Rheumatoid factors (RFs) are antibodies directed against the fragment crystallizable (Fc) portion of immunoglobulin G (IgG). RF, as initially described by Waaler and Rose1 in 1940, is most commonly measured in clinical practice as an isotype of immunoglobulin M (IgM), although it can be an isotype of other immunoglobulin types including IgG and immunoglobulin A (IgA).
Despite its notoriety, the origin of RF is not fully understood.2,3 An abnormal immune response appears to select, via antigenic stimulation, high-affinity RF from the host’s natural antibody repertoire.4 This may occur in rheumatic diseases, such as rheumatoid arthritis (RA), and in a number of inflammatory diseases characterized by chronic antigen exposure, such as subacute bacterial endocarditis. The fact that RFs can be detected after such infections suggests that they represent an antibody response to microbes. This possibility is supported by experimental evidence showing that mice immunized with IgM-coated vesicular stomatitis virus develop RFs.5
Normal human lymphoid tissue commonly contains B-lymphocytes that express RF on the cell surface. However, RF is not routinely detectable in the circulation in the absence of an antigenic stimulus. Modified IgG variants (e.g. antigalactosyl IgG) could act a stimulus to RF production and could be an important component of RA pathogenesis.6,7
Costimulation of B-cells, perhaps mediated by toll-like receptors (TLRs), may allow B-cells with low-affinity receptors for IgG to become activated. TLRs are components of the innate immune system that provide signals after engaging various bacterial and viral products.8,9
How chronic infections and rheumatic diseases lead to increased RF in serum is uncertain, but one attractive hypothesis suggests that costimulation of B-cells with low-affinity receptors for IgG through TLRs by DNA- or RNA-containing immune complexes may play a role.
The following studies in patients with RA have contributed to the understanding of the origin of RF:
Synovial fluid RF may be produced by synovium-derived CD20-negative, CD38-positive plasma cells.12
Circulating B-cells require interleukin 10 for RF production.13
In RF-negative patients with RA, B-cells capable of RF production are fewer in number and less responsive to T-cell help than in RF-positive patients with RA. In one study, the frequency of RF-positive, IgM-positive B-cells was increased more than 50-fold in seropositive patients (7–20% of IgM-positive B-cells in seropositive patients vs. < 1% of IgM-positive B-cells in normal patients); patients with seronegative RA had intermediate values (1.5–6% of IgM-positive B-cells).14
Cigarette smoking, a risk factor for more severe RA, is associated with an increased prevalence of RF.15
In RA, as in most autoimmune diseases, evidence of autoimmunity often precedes clinical disease. RF has been found up to 30 years prior to clinical diagnosis of RA, and higher titres of RF are more likely to be associated with future development of RA.16–20
Rheumatoid factors may also react against a variety of cellular and tissue antigens but may have different biological activity in different hosts and anatomical locations. For example, one report of RF derived from synovial tissue lymphocytes in a patient with RA found specificity for gastric gland nuclei and smooth muscle; in contrast, RF derived from a control patient’s peripheral blood did not show this pattern of reactivity.23
The function of RF is poorly understood. Possible functions include:
binding and processing of antigens embedded in immune complexes;
presentation of antigens to T-lymphocytes in the presence of human leucocyte antigen (HLA) molecules;
amplification of the humoral response to bacterial or parasitic infection;
immune complex clearance.
The role of RFs in the pathogenesis and perpetuation of RA or other rheumatic diseases is unknown. A high correlation for RF has been noted among identical twins with RA, suggesting that genetic factors influence both RF function and disease development.24 However, some studies have shown that patients with RF-negative RA have HLA susceptibility alleles similar to those in RF-positive patients.25,26 Thus, these patients may exhibit a similar immunogenetic predisposition to RA but which is independent of RF.
Bearing these concepts in mind, clinical disorders associated with RF positivity are known as seropositive arthropathies. Patients may have detectable serum RF in a variety of rheumatic disorders, many of which share similar features, such as symmetrical polyarthritis and constitutional symptoms. These include:
RA – 26–90%;
Sjögren’s syndrome – 75–95%;
mixed connective tissue disease (MCTD) – 50–60%;
mixed cryoglobulinaemia (types II and III) – 40–100%;
systemic lupus erythematosus (SLE) – 15–35%;
polymyositis or dermatomyositis – 5–10%.27
The reported sensitivity of the RF test in RA has been as high as 90%. However, population-based studies which include patients with mild disease have found much lower rates of RF-positive RA (26–60%).28–31 This difference may reflect classification criteria that led a published series of patients with RA to be biased towards more severe (and more seropositive) disease, thereby overestimating the sensitivity of RF in RA.
Other autoantibodies, including anticitrullinated peptide antibodies [ACPAs, which include anticyclic citrullinated peptides (anti-CCPs)], may be present in patients with suspected or established RA who are RF negative, as well as in those who test positive for RF.32 The optimal clinical use of ACPA testing and its relationship to RF testing remains uncertain.33–36 Although ACPAs and RF have similar sensitivity for the diagnosis of RA, ACPA is a more specific marker for RA.37 Of note, the 2010 revised classification criteria for RA include both RF and ACPA.38
Immunoglobulin G and IgA RFs are occasionally present in patients with RA in the absence of IgM RF.39,40 Measurement of these non-IgM RFs is not widely available in the USA and it is unclear whether it is routinely used in other countries. However, they may be of prognostic value, as there is evidence suggesting that IgG, IgA and 7S IgM RFs are associated with more severe disease.41–45 This risk appears to be independent of HLA alleles associated with severe disease.46
Therefore, knowing when is it useful to measure RF may be relevant in terms of the diagnostic tools for patients presenting with inflamed joints. Population-based studies have shown that some healthy people who are RF positive develop RA over time, especially if more than one isotype is persistently elevated and if they have high levels of RF.16,17 A retrospective study of stored blood samples collected as part of routine blood donation found that, in nearly 30% of those who later developed RA, RF was present in serum for 1 year or more (median 4.5 years) prior to diagnosis.18 Likewise, among military recruits who are diagnosed with SLE, the presence of IgG or IgM RF in banked serum often precedes the development of arthritis.19 The presence of both RF and ACPAs in apparently healthy people substantially increases the probability of developing RA, with the presence of the two autoantibodies associated with a relative risk of approximately 70 in a population study.47 However, most asymptomatic persons who are RF positive do not progress to RA or SLE and, as a result, measurement of RF is a poor screening test for future rheumatic disease.17
Estimates of the sensitivity and specificity of RF vary depending on the populations being examined, which will also affect the calculated predictive value. As noted, the sensitivity of RF in RA (i.e. the proportion of patients with RA who are RF positive) has ranged from 26% to 90%. One specific meta-analysis reported the overall sensitivity to be 69% (95% confidence interval 65–73%).48
The specificity (i.e. the proportion of a control population without RF) depends substantially on the choice of the control group. The overall specificity of RF reported in heterogeneous publications analysed as part of a meta-analysis by Nishimura et al.49 was 85%. The prevalence of RF in a young, healthy population is approximately 4%; therefore, the specificity with respect to a young, healthy population is approximately 96%. The specificity with respect to disease control populations is substantially lower, especially if the disease control population includes patients with rheumatic and other diseases associated with RF.
As with any diagnostic test, the predictive value is also affected by the estimated likelihood of disease prior to ordering the test (i.e. the pretest probability) and, in the case of RF, by the proportion of patients with a non-rheumatic disorder associated with RF production.
In a study of consecutive tests ordered by health care providers in a large academic medical centre in the USA, the prevalence of RA was approximately 13%.50 The positive predictive value of RF (the likelihood of having RA if RF is positive) was only 24% for RA and 34% for any rheumatic disease. Thus, RF has a low positive predictive value if the test is ordered among patients with a low prevalence of RF-associated rheumatic disease or with few clinical features of systemic rheumatic disease.
In contrast, RF testing, while imperfect, is often useful in evaluating patients presenting with inflammatory arthritis and with a high pretest probability of RA. RF testing in a rheumatology clinic practice with a prevalence of RA of 16.4% demonstrated a specificity of 97–98% in patients with non-inflammatory disorders and of 95–97% in all patients, yielding a positive predictive value of approximately 80%.47 In patients with ‘undifferentiated’ inflammatory arthritis, the presence of RF was helpful in predicting the ultimate diagnosis of RA, with a positive odds ratio of approximately 30.51 While an earlier study found no predictive value for RF testing in such patients,52 most studies of patients in early arthritis clinics demonstrate that RF, as well as ACPAs, contributes to the ability to predict persistent synovitis and/or RA.53
Thus, the positive predictive value of RF testing is poor in a population with a low pretest probability of RA, including those without inflammatory arthritis. Testing patients with non-specific arthralgia, fibromyalgia or osteoarthritis is not recommended because a positive test result is likely to be a false positive. In contrast, RF testing for patients with inflammatory arthritis (which is much less common than the non-inflammatory conditions) has high positive predictive value and clinical utility.
The calculated negative predictive value of RF (the likelihood of not having disease if RF is negative) tends to be high in a population with a low pretest probability of RA. However, a negative RF test result in this setting may not be particularly useful clinically. In the study undertaken by Shmerling et al.,50 the negative predictive value for RA and for any rheumatic disease was 89% and 85%, respectively. It is important to appreciate, however, that the value of a negative test depends on the clinical setting. Suppose, for example, that a patient has an estimated 10% chance of developing RA; a negative RF test (assuming a sensitivity of 70% and specificity of 85%) will decrease the likelihood of developing RA from 10% to 4%. This small benefit may not justify performing the RF test.
The negative predictive value of RF in patients with a high pretest likelihood of RA is limited by the relatively high frequency of seronegative RA. For example, if the pretest probability is 50% and if the sensitivity is 70%, the post-test probability is 27% (i.e. not low enough to feel confident that RA is not the diagnosis). In fact, patients with RF-negative RA (constituting 20–50% of RA sufferers) present in this fashion.
The presence or absence of RF may have some value in predicting response to treatment. As an example, the anti-CD20 B-cell-depleting monoclonal antibody rituximab may be less effective for patients with seronegative RA than for those with seropositive RA.54
The term seropositive does not encompass the concept of how positive the RF is. The titre of RF should be considered when analysing its utility. The higher the titre, the greater is the likelihood that the patient has a rheumatic disease. There are, however, frequent exceptions to this rule, particularly among patients with one of the chronic inflammatory disorders noted above. Furthermore, the use of a higher titre for diagnosis decreases the sensitivity of the test as it increases the specificity (by decreasing the incidence of false-positive results). In the study carried out by Shmerling et al.,50 for example, an RF titre of 1:40 or greater was 28% sensitive and 87% specific for RA; in comparison, a titre of 1:640 or greater increased the specificity to 99% (i.e. almost no false-positive results) but reduced the sensitivity to 8%.
The paradoxical nature of the term ‘seropositive arthritis’ often leads patients to believe, erroneously, that it refers only to rheumatoid arthritis; however, this destructive arthritis can be made diagnosed in the absence of RF. Hence, there is tendency to consider the presence of RF as a prognostic indicator in such conditions. RF-positive patients with RA may experience more aggressive and erosive joint disease and extra-articular manifestations than those who are RF negative.55,56 Similar findings have been observed in patients with juvenile idiopathic arthritis.57 However, these general observations are of limited utility in ant individual patient owing to wide interpatient variability. In this setting, it is not possibly to accurately predict the disease course from the presence or absence of RF alone.
Some have suggested that erosive disease may be accurately predicted by analysing the combination of HLA-DRB1 and RF status among patients with RA.58 However, these tests are of limited value in an individual patient as almost one-half of ‘high-risk’ patients have no erosions at 1 year.58
Rheumatoid factor status may be useful in combination with other indicators, including C-reactive protein, the erythrocyte sedimentation rate and severity of synovitis on physical examination, to predict progression of radiographic changes in RA patients and to guide treatment.59,60
Repeat testing of RF may be useful if a patient’s diagnosis remains uncertain. However, serial testing is of no clear benefit in a patient with established RA. In cases of Sjögren’s syndrome, the disappearance of a previously positive RF may herald the onset of lymphoma,61 therefore, some clinicians check RF repeatedly in their patients with this disease. However, the clinical utility of this practice has not been critically assessed.
Effective treatment of RA may result in a fall in RF titre in patients who are RF positive.62 Nevertheless, the fluctuation in level of RF does not correlate closely with disease activity. RF testing should not be used routinely to monitor RA disease activity in clinical practice.
Anticitrullinated peptide antibodies
Traditionally, the concept of seropositive arthropathy is related to RF; however, this can now be expanded to include the presence of other biological indicators or antibodies with the advent of testing for ACPA. This has become common in the evaluation of patients for diagnosing RA. The sensitivity of ACPA assays for RA varies from about 50% to 75%, depending on the assay and study population, while specificity of ACPA for RA is relatively high, usually over 90%.32,49,63–68 Enzyme-linked immunosorbent assays (ELISAs) for antibodies against CCP are the most commonly used assays for ACPA. Another ACPA assay that is commercially available detects antibodies against mutated citrullinated vimentin. As with RF, ACPA may be present prior to the appearance of symptoms of RA.
A decrease in ACPA titres can be seen in patients treated effectively, particularly if treated early with non-biological or biological disease-modifying antirheumatic drugs, but is less frequent and of a lower magnitude than the decrease in IgM RF.70
In an ELISA for ACPA, arginine residues are replaced by citrulline in a mixture of CCPs, increasing the sensitivity of the assay for ACPA. These anti-CCP assays have become the principal commercially available assay kits for the detection of ACPA.
The newer-generation assays, including the second-generation anti-CCP antibody assays (anti-CCP2), have improved sensitivity and specificity compared with the original anti-CCP assays.67,69,71–73 The best data regarding test performance characteristics of ACPA come from a 2010 systematic review and meta-analysis of 151 studies.69 The pooled results in this 2010 meta-analysis confirmed the results of a 2007 meta-analysis, which did not stratify by study design or disease duration.49 Sensitivity of ACPA in both analyses was 67% and the specificity in the 2010 and the 2007 analysis was 96% and 95%, respectively.
The 2010 analysis included the following findings:
There was substantial heterogeneity in test performance, which resulted from differences in study design, stage of RA and type of ACPA used in a given study. Cross-sectional and case–control studies of patients with established RA overestimated sensitivity.
In the cohort studies of patients with RA for < 2 years, the sensitivity of anti-CCP2 and RF was nearly the same (58% vs. 56%, respectively), but specificity was significantly higher for anti-CCP2 (96% vs. 86%, respectively).
There was insufficient evidence to determine whether or not the combination of anti-CCP2 and RF provides greater benefit than the use of anti-CCP2 alone.
Although ACPA testing is more specific for RA than testing for RF,49,69 positive results can occur in other diseases, including several autoimmune rheumatic diseases, tuberculosis (TB) and sometimes chronic lung disease.68,74–78 For example:
Anti-CCP antibodies have been reported in SLE and primary Sjögren’s syndrome, usually in association with deforming or erosive arthritis.79–84 For example, 17% of a cohort of 335 patients with SLE were anti-CCP positive in one study80 and, in another, 10% of 155 consecutive patients with primary Sjögren’s syndrome were anti-CCP positive.83 However, in such cases, some experts suggest that such patients should be reclassified as having SLE-RA and primary Sjögren’s syndrome/RA overlap syndromes, respectively.84 Similar findings have also been reported among patients with psoriatic arthritis.85
An increased prevalence of anti-CCP antibodies has been noted in patients with active TB. The rate in different studies has ranged from as high as 32–39%74,75 to as low as 7%.76 Many patients with TB and anti-CCP also have antibodies to a cyclic peptide that contains an unmodified arginine residue.75 This suggests that binding of the antibodies from patients with TB is determined by portions of the CCP other than the citrulline moiety.
In contrast to RF, anti-CCP antibodies are rarely present in the serum of patients with hepatitis C virus (HCV) infections.
In a study of 257 patients with α1-antitrypsin deficiency and 113 patients with chronic obstructive pulmonary disease, anti-CCP antibodies were detected in 3% and 5% of patients, respectively.78 In the patients with α1-antitrypsin deficiency there was no difference in the presence of anti-CCP antibodies between cigarette smokers and non-smokers.
Anti-citrullinated peptide antibody-positive patients with early RA are at increased risk of progressive joint damage,64,86–89 but ACPA testing may predict erosive disease more effectively than RF.49,90 This was illustrated in a series of 145 patients, among whom the extent of radiographically apparent damage after 5 years of observation was higher in those with detectable ACPA than in those who were RF positive.88 The presence of ACPA was also predictive of more rapid radiographic progression in two studies of 183 and 279 Swedish patients with early RA86,89 and in the BEST [Behandelstrategieën voor Reumatoide Artritis (Treatment Strategies for Rheumatoid Arthritis)] trial of 508 Dutch patients.86
A positive ACPA test also appears to predict an increased risk of radiographic progression in patients with early oligo- or polyarthritis who are IgM-RF negative; this was demonstrated in a prospective study that included 178 patients.91 Radiographic progression (more than 5 units by Sharp score) was more frequent in the ACPA-positive patients than in those with a negative test result (40% vs. 5%, respectively). The anti-CCP test for ACPA correctly predicted whether or not there would be worsening radiographic damage in 83% of these patients.
Having defined the concept of seropositivity, it is appropriate to describe briefly the associated arthropathies.
Rheumatoid arthritis is a common disorder recognized with increased frequency since the nineteenth century,92 although it may have been ubiquitous since ancient times. Alfred Baring Garrod was the first to distinguish RA from gout and rheumatic fever in the mid-nineteenth century. The incidence of RA has continued in recent decades, demonstrating an apparent increase into this millennium.93
The cause of RA is not known, but many possible aetiologies have been explored. Important aetiological clues have been suggested, but an influence of environmental factors interacting with hormonal, genetic, infectious and other variables has been postulated. Complex interactions of these factors, perhaps in genetically susceptible hosts, leads to the final outcome of polyarticular synovitis, which defines RA. Table 1 sets out the latest classification criteria to assist RA diagnosis.
|A. Joint involvement:|
|1 large joint||0|
|2–10 large jointsa||1|
|1–3 small joints (with or without involvement of large joints)||2|
|4–10 small jointsb (with or without involvement of large joints)||3|
|> 10 joints (at least 1 small joint)||5|
|B. Serology (at least 1 test result is needed for classification):|
|Negative RF and negative anti-CCP antibodies||0|
|Low-positive RF or low-positive anti-CCP antibodiesc||2|
|High-positive RF or high-positive anti-CCP antibodiesd||3|
|C. Acute phase reactants:|
|Normal CRP level and normal ESR||0|
|Abnormal CRP level or abnormal ESR||1|
|D. Duration of symptoms:|
|< 6 weeks||0|
|≥ 6 weeks||1|
ACR, American College of Rheumatology; anti-CCP, anti-cyclic citrullinated protein; CMC, carpometacarpal; CRP, C-reactive protein; DIP, distal interphalangeal; ESR, erythrocyte sedimentation rate; EULAR, European League Against Rheumatism; MCP, metacarpophalangeal; MTP, metatarsophalangeal; PIP, proximal interphalangeal; RA, rheumatoid arthritis; RF rheumatoid factor.
The annual incidence of RA has been reported to be around 40 per 100 000 population. The disease prevalence is approximately 1% in Caucasians, but varies between 0.1% in rural Africans and 5% in Pima, Blackfeet and Chippewa Indians.94,95 Women are affected two to three times more often than men.
Rheumatoid arthritis can occur in patients at any age. The peak onset is between the ages of 50 and 75 years, and the prevalence of RA in women over 65 years is up to 5%. Implications such as loss of employment, productivity and function, and the associated socioeconomic impact, have been explored.96 The lifetime risk of RA in adults is 3.6% (1 in 28) for women and 1.7% (1 in 59) for men.97
The contribution of HLA and other genes to disease susceptibility, severity and treatment response in RA is briefly described here. Among the growing list of genetic risk factors for RA, significant overlap exists with genes identified as risk factors for other autoimmune diseases, including SLE, inflammatory bowel disease, multiple sclerosis and ankylosing spondylitis.98 Newly recognized genetic markers associated with RA include an allele of the Fc-gamma receptor, a polymorphism marker in the β2-adrenergic receptor, and a low-inducible allele of the cytochrome P450 subtype 1A2.99
The firmest link between a genetic susceptibility factor and RA is the association of the disease with an epitope in the third hypervariable region of the HLA-DR beta-chains, known as the ‘shared epitope’. Individuals with the sequence Leu-Glu-Lys-Arg-Ala at residues 67, 70, 71, 72 and 74 have a much higher incidence and prevalence of RA than those who do not have this epitope.100 This sequence is found in DR4 and DR14 and some DR1 beta-chains. The DR4 beta-chains with the strongest associations with RA are DR-beta*0401, DR-beta*0404, DR-beta*0101 and DR-beta*1402.
Although many other genetic risk factors are yet to be defined, these appear to contribute significantly to disease susceptibility and severity.
Rheumatoid factor was discovered in 1940, and much research has linked this autoantibody to the pathophysiology of severe RA. Although it is clear that the presence of RF alone does not cause RA, there is no doubt that patients with significant RF titres have a greater likelihood of extra-articular disease than seronegative patients. Classic experiments by Hollander and colleagues101 showed that injection of RF into the joint of a patient with RA led to a marked inflammatory response, which did not occur following injection of IgG.101
In a cohort study of healthy individuals from Finland, 9 out of 129 patients who were RF positive subsequently developed seropositive RA over a 10-year investigation period, compared with only 12 out of 7000 patients with negative tests (7% vs. 0.2%, respectively).102
In a study of 83 patients with RA for whom stored blood samples were available as a result of blood donation or prenatal testing, the prevalence of ACPA was significantly higher in patients preceding diagnosis than in control subjects (34% vs. 5%, respectively).103 There was also a significant increase in the prevalence of RFs of all isotypes (17%, 19% and 33% for RF of IgG, IgM and IgA isotypes, respectively).
In a case–control study of 79 patients with RA for whom stored serum was available from blood donations prior to the development of RA (1–51 samples per patient, dating back up to 15 years before the onset of RA), 49% had detectable ACPA and/or anti-IgM RF on at least one occasion, and 41% had ACPA detectable when symptoms first developed.18
In a nested case–control study, the levels of several cytokines, cytokine receptors and chemokines were markedly higher in pre-diagnosis stored blood samples from 69 patients who subsequently developed RA than in blood from control subjects.104 Notably, levels were particularly increased in individuals also positive for anti-CCP and RF and, after disease onset, immune system activation was more generalized.
Involvement of the musculoskeletal system is extremely common in patients with SLE. Arthralgia, arthritis, osteonecrosis (avascular necrosis of bone) and myopathy are the principal manifestations. Osteoporosis, often due to glucocorticoid therapy, may increase the risk of fractures.
Arthritis and arthralgia have been noted in up to 95% of patients with SLE. These symptoms may be mistaken for another type of inflammatory arthritis and can precede the diagnosis of SLE by months or years.
The arthritis and arthralgia of SLE tend to be migratory; symptoms in a particular joint may be gone within 24 hours.
Involvement is usually symmetrical and polyarticular, with a predilection for the knees, carpal joints and joints of the fingers, especially the proximal interphalangeal joint. The ankles, elbows, shoulders and hips are less frequently involved. Involvement of the sacroiliac joints and cervical spine may occur, but this is rare. Monoarticular arthritis is unusual and suggests an alternative cause, e.g. infection.
Morning stiffness is usually measured in minutes and is not prolonged as it is in RA.
The degree of pain often exceeds objective physical findings, and tenderness may be difficult to assess because of increased pain sensitivity in some patients, which can be associated with coexisting fibromyalgia.
Although the arthritis of SLE is generally considered to be non-deforming, flexion deformities, ulnar deviation, soft tissue laxity and swan neck deformities, as seen in RA, have been noted in 15–50% of patients with SLE.105–110
However, in contrast to RA, erosions in the hand are rarely noted on plain radiographs of the hands of patients with SLE.108,111 Similarly, in one study of 26 patients with SLE using ultrasonography, synovial proliferation and effusions were detected in the knees in 23% of patients and synovitis was seen in 40%, although no erosions were detected.112 More sensitive imaging methods, such as magnetic resonance imaging, often reveal erosive changes and abnormalities of the soft tissues, including capsular swelling, proliferative tenosynovitis and synovial overgrowth.110 The presence of antibodies to anti-CCP antibodies (present in 8% of patients with SLE) is strongly associated with erosive arthritis (odds ratio of 23 for anti-CCP-positive patients vs. -negative patients).84,88
Hand deformities tend to occur in patients who have been receiving glucocorticoids, who have anti-Ro and/or anti-La antibodies,109 or who have longstanding disease.110 In contrast to the typical findings in RA, these deformities are usually easily reducible; they are thought to be caused by lax joint capsules, tendons and ligaments that cause joint instability.110 Thus, the hand deformities in patients with SLE resemble Jaccoud’s arthritis, a non-erosive chronic deforming arthritis that may follow acute rheumatic fever.113
Tendons may also be involved in SLE. Tenosynovitis has been noted in 10–44% of patients, including epicondylitis, rotator cuff tendinitis, Achilles tendinitis, tibialis posterior tendinitis and plantar fasciitis.108,110,114,115 Infrapatellar and Achilles tendon ruptures are rare.116
Synovial effusions are infrequent in patients with SLE. When they do occur, they are usually small and the fluid is clear or slightly cloudy.117 In contrast to the highly inflammatory exudates of RA, the synovial fluid is only mildly inflammatory, with low protein levels and white blood cell counts (similar to a transudate).108 Antinuclear antibodies (ANAs) and lupus erythematosus cells have been observed in synovial fluids and, if present, have been thought to be useful diagnostically. However, they add little to positive ANA in the serum.
Synovial histopathology tends to be non-specific, with superficial fibrin-like material and local or diffuse lining cell proliferation.108 Vascular changes have included perivascular mononuclear cells, lumen obliteration, enlarged endothelial cells and thrombi; fibrinoid necrosis (e.g. vasculitis) is uncommon.118 Synovitis in SLE has a molecular synovial signature distinct from that seen in osteoarthritis or RA in that interferon-inducible genes are upregulated.119
Septic bacterial arthritis is uncommon, although patients with SLE may be intrinsically immunosuppressed and may become more so with medication. When septic arthritis occurs, it can be secondary to infections with Salmonella, Neisseria gonorrhoeae, Neisseria meningitidis and other organisms.120
Mixed connective tissue disease
Mixed connective tissue disease is defined as a generalized connective tissue disorder characterized by the presence of a high titre of anti-U1 ribonucleoprotein (RNP) antibodies in combination with clinical features commonly seen in SLE, scleroderma and polymyositis.121,122 It often takes several years before enough overlapping features have appeared to be confident that MCTD is the most appropriate diagnosis.123 The distinctive overlap features of SLE, scleroderma and polymyositis commonly appear sequentially over time. Thus, in its early stages, MCTD is often referred to as an undifferentiated connective tissue disease (UCTD).124,125
The early clinical features of MCTD are non-specific and may consist of general malaise, arthralgias, myalgias and low-grade fever.126,127 A specific clue that these symptoms are caused by a connective tissue disease is the discovery of a positive ANA in association with Raynaud’s phenomenon.
Almost any organ system can be involved in MCTD. There are, however, four clinical features that suggest the presence of MCTD rather than another connective tissue disorder such as SLE or scleroderma:
autoantibodies whose fine specificity is anti-U1 RNP, especially antibodies to the 68-kDa protein.134
Most patients with MCTD present in the second or third decade of life. Unlike SLE, however, sun exposure is not a precipitating factor. Drug-induced MCTD is a rare occurrence but may be an occasional feature of anti-tumour necrosis factor therapy.135,136 Vinyl chloride137 and silica138 are the only environmental agents that have been associated with MCTD.
In the early phases of MCTD, many patients complain of easy fatigability, poorly defined myalgias, arthralgias and Raynaud’s phenomenon, and diagnostic considerations include the early stages of RA, SLE or UCTD.139,140 Most, if not all, of the major organ systems may be involved at some time during the course of MCTD, including the skin, joints, muscles, heart, lungs, gastrointestinal tract, kidneys, central nervous system and haematological system. A high titre of anti-RNP antibodies in a patient with UCTD is a powerful predictor for later development of MCTD.141
Joint involvement in MCTD is more common and frequently more severe than in classical SLE. Approximately 60% of patients with MCTD develop an obvious arthritis, often with deformities characteristic of rheumatoid disease, such as boutonniere deformities and swan neck changes.137,142 The radiographic appearance often resembles Jaccoud’s arthropathy.142
About 70% of patients with MCTD test positive for RF.148 Anti-CCP antibodies are found in about 50% of patients with MCTD, especially in those MCTD patients who also fulfil the American College of Rheumatology diagnostic criteria for RA.149
Sjögren’s syndrome is a chronic inflammatory disorder characterized by diminished lacrimal and salivary gland function. Sjögren’s syndrome occurs in a primary form not associated with other diseases and in a secondary form that complicates other rheumatic conditions, with the most common being RA.
In primary or secondary Sjögren’s syndrome, reduced exocrine gland function leads to the ‘sicca complex’, a combination of dry eyes (xerophthalmia) and dry mouth (xerostomia).150,151 In addition, a variety of other disease manifestations can occur in Sjögren’s syndrome. The clinical manifestations of Sjögren’s syndrome are divided into the exocrine gland features and the extraglandular disease features.152
The extraglandular disease manifestations and prognosis of Sjögren’s syndrome will be reviewed here.
Approximately 50% of patients with primary Sjögren’s syndrome complain of arthralgia, with or without evidence of arthritis.153 The arthropathy is usually symmetrical, intermittent and affects hands and knees. Joint disease in Sjögren’s syndrome is typically non-erosive and non-deforming.
Rheumatoid factor is reported in approximately 40% of patients with Sjögren’s syndrome and is associated with a significantly higher prevalence of articular symptoms (45% vs. 33% without articular complaints).154 Anti-CCP antibodies are less commonly reported.155
In one particular study,156 14.5% of the 509 RA patients fulfilled the Sjögren’s syndrome criteria. The data suggest that patients with RA/Sjögren’s syndrome are older than patients with primary Sjögren’s syndrome. RA/Sjögren’s syndrome patients had a longer duration of disease than those with RA only or primary Sjögren’s syndrome. Patients with RA/Sjögren’s syndrome have a more severe form of arthritis.156
Cryoglobulinaemia refers to the presence of cryoglobulins in a patient’s serum. The term is often used to refer to a systemic inflammatory syndrome that generally involves small- to medium-vessel vasculitis resulting from cryoglobulin-containing immune complexes. The terms ‘cryoglobulinaemic syndrome’ and ‘cryoglobulinaemic vasculitis’ are sometimes used to make a distinction between the clinically apparent disorder and the asymptomatic presence of cryoglobulins.157–159
Although the phenomenon of cryoprecipitation and associated hyperviscosity was described in the 1930s, the association of cryoglobulin with the triad of palpable purpura, arthralgia and myalgia was not described until the 1960s.160,161 It is important to note that these two different, yet highly representative, clinical syndromes generally reflect different types of underlying cryoglobulin:
Hyperviscosity is typically associated with cryoglobulin owing to haematological malignancies and monoclonal immunoglobulins.
‘Meltzer’s triad’ of palpable purpura, arthralgia and myalgia is generally seen with polyclonal cryoglobulins.
The prevalence of clinically significant cryoglobulinaemia has been estimated to be approximately 1 in 100 000, although detectable levels of circulating cryoglobulins have been seen in a significant proportion of patients with chronic infections and/or inflammation: 15–20% of human immunodeficiency virus (HIV)-infected individuals, 15–25% of patients with connective tissue diseases, 40–65% of hepatitis C-infected individuals and as many as 64% of people with HIV and hepatitis C coinfection, perhaps especially among patients with HCV genotype 1.162–166
One frequently used classification scheme is that of Brouet, which uses the immunological analysis of the cryoglobulin to delineate the clonality of the responsible cryoglobulin, particularly with regard to RF binding activity:167
Type I. The presence of isolated monoclonal immunoglobulin (typically IgG or IgM, less commonly IgA or free immunoglobulin light chains) is the criterion for classifiying cryoglobulinaemia as type I. The proportion of patients with a type I cryoglobulinaemia varies substantially among case series but is approximately 5–25%. The haematological diagnoses are typically Waldenström’s macroglobulinaemia and multiple myeloma.
Type II. A mixture of polyclonal immunoglobulin in association with a monoclonal immunoglobulin, typically IgM or IgA, with RF activity defines type II cryoglobulinaemia. This type of cryoglobulinaemia, also called essential mixed cryoglobulinaemia, accounts for approximately 40–60% of cases. Type II cryoglobulinaemia is often the result of persistent viral infections, particularly HCV and HIV.
Type III. Mixed cryoglobulins consisting of polyclonal immunoglobulins characterize type III cryoglobulinaemia. Cryoglobulinaemia type III accounts for approximately 40–50% of all cases and is often secondary to connective tissue diseases.
While generally useful, this schema does not account for cryoglobulinaemia with characteristics, such as oligoclonal IgM components with or without trace polyclonal immunoglobulin responses, often referred to as type II–III,168,169 or biclonal cryoglobulinaemia.170,171 Many speculate that such phenomena reflect an intermediate stage of transition between types II and III, akin to the malignant transformation of gammopathies of unknown significance (e.g. monoclonal gammopathy of unknown significance) to multiple myeloma or other lymphoproliferative disorders,172 but a direct pathogenic link has not been clearly demonstrated.
Otherwise, cryoglobulinaemia syndromes are often classified as essential or secondary, based upon the presence of underlying diseases, particularly chronic HCV infection or connective tissue diseases.173 At the same time, the clinical features of both essential and secondary cryoglobulinaemia overlap, with respect to Brouet subtype and organ involvement, making most distinctions among cryoglobulinaemia syndromes difficult to interpret consistently.
The mixed cryoglobulinaemias (types II and III) generally result from chronic inflammatory states, such as connective tissue diseases, e.g. SLE or Sjögren’s syndrome, or viral infections, e.g. HCV, although lymphoproliferative disorders have rarely been associated in addition. Many of these disease states share B-cell hyperactivation and/or hyperproliferation,174,175 which seem to predispose to the selective expansion of cryoglobulin-producing B-cell clones,176 but a precise ontology of pathogenic cryoglobulins in mixed cryoglobulinaemia has not been delineated. This paradigm of pathogenesis has been most intensively studied for chronic HCV infection, in which B-cell hyperactivation may result upon direct infection via the cell surface protein CD81,177 via chronic, antigen-non-specific stimulation by macromolecular serum complexes containing HCV – including HCV-IgG and HCV-lipoprotein178,179 – or via an HCV antigen-specific mechanism180 – resulting in expansion of specific B-cell clones, such as those expressing the WA idiotype181 or V(H)1-69.182 HCV particles are often found in such patients’ serum cryoglobulin complexes, but at the same time, cryoglobulinaemia development in HCV does not directly require the hepatitis C virion or its components.183 In this sense, cryoglobulin development may, in fact, reflect a normal, expected response to regulate immune complexes in states of chronic immune activation.
Mixed cryoglobulinaemias most often produce constitutional and non-specific symptoms, such as arthralgias, fatigue and myalgias, as well as palpable purpura resulting from cutaneous vasculitis and sensory changes or weakness due to peripheral neuropathy. Nevertheless, the classical ‘Meltzer’s triad’ of purpura, arthralgias and weakness is seen in as few as 25–30% of patients.166,179
Musculoskeletal complaints such as arthralgias and myalgias are common in the mixed cryoglobulinaemias, but frank arthritis or myositis is rare.184,185 Arthralgias are typically described in over 70% of patients, especially affecting the metacarpophalangeal joints, proximal phalangeal joints, knees and ankles, often in type III cryoglobulinaemia and often exacerbated by cold exposure. Many suggest that such complaints typically accompany cryoglobulinaemia syndromes in association with connective tissue diseases,173 but these rarely accompany type I cryoglobulinaemia.