Rheumatoid arthritis (RA) is a type of arthritis (joint inflammation) in which the joints in the limbs, and sometimes in other parts of the body become painful, swollen, stiff, and, in severe cases, deformed. Tissues that are outside the joints, for example the heart, can also be affected.
Causes and incidence
Rheumatoid arthritis is an autoimmune disorder (in which the body’s immune system attacks its own tissues) that usually starts in early adulthood or middle age but can also develop in children or elderly people. Women are affected more commonly than men. There are usually recurrent attacks.
Symptoms and signs
Early symptoms of rheumatoid arthritis include mild fever and aches followed by swelling, redness, pain, and stiffness in the joints, particularly in the morning. Ligaments, tendons, and muscles around the affected joint may also be involved. Raynaud’s phenomenon (a condition in which the fingers turn white on exposure to cold) may occur and there may also be rheumatoid nodules (small lumps under the skin) over pressure points, such as the elbows or finger joints. Swelling of the wrist may cause carpal tunnel syndrome (tingling and pain in the fingers due to pressure on the median nerve) and also tenosynovitis (inflamed, painful tendon sheaths).
Complications of severe rheumatoid arthritis may include anaemia (a reduced level of the oxygen-carrying pigment haemoglobin in the blood), pericarditis (inflammation of the membranous heart covering), ulcers on the hands and feet, pleural effusion (accumulation of fluid around the lungs), pulmonary fibrosis (scarring and thickening of lung tissue), and Sjögren’s syndrome (in which the eyes and mouth become excessively dry).
Diagnosis and treatment
A diagnosis can be confirmed through X-rays of the affected joints and blood tests (including a check for specific antibodies known as rheumatoid factor). Initially, rheumatoid arthritis may be treated with nonsteroidal anti-inflammatory drugs (NSAIDs) to relieve joint pain and stiffness; corticosteroid drugs may be given in the short term if NSAIDs are not effective.
Disease-modifying antirheumatic drugs are then prescribed to slow or arrest the progress of the disease; these include antirheumatic drugs (such as sulfasalazine, gold, and penicillamine) and immunosuppressant drugs (such as methotrexate), which suppress the body’s immune system if the antirheumatics have not proved effective. Physiotherapy is needed to prevent or limit deformity and to help relieve symptoms and maintain mobility.
People who are disabled by arthritis can be helped to cope with everyday tasks through occupational therapy. In severe cases, surgery may be performed to replace damaged joints with artificial ones (see arthroplasty). Hip and knee replacements are the most common operations of this type. Most sufferers must continue drug treatment for life, but many can achieve a near-normal level of activity with effective control of their symptoms.
Rheumatoid arthritis in detail -technical
Rheumatoid arthritis (RA) is a common, painful, and disabling disease affecting 0.8 to 1.0% of the adult population worldwide, with a female:male ratio of 3:1.
The predominant feature of the disease is a deforming and destructive polyarthritis, with (less commonly) extra-articular manifestations such as subcutaneous nodule formation, serositis, vasculitis, fibrosing alveolitis, amyloidosis, and Felty’s syndrome. Patients also exhibit systemic features of inflammation, including fatigue, anaemia, weight loss, a raised ESR, and elevated concentrations of acute phase proteins.
Severe disease is associated with comorbidities including cardiovascular disease, serious infections, and B-cell lymphomas; although iatrogenic factors are contributory factors to some or all of these, there is good evidence that endogenous mechanisms involved in rheumatoid disease also play an important part. Severe disease is a cause of premature death.
Aetiology, pathogenesis, and diagnosis
Aetiology is multifactorial, with a role for genetic factors, such as HLA DR genes coding a pentapeptide sequence in the antigen-binding region, and PTDN22, a signalling molecule in T cells. Genetic factors interact with environmental and host factors, including smoking and sex hormones, and initiate a biological response to an unidentified trigger that results in the recruitment of cells derived from the bone marrow into the joints and other sites of disease, where a chronic immune-inflammatory reaction ensues. Production by B cells of autoantibodies, such as rheumatoid factors and antibodies to citrullinated peptides, as well as interactions between activated T cells and monocyte-macrophages and synoviocytes mediated by cell contact and cytokines, sustain and amplify the inflammatory reaction. Invasion and enzymatic degradation of cartilage by synoviocytes and of bone by osteoclasts follows and leads to irreversible structural damage and joint failure.
Diagnostic features include: (1) the presence for at least 6 weeks of an observable symmetrical, soft tissue swelling of three or more joints, which usually include metacarpophalangeal, interphalangeal, or wrist joints; (2) stiffness of joints in the morning for more than 1 h; (3) subcutaneous nodules; (4) the presence of IgM rheumatoid factor and/or anticitrullinated protein antibody (ACPA); and (5) radiographic erosions and/or osteopenia of bones in the hand. Diagnostic specificity increases with the presence of increasing numbers of these features.
Prognosis and management
The disease follows a relentless progressive course of variable trajectory in individual patients if untreated. Symptoms, signs, laboratory tests, and imaging are used to monitor inflammatory disease activity, damage to joints, and extra-articular disease.
Mild disease—is treated with judicious use of analgesics and nonsteroidal anti-inflammatory drugs (NSAIDs). Corticosteroid injections into individual affected joints, tendon sheaths, and bursas for persistent swelling, tenderness, or loss of normal range of movement can be very effective.
Moderate or severe disease—this occurs when there is an unremitting pattern of polyarthritis with evidence of significant functional impairment and joint damage in early stages of presentation. In these circumstances, the aim of drug treatment is to achieve rapid control of disease activity and, if possible, remission. This usually requires simultaneous or sequential use of NSAIDs, disease-modifying antirheumatic drugs (DMARDs), corticosteroids, and biological therapies.
- NSAIDs are used at optimal doses for control of pain and stiffness, with naproxen, low-dose celecoxib, and low-dose ibuprofen being relatively free of cardiovascular risk. These alleviate symptoms and signs of inflammation but have no effect on preventing structural damage.
- DMARDs should be used in all patients with moderate or severe disease. The two most commonly employed are sulfasalazine and methotrexate, the latter being the drug of choice in most centres because of its efficacy and superior effect in the long term. Methotrexate can be used alone or concomitantly with other traditional DMARDs or anti-tumour necrosis factor (anti-TNF) drugs.
- Corticosteroids are required in practice in over 50% of patients with moderate or severe disease. If continuing long-term use appears necessary, the aim should be to reduce the dose to the equivalent of 5 to 7.5 mg of prednisolone daily by more aggressive use of DMARDs or instigation of anti-TNF therapy.
- Biological monoclonal antibody and recombinant protein drugs that have proved efficacious in rheumatoid arthritis include: (a) three anti-TNF drugs (two monoclonal antibodies, infliximab and adalimumab, and etanercept, a TNF receptor fused to Fc IgG); (b) an interleukin-1 receptor antagonist (anakinra); (c) an anti-CD20 B-cell-depleting monoclonal antibody (rituximab); and (d) CTLA4-Ig, a receptor-fusion-FcIgG recombinant protein (abatacept). Anti-TNF therapy should ideally be instituted as soon as it becomes apparent that remission or near-remission is not induced by the best use of traditional DMARDs in the early stages of an established diagnosis, but cost constraints or safety concerns may limit the institution of this therapeutic approach.
Systemic rheumatoid vasculitis is potentially a life-threatening complication: therapy with high-dose corticosteroids and cyclophosphamide is favoured by many specialists.
Long-term outcome: with a holistic approach, judicious use of drugs and nonpharmacological measures such as patient education, physiotherapy, aids, appliances, and surgical treatment, mobility and pain-free quality of life can be maintained in most patients for many years. Effective control of disease prevents progressive disability, comorbidity, and premature death. The goal of reliable cure of disease has not yet been achieved.
The first clinical description of rheumatoid arthritis (RA) in the medical literature is generally accorded to Landry-Beavais (1800), although Garrod was the first to use the term in his book published in 1859. Whether RA existed in western Europe in ancient times is debated by scholars of medical history: descriptions of chronic deforming arthritis suggestive of rheumatoid arthritis in classical writings of Galen and others have, for example, been ascribed to chronic polyarticular gout. The suggestion has been made that rheumatoid arthritis was imported to Europe after the discovery of the New World in the 15th century, where it pre-existed, as evidenced by examination of clusters of archaic Amerindian skeletal remains. The possibility that rheumatoid arthritis spread from the New World in modern times is not only of historical interest but has also led to speculation suggesting the importance of environmental factors in its causation.
The concept of rheumatoid arthritis as a disease entity continues to evolve with advances in knowledge of the multiple causes of chronic inflammatory joint diseases. Thus, improved microbiological, immunological, and epidemiological methods have led to a reclassification of certain forms of chronic arthritis, which in the past may have been labelled as rheumatoid arthritis. These include arthritis caused by infections, such as rubella, parvovirus, and borrelia (Lyme disease), or resulting from biological responses to nonviable products of microorganisms (reactive arthritis) such as yersinia, salmonella, and chlamydia. Diseases of uncertain aetiology, e.g. the spondyloarthropathies, sarcoidosis, and chronic arthritis associated with systemic lupus erythematosus (SLE), primary Sjögren’s syndrome, and other connective tissue diseases, have all been recognized as distinct from rheumatoid arthritis in the relatively recent past.
Rheumatoid arthritis is a chronic systemic inflammatory disorder characterized by deforming symmetrical polyarthritis of varying extent and severity, associated with synovitis of joint and tendon sheaths, articular cartilage loss, erosion of juxta-articular bone, osteopenia, and—in most patients—the presence of IgM rheumatoid factor and/or anticitrullinated protein antibodies (ACPAs) in the blood. In some patients systemic and extra-articular features may be observed during the course of the disease, and rarely before joint disease. These include anaemia, weight loss, vasculitis, serositis, mononeuritis multiplex, interstitial inflammation in lungs and exocrine salivary and lacrimal glands, as well as nodules in subcutaneous, pulmonary, and scleral tissues.
The American College of Rheumatology (ACR) has developed and revised criteria for the classification of RA based on a hospital population of patients with established active disease (Bullet list 1). These criteria distinguish active RA from other forms of inflammatory arthritis, with a diagnostic sensitivity and specificity of about 90%. However, they are of less value in prevalence studies, which should include patients with inactive rheumatoid arthritis.
The classification criteria are too restrictive to diagnose RA reliably early in its presentation, as not all the required features may be present at this stage. Moreover, a few patients presenting with polyarthritis may later differentiate into other disease types or follow a self-limiting course.
Bullet list 1 American College of Rheumatology criteria for the classification of rheumatoid arthritis
Criteria 1 to 4 of at least 6 weeks’ duration. Rheumatoid arthritis is defined by the presence of four or more criteria.
- Morning stiffness in and around joints for at least 1 h
- Soft tissue swelling of three or more joints observed by a physician
- Swelling (arthritis) of proximal interphalangeal, metacarpophalangeal, or wrist joints
- Symmetrical swelling of joints
- Subcutaneous rheumatoid nodules
- Presence of IgM rheumatoid factor in abnormal amounts
- Radiographic erosions and/or periarticular osteopenia in hand and/or wrist joints
From Arnett FC, et al. (1988). The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum, 31, 315–34. © 1988.
Criteria and methods for diagnosis of rheumatoid arthritis have varied in different epidemiological studies: some have been based on retrospective analysis of hospital records and others on prospective observation of patients attending hospitals where clinical examination, rheumatoid factor tests, and radiography have been employed. However, in recent years the more widespread use of ACR criteria, including a version with a modified format for use in population studies, has introduced a measure of standardization.
Given the inherent variability in the methodology employed, it is not surprising that estimates of the incidence of RA in the United States of America and Europe vary. In a study in the United Kingdom the incidence was 54 per 100 000 in women and 24.5 per 100 000 in men. The incidence increases sharply from the age of 18 to a maximum in women over the age of 45, and in men continues to rise into the seventh decade. A declining trend in the incidence of RA among women has been observed in recent years.
The prevalence of RA has been consistently assessed as being between 0.8 and 1.1% of the adult population in cross-sectional studies in the United States of America and western Europe, with a female:male ratio of 3:1 and higher prevalence rates in older people. Lower rates of 0.2 to 0.3% have been reported in China and Japan. The prevalence of rheumatoid athritis among the black population is low in rural South Africa (c.0.2%) and it is virtually nonexistent in parts of Nigeria. By contrast, prevalence rates of almost 1% have been observed in black populations in urban South Africa and in the United States of America. A strikingly high prevalence rate of over 5% has been noted among certain American Indian tribes in the United States of America, e.g. the Pima and Chippewa Indians. Differences in both genetic and environmental factors are likely to impact on these variations in incidence and prevalence rates, as are differing access to medical facilities, population age structures, and mortality.
The initiating cause of rheumatoid arthritis remains unknown. A prevalence of 12 to 15% in genetically identical (monozygotic) twins observed in Finland and the United Kingdom, compared with 4% in nonidentical (dizygotic) twins, and between 0.5 to 1% in the general population, strongly favours multigenic influences. It also argues for an environmental trigger. Genetic factors are estimated to contribute around 60% of the liability to disease.
Genotyping confirms an association between the occurrence of RA and allelic polymorphisms of genes on the short arm of chromosome 6 that code for a hypervariable region of the β chain of HLA DR molecules. The critical expressed pentapeptide sequence (glutamine–arginine or lysine–arginine–alanine–alanine) of amino acid residues 70 to 74 has been located to the helical wall of the antigen-binding cleft of the HLA DR β chain by molecular structural studies. This pentapeptide region is also referred to as the ‘shared epitope’ because of its detection by a specific monoclonal antibody. The sequence is present, e.g., on HLA DR4 subtypes Dw4 and Dw14, and HLA DR1 subtype Dw1, coded by DRB1*0401, *0404, and *0101 genes, respectively. The shared sequence and corresponding allelic genes have been detected in a frequency of up to 90% in patients with RA of western European descent. Their association with RA supports the hypothesis that these particular HLA DR molecules present antigens to T-cell receptors and activate pathogenic reactions.
By contrast, HLA DR4 subtypes, of which Dw10 and Dw13 are examples (coded by DRB1*0402 and *0403 genes, respectively), are negatively associated with RA. These subtypes are characterized by a substitution of the basic amino acids glutamine and arginine in positions 70 and 71 by acidic amino acids aspartic and glutamic acid. These alterations are sufficient to alter the specificity of binding such that a different set of antigens binds to the HLA cleft. Another possibility is that specific shared epitope sequences influence T-cell receptor interactions with the HLA-DR β chain α-helix independently of peptide. It is proposed that under these circumstances, signals delivered to T cells lead to the activation of regulatory pathways that elicit a protective response.
It is suggested that major histocompatibility (MHC) genes exert a major influence in the genetic component of susceptibility to RA. However, the presence of DRB1*04 susceptibility genes also correlates with seropositive, erosive, and extra-articular disease. Homozygosity for DRB1*0401 or DRB1*0404, or when they are combined with each other or with DRB1*0101 (as compound homozygotes), appears to correlate with more severe disease. These data have been interpreted as indicating that the shared epitope-encoding genes may be more useful as markers of disease severity in established RA, rather than as markers of disease susceptibility.
It is intriguing to note that the HLA DRB1 allele *0405—coding the shared epitope in a different HLA DR4, Dw15 subtype—is increased in Japanese patients, whereas the DRB1*1402 gene coding HLA DR6, Dw16 is increased in Yakima American Indians. However, other population studies, e.g. in black Americans with rheumatoid arthritis, show no increase in frequency of the gene coding the shared epitope, thus casting some doubt on the hypothesis that it is an essential aetiological factor.
Recent studies have sought positive or negative correlations between susceptibility or disease severity and non-HLA gene polymorphisms detected by genome-wide screens and single nucleotide sequencing. Associations with polymorphic alleles of candidate genes have been described, and a single nucleotide polymorphism (SNP) of a protein tyrosine phosphatase-22 (PTPN22) gene has now been confirmed in several studies as being associated with rheumatoid-factor-positive RA. PTPN22 is expressed by haemopoietic cells and encodes a cytoplasmic signalling tyrosine phosphatase in T cells, thus supporting the concept that the gene product may be implicated in T-cell-mediated pathogenesis of rheumatoid arthritis. SNPs of cytokines and receptors that positively or negatively regulate inflammation, such as the promoter region of tumour necrosis factor α (TNFα) and its type II receptor, interleukin 10 (IL-10), and IL-4 receptor (IL-4R) have also been described in severe subtypes of RA but need confirmation. In Japanese but not in white populations, an increased susceptibility to RA is associated with genetic polymorphism of peptidylarginine deiminase-14 (PAD14), an enzyme that citrullinates proteins. This is of interest, as antibodies to citrullinated proteins occur in rheumatoid arthritis patients with high specificity.
Infectious agents such as rubella, parvovirus B19, and Epstein–Barr virus (EBV) have been implicated in initiating rheumatoid disease. One study, for example, described the parvovirus B19 antigen VP-1 as being expressed in active lesions in synovium with rheumatoid arthritis, but not in osteoarthritis or controls. An increase in EBV-viral DNA load and isolation of rubella virus or sequences has been described in synovial cells from rheumatoid patients. In other studies, EBV-specific or rubella-specific lymphocytes have been detected in joints with rheumatoid arthritis. However, the arthritis caused by such known infections is almost always sporadic and self-limiting, and the clustering of new cases that one might expect if RA was an infectious disease caused by an unknown virus has not been reported. Moreover, corroboration of claims by independent studies is still lacking; hence these theories of causation remain speculative.
In attempts to define a role for environmental factors in the aetiology of rheumatoid arthritis, epidemiological studies have sought differences in the incidence or prevalence of the disease in genetically similar populations exposed to urbanization, different socioeconomic conditions, lifestyles, and known industrial noxious agents. Some differences have been found in genetically homogeneous populations exposed to urbanization, as well as exposure to silica dust, organic solvents, and mineral oils. A gene–environment interaction is strongly suggested by a study from Sweden showing that the risk of rheumatoid arthritis with smoking is associated with the occurrence of rheumatoid factors and antibodies to citrullinated proteins in the presence of the HLA DR shared epitope. However, the link between smoking and the shared epitope did not extend to the presence of rheumatoid factor in the absence of antibodies to citrullinated proteins. It is therefore proposed that the HLA DR shared-epitope gene, in a dose-dependent manner, primarily determines risk in ACPA-positive, but not antibody-negative, rheumatoid arthritis.
It is possible that a decline in the incidence of RA among women noted in Rochester, United States of America, in the period 1950 to 1975, and in a general practice register in the United Kingdom in the decade following 1976, are indicative of a change in environmental pressures or in lifestyle.
A number of observations implicate sex hormones and prolactin in susceptibility to, or protection from, RA. Thus nulliparous females have a higher incidence of rheumatoid arthritis and the highest female:male predominance occurs in the premenopausal period. Exposure to the oral contraceptive pill confers a level of protection and postpones the onset of RA. Pregnancy is associated with suppression of disease, and the incidence of RA is increased following parturition and during lactation. Testosterone levels are reported to be low in men with RA, and the incidence of disease increases with advancing age when levels of male sex hormones are on the decline. Interconnections between the hypothalmic–pituitary axis, hormones, and cytokines have been described, suggesting possible mechanisms whereby these may influence the evolution of rheumatoid arthritis.
The rheumatoid disease process in the joints is characterized by synovitis, an inflammatory effusion and cellular exudate into the joint space, and by damage to tendons, ligaments, cartilage, and bone in and around articulating surfaces of the joint. Long tendons with sheaths lined by synovial membrane—such as in the palms, wrists, ankles, and feet—may also be involved by the inflammatory process and cause malfunction due to damage, rupture, and fibrosis.
In health the synovial membrane (the intima) is a film of one or two cells lining the capsule and its circumferential attachment to the periosteum at the cartilage–bone junction of the joint. The normal synovial membrane consists of type A and B cells, without a basement membrane and lying on a bed of loose connective tissue and a network of small blood vessels (the subintima). Type A cells have morphological and phenotypic features of macrophages. Type B cells are of mesenchymal origin and share many, but not all, of the phenotypic features of typical tissue fibroblasts and are hence referred to as fibroblast-liken synoviocytes.
In established rheumatoid arthritis the synovial membrane typically becomes enormously thickened and assumes a villous appearance. The diseased tissue now consists of an intima that is several (2–10) cell layers deep and coated by a film of fibrin. Type A cells predominate over type B cells and tend to lie in the more superficial part of the intima. The sublining layer (subintima) is also greatly expanded by newly formed blood vessels and infiltrating activated mononuclear cells, including T lymphocytes, lymphoblasts, B cells, plasma cells, monocytes, macrophages, dendritic cells, synoviocytes, and mesenchymal stem cells. The cellular infiltrate usually has a recognizable architecture, comprising perivascular aggregates of CD4+ T cells. Interaggregate areas show a mixed inflammatory cell population, including dendritic cells and macrophages expressing HLA class II, CD8+ T cells, activated B cells, and plasma cells. The aggregates may be organized into prominent lymphoid follicles, some of which display germinal centre formation.
The surface of the thickened synovial membrane is bathed in an inflammatory synovial fluid containing a predominance of polymorphonuclear cells, but also CD4+ and CD8+ lymphocytes, dendritic cells, macrophages, and synoviocytes. The synovial fluid is rich in proinflammatory cytokines and immune complexes containing rheumatoid factor. It is a site of local complement consumption, resulting in low haemolytic complement activity, low C3 and C4, and increased complement breakdown products.
The destructive lesion in the joint typically occurs at the circumferential attachment of the joint capsule, just below and adjacent to the articular cartilage and subchondral bone. Here the intima of the adjacent hypertrophic synovial membrane creeps over the cartilage, and hypoxic tissue rich in newly formed blood vessels, macrophages, osteoclasts and synoviocytes (termed pannus) invades and destroys variable parts of articular cartilage and subchondral bone. The cells at the cartilage–pannus junction are synoviocytes and macrophages, whereas the pannus invading subchondral bone is enriched in osteoclasts. The connective tissue matrix of cartilage adjacent to pannus tissue becomes depleted of proteoglycans and collagen type II as a result of enzymatic degradation and lack of regeneration. A number of enzymes responsible for degradation of cartilage matrix have been demonstrated in the rheumatoid joint, including the collagen-degrading matrix metalloproteinases I, III, and XIII, neutrophil-derived cathepsins L and D, and collagenase, as well as aggrecanase, which degrades proteoglycans. The matrix in which chondrocytes are embedded becomes depleted, with loss of chondrocyte numbers, suggesting that matrix degeneration is secondary to degradative effects mediated by both pannus and chondrocyte activity and cell death. Bone erosion is mainly mediated by activated osteoclasts. Enzymatic degradation of bone matrix beneath the microenvironment of osteoclasts is implicated in the evolution of bone erosions. There may also be an ineffective reparative response in the later stages of disease, as suggested by the presence of fibrous tissue replacing areas of destroyed cartilage and bone in joints removed at surgery.
Extra-articular features associated with rheumatoid arthritis include four types of tissue pathology: (1) diseased arterial walls, (2) the formation of extravascular nodules consisting of fibrotic lymphocyte–macrophage granulomas, (3) chronic inflammation of pleuro-pericardial surfaces, and (4) chronic immune inflammatory reactions in organs such as the lungs and salivary glands.
Two types of pathology are described in lesions involving arterial walls. The first is a bland fibro-intimal hyperplasia, without obvious inflammatory changes, resulting in vascular occlusion. This is typically observed in digital vessels of patients with longstanding disease and is associated with collateral blood vessel formation. It correlates with a history of benign, intermittent nailfold infarcts that develop in winter months. By contrast, the second type of lesion has polyarteritic pathology and is observed in patients with rheumatoid systemic vasculitis and a poor prognosis. Medium-sized and small arteries of the limbs, peripheral nerves, and organs are involved, but renal vessels are spared. Histopathological examination of involved vessels reveals lymphocytic, histiocytic, and inflammatory cell infiltration of the medial and perivascular area, disruption of the internal elastic lamina by fibrinoid necrosis, and proliferation of the vessel wall intima with intravascular thrombosis and occlusion.
Extravascular nodule formation in areas subject to pressure or friction, such as the elbow, is the characteristic granulomatous lesion of RA. Nodules have a central core of fibrinoid eosinophilic material surrounded by a palisade of histiocytes, occasional giant cells, and an outer layer of lymphocytes, fibroblasts, and fibrous tissue. Extravascular granulomatous inflammation, with or without nodule formation, has also been documented on the surface of the pleura, pericardium, sclera, and endocardial valves. As is the case with systemic vasculitis and Felty’s syndrome, the occurrence of nodules correlates with seropositive disease and the carriage of HLA DRB1*04 alleles. Notable among organ-based chronic inflammatory diseases are fibrosing alveolitis and/or bronchiolitis, and salivary gland exocrinopathy (secondary Sjögren’s syndrome).
Although the initiating cause of rheumatoid arthritis remains uncertain, there has been considerable progress in understanding the cellular and molecular mechanisms involved in chronic immune hyperreactivity, inflammation, and tissue damage. The immune-inflammatory reaction is characterized by expression of cytokines, adhesion molecules and chemokines that result in recruitment of lymphoid and inflammatory cells. These recruited cells interact with each other and local resident cells via cytokines and contact-mediated interactions, resulting in chronic inflammation and tissue damage.
B cells and autoantibodies
The discovery of rheumatoid factor in the blood of patients with rheumatism over half a century ago led to the immunological hypothesis of disease pathogenesis. As rheumatoid factor is an autoantibody directed against epitopes on the constant domains of the Fc portion of IgG1, the concept that rheumatoid arthritis is an autoimmune disease gained credibility. However, IgM rheumatoid factor occurs in a variety of other diseases in the absence of joint pathology. In the case of RA, rheumatoid factor complexes are present in synovial fluids, and IgG-producing B cells, whose rearranged immunoglobulin gene sequences implicate antigen stimulation, are present in inflamed synovium. B cells in rheumatoid joints also synthesize antibodies to some cartilage components such as collagen type II, but these are not disease specific.
By contrast, recent research has shown a highly disease-specific antibody directed against citrullinated peptides in the serum of patients and some experimental models of rheumatoid arthritis. The presence of such IgG ACPAs has been detected in healthy individuals several years before the onset of symptoms of RA and correlates with a severe disease course. Antibodies also play a crucial role in an experimentally produced, genetically modified, strain of KBN T-cell receptor transgenic mice, which develop antibodies toglucose-6-phosphatase and a rheumatoid-like disease. This disease can be transferred by the antibody to healthy mice, requiring complement and mast cells for its expression. The presence of antibodies to glucose-6-phosphatase in RA has been reported in one study, but subsequent studies failed to show diagnostic sensitivity and specificity of classical significance.
It seems possible that autoantibodies could interact with complement and Fc receptors expressed on cells in the rheumatoid joint and so contribute to inflammation. Alternative hypotheses include a role for B cells as producers of pathogenic cytokines and as cells that present antigens to T cells. The proven therapeutic efficacy of a B-cell-depleting monoclonal antibody supports a role for B cells in the pathogenesis of rheumatoid arthritis.
The predominance of CD4+ T cells in proximity to antigen-presenting cells in the rheumatoid joint suggests their involvement in perpetuating the immune response. This is also supported by the association of RA with HLA class II genes and the beneficial response to T-cell-directed therapies (lymphophoresis, ciclosporin, and the biological drug CTLA4-Ig, abatacept, which blocks interactions between antigen-presenting cells and T cells). In animal models, rheumatoid-like disease can be induced by immunizing with collagens and proteoglycans restricted to cartilage, administered with mycobacterial adjuvant. These cartilage antigens and citrullinated peptides such as fibrinogen and α-enolase that are present in inflamed joints are candidate autoantigens under current investigation in rheumatoid arthritis.
An alternative proposal envisages that disease is initiated as a result of molecular mimicry between epitopes on antigens present on infectious agents and autoantigens. Molecular mimicry, by a phenomenon known as epitope spreading, overcomes tolerance mechanisms and results in autoimmunity. Candidate exogenous antigens proposed include Escherichia coli bacterial heat shock protein DNAj and EBV gp110 envelope protein, which share a peptide sequence with the HLA DR shared epitope.
The predominant CD4+ T cells in joints appear to be of the Th1 type, bearing memory and activation markers such as CD45 RO+, CD45B dim+, VLA-4+, CD69+, and HLA class II+. However, whether these T cells are activated by a dominant antigen is debated, as they show variable T-cell receptor oligoclonality and antigen specificity. Moreover, T cells in joints do not proliferate but increase in number by recruitment and accumulation.
Recent data have highlighted a deficiency in a naturally occurring subset of CD4+ T cells that regulate the effector function of CD4+ T cells in vitro. This regulatory T-cell subset (Tregs) shows a characteristic phenotype and is CD4+, CD25bright, and FOXP3+ve. Tregs not only regulate responses to exogenous, potentially pathogenic antigens, but also promote self-tolerance to autoantigens, thereby preventing autoimmune disease. Treg function has been found to be defective in rheumatoid arthritis. As the abnormal function of T cells can be replicated in vitro by exposure to TNFα, it has been postulated that T cells in rheumatoid arthritis are conditioned by TNFα to promote T-cell-mediated inflammation. Other cytokines such as IL-15 and IL-6 produced in excess, and IL-10 produced in insufficient quantity, may also promote the loss of homeostasis of the immune-inflammatory reaction. Cell membrane contact between macrophages and cytokine-activated T cells may also be a key event in driving production of the pivotal cytokine TNFα (see below).
Cytokines are protein messenger molecules that transmit signals from one cell to another by binding to specific receptors on the surface of cell membranes. Their activity is usually restricted to adjacent cells in the local milieu. Cytokines are normally produced and exported as soluble molecules into the fluid phase, although some cytokines, such as TNFα, are also active as molecules displayed on the surface of the producer cells. Expression of mRNA and protein of a large number of cytokines is reported in rheumatoid synovial tissue, and these molecules regulate a diverse range of functions relevant to an understanding of the pathogenesis and clinical features of rheumatoid disease. Both pro- and anti-inflammatory cytokines, chemokines, and mitogenic factors are produced, but proinflammatory mediators predominate during active phases of disease.
Of the proinflammatory cytokines, TNFα, IL-1, and IL-6 are of key importance in the pathogenesis of inflammation in rheumatoid arthritis. Experimental evidence in vitro and in animal modelssuggests that they are intimately involved in activation of the cytokine network, leucocyte recruitment and activation, the local immune response, angiogenesis, and fibroblast proliferation in joints. IL-1 and TNFα also regulate production of a number of mediators of connective tissue damage by synoviocytes, including matrix metalloproteinases and prostaglandins. Furthermore, these cytokines in combination with the cytokines monocyte colony stimulating factor (M-CSF) and receptor activator of NF-κβ ligand (RANKL), activate osteoclasts that are implicated in bone damage.
TNFα is produced mainly by type A cells of the macrophage lineage in the intima, subintima, and cartilage–pannus junction. The p55- and p75-TNF receptors are coexpressed by cells in the vicinity. The hypothesis that TNFα is a dominant proinflammatory mediator in the cytokine disequilibrium observed in the rheumatoid synovium has gained considerable support. In particular, TNFα regulates production of IL-1 and together these two cytokines orchestrate rheumatoid inflammation and damage. The identification of TNFα as a molecular target for therapy has been validated by clinical trials of biological inhibitors of TNFα—a monoclonal chimeric and a human anti-TNF antibody, as well as a soluble TNF receptor-IgG Fc fusion protein—that bind to TNFα, thereby neutralizing its activity. The IL-1 antagonist, human recombinant IL-1 receptor antagonist (IL-1RA), is relatively less effective, but its efficacy supports a role for IL-1 in disease pathogenesis.
The importance of TNFα as a mediator of rheumatoid disease is also supported by a number of observations on blood and joint tissues in clinical studies. For example, following treatment with infliximab, a monoclonal anti-TNFα antibody, there is a: reduction in the expression of adhesion molecules and chemokines involved in the recruitment of immune inflammatory cells; reduction in angiogenesis; reduction in the production of proinflammatory cytokines; and an inhibition of molecular and cellular pathways of cartilage and bone destruction.
Many other cytokines and chemokines have been described in rheumatoid joints and implicated in the pathogenesis of disease, and interventional studies have proved to be a powerful tool for investigating their role in disease. In clinical trials in RA, blockade of IL-6 and IL-15 has shown promise, supporting their proinflammatory role. Many other cytokines expressed in disease may be implicated, but because of overlapping function with other cytokines they may be redundant in this respect. Thus far the blockade of chemokines, e.g. MCP-1 and IL-8, by biological agents has not proved successful, and similarly the administration of anti-inflammatory cytokines such as human-recombinant IL-4, IL-10, and IL-11 in clinical trials has proved to be ineffective at the doses used.
Clinical features of rheumatoid arthritis
The onset of rheumatoid arthritis is frequently insidious, and the principal symptoms are pain and stiffness, mainly of peripheral joints, with associated swelling. Prolonged stiffness of joints on waking and following inactivity is usual and may last an hour or more. There is progressive decline in physical function and ability to perform daily activities. Fatigue and lethargy are common and there may also be low-grade fever and weight loss. Symptoms are persistent in affected joints, although there may be some day-to-day variation in severity. As the disease evolves, further joints may become involved and some may remit, but ultimately the distribution of arthritis becomes permanently established.
Other patterns of disease presentation are also recognized. Up to one-third of patients present with an explosive or subacute onset of arthritis, leading to severe immobility. In a minority of patients a migratory polyarthritis flitting from joint to joint is observed. This is referred to as palindromic rheumatism and may be a recurring pattern over months before chronic polyarthritis becomes established. About 10% of patients present with features of the syndrome of polymyalgia rheumatica, characterized by prominent limb-girdle pain, stiffness, and painful movement of the neck, shoulders, and hips. Persistent inflammatory arthritis of a single joint such as the knee, wrist, ankle, shoulder, or hip may be the only rheumatological symptom and can antedate the onset of polyarthritis by months or years.
In some patients bilateral diffuse swelling of the fingers and hands may be a presenting complaint, often associated with symptoms of carpal tunnel syndrome. Synovitis of tendon sheaths of the dorsal extensors of the wrist and of flexor tendons in the palm and wrist may be present with concurrent joint signs, but may also occur as a prominent clinical feature in the absence of polyarthritis. Swelling of the ankles with pitting oedema is commonly seen in active rheumatoid arthritis. Lymphoedema of the forearm or lower limb is observed less frequently.
Rarely, the initial manifestations of rheumatoid arthritis are confined to extra-articular disease. Examples include subcutaneous nodules, one or more nodules in the thorax presenting as pulmonary lesions on a chest radiograph, pleurisy with pleural effusion, pericarditis, episcleritis, and vasculitis.
The expression of rheumatoid arthritis shows interindividual variation with respect to the anatomical sites and numbers of involved joints. For example, some patients have mainly small joints affected, whereas others show simultaneous involvement of small and large joints. The hip and shoulder joints may be spared in some, while in others they bear the brunt of the disease. The actual numbers of diseased joints can vary from three or four to over fifty. Diseased neck joints may be asymptomatic until, in the late stages, neurological complications alert the physician to subluxation of the cervical spine or the atlantoaxial joint.
In over 80 to 90% of patients, one or more of the metacarpophalangeal and proximal interphalangeal joints of the hand and the metatarsophalangeal joints are involved. Other frequently involved sites include the wrists, glenohumeral joints of the shoulders, knees, and the elbow joints, followed by the midtarsal, acromioclavicular, interfacetal, and atlantoaxial joints of the cervical spine and hip joints. The temporomandibular, sternoclavicular, and cricoarytenoid joints are involved in about one-third of patients.
Symmetrical involvement of the joints is usual, but joint damage and deformity may be asymmetrical and related to overuse or traumatic injury. Conversely, neurological paralysis of a limb results in joint protection.
In addition to involvement of diarthrodial joints, the rheumatoid process frequently involves tendon sheaths of hands, wrists, shoulders, and ankles.
Features of joint disease
Hands and wrists
In active rheumatoid disease, soft tissue swelling and tenderness of metacarpophalangeal and proximal interphalangeal joints is observed. Thickening and nodularity of flexor tendons in the palms may be palpable and tenosynovitis can be a cause of ‘triggering’ of the fingers. Wasting of the interossei is prominent and fist closure restricted. Flexor tendonitis and wrist synovitis may be associated with signs and symptoms of median nerve compression (carpal tunnel syndrome).
Ulnar deviation and volar subluxation of the digits and wrists may develop later. Other recognized deformities include boutonnière (buttonhole) flexion deformity of the proximal interphalangeal joint and ‘swan-neck’ deformities of fingers due to hyperextension of the proximal interphalangeal joint and flexion at the distal interphalangeal joint.
Diffuse synovial swelling may be pronounced at the dorsal aspect of the wrist and the ulnar styloid may become dorsally subluxed. The carpus may drift in a volar direction such that supination of the hand is restricted. In this late stage, the extensor tendons appear stretched across a shrunken carpus (‘the bowstring’ sign). Extensor tendons may occasionally rupture, most commonly affecting the little or ring fingers.
Nail fold and fingertip infarcts and splinter haemorrhages indicate digital vascular occlusive disease. Palmar erythema is common but not specific for rheumatoid arthritis.
Elbows and shoulders
Physical signs in early stages include swelling, limitation of movement, and inability to flex or extend the elbow. Later, pronation and supination are restricted, and the head of the proximal radioulnar joint may dislocate. Olecranon bursitis and subcutaneous nodules around the elbow are common. In the shoulder, aside from glenohumeral joint synovitis, there may be accompanying subacromial bursitis and rotator and biceps tendon involvement.
Rheumatoid involvement of the apophyseal joints of the neck can cause pain, stiffness, and restricted movement. Loss of stability in the spine may occur at several levels and be associated with symptoms and signs of radicular or cord compression. Subluxation of the atlantoaxial joint diagnosed by plain radiography or MRI occurs in 6% of the rheumatoid population and up to 30% of patients who are admitted to hospital. It may be asymptomatic, but when severe tends to occur in patients who also suffer from severe generalized disease and advanced disability, and is a recognized cause of quadriplegia and sudden death.
Involvement of the knees is common, and chronically active synovitis is associated with irreversible destruction and rapid deterioration in functional capacity. In early stages especially, high pressure in the knee joint on active flexion, e.g. during squatting, can lead to joint rupture and leakage of inflammatory fluid into the calf. This complication simulates signs and symptoms of a calf deep vein thrombosis: it can be diagnosed by arthrography using contrast medium, or by ultrasonography of the knee. A chronic effusion in the knee joint may also be associated with a posterior popliteal (Baker’s) cyst and occasionally this extends into the medial aspect of the calf.
Ankles and feet
Inflammation of the metatarsophalangeal joints is common and results in subluxation of the metatarsal heads and, ultimately, claw- or hammer-toe deformities. The soft-tissue pad that is normally positioned underneath the metatarsal heads becomes displaced such that the heads of the metatarsal bones become painful to walk on. Patients may describe this as feeling as if they were walking on marbles or stones. Involvement of the tarsal and subtalar joints may result in flattening of the arches of the foot and valgus deformity of the hindfoot. These deformities cause difficulties with footwear, and where shoes rub the feet there is a tendency for callosities to form.
The hips are less often involved, but there may be erosions in severe cases with remodelling of the acetabulum (protrusio acetabuli). There may also be secondary degenerative disease at the hip.
Nodules occur in 25 to 30% of patients with rheumatoid arthritis and are associated with seropositive disease. Common sites for subcutaneous nodules include the elbow, ischial tuberosity, heel, and dorsum of fingers. Multiple, small, rapidly evolving nodules can occur in those on methotrexate treatment. Nodules in the pleura may present as single or multiple round shadows on a routine chest radiograph.
Rheumatoid vasculitis occurs in patients with seropositive and nodular disease. It presents with a severe systemic illness characterized by fever and weight loss. Associated clinical features are consequent upon occlusion of medium- to small-sized arteries. These include Raynaud’s phenomenon, nail fold and digital infarcts, and gangrene, skin ulceration, mononeuritis multiplex, scleromalacia perforans, and occlusion of arteries to visceral organs. The latter include coronary, pulmonary, coeliac axis, and cerebral vessels. In some patients vasculitis may present as a skin rash associated with necrotizing polyangiitis of small cutaneous blood vessels.
Fibrosing alveolitis and obliterative bronchiolitis
Physiological abnormalities in lung function tests indicative of airways and interstitial disease may be present without symptoms. In some patients with RA, more frequently men than women, dyspnoea of insidious onset, physical signs, characteristic lung function abnormalities, a chest radiograph, and high-resolution CT may reveal characteristic features of chronic fibrosing alveolitis. More rarely, acute pneumonitis may be the presenting feature, with rapid deterioration and development of respiratory failure. Patients with fibrosing alveolitis are usually seropositive, have a high frequency of antinuclear antibodies, and may also exhibit evidence of multisystem disease, including vasculitis.
Obliterative bronchiolitis can be associated with rheumatoid arthritis. It is usually rapidly progressive, but some patients follow a chronic protracted course that may respond to corticosteroid and immunosuppressive therapy.
Evidence of previous pericardial and pleural inflammation is common at autopsy and may be discovered by imaging techniques in asymptomatic patients. Both may present with clinical symptoms, generally following a benign course with resolution associated with disease-modifying antirheumatoid drugs (DMARDs) or corticosteroid therapy. Rare cases of constrictive pericarditis have been reported. Typically, pleural effusions are exudates with a high protein content and cellular exudate enriched in lymphocytes, but also containing polymorphonuclear cells and macrophages. A low level of complement activity relative to blood concentrations and a low glucose concentration (usually <1.4 mmol/litre) is of diagnostic value.
Scleritis, episcleritis, scleromalacia perforans, corneal melt, and keratoconjunctivitis sicca have all been described and need evaluation and treatment by a specialist.
Secondary amyloidosis due to deposition of amyloid AA fibrils in blood vessels and parenchyma of kidneys, liver, spleen, and gastrointestinal tract has been described in the tissues of 10 to 15% of patients examined at autopsy, or in the blood vessels in the submucosa of rectal and gingival biopsies. Proteinuria, nephrotic syndrome, or renal failure are less common and have a poor prognosis unless detected and treated before irreversible renal failure has occurred.
Juxta-articular osteoporosis is a common feature of radiographs of affected joints and is related to local disease activity. However, decreased bone mineral density of the spine and pelvis has been described in patients with active severe rheumatoid arthritis. This is likely to reflect the response of bone metabolism to prostaglandins and catabolic cytokines such as IL-6, IL-11, and the receptor for activation of NF-κβ (RANK) ligand, which increase osteoclast activity. This is distinct from immobility-associated or corticosteroid-induced osteoporosis, although these factors may be additive in individual patients. It has been suggested that increased mobility following low-dose prednisolone may be beneficial and reverse, rather than aggravate, corticosteroid-induced osteopenia.
Felty’s syndrome is characterized by a combination of seropositive rheumatoid arthritis, neutropenia, and splenomegaly. Lymphadenopathy, leg ulcers, and nodular hyperplasia of the liver have been described. Patients with severe neutropenia are liable to bacterial infections. Some patients also develop anaemia and thrombocytopenia. In a variant of Felty’s syndrome, an expansion of large granular lymphocytes is found in the blood: these are cytotoxic CD8+ lymphocytes and may present as clonally expanded cell populations.
Myocardial disease due to diffuse fibrosis or granulomatous lesions is recognized in rheumatoid arthritis, although the more frequently recognized association is with coronary artery disease. Systemic vasculitis may also involve coronary vessels. Aortic incompetence due to valvular thickening and nodule formation or dilation of the ascending aorta have been described.
A number of compression neuropathies may occur in rheumatoid arthritis. These include compression of the median nerve at the wrist, the ulnar nerve and posterior interosseous branch of the radial nerve at the elbow, and posterior tibial nerve at the level of the knee or ankle. It is important to recognize and confirm these neuropathies by nerve conduction studies, as surgical decompression usually cures symptoms.
A mild, symmetrical, sensory peripheral neuropathy involving the hands and legs in a ‘glove and stocking’ distribution also occurs in rheumatoid arthritis. This is distinct from the rarer and more severe sensorimotor mononeuritis multiplex associated with wrist and foot drop and usually due to vasculitis of vasa nervora, when other features of a systemic vasculitis and extra-articular disease may be present. In some patients, however, no vascular pathology is demonstrable and the cause of axonal degeneration is not understood.
Rheumatoid involvement of the transverse ligament and odontoid process of the atlantoaxial joint may lead to posterior subluxation or upward movement of the odontoid and cause cervical cord compression. Cord compression may also occur because of rheumatoid damage at lower levels of the cervical spine. Compression is a recognized cause of tetraparesis and sudden death. Surgical stabilization of the neck can be successful but cannot always be recommended in patients with associated severe disability and poor health status.
Patients with rheumatoid arthritis are susceptible to local and systemic bacterial, and opportunistic infections. Infections of joints, respiratory and urinary tracts, skin ulcers, and septicaemia are all described, and infections are one of the causes of increased mortality in the condition. Endogenous disease-related and iatrogenic immunosuppressive mechanisms are thought to play an important part. Neutropenia compromises host defence in Felty’s syndrome.
The incidence of lymphoproliferative disease, mainly B-cell lymphoma, is significantly increased in rheumatoid arthritis. In a case–control retrospective study from Sweden in an era when cytotoxic agents were not in use, the risk of lymphomas was increased several fold in patients with severe inflammatory disease. Confounding variables such as smoking are thought to play a part in the increased risk of lung cancer reported in some studies.
The course of the disease activity fluctuates over time, partly due to the endogenous mechanisms of disease and partly as a result of effective therapy. Recurring periods of weeks or months of exacerbation of symptoms, described as ‘flares’, alternate with periods of relative quiescence of disease. In about 10 to 20% of patients the disease continues unabated throughout.
The key clinical features of disease activity in rheumatoid arthritis are pain, fatigue, stiffness of joints on waking, swelling, tenderness of joints on palpation, restriction of joint motion, and loss of physical functional capacity (see below). Joint deformities become apparent as the disease progresses. Symptoms are assessed by taking a history in descriptive terms, but also by attempting to quantify their severity by a visual analogue scale. These measurements have been incorporated into various criteria for assessment of disease activity, remission, and response to therapy, developed and validated e.g. by the ACR and the European League Against Rheumatism (ELAR).
Swelling of joints caused by synovial thickening may be detected by palpation as a ‘spongy’ or ‘boggy’ feel. Concomitant effusion can be demonstrated by fluctuation. In later stages of disease, subluxed surfaces of bones (such as the heads of metacarpals in the hands, the styloid of the ulna, and distal radius at the wrist) can give the appearance of bony swelling. Tenderness is elicited by digital pressure or squeezing of a joint. The classic signs of inflammation, such as redness and increased temperature overlying joints, are not usually prominent, although readily demonstrable by thermography. Active and passive movement of joints through their anatomical range of motion elicits restriction of movement associated with pain.
Functional capacity can be assessed by testing grip strengths using an inflatable bag attached to a sphygmomanometer, walking time over a standard distance, and by standard health assessment questionnaires (such as the Stanford questionnaire). In the early and mid stages, disease activity is the major determinant of impairment of physical function. The degree, quantity, and severity of pain is recorded as experienced by the patient, graded on a visual analogue scale of 1 to 10. The duration of morning stiffness is recorded in minutes. A ‘global assessment’ of disease activity on a visual analogue scale of 1 to 10 as judged by the patient and physician may also be used as a quantitative measurement of disease activity over time.
The rheumatoid disease process leads to structural damage to the cartilage, bone, and associated joint structures. Serial radiographs of the hands and feet are employed to assess structural damage to joints, with more sensitive techniques including ultrasonography and MRI employed for research studies. Damage is cumulative and irreversible and appears to be related to the severity of inflammatory activity over time. In later stages of disease, loss of normal joint architecture and mechanical derangement are the major cause of disability.
The longer-term health status of patients presenting to hospital clinics and in the community with recent-onset rheumatoid arthritis has been documented in a number of studies. Functional deterioration occurs rapidly, but the trajectory of the course of disease varies considerably in individual patients. In one hospital-based study, half were moderately disabled in 2 years and severely disabled by 10 years, with a severe impact on employability.
Patients with rheumatoid arthritis have a higher than expected prevalence of other serious illnesses and an increased mortality compared with the general population. Survival rates of about 50% at 5 years have been recorded in a subset of patients with severe polyarticular disease, poor functional status, or extra-articular disease. Rheumatoid arthritis itself may contribute to premature death in patients as a result of recognized extra-articular complications such as fibrosing alveolitis, vasculitis, secondary amyloidosis, cardiac disease, or transection of the cord due to cervical spinal subluxation. More frequently, death is the consequence of comorbid conditions, e.g. premature atheromatous coronary and cerebrovascular disease, predisposition to infections and lymphoproliferative diseases. Most of these complications and comorbidities are now considered to be a result of the systemic nature of molecular pathways involved in inflammation and a dysregulated immune response characteristic of the disease. Iatogenic causes also play an important role, e.g. causing gastrointestinal haemorrhage or perforation, renal failure, liver damage, and bone marrow suppression. However, treatment with DMARDs, especially methotrexate, is now recognized to improve the quality of life and life expectancy of rheumatoid patients.
A number of prognostic factors that herald rapid functional deterioration and premature death have been identified in cohorts of patients with rheumatoid arthritis. These include a large number of affected joints, a persistently raised level of acute phase proteins, detection of bony erosions on radiographs, lower socioeconomic status, early development of functional incapacity, a positive rheumatoid factor and ACPA tests, cryoglobulinaemia, and (in northern European patients) the presence of the HLA DRB*04 genes. Their presence in combination is predictive of progressive, severe disease. However, on an individual patient basis the reliability and weighing attributed to each factor requires validation.
Criteria for defining remission have been developed by the ACR and ELAR. The former are very rigorous and require absence of specific symptoms and signs and a normal ESR for two consecutive months; the latter is a composite disease activity score, a continuous variable measurement, derived from a mathematical formula based on specified signs, symptoms, and ESR. These criteria do not give concordant results and have shortcomings, the former criteria being achieved by a smaller proportion of patients than the latter. They have been variably applied in epidemiological studies and clinical trials. However, there is general agreement that spontaneous remission of established disease is rare (<15% of patients). Drug-induced remission at single time points following therapeutic interventions is more common, being documented in up to about 50% patients, especially with early disease, but is not frequently long lasting. Progressive structural damage of joints has been observed despite remission by these criteria in some studies, suggesting that more sophisticated biomarkers and imaging techniques are required to define true disease remission.
Diagnosis and stages of disease
A diagnosis of rheumatoid arthritis is likely if three or more symmetrically distributed joints are found to be swollen and tender for more than 6 weeks in a patient with a positive rheumatoid factor and/or ACPA test and elevated ESR or serum concentration of C-reactive protein. However, not all these features are necessarily present at the early stages of disease, at which point arthritis is best termed ‘undifferentiated’. The broad spectrum described in the ‘presentation’ section may cause considerable difficulty in making a definitive diagnosis.
With the passage of time, the emergence of other features, such as subcutaneous nodules, radiographic evidence of joint space narrowing, juxta-articular osteopenia, and bony erosions add further certainty to the diagnosis.
The ACR criteria in Bullet list 1 above may not be fulfilled for 6 to 12 months, by which time the pattern of joint involvement and a chronic disease course are usually evident. Prognostic factors declare themselves and the patient is regarded as having reached the stage of established disease.
In patients with recent onset of symptoms of arthritis, the following disorders should be considered in the differential diagnosis.
Polyarthritis associated with connective tissue disease
Systemic lupus erythematosus (SLE) may present with chronic nondeforming polyarthritis, but features such as Raynaud’s phenomenon, photosensitivity, rashes, alopecia, haemolytic anaemia, leucopenia, thrombocytopenia, and renal or neurological involvement are detectable sooner or later, and diagnostic antinuclear antibodies (anti-ds-DNA, anti-Sm and others) are present. Other connective tissue diseases such as systemic sclerosis, polymyositis, mixed connective tissue disease, ‘overlap’ syndromes, and primary Sjögren’s syndrome may also present with marked polyarthralgia or polyarthritis that mimics rheumatoid arthritis. In many such patients, the presence of rheumatoid factor can further confuse the diagnosis. Careful clinical examination and measurement of marker autoantibodies directed against nuclear and cytoplasmic antigens will usually permit recognition of the underlying disorder. In some cases the diagnosis may only unfold after a period of weeks or months of a disorder best labelled ‘undifferentiated connective tissue disease’.
The polyarthritis of rubella or other microbial agents, such as parvovirus B19 and Borrelia burgdorferi, and reactive arthritis associated with genitourinary or gastrointestinal infections can all cause diagnostic difficulty. A positive diagnosis is made by microbiological tests on relevant body fluids and serological tests for the detection of IgM antibodies or a rising titre of IgG antibodies to the suspected microorganism in sequential serum samples taken over 2 weeks.
Peripheral joint disease can be seen in conjunction with ankylosing spondylitis, psoriasis, and inflammatory bowel disease. Clinical examination of the spine, skin, and nails, radiological examination of the bowel using double-contrast enema or small bowel enema, endoscopy, and biopsy may reveal the underlying diagnosis. Many patients are HLA B27 positive.
Osteoarthritis may present with inflammatory symptoms and signs but is readily distinguished by its different joint distribution (proximal and distal interphalangeal joints, carpometacarpal joints of the thumb) and radiographs that show joint space narrowing, subchondral new bone formation, osteophytes, and subchondral cysts. Where there is pre-existing osteoarthritis, the superimposition of RA can be difficult to distinguish.
In late middle-aged and elderly patients the clinical presentation of polyarticular chronic pyrophosphate arthropathy may be difficult to distinguish from rheumatoid arthritis. The former diagnosis may be suspected where there is an atypical distribution of synovitis together with periarticular complications. Chronic pyrophosphate arthropathy may be associated with a modest acute phase response and low-titre rheumatoid factor, but can usually be distinguished from RA on the basis of typical radiographic appearances and the finding of calcium pyrophosphate dihydrate (CPPD) crystals in synovial fluid aspirates. Rarely, RA and chronic pyrophosphate arthropathy may coexist.
Other diagnoses to be considered include hypermobility syndrome, polyarticular gout, psoriatic arthritis, haemochromatosis, sarcoidosis, sickle-cell disease, primary amyloidosis, and paraneoplastic disease.
Laboratory studies are an integral part of the management of patients with rheumatoid arthritis and are employed for diagnosis, evaluation of prognosis, assessment of disease activity, response to therapy, and monitoring toxic effects of drugs. Only routinely used tests are considered here.
The measurement of IgM rheumatoid factors is useful in the early stages of assessment of a patient with suspected rheumatoid arthritis. In a patient with recent-onset polyarthritis, a positive rheumatoid factor is moderately specific for RA, but can also be observed in patients with other connective tissue diseases such as SLE and primary Sjögren’s syndrome. A repeat test may be positive after an initial test is negative and is therefore necessary before a patient can be categorized as having seronegative rheumatoid arthritis. A significant titre of rheumatoid factor is associated with a poor prognosis and extra-articular disease.
The detection of IgG antibodies to citrullinated proteins (ACPAs) has been extensively validated as specific for rheumatoid disease and may be positive in the absence of rheumatoid factor. As routinely measured by a widely used enzyme-linked immunoassay (ELISA) ‘kit’ that detects antibodies to citrullinated synthetic peptides (anti-CCP), the sensitivity increases from around 50% in early disease to around 80% in established disease with more than 95% specificity.
Measurement of ESR (Westergren method) and serum C-reactive protein are extensively used. High values correlate with disease severity and a reduction is one criterion of response to therapy. Persistently elevated C-reactive protein concentrations correlate with deforming erosive disease.
Patients with active rheumatoid arthritis show haematological abnormalities as a consequence of disease-related mechanisms. A high level of disease activity is associated with a normocytic normochromic anaemia, polymorphonuclear leucocytosis, and thrombocytosis. An exception is Felty’s syndrome, associated with neutropenia and thrombocytopenia. These abnormal values tend to return to normal as the inflammatory component of disease responds to therapy.
Active disease may also be associated with a raised serum alkaline phosphatase and a low serum albumin. Serum chemistry is otherwise normal. Serum immunoglobulin and complement C3 and C4 levels may be elevated.
Nonsteroidal anti-inflammatory drug (NSAID) therapy may cause microcytic iron-deficiency anaemia from blood loss from the gastrointestinal tract. A low serum ferritin level suggests iron deficiency, but this is not a reliable guide in cases of rheumatoid arthritis, as serum concentrations may be elevated as part of an acute phase response. Corticosteroids may be responsible for increased polymorphonuclear cell counts and decreased lymphocyte counts. Many DMARDs show dose-related bone marrow toxicity, and sulfasalazine, D-penicillamine, azathioprine, and gold can cause unexpected agranulocytosis as a result of hypersensitivity, unrelated to the dose administered.
Of the commonly used drugs, methotrexate, sulfasalazine, azathioprine, and leflunomide are hepatoxic and can cause elevation of liver enzymes and alkaline phosphatase. Repeated monitoring is advisable: persistent or highly raised values should prompt further investigation or discontinuation.
Radiographs of hands and feet can be used to assess the presence and progression of cartilage loss and bone erosions. Standardized measurements (the Larsen or Sharp scoring methods) have been devised to quantify these measures. Changes seen in the hands and feet correlate with radiological changes in other affected joints, showing a linear progression after the initial 1 to 2 years. The erosion count correlates with physical function. Radiographs of affected joints are used for the assessment of integrity and damage. Flexion views of the cervical spine are suitable for demonstration of atlantoaxial subluxation and cervical instability. Arrest or retardation of radiographic change is considered to be a marker of good control of disease.
MRI and CT are valuable in assessing neck pathology and pressure on the cervical cord. MRI and high-frequency ultrasound examination are sensitive methods to evaluate synovitis and early change in cartilage and bone, but their place in routine management is not yet established. Dual emission X-ray absorptiometry (DEXA) scanning is in routine use for the assessment of bone mineral density.
Aims of treatment
The aims of treatment are to:
- ◆ relieve symptoms and signs of disease and induce remission
- ◆ maintain normal physical function
- ◆ prevent structural damage to joints and associated structures
- ◆ restore and maintain quality of life that permits the pursuit of normal work, domestic, and social life
- ◆ reduce the comorbidity and increased mortality associated with the disease and therapies
- ◆ correct abnormal laboratory-based values of haemotopoietic function, acute phase proteins and other markers of disease process
Considerable progress has been made in developing effective therapies for the rapid relief of symptoms and signs of disease. However, despite the best therapies in current use, the goals of durable drug-free remission halting structural damage and maintaining a normal quality of life, over the long term, have not yet been realized, although significant progress has been made. The realistic aims, therefore, are to maximize gains while minimizing toxicity of drugs (an optimum risk:benefit ratio) and to operate within the pharmacoeconomic constraints (cost–benefit and cost–utility) that apply in the setting in which the patient is being treated.
The costs of treatment of rheumatoid arthritis over the lifetime of a patient are considerable. Direct costs increase over the course of disease and include those of hospital admissions, drugs, surgery, aids, and appliances. Indirect costs include those arising from loss of economic productivity and earnings, unemployment and disability benefits, the cost of maintaining mobility, and domestic help and daily care for the severely disabled.
Once an individualized management plan has been instituted, response to therapy should be monitored to ensure efficacy, using quantifiable clinical and laboratory indices of inflammatory activity and impact on the progression of damage to joints. A lack of response to initial therapy should trigger a change in the management plan and consideration of alternative strategies at intervals of 3 to 6 months. A thorough knowledge of the scope and limitations of treatment modalities is essential in the art of management.
With aggressive and continuing use of disease-modifying chemical and biological therapeutic agents, it is possible to achieve low disease activity in most and maintain remission in some patients. As clinically defined low or absent disease activity correlates with retardation of joint damage, the long-term benefit of treatment is likely to be most marked in this early phase of disease.
Nonpharmacological measures and support
Other essential elements of the management include measures such as patient education; psychological and employment counselling; setting appropriate levels of rest and exercise; coping with tasks of daily living and maintaining mobility; access to splints, aids, and appliances for the disabled; and access to social and financial benefits. The provision of holistic care thus requires teamwork and coordination between the treating physician and other medical and healthcare professionals, including specialist nurses, physiotherapists, occupational therapists, and social workers. Fully involving patients in their management improves outcomes.
Bed rest and the use of resting splints may be helpful during the very acute stages of joint disease, but should always be accompanied by daily passive joint movements and appropriate isometric exercises to avoid contractures, muscle atrophy and osteoporosis, and to retain joint function. Exercise initiated under supervision and maintained by patients on a regular basis does not accelerate joint damage and is effective in diminishing pain and promoting a sense of well-being in those in whom fatigue is a major feature of active disease.
Dietary manipulation—including fasting and exclusion of certain foods and beverages, and a vegetarian diet—has enjoyed popularity and in some patients appears to be beneficial, but there is little evidence that most such diets are of durable value. Diets rich in fish oils and ω – 3 fatty acids appear to be of some benefit. As excessive weight accelerates joint damage and increases the risk of complications when undergoing essential surgery, obese patients should be encouraged to lose weight.
Surgical treatment plays an important and essential role in relieving intractable symptoms and restoring loss of physical function and mobility caused by damage to joints, tendons, and associated soft tissues. It is also indicated in the treatment of secondary complications such as entrapment of peripheral nerves at the wrist and elbow and cervical cord compression due to instability of the cervical spine.
Maintenance of mobility requires attention to foot care, podiatry, comfortable shoes, a walking stick or elbow crutches, and specially adapted motor vehicles to get to work and for social purposes.
In the United Kingdom disabled people have certain privileges in employment and may qualify for disability allowances. Some may benefit from retraining for suitable work. The health care team needs to recognize that chronic illness and disability places increased pressure on spouses and family, who generally end up as carers of the patient with rheumatoid arthritis: support and counselling should therefore extend to them.
Mild disease may be defined as rheumatoid arthritis with limited joint involvement and low disease activity, and without markers of poor prognosis. Such patients will typically show most of the following features: involvement of less than six or seven individual joints and sparing of weightbearing joints; pain readily controlled with NSAIDs; less than 15 min of joint stiffness on waking or following inactivity; lack of extra-articular disease; minimally elevated ESR or concentration of C-reactive protein; negative rheumatoid factor test; a normal haematological profile; little or no impairment of physical function; and ability to undertake activities of daily living, maintaining employment, and enjoying nonstrenuous social and leisure activities. Radiographs of hands and feet show a lack of significant osteopenia, joint space narrowing, and bony erosions at baseline and annual follow-up. The disease course may be punctuated by self-limiting exacerbations of symptoms and signs. Patients with mild disease are only a small proportion of those referred to specialist clinics but are more numerous in the community and in the primary care setting.
Drug treatment involves judicious use of NSAIDs and analgesic drugs. Corticosteroid injections into individual affected joints, tendon sheaths, and bursas for persistent swelling, tenderness, or loss of normal range of movement can be very effective. Follow-up assessment is necessary to ensure that the disease has not evolved to a more severe pattern. DMARDs are indicated in those with recurrent or persistent symptoms and signs, deformities, or radiographic evidence of structural damage. Hydroxychloroquine or sulfasalazine are used initially. If the decision to embark on the use of DMARDs is made, the aims and management strategy are the same as for patients with moderate or severe disease.
Moderate or severe disease
Rheumatoid disease is defined as moderate or severe when it has evolved into an unremitting pattern of polyarthritis with evidence of significant functional impairment and joint damage in early stages of presentation. With increasing severity most of the following features are present: 10 to 30 swollen and tender joints; frequent involvement of proximal joints of the arms and legs; moderate to severe pain; inactivity and morning stiffness exceeding 1 h in duration; prominent fatigue; elevated ESR and/or C-reactive protein concentrations; low haemoglobin concentration, polymorphonuclear leucocytosis, and thrombocytosis; and positive rheumatoid factor test. Deformities of joints are apparent early in the course of disease and radiographs of hands, feet, and affected joints already show loss of joint space and subchondral erosions within 2 years of presentation. Such patients have significant impairment in daily activities and restricted ability to perform domestic and work-related tasks and to enjoy social and leisure activities.
The aim of drug treatment is to achieve rapid control of disease activity and, if possible, remission of disease. This usually requires simultaneous or sequential use of drugs belonging to different classes, e.g. NSAIDs, DMARDs, corticosteroids, and biological therapies, as discussed below.
So-called ‘traditional’ NSAIDs are used at optimal doses for control of pain and stiffness (see Table 1). Many physicians prefer to administer NSAIDs with a short half-life in slow-release preparations to be taken in the morning and before the patient retires to bed at night. In patients with a history of gastric intolerance, COX-2-selective NSAIDs may be preferable, or else the simultaneous use of a gastroprotective agent, most commonly a proton pump inhibitor. In addition, simple analgesics such as 0.5 to 1 g paracetamol every 6 h may be required for relief of pain. In clinical practice, a change of NSAIDs may be required to establish which of the several NSAIDs available is most suited to an individual patient (see section on antirheumatic drugs for detailed discussion).
|Table 1 Current NSAIDs: dosage for rheumatoid arthritis in adults|
|COX inhibitor||Dose (mg)/24 h|
|Ibuprofen||400–800, 3 times|
|Diclofenac||25–50, 3 times or slow release 75 once or twice|
|Nabumetone||1000 at night; to 1000 twice|
|Fenoprofen||300–600, 3–4 times|
|Ketoprofen||50, 3–4 times|
|Indometacin||Slow release 75, once or twice or 25–50 morning and noon, and 50–100 at night|
|COX-2 selective (Cox 2>1)|
|Etodolac||300 twice or 600, slow release once|
From Arnett FC et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and Rheumatism 31 (1988), pp315–34.
DMARDs should be used in all patients (see Table 2), the two most commonly employed being sulfasalazine and methotrexate, provided there are no contraindications. These are usually commenced as a single drug given in incremental doses over 3 to 4 months to the maximum recommended or tolerated dose. In most centres methotrexate is the drug of choice because of its efficacy and superior durability in the long term; indeed it has been described as the ‘gold standard’ drug. If a clearcut reduction in disease activity (or remission) is not observed with one of these drugs, then other DMARDs are added at this stage. Commonly used DMARD combinations include: methotrexate and hydroxychloroquine; methotrexate, sulfasalazine and hydroxychloroquine; and methotrexate and leflunomide or ciclosporin. The choice of therapy is ultimately determined by evaluation of risks of toxicity, efficacy, durability, and direct and indirect costs of treatment. There is no consensus on the most effective combination regimen. Meticulous monitoring for toxic effects is necessary.
In practice, over 50% of patients with moderate or severe disease require coadministration of corticosteroid therapy (see section on corticosteroids below). If continuing long-term use appears necessary, the aim should be to reduce the dose to the equivalent of 5 to 7.5 mg of prednisolone daily by more aggressive use of DMARDs, or instigation of anti-TNF therapy. By 2006, anti-TNF drugs, especially given concomitantly with methotrexate, had become the standard of care in patients with continuing active disease despite prior exposure to DMARDs, one of which is methotrexate (for further information including safety issues see section on biological therapy below). The guidelines proposed by the National Institute of Health and Clinical Excellence (NICE) in the United Kingdom require high disease activity and nonresponsiveness to two DMARDs before anti-TNF biologicals can be prescribed. Anti-TNF drugs have become widely used, with an estimated 1 million patients exposed to them for a variety of indications, but mostly RA. High cost continues to restrict access to treatment with biologicals.
As 30 to 40% of patients do not respond to one of the three anti-TNF agents and develop resistance to therapy, switching to another (e.g. infliximab to etanercept or adalimumab or vice versa) appears to be effective in some patients. In patients with lack of efficacy or adverse events, anakinra has been used as an alternative in this setting. Two new biologicals, B-cell-depleting rituximab, or T-cell-activation-blocking abatacept (CTLA4-Ig), are efficacious in anti-TNF failures. However, the use of anakinra and abatacept simultaneously with anti-TNF biologicals does not improve outcomes and increases toxicity, hence their combined use is contraindicated. As the use of all biologicals is associated with adverse events, careful selection and monitoring of patients is mandatory (see section on biological therapy, below).
Treatment of extra-articular disease
Effective treatment of rheumatoid arthritis generally reduces the risk of developing severe extra-articular disease. Systemic rheumatoid vasculitis is potentially a life-threatening complication and may be aggravated by coincidental infection, such as through cutaneous ulcers. After due attention to confirming the diagnosis and excluding and treating infections with appropriate antimicrobial drugs, therapy with high-dose corticosteroids and cyclophosphamide is favoured by many specialists, although no randomized placebo-controlled trial data are available. One regimen recommends intravenous methylprednisolone at 1 g daily for 3 days, simultaneously with an initial single pulse of intravenous cyclophosphamide (10 to 15 mg/kg) in a fully hydrated patient to prevent bladder toxicity. Cyclophosphamide is repeated every 3 to 4 weeks, subject to a satisfactory clinical response or lack of toxicity, up to a total dose of 10 to 12 g in a cycle of treatment. Alternatively, oral cyclophosphamide at 2 mg/kg (maximum dose 150 mg daily) may be used. Oral high-dose prednisolone is continued until clinical response is observed or toxic effects occur, when it is rapidly tapered to a maintenance dose, generally about 15 mg daily. Similarly, cyclophosphamide is substituted by the less toxic azathioprine at 1.5 to 2 mg/kg daily or methotrexate at 15 mg/week.
Similar regimens have been used for severe fibrosing alveolitis and for severe scleritis and corneal melt in conjunction with local therapy. Occasional patients with Felty’s syndrome and hypersplenism that do not respond to DMARDs benefit from splenectomy, and their neutropenia may respond to recombinant human granulocyte colony stimulating factor (G-CSF). Keratoconjunctivitis sicca and dry mouth due to secondary Sjögren’s syndrome respond to local measures, including artificial tears, dental hygiene, and saliva substitute.
Effective treatment of rheumatoid arthritis with suppression of the acute phase response with DMARDs, corticosteroids, and anti-TNFα therapy prevents progression of secondary amyloidosis. In patients with a continuing acute phase response despite the standard DMARD therapy, treatment with chlorambucil is reported to be of some benefit. Imaging of radionuclide-labelled serum amyloid P protein in the spleen and kidneys may be used to monitor treatment.
|Table 2 DMARDs in adultsa,b|
|Drug||Dose and comments||Contraindications||Some side effectsa|
|Methotrexate||Oral 7.5 mg/week initially given as a single dose; usual dose 12.5–15 mg/week; increase up to 25 mg/week orally and try intramuscular route in unresponsive patients or in presence of gastrointestinal intolerance; folic acid 5 mg daily for 1–5 days/week improves tolerability||Pregnancy and planned conception (teratogenic), alcohol abuse, chronic liver disease, diabetes mellitus, moderate to severe chronic lung disease||Bone marrow suppression, hepatotoxicity, interstital pneumonitis, anorexia, nausea, stomatitis, vomiting, viral and opportunistic infections, possible increased risk of lymphoma|
|Sulphasalazine||Enteric-coated tablets 0.5 g once a day for initial week; thereafter 0.5 g increment in dose per week to total 2 g/day; maximum dose 3 g/day||Sulfonamide, salicylate allergy, glucose-6-phosphate dehydrogenase deficiency||Nausea, anorexia, rashes, blood dyscrasia (especially neutropenia), lupus-like syndrome, oligospermia on taking drug (reversible)|
|Hydroxychloroquine||200–400 mg in divided doses daily; maximum dose 6.5 mg/kg (not exceeding 400 mg/day); maintenance dose 200–400 mg daily||Glucose-6-phosphate dehydrogenase deficiency, retinal disease, psoriasis||Maculopathy; test visual acuity and fields, colour vision before commencing therapy; if abnormal, full ophthalmology examination required; patient to stop drug if any disturbance of vision noted. Annual check-ups advisable; discontinuation after 10 years recommended|
|Chloroquine sulfate||(200 mg, equivalent to chloroquine phosphate 250 mg or chloroquine base 150 mg); daily dose of chloroquine base 150 mg, daily maximum 2.5 mg/kg||As above|
|Injectable gold aurothiomalate||Deep intramuscular, upper/outer gluteal muscle; 10 mg test dose to check for hypersensitivity; 20–50 mg weekly according to tolerability and severity to a total of 1 g or until response is observed; maintenance dose 20–50 mg/month||Gold hypersensitivity, chronic liver and renal disease, psoriasis||Blood dyscrasias, aplastic anaemia, nephropathy, dermatitis|
|Azathioprine||Oral 1 to 3 mg/kg daily||Up to 1 in 200 of the population have hypersensitivity characterized by severe leucopenia on initial administration||Hepatitis, reversible dose-related bone marrow depression, possible increased risk of lymphoma|
|Ciclosporin||Oral initially 2–5 mg/kg up to 4 mg/kg provided serum creatinine is in normal range and does not increase more than 30% above baseline||Renal disease with compromised function hypertension||Nephrotoxicity, hirsurtism, hypertension, tremor|
|D-Penicillamine||Oral 250–1000 mg daily in divided doses starting at 125–250 mg initially||Penicillin hypersensitivity||Taste loss, thrombocytopenia and other dyscrasias, nephropathy, myaesthenia gravis, rashes|
|Leflunomide||Initial loading dose 100 mg orally, once daily for 3 days, followed by maintenance dose 10–20 mg daily||Pregnancy (teratogenic), planned conception; long half-life (several months): cholestyramine accelerates clearance||Increases serum concentrations of drugs metabolized by CYPZC9, including NSAIDs; diarrhoea, alopecia, hepatotoxicity, and (rarely) bone marrow suppression|
a Refer to institutional, and or/national, and/or British Society for Rheumatology Drug Monitoring Guidelines (National guidelines for monitoring second-line drugs, July 2000,
b For other contraindications and details consult product literature before use.
British Society for Rheumatology Drug Monitoring Guidelines (National guidelines for monitoring second line drugs, July 2000.
Antirheumatic drugs: clinical trials, mechanism of action, and use in the clinic
NSAIDs are widely used for treating symptoms of inflammation and pain in rheumatoid arthritis and do not modify the progression of structural damage. They act by inhibiting the enzymes COX-1 and/or COX-2, which convert lipid substrates in cells to prostanoids. Tissues such as the gastric and duodenal mucosa express COX-1, which in the gastroduodenal mucosa regulates the production of prostaglandins, including PGE2, that exert a protective effect on its integrity by reducing acid secretion and increasing the secretion of mucus and bicarbonate. COX-2 is mainly induced by proinflammatory cytokines in synovial tissue cells and monocytes at sites of inflammation, stimulating production of inflammatory prostaglandins—hence it was thought that COX-2 might be an ideal anti-inflammatory drug target for treating rheumatoid arthritis.
Aspirin blocks COX-1 activity and is an effective anti-inflammatory drug in high doses, but it is a well-recognized cause of gastric bleeding. Many of the previous generation of NSAIDs, such as indometacin, diclofenac, and high-dose ibuprofen, which inhibit both COX-1 and -2 activity, exert an anti-inflammatory effect by inhibiting the production of inflammatory mediators. However, they compromise the gastroprotective effect of COX-1, and traditional NSAIDs—apart perhaps from naproxen—do not significantly inhibit the prothrombic platelet COX-1 activity and synthesis of thromboxane A2 (which is inhibited by low-dose aspirin).
With regard to specific COX-2 inhibitors, based on laboratory experiments, it is suggested that they might suppress the production of prostacyclin by endothelial cells. As prostacyclin plays a protective role as a vasodilator and regulator of blood pressure and cardiac function, blockade of COX-2 could theoretically leave unopposed the action of COX-1-induced platelet aggregation, potentially promoting hypertension and cardiovascular thrombotic complications (see below).
The traditional NSAIDS are responsible for admission to hospital of over 1% of patients with rheumatoid arthritis per year for complications such as peptic ulceration, gastric haemorrhage, and perforation, and account for a twofold increase in death over expected rates. It is claimed that the least gastrotoxic are ibuprofen and nabumetone, with naproxen and diclofenac carrying intermediate risk, followed by drugs with a high risk, such as fenoprofen, ketoprofen, indometacin, piroxicam, and azapropazone. It is claimed that drugs such as meloxicam that act by greater selective inhibition of COX-2 and thus spare COX-1 also have fewer gastropathic effects (see Table 1 for drugs and dose ranges).
For patients who develop dyspepsia and/or NSAID-induced gastropathy, or elderly patients who have a high risk of gastroduodenal side effects, concomitant administration of prostaglandin analogues (such as misoprostol) or proton pump inhibitors (such as omeprazole or lansoprazole) is recommended, and NSAIDs are best avoided for patients with a history of peptic ulcers. Eradication of Helicobacter pylori infection results in long-term healing of preexisting gastric and duodenal ulcers, but whether it decreases dyspepsia or ulceration caused by NSAIDs is uncertain.
The highly selective or specific inhibitors of COX-2, including celecoxib, rofecoxib, valdecoxib, and etoricoxib, possess no significant COX-1 inhibitory activity at anti-inflammatory therapeutic doses. Clinical trials have demonstrated their improved safety profile in respect to endoscopically detectable gastroduodenal ulcers, upper gastrointestinal haemorrhage, and perforation when compared with conventional NSAIDs. However, rofecoxib was withdrawn in 2004 by the manufacturer after a trial for the prevention of adenomatous polyps demonstrated a twofold increase in cardiovascular complications at a dose of 25 mg daily. Similar concerns were raised in regard to celocoxib 400 mg twice daily in the same preventive setting, but at lower doses it appeared to have an acceptable risk (similar to naproxen) for cardiovascular thrombotic events. Valedocoxib was withdrawn following reports of serious adverse events in 2005.
Although the adverse risk profiles of individual drugs continues to be debated, meta-analyses of controlled and observational trials has confirmed that all COX-2 inhibitors have increased risk of thrombotic cardiovascular complications, such as myocardial infarction and stroke, most marked for rofecoxib at normal doses and at high dose for celocoxib. Meta-analyses of observational studies have also demonstrated that traditional NSAIDS such as diclofenac, indometacin, and (probably) meloxicam also carry an increased cardiovascular risk, whereas naproxen does not appear to increase or decrease this risk. Based on these observations, naproxen, low-dose celecoxib, and low-dose ibuprofen emerge as relatively free of cardiovascular risk. Regulatory authorities advise use for short periods and careful monitoring for longer-term use of all NSAIDs, and specifically advise against the use of COX-2 inhibitors in patients with ischaemic heart disease or cerebrovascular disease. Evidence from meta-analysis supports the view that the same advice is appropriate for high doses of traditional NSAIDs.
All NSAIDs can cause fluid retention and oedema by a renin–angiotensin-dependent mechanism that may also aggravate congestive cardiac failure and systemic hypertension. Patients with impaired renal function, cirrhosis of the liver, and decreased plasma volume from any cause are at risk from developing NSAID-induced renal toxicity. It is claimed that sulindac may be safer than other NSAIDs in patients with renal failure.
NSAIDs, especially indometacin, may cause central nervous system side effects such as headache, dizziness, anxiety, disorientation, and drowsiness. Use of NSAIDs may aggravate asthma and cause hypersensitivity reactions, and they may rarely be associated with aseptic meningitis. Blood dyscrasias and an increase in serum concentration of liver enzymes and alkaline phosphatase are described. Drug interactions may decrease the efficacy of some concomitantly prescribed therapies—e.g. antihypertensives and lithium—and potentiate the effects of others—e.g. anticoagulants, antiepileptics, and oral hypoglycaemics. NSAIDs decrease the excretion of methotrexate but do not appear to increase its toxicity in the dose range used to treat rheumatoid arthritis. They also increase plasma concentrations of ciclosporin and tacrolimus (FK-506) and hence may increase the risk of renal toxicity.
DMARDs modify the trajectory of progressive structural damage and disability and are the cornerstone of drug therapy for rheumatoid arthritis (Table 2). They are also classified as slow-acting antirheumatoid drugs (SAARDs) because of the lag period of some weeks before their anti-inflammatory effect becomes apparent. Drugs in this category include: the antimalarials hydroxychloroquine or chloroquine sulfate; sulfasalazine; weekly low-dose oral or parenterally administered methotrexate; weekly injections of gold aurothiomalate or gold aurothioglucose; leflunomide; ciclosporin; azathioprine; and D-penicillamine.
Drugs such as gold, antimalarials, and methotrexate were introduced for use in rheumatoid arthritis by serendipity. Others such as azathioprine, ciclosporin, and lefluomide were developed as immunosuppressive agents for preventing transplant rejections and subsequently used to curb the aberrant immunological response in rheumatoid arthritis. The mechanism of action of these drugs in RA is complex and still incompletely understood. Inhibitory effects on inflammatory pathways, immune responses, and cell activation have been described in experimental systems and clinical studies.
Clinical trials have demonstrated superior efficacy of all these drugs over placebo in controlling symptoms and signs in patients previously treated with only NSAIDs in early and established rheumatoid arthritis. In addition, sulfasalazine, methotrexate, and leflunomide retard progression of structural damage, as assessed by serial radiographs of the hands and feet in controlled trials lasting 6 to 12 months compared with placebo.
A meta-analysis of clinical trials of commonly used DMARDs, usually given to patients stabilized on NSAIDS and—in many instances—low-dose corticosteroid therapy, has been analysed for efficacy and toxicity relative to each other and to placebo treatment. Methotrexate and sulfasalazine have the best and equal efficacy in the short term compared with placebo. The antimalarials (hydroxychloroquine and chloroquine) and azathioprine appear to be less efficacious in this analysis. However, the toxicity profile shows a different rank order: antimalarials are least toxic, followed by methotrexate and sulfasalazine in an intermediate range, with injectable gold, azathioprine, and D-penicillamine at the most toxic end of the spectrum. Methotrexate and sulfasalazine emerge with the best balance between efficacy and toxicity, and injectable gold and D-pencillamine are now rarely used. Leflunomide was introduced recently and not included in this meta-analysis, but its efficacy and toxicity profile is similar to that of methotrexate and sulfasalazine.
Conclusions from short-term randomized clinical trials do not reflect the effectiveness of DMARDs in controlling disease activity in the longer term. Incomplete responses, relapses, and adverse events are common and account for discontinuation of antimalarials, gold salts, D-penicillamine, sulfasalazine, and azathioprine in most patients in 1 to 3 years. By contrast, responses to methotrexate appear to be more durable in follow-up studies of large cohorts of patients with rheumatoid arthritis, with about 50% continuing therapy at 5 years. Data on long-term effectiveness, tolerability, and toxicity of leflunomide are not yet available.
Combinations of DMARDs have been used in the expectation that their different modes of action might provide added efficacy. Differing approaches have been used: e.g. sequential change from one DMARD to another; add-on of DMARDS, referred to as ‘step-up’; commencing with a combination from the start with phased withdrawal of all except one (‘step-down’); and long-term continuing combinations of two or more DMARDS (in parallel). A meta-analysis showed that there was only marginal benefit at the doses and combinations used before 1994, especially in the reduction of number of tender joints, with increased toxicity when compared with single agents. However, several subsequent randomized controlled trials have demonstrated significantly improved efficacy of combination therapy, without increased toxicity: some examples of such trials are given below.
In a Finnish study on patients with disease of less than 2 years’ duration, the introduction of a combination of sulfasalazine, methotrexate, hydroxychloquine, and oral prednisolone not only controlled symptoms better but, at the end of 2 years, had induced remission in 37% compared with 21% of patients on monotherapy with sulfasalazine (initially) or methotrexate alone, plus prednisolone in two-thirds of patients. There was greater retardation of joint damage and a better quality of life in the combination than in the monotherapy group.
In a North American trial lasting 2 years, a combination of oral methotrexate (7.5 to 17.5 mg/week), sulfasalazine (0.5 g twice daily) and hydroxychloroquine (200 mg twice daily) showed superior control of symptoms and signs compared with methotrexate alone or a combination of sulfasalazine and hydroxychloroquine. Patients enrolled in this trial with advanced disease had already failed to respond to DMARD monotherapy.
In a Dutch study, a combination of sulfasalazine 2 g/day, methotrexate 7.5 mg per week for 40 weeks, and prednisolone in a dose of 60 mg daily for 6 weeks, rapidly tapered to 7.5 mg per day and withdrawn at 28 weeks, was compared with monotherapy with sulfasalazine 2 g/day. Clinical responses were better in the combination therapy group at 1 year, and markedly so for the first 28 weeks. In a follow-up study during which patients received similar exposure to DMARDs, the rate of radiographically assessed join damage remained slower in the group initially treated with combination of high-dose corticosteroid and DMARDs.
In a British trial, an intensive therapy group was compared with a group receiving routine care from specialist physicians. The intensively treated group was seen every month and, according to protocol if disease was active, therapy escalated with maximum tolerated doses of sulfasalazine followed by addition of corticosteroids and methotrexate, or variously of ciclosporin or leflunomide or sodium aurothiomalate. Routinely treated patients mostly received DMARD monotherapy and intra-articular corticosteroid. At 18 months the intensively treated groups had superior outcomes in all measurements of disease activity, with higher remission levels and control of structural radiographic damage than the group receiving routine care.
From these studies it may be concluded that if disease is not suppressed by monotherapy with methotrexate in maximum tolerated doses, then a combination of DMARDs with dose adjustments that achieve low disease activity is highly effective in a significant proportion of patients with moderate and severe rheumatoid arthritis.
Corticosteroids are potent anti-inflammatory agents and are most efficacious in treating symptoms and signs of rheumatoid arthritis and for amelioration of systemic features, but their use is limited by toxicity related to dose and duration of exposure. The circumstances in which use of corticosteroids has been established and those in which it is debated are described below.
In patients in whom loss of function and disease activity is restricted to a few joints, local corticosteroid therapy can be most effective. This indication may arise in those whose rheumatoid disease is limited to a few joints, or in patients with an incomplete response to NSAID and DMARD therapy. Several alternative corticosteroid preparations are available, the dose being dependent on the size of the joint. Depot methylprednisolone (dose range 4–40 mg) or triamcinolone acetonide (dose range 2.5–40 mg) are suitable alternatives. Repeat injections may be necessary, but more than three per joint per year should be avoided.
Corticosteroid administered orally in courses lasting a few weeks to months (such as prednisolone at 7.5–10 mg daily), or in the form of ‘pulse therapy’ (such as depot methylprednisolone at 80–120 mg by intramuscular injection), is a suitable adjunctive therapy in patients in whom the benefit of DMARDs is not yet established. Longer-term, more or less indefinite, treatment with low-dose prednisolone is necessary in patients with moderate to severe disease, especially if associated with refractory anaemia that is not controlled with currently used antirheumatoid drugs. Long-term low-dose prednisolone retards the progression of rheumatoid bone erosions in radiographs of hands and feet and, hence it is claimed, deterioration of physical function. Whether this benefit is outweighed by the side effects and morbidity of corticosteroid therapy is debatable. Higher doses of corticosteroids are indicated in the treatment of severe extra-articular disease.
Prevention of corticosteroid-induced osteoporosis and reduction in risk of fractures requires adequate prophylaxis with calcium and vitamin D intake (e.g. daily intake of 1000 mg of calcium and 800 IU of vitamin D). In susceptible patients, or those on doses exceeding the equivalent of 7.5 mg of prednisolone daily, measurement of bone mineral density is used to identify and monitor management. Bisphosphonates may be required in addition to calcium and vitamin D, and hormone replacement therapy is recommended in perimenopausal women.
Biological monoclonal antibody and recombinant protein drugs that have proved efficacious in double-blind randomized controlled clinical trials include three anti-TNF drugs—two monoclonal antibodies, infliximab and adalimumab, and etanercept, a TNF-receptor fused to Fc IgG—which are the most widely used biologicals, and other biologicals, namely anakinra, an IL-1 receptor antagonist; rituximab, an anti-CD20 B-cell-depleting monoclonal antibody; and abatacept (CTLA4-Ig), a receptor-fusion-FcIgG recombinant protein. More biological drugs for the treatment of rheumatoid arthritis are emerging, including 2 additional anti-TNF blocking agents (golimumab, a fully human monoclonal antibody and certolizumab, a PEGylated Fab’ fragment of a humanized anti-TNF monoclonal antibody) and tocilizumab, a monoclonal antibody which inhibits binding of IL-6 to its receptor.
Despite good initial responses to currently available DMARD treatments, about 40 to 50% of hospital patients eventually show continuing disease activity and progressive disability. A TNF-receptor-FcIgG recombinant protein (etanercept) and a chimeric monoclonal antibody specific for TNF (infliximab), followed by a humanized monoclonal antibody (adalimumab), were in the first group of biologicals to be shown to be beneficial for the treatment of RA in this setting. All three (and the newer anti-TNF drugs)inhibit the binding of TNFα to its cell surface receptors, thereby inhibiting its immune and inflammatory action. Although there are some differences in their biological properties in vitro, their efficacy in placebo-controlled randomized clinical trials and observational studies is similar.
Symptoms and signs are rapidly alleviated in approximately 60 to 70% of patients in clinical trials and confirmed by large single and multicentre observational studies. Attempts to identify the nonresponders by clinical characteristics or biomarkers have not been successful thus far. Continuing therapy is needed by those who do respond, with relapse following withdrawal except in early stages of disease (see below).
Infliximab is given intravenously at a dose of 3 mg/kg over 1 h every 8 weeks in combination with methotrexate therapy once a week. Etanercept is given at a dose of 25 mg subcutaneously twice weekly or 50 mg weekly, and adalimumab 40 mg subcutaneously weekly or fortnightly, both of these drugs being efficacious as monotherapy or in combination with methotrexate. Further details, including contraindications and some side effects, are shown in Table 3.
In patients with long-standing disease and a high level of disease activity despite DMARD and methotrexate therapy, in whom DMARDs are discontinued, randomized controlled trials have demonstrated that anti-TNF drugs are efficacious as monotherapy in approximately 70% of patients. The concomitant addition of anti-TNF drugs to methotrexate shows more marked efficacy in the control of signs, symptoms, and especially structural damage, when compared with continuing methotrexate or initiating monotherapy with the anti-TNF drugs. Improvement in quality of life measurements and work employability is well documented.
In early disease and in methotrexate naïve patients, infliximab and methotrexate given concomitantly is more effective than methotrexate alone. Similarly, etanercept or adalimumab and concomitant methotrexate is more effective than methotrexate, etanercept, or adalimumab alone.
In two observational studies in early rheumatoid arthritis it was shown that following induction therapy with a combination of infliximab and methotrexate for 6 to 9 months, anti-TNF drug-free remission could be maintained in around 50% of patients on methotrexate alone for up to 3 years, which has possible pharmacoeconomic benefits.
|Table 3 Biological therapies for rheumatoid arthritis|
|Adalimumab||Fully human IgG1 monoclonal antibody||40 mg SC alternate weeks; given with methotrexate; may be used as monotherapy||Infection, especially Mycobacterium tuberculosis, cancer, heart failure, SLE, demyelinating disease, e.g. MS||Unusual infections, diarrhoea, constipation, vomiting, gastritis rash|
|Etanercept||TNF receptor-Fc-IgG1 fusion protein||25 mg SC twice weekly or 50 mg SC once weekly, alone or with methotrexate||Infection, cancer, heart failure, SLE, demyelinating disease, e.g. MS||Vomiting/oesophagitis, cholecycystitis, hypotension/hypertension, lymphadenopathy|
|Infliximab||Mouse/human chimaeric IgG1 monoclonal antibody||3 mg/kg IV, repeated at 2 weeks and 6 weeks after initial infusion, then maintenance 8 weekly; given with methotrexate||Infection, especially Mycobacterium tuberculosis, cancer, heart failure, SLE, demyelinating disease, e.g. MS||Dyspepsia, diarrhoea, constipation, hepatitis, cholecystitis, flushing, bradycardia, arrhythmias, rash|
|Anakinra||Recombinant nonglycosylated synthetic human form of the IL-1RA protein||100 mg/day SC||Neutropenia, documented hypersensitivity to E. coli derivatives||Neutropenia, injection site reaction, headache, infections|
|Rituximab||Chimeric mouse/human monoclonal antibody against CD20+ B cells||1 g IV, day 1 and day 15; given with methotrexate||Known hypersensitivity to murine products||Hypersensitivity reaction to first infusion, fever, chills, angioedema, nausea, pruritis, rash, infusion reaction|
|Blockade of CD28 on T cells|
|Abatacept||Human recombinant CTLA-4 IgG fusion protein||500 mg–1 g IV (depending on body weight), repeated at 2 weeks and then every 4 weeks after initial administration||Cancer, infections||Headache, nasopharyngitis, dizziness, cough, hypertension, rash|
CTLA-4, cytotoxic T-lymphocyte-associated antigen 4; IL-1RA, interleukin 1 receptor antagonist; IV, intravenously; MS, multiple sclerosis; SC, subcutaneously; SLE, systemic lupus erythematosus.
Data from the British Society for Rheumatology Drug Monitoring Guidelines (National guidelines for monitoring second line drugs, July 2000)
Notable additional features of therapy with anti-TNF drugs in combination with methotrexate include radiographic changes suggestive of the healing of structural damage in some patients. A subanalysis of trials with infliximab or etanercept plus methotrexate given concomitantly also demonstrates inhibition of radiographically assessed structural damage, even in patients with no improvement in clinical signs of inflammatory disease, suggesting that this combination, unlike methotrexate monotherapy, uncouples mechanisms of inflammation and tissue damage.
Durable responses in the responder population have been demonstrated in several observational studies from single or multiple centres for 2 to over 5 years. An increase in the dose of the anti-TNF drugs may be necessary to maintain efficacy. In patients with loss of efficacy following dose adjustments (secondary nonresponders), switching to a different anti-TNF biological may be efficacious.
Anti-TNF biologicals are generally well tolerated, but concerns have been raised from reported adverse events in controlled clinical trials of serious infections, increased incidence of antinuclear antibodies, infrequent drug-induced lupus, and an increased incidence of B-cell lymphomas and cancer.
Post-marketing surveillance of adverse events has confirmed the occurrence of rare but significant numbers of cases of sepsis, tuberculosis, and fungal and opportunistic infections. These infections are compatible with the consequences of blockade of the postulated role of TNF in host defence mechanisms. Based on these reports, regulatory authorities in the United States of America and Europe have advised that anti-TNF therapy is contraindicated in the presence of active serious infections and latent untreated tuberculosis, and screening for latent tuberculosis and prophylactic antituberculosis therapy has markedly reduced complication with this infection. Other rare adverse events have included the occurrence of B-cell lymphomas, demyelinating syndromes (hence it is advisable not to treat patients with a history of multiple sclerosis), lupus syndrome, and bone marrow depression. Based on reports of an unexpected number of deaths observed in a phase II clinical trial of infliximab in the treatment of severe congestive cardiac failure, use of anti-TNF biologicals is not advisable for the treatment of patients with rheumatoid arthritis in moderate or severe congestive cardiac failure.
Prospective and retrospective studies of control populations of rheumatoid patients that have not been exposed to biologicals have demonstrated that in an aging rheumatoid population the incidence of infections, lymphomas, other cancers, myocardial infarction, and strokes is higher than the normal population. When these confounding variables are taken into consideration in analysing large databases in Europe and North America, the conclusions have been at variance from those described above. Thus it is suggested that aside from opportunistic infections, the incidence of life-threatening infections, lymphomas, cancers (apart from skin cancers), and cardiovascular comorbidity appears not to be increased in anti-TNF-treated patients. However, at least in respect of infections, these findings may reflect the exclusion from treatment of patients who are susceptible to comorbidity such as infections, and to early detection and treatment of potentially serious infections.
Provided suitable screening and monitoring practices are in place, the benefit of anti-TNF therapy exceeds harm and is indicated for the treatment of moderate to severe rheumatoid arthritis with persistent disease activity despite best available standard therapy. Ideally, anti-TNF therapy should be instituted as soon as it becomes apparent that remission or near-remission is not induced by the best use of traditional DMARDs in the early stages of an established diagnosis, but cost constraints or safety concerns may limit the institution of this therapeutic approach.
Recombinant human IL-1RA (anakinra) binds to IL-1 receptor, thereby inhibiting access of IL-1 to the receptor and initiating an IL-1-mediated proinflammatory response. In rheumatoid arthritis the endogenous production of IL-1RA is very significantly increased simultaneously with overproduction of IL-1, but insufficient to inhibit the action of excess amount of IL-1. To be clinically effective in fully inhibiting the action of IL-1, almost full engagement of IL-1 receptors by anakinra is required, and even daily administration may not achieve this goal, because anakinra is very rapidly cleared from the circulation.
In randomized controlled trials, anakinra is effective vs placebo in reducing signs, symptoms, and progressive radiographically assessed structural damage in rheumatoid patients who have failed traditional DMARDs. In clinical trials the coadministration of anakinra in patients with active disease despite methotrexate is similarly efficacious. Its use in early rheumatoid arthritis has not been demonstrated.
Anakinra is administered at a dose of 100 mg daily by subcutaneous injection. Injection site reactions occur in most patients, and severe infections have been described, although not activation of tuberculosis as seen with anti-TNF drugs. However, it is unclear whether infections occur in a higher incidence than in a matched-control population of rheumatoid patients, and postmarketing databases are not as extensive as those for anti-TNF drugs. The combined use of anakinra and etanercept showed no additional efficacy, but was associated with increase in serious infections, hence combination of anti-TNF drugs and anakinra is not recommended. It is not definitively established whether anakinra is effective in anti-TNF failures, but it has been used clinically in this setting. Anakinra is licensed for use alone or in combination in the United States of America, but only in combination with methotrexate in the European Union. According to NICE guidelines in the United Kingdom, it may only be prescribed as part of an experimental protocol.
Rituximab, a chimeric monoclonal antibody that depletes B cells, has been licensed since 1997 for the treatment of B-cell lymphomas. It binds to mature B cells and activated B cells (plasmacytoid cells) but not plasma cells. In rheumatoid arthritis, randomized placebo-controlled trials have demonstrated efficacy of rituximab in patients who have failed on DMARD and anti-TNF therapy. Combination therapy with methotrexate is more effective than monotherapy, and is administered as a course of two infusions of rituximab at a dose of 1 g separated by 2 weeks with corticosteroid co-administration (see Table 3).
Rituximab administration is associated with a loss of circulating B cells for several months and a reduction in rheumatoid factor levels, but no significant reduction in circulating immunoglobulin concentrations. Relapse in disease usually occurs after about 6 months, and there is limited experience thus far of the long-term efficacy and safety of repeat therapy. Infusion reactions occur, especially with the first infusion. Severe infections are described and it is recommended that if vaccination against influenza or pneumococcal pneumonia is deemed necessary, these should be given before commencing rituximab therapy.
Abatacept, a recombinant protein, consists of the extracellular domain of cytotoxic T-lymphocyte antigen 4 (CTLA4) linked to the Fc portion of IgG1. Abatacept binds to CD80 and CD86 on antigen-presenting cells, thereby preventing their ability to bind to CD 28 on T cells. The engagement of CD28 is the second signal that a T cell requires for activation in addition to the signal delivered by antigen–HLA complex via the T-cell receptor. It is proposed that in rheumatoid arthritis abatacept terminates the activation of T cells and subsequent pathology, e.g. as mediated by proinflammatory cytokines.
Abatacept administered as monotherapy and in combination with methotrexate is efficacious in patients who have failed DMARD and anti-TNF therapy in randomized clinical trials. However, the co-administration of abatacept and anti-TNF biologicals or anakinra significantly increases serious infections without a corresponding increase in efficacy. Abatacept is administered by intravenous injection, and following initiating therapy at week 0 and at week 2 is given thereafter repeatedly every 4 weeks. It is envisaged that abatacept will offer a useful biological alternative in the group of patients with severe rheumatoid arthritis who fail to respond to anti-TNF agents. Its long-term efficacy, safety, and cost-effectiveness are not yet established.
Guidelines for the emerging class of biological drugs for the treatment of rheumatoid arthritis are being constantly updated, e.g. by The National Institute of Health and Clinical Excellence (NICE) in England and Wales.
Anderson JJ, et al. (2000). Factors predicting response to treatment in rheumatoid arthritis: the importance of disease duration. Arthritis Rheum, 43, 22–9.
Arnett FC, et al. (1988). The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum, 31, 315–34.
Boers M. et al. (1997). Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis. Lancet, 350, 309–18.
Bongartz T, et al. (2006). Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies. J Am Med Assoc, 295, 2275–85.
Breedveld F, et al. (2006). The PREMIER study. A multicentre, randomized, double-blind clinical trial of combination therapy with adalimumab plus methotrexate versus methotrexate alone or adalimumbab alone in patients with early, aggressive rheumatoid arthritis who had not had previous methotrexate treatment. Arthritis Rheum, 54, 26–37.
Brennan M, et al. (2006). Comparing rates of dyspepsia with coxibs vs NSAID and PPI: a meta-analysis. Am J Med, 119, 448.
Brooks P, et al. (1999). Interpreting the clinical significance of the differential inhibition of cyclooxygenase-1 and cyclooxygenase-2. Rheumatology, 38, 779–88.
Chakravarty K, McDonald H, Pullar T, et al. (2008). BSR/BHPR guideline for disease-modifying anti-rheumatic drug (DMARD) therapy in consultation with the British Association of Dermatologists. Rheumatology (Oxford), 47(6), 924–5.
Choi HK, et al. (2002). Methotrexate and mortality in patients with rheumatoid arthritis: a prospective study. Lancet, 359, 1173–7.
Cohen SB, et al. (2006). Rituximab for rheumatoid arthritis refractory to anti-tumor necrosis factor therapy: results of a multicentre, randomized, double-blind, placebo-controlled, phase III trial evaluating primary efficacy and safety at twenty-four weeks. Arthritis Rheum, 54, 2793–806.
Emery P (2006). Treatment of rheumatoid arthritis. Br Med J, 332, 152–5.
Erhardt CC, et al. (1989). Factors predicting a poor life prognosis in rheumatoid arthritis: an eight year prospective study. Ann Rheum Dis, 48, 7–13.
Feldmann M, Brennan FM, Maini RN (1996). Role of cytokines in rheumatoid arthritis. Annu Rev Immunol, 14, 397–440.
Feldmann M, Maini RN (2001) Anti-TNF alpha therapy of rheumatoid arthritis: what have we learned? Annu Rev Immunol, 19, 163–96.
Feldmann M. Maini RN (2003). TNF defined as a therapeutic target for rheumatoid arthritis and other autoimmune diseases. Nat Med, 9, 1245–50.
Felson DT, Anderson JJ, Meenan RF (1994). The efficacy and toxicity of combination therapy in rheumatoid arthritis: a metaanalysis. Arthritis Rheum, 37, 487–91.
Felson DT, et al. (1995). American College of Rheumatology preliminary definition of improvement in rheumatoid arthritis. Arthritis Rheum, 38, 727–35.
Fries JF, Spitz PW, Young DY (1982). The dimensions of health outcomes: the health assessment questionnaire, disability and pain scales. J Rheumatol, 9, 789–93.
Furst DE (2000). Aggressive strategies for treating aggressive rheumatoid arthritis: has the case been proven? Lancet 356, 183–4.
Furst DE, et al. (2006). Updated consensus statement on biological agents for treatment of rheumatic diseases, 2006. Ann Rheum Dis, 65 Suppl 3, iii2–15.
Gardner DL (1992). Rheumatoid arthritis: cell and tissue pathology. In: Gardner DL, (ed.) Pathological basis of the connective tissue diseases, pp. 444–526. Edward Arnold, London.
Genovese MC, et al. (2005). Abatacept for rheumatoid arthritis refractory to tumor necrosis factor alpha inhibition. New Engl J Med, 353, 1114–23.
Goekoop-Ruiterman, et al. (2005). Clinical and radiographic outcomes of four different treatment strategies in patients with early rheumatoid arthritis (the BeSt study). Arthritis Rheum, 52, 3381–90.
Graham DJ (2006). COX-II inhibitors, other NSAIDs, and cardiovascular risk. The seduction of common sense. J Am Med Assoc, 296, 1653–6.
Gregersen PK, Silver J, Winchester RJ (1987). The shared epitope hypothesis. An approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum, 30, 1205–13.
Griffiths ID (2004). Extra-articular features of rheumatic diseases. In: Isenberg DA, Maddison PJ, Woo P, Glass DN, Breedveld F (eds) Oxford textbook of rheumatology, 3rd edition, pp. 110–7. Oxford University Press, Oxford.
Grigor C, et al. (2004). Effect of a treatment strategy of tight control for rheumatoid arthritis (the TICORA study): a single-blind randomised controlled trial. Lancet, 364, 263–9.
Kearney RM, et al. (2006). Do selective cyclo-oxygenase-2 and traditional non-steroidal anti-inflammatory drugs increase the risk of atherothrombosis? Meta-analsysis of randomized trials. BMJ, 332, 1302–8.
Kirwan JR (1995). The effect of glucocorticoids on joint destruction in rheumatoid arthritis. The Arthritis and Rheumatism Council Low-Dose Glucocorticoid Study Group. New Engl J Med, 333, 142–6.
Klareskog L, et al. (2004). Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet, 363, 675–81
Klareskog L, et al. (2006). Smoking may trigger HLA-DR (SE)-restricted immune reactions to autoantigens modified by citrullination. Arthritis Rheum, 54, 38–46.
Lawrence JS (1970). Rheumatoid arthritis: nature or nurture? Ann Rheum Dis, 29, 357–69.
Maetzel A, et al. (2000). Meta-analysis of treatment termination rates among rheumatoid arthritis patients receiving disease-modifying anti-rheumatic drugs. Rheumatology, 39, 975–81.
Maini RN, Feldmann M (2002). How does infliximab work in rheumatoid arthritis? Arthritis Res, 4 Suppl 2, S22–8.
Maini RN, et al. (1998). Therapeutic efficacy of multiple intravenous infusions of anti-tumor necrosis factor α monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis. Arthritis Rheum, 41, 1552–63.
Maini RN, et al. (1999). Randomised phase III trial of infliximab (Chimeric anti-TNFα monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate. Lancet 354, 1932–9.
Mangge H, Hermann J, Schauenstein K (1999). Diet and rheumatoid arthritis—a review. Scand J Rheumatol, 28, 201–9.
Moreland LE, et al. (1999). Etanercept therapy in rheumatoid arthritis. Ann Intern Med, 130, 478–86.
National Institute of Clinical Excellence. Guidelines on the use of biological therapies in rheumatoid arthritis. Completed appraisals of etanercept, infliximab, adalimumab; rituximab; and abatacept
O’Dell JR, et al. (2004). Therapeutic strategies for rheumatoid arthritis. New Engl J Med, 350, 2591–602.
Osiri M, et al. (2003). Leflunomide for treating rheumatoid arthritis. Cochrane Database Syst Rev, 1, CD002047.
Pinals RS, et al. (1981). Preliminary criteria for clinical remission in rheumatoid arthritis. Arthritis Rheum, 24, 1305–15.
Pincus T, Callahan LF (1993). What is the natural history of rheumatoid arthritis? Rheum Dis Clin North Am, 19, 123–51.
Prevoo MLL, et al. (1995). Modified disease activity scores that include twenty-eight-joint counts development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum, 38, 44–8.
Scott DL, et al. (2000). The links between joint damage and disability in rheumatoid arthritis. Rheumatology, 39, 122–32.
Short CL (1974). The antiquity of rheumatoid arthritis. Arthritis Rheum, 17, 193–205.
Silman AJ, Pearson JE. (2002). Epidemiology and genetics of rheumatoid arthritis. Arthritis Research and Therapy. 4 Suppl 3, S265–72.
Smolen JS, Aletaha D, Keystone E. (2005). Superior efficacy of combination therapy for rheumatoid arthritis. Fact or fiction? Arthritis Rheum, 52, 2975–83.
Suarez-Almazor ME, et al. (1998). Methotrexate for treating rheumatoid arthritis. Cochrane Database Syst Rev, 1, CD000957.
Symmons DPM, Silman AJ (2006). What determines the evolution of early undifferentiated arthritis and rheumatoid arthritis? An update for the Norfolk arthritis register. Arthritis Res Ther, 8, 214–20.
Visser H, et al. (2002). How to diagnose rheumatoid arthritis early. A prediction model for persistent (erosive) arthritis. Arthritis Rheum, 46, 357–65.
Warner TD, Mitchell JA (2004). Cyclooxygenases: new forms, new inhibitors, and lessons from the clinic. FASEB J, 18, 790–804.
Weinblatt ME, et al. (1994). Methotrexate in rheumatoid arthritis. Arthritis Rheum, 37, 1492–8.
Wiles N, et al. (1999). Estimating the incidence of rheumatoid arthritis. Trying to hit a moving target? Arthritis Rheum, 42, 1339–46.
Young A, et al. (2000). How does functional disability in early rheumatoid arthritis (RA) affect patients and their lives? Results of 5 years of follow-up in 732 patients from the early rheumatoid arthritis study (ERAS). Rheumatology, 39, 603–11.