Published: August 2010
The diagnosis of rheumatologic diseases is based on clinical information, blood and imaging tests, and in some cases on histology. Blood tests are useful in confirming clinically suspected diagnosis and monitoring the disease activity. The tests should be used as adjuncts to a comprehensive history and physical examination.
The value of a test in diagnosing a certain condition depends on its pretest probability. A positive test result with high pretest probability helps to make a diagnosis, but a negative test result with low pretest probability helps to rule out the diagnosis. However, clinicians cannot rely heavily on blood tests in making the diagnosis of rheumatologic diseases, except for certain tests that are highly specific for certain diseases. Improper application of these tests leads to misdiagnosis, inappropriate therapy, and unnecessary health care expenses. This chapter discusses blood tests that are useful in evaluating various rheumatologic diseases.
Acute-phase reactants are proteins whose plasma concentration increases (positive acute-phase proteins) or decreases (negative acute-phase proteins) by at least 25% during inflammatory states.1 Box 1 lists positive and negative acute-phase reactants. The effect of inflammatory molecules such as interleukin (IL)-6, IL-1, tumor necrosis factor α (TNF-α), interferon gamma (IFN-γ), and transforming growth factor β (TGF-β) causes a change in hepatic protein synthesis collectively known as acute-phase response. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are the most widely measured acute-phase reactants in clinical practice.
ESR is a measure of the height of erythrocytes that fall through plasma in a Westergen or a Wintrobe tube over a period of 1 hour. ESR can be greatly influenced by the shape and number of red blood cells as well as other plasma constituents like fibrinogen, globulins, and albumins. It can be spuriously high in the absence of inflammation, as in anemia, nephritic syndrome, and hypergammaglobulinemia, and it can be spuriously normal in cryoglobulinemia and hemoglobinopathy. ESR increases steadily with age, and the upper limit varies with sex; hence, ESR is difficult to interpret compared to CRP.
The concentration of CRP in serum is more sensitive than ESR to evaluate and monitor inflammation, and it is independent of factors that affect ESR. It correlates better with disease activity, and the rise in CPR level is seen much earlier than that of other acute-phase reactants, usually 4 to 6 hours after tissue injury.
|Box 1: Positive and Negative Acute-Phase Reactants|
|Positive Acute-Phase Reactants|
|Negative Acute-Phase Reactants|
Both ESR and CRP levels can be elevated in a wide variety of conditions including trauma, infection, infarction, neoplasms, and inflammatory arthritis. Usually ESR and CRP levels correlate well, but in some patients levels may be discordant for reasons that are unclear. They are very useful in monitoring disease activity in rheumatologic conditions such as rheumatoid arthritis, polymyalgia rheumatica,2 and giant cell arteritis. Some studies have shown that the pretreatment ESR value is of some prognostic value in polymyalgia rheumatica. Most patients with active lupus have normal or minimally elevated CRP levels, and markedly elevated concentrations of CRP in SLE should raise a suspicion of bacterial infection. Other causes for elevated CRP in SLE patients include serositis, synovitis, and vasculitis.
Antinuclear antibodies (ANAs) directed against a variety of nuclear antigens have been detected in the serum of patients with many rheumatic and nonrheumatic diseases as well as in healthy persons. Various immunochemistry techniques are used to detect and characterize these ANAs. These methods include immunofluorescence microscopy, hemagglutination, immunodiffusion, complement fixation, and enzyme-linked immunosorbent assay (ELISA).
Immunofluorescent microscopy performed on human epithelial-2 (Hep-2) cells is widely used for initial screening. It is a highly sensitive test and is often abnormal in patients with ANA-associated diseases, but the specificity is low and the test has many false-positive results. It is reported as positive or negative and includes a titer. ANA testing performed using ELISA technology is very sensitive and has a high incidence of false-positive results (Figure 1).
ANA testing is very useful in establishing a diagnosis of systemic lupus erythematosus (SLE). Nearly all patients with SLE have a positive ANA test, with a sensitivity of 93% to 95% and a specificity of 57%.3 However, even healthy persons can have a positive ANA test at lower titers. About 25% to 30% of healthy persons have a positive test with a titer of 1 : 40, 10% to 15% at a titer of 1 : 80, and 5% at a titer of 1 : 160 or greater. The frequency increases with age, particularly in women. ANA titer of 1 : 40 is seen in 25% to 30% of relatives of patients with rheumatologic disorders.3
A positive ANA in a patient with low pretest probability is often of unclear significance. Due to the low prevalence of SLE (40-50/100,000), most people with positive ANA do not have lupus (positive predictive value [PPV], 11%). However, a high ANA titer (>1 : 640) should increase the suspicion for an autoimmune disorder, although not diagnostic of a disease, and patients with high titers should be carefully followed for the development of connective tissue disorder. ANA titer is not routinely used for assessing the disease activity in lupus, and serial ANA testing is therefore not useful.
In addition to lupus, ANA testing is helpful in diagnosing other rheumatic diseases such as systemic sclerosis and Sjögren's syndrome (Table 1). The sensitivity of ANA in diagnosing systemic sclerosis is 85% and the specificity is 54%.3 Although ANA is not included in the 2002 classification criteria for Sjögren's syndrome, it is found in 80% of patients with primary Sjögren's syndrome and at high titers (>1 : 320) in nearly one half of the patients.4 Patients presenting with Raynaud's phenomenon should also have ANA testing because a positive ANA test indicates an increased risk of developing an associated systemic rheumatic disease from 19% to 30%, whereas a negative test indicates a risk of 7%.5 Additionally, ANA testing helps to stratify the risk of uveitis in patients with juvenile idiopathic arthritis.
|Disease||Sensitivity (%)||Specificity (%)|
|Systemic lupus erythematosus||93-95||57|
|Juvenile chronic arthritis||57||39|
|Juvenile chronic arthritis with uveitis||80||53|
ANA can also be positive in many autoimmune disease states not associated with connective tissue diseases, such as autoimmune hepatitis, primary autoimmune cholangitis, primary biliary cirrhosis, and Crohn's disease. Other disorders associated with positive ANA titer include such chronic infectious diseases as mononucleosis, subacute bacterial endocarditis, tuberculosis, and lymphoproliferative diseases. Hence, ANA testing should be reserved for patients with high suspicion for systemic autoimmune disease, such as young women with fatigue, joint pain, and rash, and should not be used as a screening test in patients complaining of generalized fatigue and musculoskeletal pain, particularly elderly patients.
There are different types of ANAs based on their target antigen, including single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA), nuclear histone and nonhistone nuclear proteins, and RNA protein complexes. The staining pattern seen on indirect immunofluorescence (IIF) gives some indication of the specificity of the antibodies in the sample (Table 2 and see Figure 1). Identification of the specificity for extractable nuclear antigens (ENA) is warranted because this can further differentiate between the distinct types of autoimmune connective tissue diseases. Hence, a positive ANA test should be followed by an anti-DNA antibodies assay.
|Homogenous and Diffuse||DNA-histone complex (nucleosome)||SLE (60%)
Drug-induced lupus (95%)
|Speckled||RNA polymerase types II and III||Systemic sclerosis|
|RNP||MCTD (100%)||Scl-70||Systemic sclerosis (15%-70%)|
|Sm||SLE (25%-30%)||SS-A||Sjögren's syndrome (8%-70%)||SLE (35%-40%)|
|SS-B||Sjögren's syndrome (14%-60%)|
|Nucleolar RNA, RNA polymerase 1||Systemic sclerosis|
MCTD, mixed connective tissue disease; SLE, systemic lupus erythematosus.
Antibodies to dsDNA are often measured in SLE and are commonly referred to as anti-DNA antibodies. They are very useful in the diagnosis of SLE and assessment of disease activity, and they are associated with lupus nephritis.
The most commonly used techniques for measuring anti-dsDNA antibodies are ELISA and immunofluorescence (e.g., using Crithidia luciliae as substrate). Radioimmunoassay (e.g., the Farr assay) is still available but its use has declined. The hemoflagellate Crithidia luciliae contains in its kinetoplast pure dsDNA, not complexed to proteins. Serially diluted serum samples are added to the slide carrying Crithidia cells. Binding of antibodies is visualized by fluorescinated anti-immunoglobulin (Ig) G antibodies. This method can be used to confirm the presence of anti-dsDNA antibodies when the ELISA results are discrepant.
The sensitivity of anti-dsDNA antibody for diagnosis of SLE is 57.3% and the specificity is 97.4%.6 These antibodies are present at some time in the course of the disease as the levels fluctuate and may be absent at times. Anti-DNA antibodies have been reported in patients with a variety of other rheumatologic and nonrheumatologic diseases including rheumatoid arthritis, Sjögren's syndrome, scleroderma, drug-induced lupus, Raynaud's phenomenon, mixed connective tissue disease, discoid lupus, myositis, chronic active hepatitis, uveitis, Graves’ disease, and anticardiolipin antibody syndrome and in women with silicone breast implants. Not all patients with SLE have positive anti-dsDNA antibodies; therefore, a negative test does not exclude the diagnosis of SLE. The prevalence of patients with a positive anti-DNA assay despite a negative ANA has been reported to be 0% to 0.8%. Therefore, unless there is a reasonable suspicion that the ANA is falsely negative, anti-DNA antibody testing is not generally indicated in ANA-negative patients.
Antibodies directed against small nuclear riboprotein include anti-Smith (anti-Sm) antibody and antiribonuclear protein (anti-RNP) antibodies. They bind to related but distinct antibodies.
Anti-Sm antibodies are very useful for confirming the diagnosis of SLE. A positive test result strongly supports the diagnosis, although a negative test result cannot exclude it. The sensitivity of anti-SM antibody for diagnosis of lupus ranges from 24% to 30%, and specificity ranges from 96% to 98%.7
Anti-RNP antibodies bind to protein containing U1-RNA. They coexist with anti-Sm antibodies in many patients with SLE. They have a very low sensitivity and moderate specificity for diagnosing SLE, but they are very useful in diagnosing mixed connective tissue disease. The sensitivity of anti-RNP antibodies for diagnosing mixed connective tissue disease is 71% to 100% and the specificity is 84% to 100%.
The Sjögren's syndrome A antigen (anti-SSA/Ro) consists of 52- and 60-kDa proteins (called Ro52 and Ro60) complexed with Y1-Y5 RNAs. Anti-SSA/Ro and anti-SSB/La antibodies are detected by counterimmunoelectrophoresis, ELISA, and Western blot. ELISA and Western blot are the most sensitive assays but are less specific.
Anti-SSA antibodies are seen in 50% to 60% of patients with primary Sjögren's syndrome and rarely in healthy persons.8 These antibodies are also detected in other autoimmune diseases including rheumatoid arthritis, SLE, and polymyositis. About 10% to 15% of patients with secondary Sjögren's syndrome and 35% to 40% of patients with SLE have anti-SSA antibodies. Moreover, their presence is associated with development of extraglandular features including vasculitis, lymphadenopathy, nephritis, and leukopenia in patients with Sjögren's syndrome4 and with features including photosensitivity, subacute cutaneous lupus, cutaneous vasculitis, neonatal lupus, and congenital heart block.
Anti-SSB antibodies are present in 40% to 50% of patients with primary Sjögren's syndrome and in 15% of patients with SLE but rarely in other connective tissue diseases. These antibodies are sometimes seen in patients with autoimmune hepatitis.
Antibodies to both SSA and SSB antibodies should be checked in patients with sicca symptoms. Patients with primary Sjögren's syndrome and positive SSA and SSB antibodies represent a subset of patients who have more active disease and who need very close follow-up to watch for the development of extraglandular features.4
Some studies have shown antiribosomal P protein antibodies to be highly specific for lupus and associated with neuropsychiatric lupus. They are detected in 10% to 20% of patients with SLE. However, the diagnosis of neuropsychiatric lupus is still based on clinical grounds.
Antihistone antibodies are present in more than 95% patients with drug-induced lupus and up to 80% of patients with idiopathic lupus. However, the mere presence of antihistone antibodies does not indicate drug-induced lupus. Up to 80% of patients taking procainamide for 1 to 2 years develop positive ANAs, but most do not develop drug-induced lupus.
Anticentromere antibodies are associated with limited cutaneous systemic sclerosis, previously called CREST (calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, telangectasia] syndrome).They are rarely found in patients with other connective tissue diseases or in healthy persons, making them highly specific for diagnosing systemic sclerosis. The sensitivity of anticentromere antibodies for the diagnosis of limitied cutaneous systemic sclerosis is 31%, and specificity is 97%.9 They are very useful in distinguishing patients with limited systemic sclerosis from patients with diffuse systemic sclerosis or with primary Raynaud's phenomenon. Anti-centromes antibodies are predictive of limited cutaneous involvement or decreased likelihood of investital being displayed in systemic sclerosis.
Anti-Scl-70 antibody is also very useful in diagnosing systemic sclerosis. This antibody is seen in 20.2 % of patients with systemic sclerosis and is highly specific (100%) for diffuse disease.9 Anti Scl-70 and anticentromere antibodies rarely coexist in the same person. The presence of anti-Scl-70 antibodies is useful in predicting a greater likelihood for the development of diffuse cutaneous involvement and radiographic pulmonary fibrosis with an abnormal pulmonary function test.
The antinucleolar antibodies (such as RNA-polymerase I, II, and III; anti-PM Scl; and anti-Th/To) are infrequently present in patients with systemic sclerosis, which limits their predictive value in the diagnosis. However, they are highly specific for the diagnosis of systemic sclerosis.
Polymyositis and dermatomyositis are associated with autoantibodies against a group of aminoacyl tRNA synthetases. These include Jo1,PL-7,PL-12,EJ, and OJ. The anti-Jo1 antibodies are present in 20% to 25% of adult myositis patients and are highly specific for myositis associated with a constellation of symptoms including skin involvement, lung disease, Raynaud's phenomenon, inflammatory arthritis, and fever. Anti-Mi-2 antibodies are more specific for dermatomyositis and are associated with a favorable long-term prognosis.
Rheumatoid factor (RF) autoantibodies are directed against the Fc portion of IgG. The most commonly measured RF is IgM. The other RFs include IgG, IgE, and IgA. Box 2 shows RF positivity in different diseases.
|Box 2: Rheumatoid Factor Positivity in Different Diseases
|Rheumatic Conditions (Sensitivity)|
RF is detected in a wide variety of rheumatic and nonrheumatic conditions, as shown in Box 2. It is commonly used in diagnosing rheumatoid arthritis. The sensitivity of RF for diagnosing rheumatoid arthritis is around 50% to 80%, and specificity is 85% to 90%, as reported by some studies where patients with advanced disease were tested. RF may be negative in the early stages of rheumatoid arthritis, and positivity increases over time.
RF alone cannot be used for diagnosis of rheumatoid arthritis. Around 15% to 20% of patients with rheumatoid arthritis never have RF positivity, and 2% to 10% of healthy persons are RF positive. Hence, positive RF alone does not confirm rheumatoid arthritis and negative RF does not exclude it.10 RF testing must be ordered more selectively, and the best time to obtain the test may be when the suspicion of rheumatoid arthritis is low and a negative test would provide significant reassurance.10 There is a correlation between higher RF concentrations and more-severe disease and poor prognosis, but the use of RF in monitoring disease activity is unclear.
Anti–cyclic citrullinated peptide (anti-CCP) antibodies are directed against citrulline residues formed in post-translational modification of arginine. They are often elevated in patients with rheumatoid arthritis. They have a reported sensitivity of 30% to 60% and a specificity of 95% to 98% among patients meeting the criteria for rheumatoid arthritis.
Two of the most important clinical uses of this test are its high specificity for the disease and the presence of anti-CCP antibodies in early-phase rheumatoid arthritis. Some studies have shown that anti-CCP antibodies can appear in the circulation several years before the onset of rheumatoid arthritis. The presence of anti-CCP antibodies in early disease is highly predictive for more-rapid radiographic progression of disease, meaning patients with anti-CCP antibodies have significantly more joint damage than patients without this antibody.11 Hence, anti-CCP antibody should be checked in patients in whom rheumatoid arthritis is suspected on clinical grounds.
Patients with chronic hepatitis C virus infection sometimes have high titers of RF and a variety of rheumatic symptoms, but anti-CCP antibody is rarely found in these patients.12 Presence of anti-CCP antibody in such cases supports the diagnosis of concomitant rheumatoid arthritis, although a negative test does not exclude it.
Antineutrophil cytoplasmic antibodies (ANCAs) are useful adjuncts to the clinical diagnosis of certain specific diseases, such as Wegener's granulomatosis and microscopic polyangiitis. These antibodies are directed against several neutrophilic cytoplasmic components. The two main techniques for detecting ANCAs are IIF and ELISA.
IIF assay does not identify the specific antigen responsible for the immunofluorescence. ELISA is more specific and helps to identify the specific target antigen. The two relevant target antigens are proteinase 3 (PR3) and myeloperoxidase (MPO). Both PR3 and MPO are located in the azurophilic granules of neutrophils and peroxidase-positive lysosomes of monocytes. Antibodies with target specificities for PR3 and MPO are called PR3-ANCA and MPO-ANCA. IIF without identification of MPO and PR3 antigen is incomplete and of no value.
When serum from patients with vasculitis is incubated with ethanol-fixed human neutrophils, two major immunofluorescence patterns are observed: the c-ANCA pattern and the p-ANCA pattern (Figure 2). The accuracy of the results is based on the experience of the laboratory personnel interpreting ANCA immunofluorescence results.13
The c-ANCA pattern indicates diffuse staining throughout the cytoplasm, and in most cases the antibodies responsible for this pattern are directed against PR3. However, autoantibodies against other defined and undefined cytoplasmic agents, including bactericidal/permeability-increasing protein (BPI) and, rarely, MPO, can cause cytoplasmic fluorescence.
The p-ANCA pattern indicates staining around the nucleus, and the antibody responsible for this pattern is usually against MPO. However, autoantibodies against elastase, cathepsin G, lactoferrin, lysozyme, and azurocidin have also been identified as causing the p-ANCA pattern.
Atypical ANCA patterns are sometimes observed on immunofluorescence in patients with immune-mediated conditions other than systemic vasculitis, such as inflammatory bowel disease, autoimmune liver disease, malignancies, and other rheumatic diseases. Such patterns are often confused with the p-ANCA pattern.
Clinicians interpreting ANCA results should bear in mind that the results can vary from laboratory to laboratory, and hence serial determinations should be performed by the same laboratory.13 The predictive value of ANCA testing depends heavily on the clinical presentation of the patient in whom the test is performed. ANCA assays should be ordered only when the pretest probability for Wegener's granulomatosis and microscopic polyangiitis is very high.
Wegener's granulomatosis is almost always associated with ANCA positivity. Between 70% and 90% of patients with Wegener's granulomatosis are ANCA positive, with the c-ANCA pattern and antibodies directed against PR3. About 5% to 20% of patients with Wegener's granulomatosis have the p-ANCA pattern, with antibodies directed against MPO. The sensitivity of c-ANCA and PR3-ANCA is related to the extent, severity, and activity of disease at the time of testing. Patients with mild Wegener's granulomatosis may be ANCA negative. Diagnosis of Wegener's granulomatosis is based on the clinical picture; PR3-ANCA just assists in diagnosis.
Between 40% and 80% of patients with microscopic polyangiitis are ANCA positive. These patients usually have the p-ANCA pattern with MPO specificity.
ANCAs, both with PR3 and MPO specificity, have been detected in Churg-Strauss syndrome, but MPO specificity is more common. Antiglomerular basement membrane (anti-GBM) antibody disease may be associated with ANCAs. The clinical significance of anti-GBM antibodies and ANCA is uncertain.
Certain drugs are reported to induce ANCA reactivity with varying symptoms. Minocycline-associated arthritis, fever, and livedo reticularis can be ANCA positive, with anti-MPO antibodies. Propylthiouracil-induced vasculitis is also ANCA positive, with specificity to several different target antigens including PR3, MPO, and elastase. Hydralazine-associated vasculitis may be ANCA positive, with anti-MPO antibodies.
A rise in ANCA titers may be associated with relapses, but not always. Elevated ANCA titer should not be used as a sole parameter for preemptive therapy.14 Therapy should only be based on clinical or pathologic evidence of relapse. The role of sequential ANCA titers after the diagnosis is established is unclear. One study showed a weak association between disease activity and ANCA levels.
The complement system consists of plasma and membrane proteins that provide innate defense against microbial pathogens. Complement activation is usually assessed by determining the levels of individual complement components such as C3 and C4 and by quantifying the CH50 (total hemolytic complement) activity. Complement levels are measured by either functional or antigenic assays. CH50 is a useful tool for assessing all nine components of the classic pathway (C1, C2, C3, C4, C5, C6, C7, C8, and C9). CH50 is undetectable when there is complete deficiency of any individual complement component. Classic pathway activation is indicated by low levels of C3 and C4. Alternate pathway activation is indicated by low levels of C3 but normal C4.
Complement measurement is an important diagnostic tool in many connective tissue disorders. Hypocomplementemia is present in disorders associated with excessive levels of immune complexes such as SLE and cryoglobulinemia. There is a significant association between low complement levels and lupus nephropathy.15 The utility of low complement levels as predictors of lupus flares is controversial; some studies have found a clear association with lupus activity and others show no correlation. A high frequency of positive ANA and anti-dsDNA in patients with primary antiphospholipid antibody syndrome with hypocomplementemia probably suggests that these patients might develop a lupus-like illness.
Low complement levels are also seen in inherited complement deficiency. Inherited complement deficiencies of C1, C2, and C4 predispose to SLE. Complete deficiency of C3 is rare and manifests in childhood as severe recurrent infections with pyogenic organisms. Complete deficiency of C4 is rare because four genes encode the C4 protein. Partial deficiency due to the presence of one, two, or three null alleles can produce persistently low levels of C4 and predispose to SLE. Deficiency of C1 esterase inhibitor leads to unregulated C1 esterase and to depression of C4 levels.
Antiphospholipid antibodies include antibodies directed against phospholipid-associated proteins such as cardiolipin, β2-glycoprotein 1, and prothrombin. These antibodies are usually measured in patients with SLE, recurrent thrombosis, and recurrent fetal loss, raising the possibility of antiphospholipid antibody syndrome. The antiphospholipid syndrome is characterized by venous thrombolism, arterial thrombosis, or pregnancy morbidity (individually or in combination), together with antiphospholipid antibodies and lupus anticoagulant.
The anticardiolipin antibodies are measured by ELISA and usually include three serotypes: IgG, IgM, and IgA. International criteria for negative, low, medium, and high levels have been set, and standards are available for laboratory calibrations. These antibodies should be present in medium to high concentrations on at least two occasions about 12 weeks apart to establish a diagnosis of antiphospholipid antibody syndrome, along with some clinical criteria. A number of studies have shown that acute medical illness and infections can lead to a transient increase of the antibodies.