limitation of activities. It is important to enquire about the impact of the disease on the patient’s activities of daily living (ADLs) and occupational/recreational participation, so that these areas can be addressed appropriately. It is also useful to explore other patient-specific factors that could impact on a patient’s disability (Fig. 16.1). Extra-articular manifestations usually occur later in the disease course. However, they can occur early and a thorough review of all systems is also required when taking a history.

FIGURE 16.1 A model of disability

Examination

If allowed appropriate opportunity, the patient will usually have given the clinician enough information to make the diagnosis of inflammatory arthritis from the history. Examination then allows the clinician to confirm the diagnosis. Chapter 1 describes a system for comprehensive history-taking and screening examination of the musculoskeletal system and allows the clinician to identify involved joints that can then be assessed further with appropriate regional examination. The typical examination features in a patient presenting with RA are of swelling, warmth and joint-line tenderness of affected joints. Erythema is unusual. The pattern of joint involvement (Fig. 16.2) most frequently includes the wrists (carpi and distal radio-ulnar joints), the proximal interphalangeal joints (particularly index and middle fingers) and the metacarpophalangeal joints (again index and middle fingers). The feet should be examined because early foot involvement is associated with a worse prognosis and could indicate a need for more aggressive early treatment. Metacarpophalangeal and metatarsophalangeal squeeze tests, in which the heads of the metacarpals and metatarsals are squeezed gently (until the examiner’s fingernail just blanches) between thumb and middle finger, are sensitive indicators of underlying synovitis. The characteristic changes (including boutonnière and swan neck deformities of the fingers, ulnar deviation of the metacarpals or radial deviation of the wrist) are usually only seen in established disease.

FIGURE 16.2 Symmetrical polyarthritis in a patient with rheumatoid arthritis. The wrists, metacarpophalangeal and proximal interphalangeal joints are symmetrically involved

A systemic examination looking for rheumatoid nodules (most often seen on the elbows) and extra-articular features should also be undertaken. RF-positive patients and those who are also homozygous for the HLA-DR shared epitope are at increased risk of extra-articular manifestations of the disease; these are also associated with worse joint disease and an increased mortality. These are discussed in Chapter 18 (Systemic complications of rheumatic diseases).

Early rheumatoid arthritis

The 2010 EULAR/ACR classification criteria for RA allow a greater proportion of patients presenting with early inflammatory arthritis to be appropriately classified as having RA than was the case with the 1987 ARA criteria. Aggressive early treatment can be effective in preventing irreversible loss of joint function and subsequent disability. At first presentation, many patients who go on to develop chronic RA will have normal radiographs and a negative RF. Some patients may not have an increased systemic inflammatory response (C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR)). Determination of ACPA is more sensitive and specific in early disease and is a very useful investigation to guide classification. However, a diagnosis of RA should be a clinical one; approximately 30% of patients with RA do not have detectable levels of ACPA.

There is robust clinical evidence that early aggressive treatment of patients with early RA (with combination disease-modifying anti-rheumatic drug (DMARD) treatment including methotrexate) improves the long-term outcome. A fast-track referral is recommended for patients with possible inflammatory arthritis, to allow early specialist involvement. Unfortunately, owing to limited resources, this is not always possible. The pattern of disease, involvement of the feet, a rising CRP level, positive RF and the presence of ACPA have all been shown to have predictive value in determining prognosis. There is also evidence that the presence of the shared epitope (especially homozygosity) predicts the development of more severe disease (see Chapter 3). Currently these are research findings and routine investigation does not include HLA typing in suspected RA.

Investigation

Although the diagnosis of RA can be made largely through appropriate history-taking and examination, investigations are frequently confirmatory and contribute to subsequent management decisions. Blood tests should include a full blood count, renal and hepatic biochemistry, inflammatory markers (ESR and CRP), RF, ACPA and, in some cases, anti-nuclear antibodies (ANA). A full blood count may show a normochromic normocytic anaemia and less frequently a raised platelet count, in keeping with a systemic inflammatory response. Patients may have low serum albumin and increased alkaline phosphatase levels (again usually due to the systemic inflammatory response). RF is not diagnostic of RA, being found in otherwise healthy individuals, and not infrequently being negative in RA (particularly in early disease). However, a positive RF result correlates with more severe disease and is seen much more frequently in patients with extra-articular manifestations. ACPA is more specific for RA in early disease. Inflammatory arthritis is a feature of SLE and, although the pattern of joint involvement is often different, ANA should be checked in cases where there is diagnostic uncertainty. If the ANA result is positive, there is some evidence that sulfasalazine is a less favourable choice of treatment.

In early disease a viral arthritis screen should also be undertaken, looking particularly for evidence of recent exposure to parvovirus B19, which can cause a transient symmetrical polyarticular synovitis clinically indistinguishable from RA.

Imaging studies should include hand and foot radiographs (looking for peri-articular osteopenia or erosive changes; Fig. 16.3), a chest X-ray (para-neoplastic arthritis can be similar to RA, sarcoid can cause an inflammatory arthritis, particularly of the ankles, and latent tuberculosis and undiagnosed interstitial lung disease are relative contraindications to certain treatments). Other imaging modalities (ultrasonography and magnetic resonance imaging (MRI)) can also be helpful in supplementing clinical examination of the joints in early inflammatory arthritis. This can be particularly helpful where the findings are equivocal. Ultrasound imaging is increasingly used routinely by many rheumatologists in their clinical practice for this reason. MRI is more sensitive than ultrasonography for detecting both mild synovitis and early erosive changes, but is more costly and time-consuming. See Chapter 4 (Investigations) for further discussion of imaging modalities.

FIGURE 16.3 Radiographs of hands to show erosive arthritis affecting metacarpophalangeal joints and wrists

Treatment

Once a diagnosis has been confirmed, treatment should be initiated. The traditional approach to RA management was to treat symptomatically with NSAIDs, then, if required, to add weaker, slow-acting DMARDs, gradually escalating the dose to try to achieve control before moving on to more potent therapies. The paradigm has been challenged by many high-quality studies that have shown benefit from early aggressive treatment of RA. It is clear that both erosive damage and functional impairment occur early and can be controlled more effectively early in the disease. This emphasizes the need for a fast-track referral system and accurate early diagnosis for patients with possible RA.

NSAID responsiveness is frequently seen in RA, but this is only symptomatic therapy. This can be helpful diagnostically. However, monotherapy with NSAIDs has no impact on the progression of rheumatoid damage. Therefore, even if symptoms are controllable with NSAIDs, additional therapy with DMARDs is still required. The mechanism of action of NSAIDs is discussed in Chapter 5 (Medical management of arthritis). There is no difference in efficacy between conventional and cyclo-oxygenase (COX)-2-specific NSAIDs, although the side-effect profile of COX-2-specific drugs is preferable, particularly with regard to the gastrointestinal tract (in younger patients with low risk of cardiovascular disease). Given the requirements in RA for long-term therapy at high doses, COX-2-selective treatments are frequently favoured in patients with gastrointestinal risk factors. All NSAIDs (with the possible exception of naproxen) are associated with an increased cardiovascular risk and should be used with caution in patients with other cardiovascular risk factors.

Corticosteroids can, like NSAIDs, significantly improve inflammatory symptoms. Intramuscular pulses of steroids can be helpful in controlling symptoms but have not been shown to modify disease progression. By contrast, and unlike NSAIDs, low-dose prednisolone (7.5 mg per day) has been shown to reduce both the rate of progression of erosive disease and the chances of developing erosive damage in the early years of RA. Prednisolone is therefore both symptom-relieving and disease-modifying. However, although in studies there were no significant adverse events on treatment, osteoporotic prophylaxis is required and some clinicians have understandable reservations about the risks of long-term steroid use. The situation is compounded because the duration of steroid treatment required, and whether the benefits are lost on discontinuation, remains unknown.

DMARD therapy should be initiated promptly. Conventional DMARDs are a diverse group of drugs, probably having an effect at a number of levels of the inflammatory cascade (for further details see Chapter 5, Medical management of arthritis). Weekly methotrexate (up to 25 mg/week) is a common first-line treatment, although daily sulfasalazine (up to 3 g/day) is another option. Both of these treatments are known to be efficacious, although the evidence for methotrexate is stronger and longitudinal studies show it is better tolerated. Folic acid (5 mg weekly or more frequently) has been shown to reduce the side-effects of methotrexate (mouth ulcers, gastrointestinal upset, and possibly hepatic and haematological disturbance). The authors’ practice is to prescribe Methotrexate on a Monday and Folic acid on a Friday to aid compliance.

Other DMARDs include leflunomide, hydroxychloroquine, intramuscular gold, ciclosporin and azathioprine. All DMARD treatments take between 6 and 12 weeks to take effect, and all patients need to be educated about the need to take the therapy regularly and the risk associated with the different treatments. With the exception of hydroxychloroquine, all of these treatments require regular blood test monitoring. There is also robust evidence that combination therapy with two or more DMARDs (including methotrexate as one of these) is more efficacious than monotherapy. The authors recommend combination treatment with methotrexate and hydroxychloroquine in the first instance, together with a short course of systemic steroids (typically monthly intramuscular injections of methylprednisolone on three occasions or low-dose oral prednisolone for the first few months) to patients with newly diagnosed RA.

Biological treatments specifically target points in the pathogenic process and are used in patients who fail to respond to the conventional DMARDs. TNF had been shown to be a critically important cytokine in RA disease pathogenesis, and was an attractive first target for anti-cytokine therapy. Five anti-TNF therapies are currently licensed – infliximab, adalimumab, golimumab, certolizumab (all are chimeric antibody-based treatments targeting TNF) and etanercept (recombinant soluble TNF receptor) – and all have shown efficacy in randomized controlled trials. They reduce signs and symptoms of disease, preserve function, reduce radiological progression and improve patients’ quality of life. The major risk of anti-TNF treatment seems to be reactivation of latent infection (particularly tuberculosis), because TNF is an important cytokine in maintenance of the granulomata that confine mycobacteria. Pre-treatment screening and vigilance during treatment is therefore critical. Not all patients respond to anti-TNF therapy, perhaps reflecting the heterogeneous nature of what is classified as RA. Studies suggest that approximately two-thirds of patients will derive some benefit from anti-TNF treatment. In patients who have failed to respond to anti-TNF therapy, other biological treatments such as rituximab (an anti-B-cell antibody), tocilizumab (an anti-IL-6 receptor antibody) and abatacept (cytotoxic T-lymphocyte antigen (CTLA)-4 immunoglobulin, which reduces co-stimulation of T cells and may affect antigen presenting cells) are used. Most biological treatments have greater efficacy when used in combination with conventional DMARDs (particularly methotrexate), but there is no evidence currently that combining biological treatments together provides any benefit. Biological treatments are not curative and they must be administered indefinitely. However, the possibility that their use early in disease might result in long periods of remission (particularly using anti-TNF treatment) is under investigation.

Conventional analgesic pain management is also important, particularly in patients with secondary damage from previously active inflammation, as an adjunct to the anti-inflammatory approaches discussed above. Further discussion on the medical management of RA can be found in Chapter 5.