A Brief, Practical Review of Serological Tests in Rheumatic Diseases
Authors: Keith K. Colburn, MD, Chief of Rheumotology, Loma Linda University; Division of Rheumatology, Jerry L. Pettis VA Medical Center, Loma Linda, CA; Myron E. Chu, DO, Fellow in Rheumatology, Loma Linda University, Jerry L. Pettis VA Medical Center, Division of Rheumatology, Loma Linda, CA.
Editor’s NoteAfter a thorough history and physical examination, a physician normally formulates a differential diagnosis. If the patient has symptoms and signs suggestive of a connective tissue disease, which tests should the doctor order? What do the test results mean? How does one use the information provided by these tests? This article discusses how to use laboratory tests in rheumatic diseases. Put in their proper perspective, laboratory tests help rule out, confirm, or sub-classify a disease process. Therefore, it is extremely important to have a clear knowledge of the patient’s history and an appropriate physical examination before one can interpret laboratory tests; otherwise, information from these tests is relatively useless. To make sense of test results, it is essential that the doctor know the rheumatic disease process, criteria for the diagnoses of these diseases, and what the test results mean in the context of the disease one is suspecting. In using a test to rule out or confirm a diagnosis, one must understand "sensitivity" (the percent of positive test results in patients with a given disease), "specificity" (the percent of negative test results in patients without the disease in question), and "positive and negative predictive values" in the setting of one’s reference laboratory. The authors deal with the context of ordering clinical laboratory tests on patients with suspected connective tissue diseases.
Immunologic tests assist the clinician in the diagnosis and management of patients with rheumatic diseases. Recently the thrust of managed care is to have primary care physicians order and interpret test results that are usually obtained by subspecialists. This article will help to illustrate the clinical application and proper interpretation of the more commonly ordered rheumatologic laboratory tests.
Before a test is ordered, the physician must have in mind a course of action to follow in response to the outcome of the test. In general, immunologic tests are usually obtained for disease screening, help in confirming a diagnosis based on the history and clinical findings, assessing specific organ involvement, monitoring disease activity, and/or monitoring drug toxicity.1 It should be remembered that if a certain test does not have the potential to significantly contribute to or alter the diagnosis or course of therapy, it might be better left undone.
The use of a test to confirm or exclude a diagnosis is determined by its sensitivity, specificity, and positive or negative predictive value.2 Sensitivity is defined as the percent of positive test results in patients with a given disease. Specificity is defined as the percent of negative test results in patients without the disease in question. Tests that are very sensitive are used to exclude a diagnosis. An extremely sensitive test is positive in almost all patients who have a certain disease; therefore, a negative test is strong evidence against the presence of that disease, but it is not necessarily diagnostic. A good example of a highly sensitive test with very little disease specificity is an anti-nuclear antibody (ANA) test, which, if negative, nearly always rules out a diagnosis of systemic lupus erythematosus (SLE).
Tests with high specificity can help confirm a diagnosis. A good example of a very specific test is the anti-Smith antibody test, which, if positive, is almost always associated with the diagnosis of SLE. However, in immunologic tests, high specificity is frequently coupled with low sensitivity; therefore, a negative highly specific test does not exclude the presence of a specific disease. Anti-Smith antibodies, although very specific for SLE, are present in the sera of only 20% of SLE patients. If a test has a low specificity for a disease, a positive test result does little to confirm a diagnosis.
The positive predictive value is the probability that a patient with a positive test result actually has the disease, whereas the negative predictive value is the probability that a person with a negative test result does not have the disease. The positive or negative predictive values of a test will vary with the prevalence of the disease in the particular population tested. As the prevalence of a disease decreases in the population, it becomes less likely that the test represents a false-positive result. Prevalence is simply the fraction of the population who have the disease. Table 1 shows how sensitivity, specificity, and predictive values are calculated.
It is essential for clinicians to remember that a diagnosis of a rheumatologic disease should be made to a large extent based on a knowledgeable history, then on the physical examination, and, to a much lesser extent, on laboratory and other tests. The laboratory tests should be used to help support a clinical judgment. It is important to also know the limits of the tests that are ordered to fully appreciate their value.
The following is a discussion of the use of the laboratory for rheumatic diseases.
Systemic lupus erythematosus (SLE) is an autoimmune disease that can affect most organ systems. Myriad clinical manifestations can result, as well as a variety of autoantibodies to various components of the cell. A few of these antibodies may be pathogenic in certain features of SLE disease processes.
LE Cell Phenomenon. In 1948, Hargraves noticed that the bone marrow smears of patients with SLE had white blood cells containing peculiar intracytoplasmic inclusions.3 These white blood cells were termed LE cells and were at first considered to be diagnostic of SLE. The LE cell, usually a polymorphonuclear white blood cell, phagocytizes a homogeneously staining globular material called a hematoxylin body. The hematoxylin body consists of modified nuclear material, IgG, and complement. The LE cell phenomenon usually occurs in vitro but occasionally can be seen in vivo, as in SLE pleural, ascitic, pericardial, and synovial fluids. It was initially thought that the LE cell test was exclusively seen in SLE. However, specificity of this test is relatively low, since the LE cell phenomenon can be seen in rheumatoid arthritis, lupoid hepatitis, drug reaction, and in some cases of scleroderma and dermatomyositis. The LE cell test is positive in only 70-80% of patients with active SLE and in fewer patients with SLE in remission.2 A positive test, because of its limited sensitivity, does not confirm the presence of SLE, and, by the same token, a negative test, because of the limited sensitivity, does not exclude the diagnosis of the disease. The LE cell preparation is now considered to be outdated, insensitive, and nonspecific and is rarely ordered by knowledgeable clinicians. Interest in the LE cell preparation test seems to have taken on more of an historical nature.
Antinuclear Antibody Test. Antinuclear antibodies (ANA) belong to a large family of autoantibodies that react with antigens present in cell nuclei. The method most commonly used in the clinical laboratory for determining ANAs is the indirect immunofluorescent test using human tissue culture cells, such as human epithelial cells (Hep-2 cells), as substrate. By comparison, the ANA test done by enzyme-linked immunoassays (EIA) has an unsatisfactory sensitivity and specificity and is clinically less useful.4
The serum ANA is the most sensitive test done on patients with SLE. It is positive in 95-99% of untreated SLE patients.5 The degree of positivity of the ANA test may also be important, because the positive predictive value of the test increases with higher titers.6 Although the sensitivity of the ANA test is very high, the specificity is poor. The positive ANA can be found in patients with various rheumatic diseases, including rheumatoid arthritis (RA), scleroderma, juvenile rheumatoid arthritis (JRA), Sjögren’s syndrome, and mixed connective tissue disease (MCTD). The positive ANA may also be found in the serum of patients with hepatitis C infection, interstitial pulmonary fibrosis, pneumoconiosis, silicosis, chronic active hepatitis, malaria, leprosy, certain drugs, and neoplasms of various organs. A serum ANA is present in some healthy people, especially the elderly.
Because of its high sensitivity, the ANA is most useful when it is negative. A negative test result nearly excludes the diagnosis of SLE. Although ANA-negative SLE occurs, it is extremely rare, found in only 1% of patients with active disease.7 Some patients with SLE, however, develop a negative ANA as their disease remits, either spontaneously or from therapy.8 Disappearance of a positive ANA test occurs in approximately 10-20% of SLE cases, but this percentage can increase to 30-50% for those SLE patients who develop renal failure and are treated with dialysis.8 It should be noted that some patients with SLE who exhibit clinical remission will maintain high ANA titers. This persistent elevated ANA titer should not be used as an indicator for initiating therapy. ANA titers of less than 1:40 or 7 IU/mL (using the new international units) are considered normal.
When the ANA test is positive, the fluorescent pattern may be somewhat helpful in subclassifying the rheumatic diseases. Four types of patterns of nuclear fluorescence have been described in which the titer is also provided. No single pattern has been found to be diagnostic of any specific disease, but there have been strong associations. Additionally, all or any combination of patterns may appear simultaneously, and any one pattern may somewhat obscure the others.
The four patterns of ANA fluorescent nuclear staining are homogeneous (diffuse), speckled, peripheral (rim), and nucleolar. (See Table 2.) The most common ANA pattern is the non-specific, homogeneous pattern seen in patients with SLE as well as RA, Sjögren’s syndrome, and scleroderma. The speckled pattern is commonly seen in patients with SLE, MCTD, scleroderma, RA, liver disease, ulcerative colitis, and in some healthy people.2 The peripheral pattern is often detected in SLE patients who have active nephritis. The nucleolar pattern is less frequently seen but is more commonly associated with scleroderma and rheumatologic overlap syndromes with sclerodermatous features.7
Anti-Double Stranded DNA. If the clinician suspects a diagnosis of SLE or a disease process with clinical features of SLE in which the ANA result is positive, then further antibody tests are warranted. The ANA includes antibodies to both single- and double-stranded DNA. Anti-single stranded DNA (anti-ssDNA) antibodies are not specific for SLE because they occur in patients with different autoimmune diseases. Anti-double-stranded DNA (dsDNA) antibodies, or anti-native DNA (nDNA) antibodies, are considered to be characteristic of SLE and occur in as many as 70% of patients with active SLE and, especially, with lupus nephritis. The anti-dsDNA antibody test does not predict lupus nephritis nor other particular SLE disease manifestations, but because these antibodies often fluctuate in parallel with disease activity, test results are used as an SLE disease monitor.7 The specificity for a positive anti-dsDNA test is about 95%.
Anti-Smith Antibody. Anti-Smith antibodies (anti-Sm) were identified in 1966 and named after SLE patient Shirley Smith. Anti-Sm antibodies are 99% specific for SLE. The sensitivity, however, is approximately 20%. The anti-Sm antibody test does not correlate with any particular feature of SLE nor does it vary greatly with time.9 A word of caution is indicated here, as it is a common clerical error to confuse "anti-Sm" with "anti-smooth muscle" when ordering tests. The anti-smooth muscle antibody test is associated with autoimmune chronic active hepatitis and other liver diseases. Therefore, writing out "anti-Smith" instead of "anti-Sm" may ensure that the correct test is performed.
Anti-Ribonucleoprotein Antibody. In 1972, Sharp et al described a group of patients with overlapping clinical features of SLE, polymyositis, and scleroderma and coined the term "mixed connective tissue disease" or MCTD.10 Serum antibodies to ribonucleoprotein (anti-U1snRNP) are a feature of patients with MCTD. These U1snRNP antibodies were previously referred to as ribonucleoprotein (RNP) or nRNP (nuclear RNP) antibodies. Patients also have high serum titers of speckled ANA. Antibodies to ribosomal RNP most often account for cytoplasmic staining in immunofluorescence. These antibodies are called "extractable nuclear antigens" or ENAs, because they are easily extractable in normal saline solution. There are more than 20 different anti-ENA antibodies described, but the two major ones are anti-U1snRNP, and the anti-Sm, which has been previously discussed above. Anti-U1snRNP is a small nucleoprotein particle containing U1RNA (uridine-rich RNA). Antibodies to U1snRNP are present in 35-45% of patients with SLE and, by definition, virtually all patients with MCTD.11 When SLE is present, anti-Sm antibodies are usually accompanied by anti-U1snRNP antibodies. The anti-U1snRNP antibody test is not useful in monitoring disease activity or predicting SLE flares. High titers do not seem to be associated with worse disease.
The clinician should suspect MCTD in a patient with overlapping clinical features of SLE, scleroderma, and polymyositis with a high titer speckled ANA. If the patient’s serum contains anti-U1snRNP alone, then the diagnosis of MCTD is likely. However, in patients with these overlapping features, if the serum is negative for anti-RNP antibodies, the term "undifferentiated connective tissue disease" is used for the diagnosis.
An ENA panel is often ordered, which usually includes the anti-Sm and the anti-U1snRNP. A laboratory should have a listing of the component tests offered on each panel. However, rather than ordering a panel, it is better to order the necessary tests individually.
Anti-ribosomal P Antibody. Antibodies to ribosomal P proteins (anti-ribosomal P) are detected in 12-16% of patients with non-psychotic SLE and are quite rare in other rheumatic diseases.12 These antibodies, however, are detected in 45-90% of patients with severe depression associated with psychosis due to SLE.13 Yet, there has been great controversy regarding the clinical correlation of these antibodies with neuropsychiatric SLE. A positive test for anti-ribosomal P protein antibodies is not diagnostic of lupus psychosis because almost 50% of patients with anti-ribosomal P antibodies have no severe behavioral problems.13 If CNS lupus is suspected, the history and clinical physical examination ought to be used as the basis for the diagnosis. The clinician considering ordering the anti-ribosomal P antibody test should remember that the results obtained may not help to support or rule out CNS lupus and consequently will not alter the diagnosis nor the planned course of treatment.
Sjögren’s Antibodies: Anti-Ro/SSA and anti-La/SSB. Anti-SSA(anti-Ro) and anti-SSB(anti-La) are circulating antibodies that are frequently seen together in patients with primary Sjögren’s syndrome and subacute cutaneous lupus. However, they may occur in other connective tissue diseases with or without associated primary Sjögren’s syndrome. Anti-SSA antibodies are found in approximately 25-35% and anti-SSB antibodies in about 25% of patients with SLE.1,14 The serum of less than 1% of normal individuals has the anti-SSA antibody, and those that do have the antibody have low levels. Children born to mothers with a positive serum anti-SSA antibody test run a significant risk of having neonatal heart block, which will be discussed below under "Neonatal Lupus." There is a high frequency of subacute cutaneous lupus with the presence of circulating anti-SSA and anti-SSB antibodies. This subset of SLE patients with a photosensitive, non-scarring skin rash with either annular or psoriasiform morphology tends to have less frequent visceral involvement. Patients with the so-called ANA-negative SLE often have subacute cutaneous lupus and may test positive for anti-SSA antibodies and rheumatoid factor (RF).
Neonatal Lupus. Neonatal lupus syndrome present at birth, or shortly thereafter, is characterized by a lupus skin rash, transient or permanent congenital heart block, cytopenias, and liver inflammation. The mother of the infant has SLE or some form of systemic connective tissue disease and tests positive for the anti-SSA antibody. Because of transplacental transfer of maternal IgG, the infant tests positive for anti-SSA antibodies. The risk of having a child with the syndrome is higher if the mother has both antibodies to SSA and SSB, but the risk is lower if the mother has isolated SSA in low titers.7 The presence of anti-SSA antibodies increases the risk of giving birth to an infant with congenital heart block to about 5%.14 Since anti-SSA and anti-SSB antibodies readily cross the placenta, they can be used as markers for neonatal lupus. Careful monitoring of the fetus should be done before as well as after delivery. Laboratory tests on the infant should include a CBC and platelets to evaluate cytopenias as well as an ANA, anti-dsDNA, anti-SSA, and anti-SSB antibodies.
Anti-Histone Antibodies in Drug-Induced Lupus. A negative anti-histone antibody test is used to help rule out a clinical diagnosis of drug-induced lupus. The clinical features of drug-induced lupus are often less severe than those of SLE, of which the most commonly reported symptoms are fever, arthritis, and serositis.15 CNS and renal involvement are rare in drug-induced lupus. Anti-histone antibodies occur in more than than 90% of cases. However, anti-histone antibodies are not specific for drug-induced SLE, since they are also found in 25% of patients with SLE.16 Antibodies to single-stranded DNA may be present, but antibodies to double-stranded DNA are typically absent.15 (See Table 3.)
Complement. The complement system is a group of proteins that functions as a mediator of inflammation. The protein components interact with each other and with other aspects of the immune system. Complement activation is a cascade reaction. Patients with active SLE, especially with nephritis, frequently show evidence in their serum of activitation of the complement pathway by depression of the levels of C3, C4, and CH50. Lowered serum complement levels may indicate both consumption of components and a decrease in complement synthesis as in lupus nephritis. It is not practical and cost-effective to measure most of the complement components because they are unstable. CH50 and C3 are the most useful initial measures of the complement system. The CH50 is a measure of the integrity of the entire classical complement cascade, but not the alternative pathway. C3 is the most stable complement component, and its depression indicates activity in either or both classical and alternative pathways. CH50 and C3 should be assessed together, because low levels of either may indicate an SLE flare. C4 is also a measure of the classic complement pathway and may also be low in active SLE. However, C4 by itself may not be an accurate indication of SLE activity, as it is commonly low in people who have an inherited deficiency of C4. The prevalence of complement deficiency in the general population is estimated at 0.03%, and it appears that C2 deficiency is the most common inherited deficiency in the white population.17 In rheumatologic disorders such as SLE, the incidence of C2 deficiency may be as high as 5.9%.18 SLE in C2-deficiency individuals tends to be different than the general population in that the age of onset is earlier and there is a higher incidence of males.17 Milder renal involvement is noted in C2-deficient SLE individuals as well as lower ANA titers, but the prevalence of anti-SSA antibodies is increased.17
Patients with the anti-phospholipid antibody syndrome (APS) may have recurrent vascular thrombosis, pregnancy loss, and thrombocytopenia associated with a persistantly positive lupus anticoagulant (LAC) test and/or moderat-to-high levels of anticardiolipin antibody (ACL) levels. (See Table 4.) There is a strong correlation of APS and SLE, although APS can be seen alone. Cardiolipin is a phospholipid that was isolated in 1941 from beef heart by Mary Pangborn during investigations of serologic tests for syphilis.19 Patients with APS but without SLE are often referred as having primary APS (PAPS). These individuals usually do not develop SLE.
Antiphospholipid (APL) antibodies can either be drug induced or autoimmune. Infections such as syphilis, Lyme disease, and HIV-1 (HTLV-1) can induce APL antibodies, while chlorpromazine and other drugs that can cause a drug-induced lupus can also produce APL antibodies. However, clinical APS occurs with autoimmune-, not drug-induced, APL antibodies.20 Autoimmune-induced APL antibodies are frequently present in sustained high serum titers most often of IgG isotype.20 Infection-induced APL antibody titers tend to be of IgG and IgM isotypes and of lower, transient titers. Drug-induced APL antibodies are primarily of IgM isotype.20
In the presence of the appropriate clinical manifestations, a positive serum LAC, ACL, or both help to confirm a diagnosis of APS. A positive LAC test is suggested by the following clinical findings: 1) A prolonged partial thromboplastin time (PTT); 2) failure to correct the abnormal clotting time by mixing patient serum with normal serum (suggesting the presence of a clotting inhibitor); and 3) normalization of the test with freeze-thawed platelets or phospholipids.21 Serum ACL antibody tests are usually measured with a standardized EIA for each of the isotypes. IgG results are reported in GPL units, IgM in MPL units, and IgA in APL units. IgG values above 20 GPL units are considered positive and above 40 are most specific for APS.22
In the absence of APL antibodies, the differential diagnosis for unexplained arterial or venous thrombosis or pregnancy loss includes, among other conditions, use of birth control pills by a smoker, nephrotic syndrome, vasculitis, accelerated atherosclerosis, factor V Leiden activated protein C resistance, homocystinuria, Buerger’s disease, malignancy, protein C and S deficiency, and anti-thrombin III deficiency.22
Antineutrophil Cytoplasmic Antibodies (ANCA). Wegener’s granulomatosis (WG) is an uncommon, multisystem disease of unknown etiology that occurs in young or middle-aged adults with a slight preponderance toward men and is almost always fatal if untreated. The major features of WG include a triad of necrotizing granulomatous vasculitis of the upper and lower respiratory tracts and glomerulonephritis. To confirm the diagnosis in patients with suspected WG and rule out other processes, such as systemic infections and malignancies, it is essential to have tissue evidence, preferably from an open lung biopsy.23
Antineutrophil cytoplasmic antibodies (ANCA) were first described in 1982 in the serum of patients with systemic vasculitis and glomerulonephritis and were later identified as a marker for WG.23 On immunofluorescent staining of alcohol-fixed PMNs, two patterns of ANCA can be recognized. C-ANCA is identified by a coarse, granular, centrally-accentuated pattern that decreases in intensity toward the periphery of the cell. The C-ANCA is specific for proteinase 3. Proteinase 3 is a 29-kd serine protease found in the azurophilic granules of the neutrophil and is the major antigen for C-ANCA. The other recognized ANCA pattern is P-ANCA, an artifact of ethanol fixation. Ethanol causes the rearrangement of positively charged granule constituents around and on the negatively charged nuclear membrane, resulting in an artifactual perinuclear pattern. The major antigen for P-ANCA is myeloperoxidase, a lysosomal enzyme found in neutrophils.
C-ANCA is reported in the sera of as many as 90% of patients with active WG.23 In active, limited WG, the sensitivity of C-ANCA tests is 60-67%24 and can decrease to 41-43% in patients with inactive disease.25 Thus, although the sensitivity of a positive C-ANCA is high, a negative result does not necessarily exclude the diagnosis of WG, especially in patients with limited or inactive WG. Using the C-ANCA to monitor disease activity was suggested, since C-ANCA titers often parallel WG disease activity,24 but its usefulness is still not clear.
P-ANCA can be found in the sera of 80% of patients with active pauci-immune necrotizing and crescentic glomerulonephritis.26 A positive P-ANCA can be a useful marker for vasculitis-associated cresentic glomerulonephritis as well as Churge-Strauss syndrome.27 In patients with ANCA-associated glomerulonephritis, approximately 90% of the positive P-ANCA results are secondary to anti-myeloperoxidase antibodies.27 P-ANCA tests that are positive but are myeloperoxidase-negative have poor specificity and are associated with inflammatory bowel disease, chronic active hepatitis, primary sclerosing cholangitis, and normal controls.28
WG is a rare disease, and the prevalence in the population may markedly influence the predictive value of a positive C-ANCA test. If used as a screening test, a positive C-ANCA test result may often be falsely positive.23 The C-ANCA test is most valuable when the clinical situation suggests a diagnosis of vasculitis and the clinician is considering several diagnoses, including WG.23 However, a positive C-ANCA test does not substitute for a tissue biopsy to make the diagnosis of WG, because one does not want to start a patient on a two- or three-year course of cyclophosphamide without an accurate diagnosis.
Scleroderma or systemic sclerosis (SSc) is a connective tissue disease in which there is fibrosis of the skin and visceral organs. The disease is rare in childhood and peaks at 50-60 years of age. There is a preponderance for females, although the female-to-male ratio varies with age, the highest being 15:1 during childbearing years.29 Little is known about the etiology of SSc. Isolated cases of SSc may be secondary to exposure to vinyl-chloride, benzene, or the chemotherapeutic agent, bleomycin. Diffuse scleroderma includes the distal and proximal extremities, the internal organs, the face, and the trunk of the body. CREST syndrome (calcinosis, Raynaud’s, esophageal dysmotility, sclerodactyly, telangiectasis) is a limited form of scleroderma involving the distal extremities and the face. The ANA is positive in 70-90% of individuals with SSc and 60-90% of those with CREST.
Scl-70 antibodies or anti-DNA-topoisomerase I are almost exclusively seen in SSc, although only 20-40% of patients with SSc have a positive test. Anti-centromere antibodies are present in 60-90% of patients with CREST, but only 10-15% of patients with SSc. Anti-centromere antibodies are occasionally detected in patients with Raynaud’s syndrome.30
Polymyositis and dermatomyositis are idiopathic inflammatory myopathies that are characterized by proximal limb weakness, neck flexor weakness, and, occasionally, muscle pain. The laboratory hallmarks of these muscle diseases are elevated serum creatine kinase (CK) and aldolase levels. The LDH and transaminases may also be elevated. There are reported cases of patients with normal CK and serum aldolase levels but with muscle biopsies positive for the presence of an inflammatory myopathy. An electromyelogram (EMG) is a sensitive test for evaluating an inflammatory myopathy. Over 90% of patients with polymyositis/dermatomyositis will have the typical findings of irritability of myofibrils (fibrillation potentials) on needle insertion at rest and short duration, low amplitude, complex (polyphasic) potentials on contraction. However, there is little correlation between the amount of weakness or functional disability and the EMG findings.31 Reviewing medications of a patient with suspected myopathy may also be rewarding, because several drugs have been associated with a myopathy, including corticosteroids, zidovudine, and cimetidine.31 (See Table 5.)
Antibodies to the Jo-1 antigen (histydyl-tRNA synthetase) are present in the sera of 18-31% of adult patients with polymyositis and rarely in dermatomyositis.30 Anti-Jo-1 antibodies also appear to be a marker for interstitial lung disease in polymyositis.
The anti-Mi-2 antibody is an antinuclear antibody found in the sera of only 5-10% of myositis patients, but it is strongly associated with the rash of dermatomyositis, especially in juvenile dermatomyositis.31 The anti-PM-Scl (originally anti-PM-1) antibody is only occasionally detected in myositis patients and identifies a small subset of individuals with features of systemic sclerosis with myositis. See Table 6 for a brief differential diagnosis of muscle weakness.
Sjögren’s syndrome (sicca syndrome) is a chronic, slowly progressive disease that primarily affects the exocrine glands by lymphocytic infiltration, destroying and replacing functional epithelium. Patients present with characteristic symptoms of dry eyes (xerophthalmia), dry mouth (xerostomia), and, often, parotid gland swelling and vaginal dryness. The Schirmer’s tear test and Rose Bengal staining of the corneal epithelium as well as sialometry and sialography can be used to help make the diagnosis of Sjögren’s syndrome. A minor salivary gland biopsy may also be useful.
Two antibodies, anti-SSA and anti-SSB, are common in the sera of patients with Sjögren’s syndrome. The antibodies are not specific for Sjögren’s syndrome, and, as previously mentioned, may be found in other autoimmune diseases, such as SLE. Anti-SSA(Ro) antibodies are present in approximately 40-45% of patients with Sjögren’s syndrome and 25-35% with SLE.14,32 Anti-SSB(La) antibodies are detected in approximately 50% of patients with Sjögren’s syndrome and in up to 25% with SLE.32
Rheumatoid Factor. In the 1940s, Waaler and Rose and colleagues discovered that the majority of sera of patients with rheumatoid arthritis (RA) agglutinated sheep red blood cells sensitized with rabbit anti-erythrocyte antibodies.33,34 They referred to these antibodies as rheumatoid factors (RF). RFs are autoantibodies directed against antigenic determinants on the Fc fragment of IgG. The development of radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (EIA) methods have facilitated the more precise quantification of IgM, IgG, and IgA rheumatoid factors. The most commonly measured RF is IgM antibody to IgG by latex agglutination or, more recently, by nephelometry. IgM RFs are multivalent and are efficient agglutinators of antigen-coated particles. Studies indicate that RFs are present in the serum of 75-80% of patients with RA and at some time during the disease course. Diseases other than RA in which positive RF test results are frequently found include rheumatic diseases (Sjögren’s syndrome, SLE, cryoglobulinemia, etc.), viral infections, chronic inflammatory diseases, and neoplasms after chemotherapy or radiotherapy.
RF is found in the sera of 5% of healthy persons in the general population and increases with age to 10-25% of individuals over 65 years old.35 Therefore, the positive predictive value remains poor at about 6% and must be interpreted in light of other clinical features. RF tests can be performed by nephelometry or agglutination. There is some variation in the sensitivity of RF detection depending on which technique is used. Agglutination using latex particles or bentonite is more sensitive but less specific than tests that employ sheep red cells.36 The sensitivity of RF for RA patients by nephelometry is between 61-75%.37 See Table 7 for a more complete list of diseases associated with RF.
The pathologic role of RF in RA remains unclear. The RF titer does not correlate with disease activity, but it is generally considered that high titer RFs are associated with more severe disease, including the presence of rheumatoid nodules and extra-articular features. The RF test sub-classifies RA patients into seropositive and seronegative disease. Patients with "RA" but negative for RF need to be watched more closely for the development of a different diagnosis, such as psoriatic arthritis or one of the spondyloarthropathies.
Erythrocyte Sedimentation Rate (ESR). The erythrocyte sedimentation rate (ESR) is a general indicator of inflammation and is a measure of the speed at which red blood cells (RBCs) settle in uncoagulated blood. The ESR is considered to be a sensitive test for chronic inflammation, but it is not very specific. Elevated ESR values can be seen in conditions such as infection, malignancy, oral contraceptive use, pregnancy, hyperthyroidism, hypothyroidism, etc.
The mechanism of the ESR has to do with the reaction of proteins in the presence of inflammation. With the presence of acute phase reactants (the most abundant being fibrinogen), the RBCs tend to aggregate in a stacked coin fashion, which is referred to as "rouleau" formation. Other serum proteins, such as alpha- and gamma-globulins, also influence the ESR but to a lesser extent. The higher the intensity of inflammation with more fibrinogen production and increased rouleau formation, the higher the ESR.
There are two common methods of measuring ESR: Wintrobe and Westergren. The Wintrobe method is limited by a short tube (100 mm) and, consequently, limits the high readings obtained with the Westergren tube (200 mm) in some of the more serious rheumatic diseases. The ESR is a measurement in millimeters (mm) of the distance the RBCs fall in a column of uncoagulated blood in one hour. A Westergren ESR of less than 10 mm in males and 20 mm in females is considered normal. An increased ESR is often seen with aging. Anemia has a profound effect on the Wintrobe and Westergren ESR. The blood of anemic patients will markedly increase the ESR because there are fewer RBCs, leaving relatively more space for the rouleau to rapidly settle to the bottom. A rule of thumb for correcting the increased ESR due to anemia is for every 1% Hct below normal, substract 2 mm from the reported ESR. A less frequently used test, the zeta sedimentation rate, does not need to be corrected for anemia.
The ESR is often helpful in following the course of an inflammatory process and may help in distinguishing inflammatory RA from tendonitis, bursitis, and conditions like non-inflammatory osteoarthritis. The ESR is a cornerstone in the diagnosis of polymyalgia rheumatica and temporal arteritis in which the ESR is often very high. However, the ESR is of limited value in diagnosing joint symptoms. A careful history and joint examination for synovitis is far more important than an ESR value. Additionally, changes in ESR values do not always correlate with clinical disease activity in RA. The ESR should not be used to screen asymptomatic persons for diseases, as fewer than six in 10,000 persons in the general population will benefit from the test even after a history and physical examination is done.38 Table 8 shows factors that influence the ESR.
The CBC is important for monitoring patients with autoimmune diseases and their treatment of those diseases. These patients often have a normochromic, normocytic anemia (anemia of chronic disease). Patients with arthritis often take aspirin and various NSAIDs and may be prone to blood loss from the GI tract. A Coomb’s positive hemolytic anemia should be suspected in an SLE patient with a low hemoglobin and hematocrit. Patients taking cytotoxic medications for diseases such as active SLE, RA, or vasculitis need to be screened with a CBC because they may develop bone marrow suppression. Active SLE or Felty’s syndrome may cause a very low WBC, whereas treatment of a disorder with corticosteroids or infection may increase the WBC count. The eosinophil count in the differential component of the CBC is useful in certain diseases such as Churg-Strauss, Wegener’s granulomatosis, and eosinophilic fasciitis.
A low platelet count may be seen in active SLE, APS, idiopathic thrombocytopenic purpura, or heparin therapy. Platelets are acute phase reactants and may be increased in RA and other chronic connective tissue diseases.
With advancing technology, the chemistry profiles often replace the selection of appropriate individual tests; however, specific tests in these panels are often needed to monitor features of certain diseases and toxicities of treatment. Regular liver and kidney function tests are mandatory for patients with rheumatic diseases who are being treated with cytotoxic agents such as cytoplasphomide, azathioprine, and methotrexate. The blood urea nitrogen and creatinine (BUN and Cr) are important for detecting and monitoring renal involvement and drug toxicity in diseases such as SLE, scleroderma, and uric acid nephropathy in gout patients. Periodic serum Cr tests are mandatory for patients taking NSAIDs and cytotoxic drugs like cyclosporin. Serum uric acid levels are important in monitoring allopurinol dosage in gout, as the goal of long-term therapy is to achieve and maintain a serum uric acid level of 4.0-4.5 mg/dL. An elevated serum calcium level may signal possible hyperparathyroidism as a cause of pseudogout or of osteoporosis, which can be managed by treating the underlying cause.
The basic urinalysis is very helpful for monitoring disease activity in many rheumatic diseases. A 24-hour urine collection will quantitate nephrotic range proteinuria in active SLE. A routine UA may help in monitoring the safety of NSAID use and gold therapy. Proteinuria of less than 150 mg in a 24-hour period is considered normal. Less than 500 mg of protein over 24 hours in patients with SLE is often tolerated but must be followed closely. A higher degree of proteinuria may indicate the need for further investigation and treatment. Red and white blood cells and/or casts in the urine may suggest glomerulonephritis, as seen in patients with SLE and vasculitis.
Synovial fluid analysis can be very helpful in establishing a diagnosis of gout, pseudogout, or infection. Immediate synovial fluid analysis is mandatory in active monoarticular arthritis because the differential diagnosis includes diseases with potentially disastrous outcomes (i.e., infection) if undetected and treatment is delayed.
Laboratory tests assist the clinician in confirming or excluding and classifying diagnostic impressions based on a thorough history and examination of a patient. The immunologic laboratory tests are helpful in the connective tissue diseases only if the clinician is aware of the limitations of each test, including their sensitivity and specificity and the reliability of the laboratory that is performing the tests. If the clinician does not take all available information into account, the laboratory test results obtained could be misleading. In today’s environment of cost containment and managed care and, more importantly, to practice good medicine, a clinician should know the question that the test ordered is supposed to answer regarding confirming, excluding, or classifying diagnostic impressions.
1. Moder KG. Use and interpretation of rheumatologic tests: A guide for clinicians. Mayo Clin Proc 1996;71:391-396.
2. Wong LG, Colburn KK. Laboratory tests in rheumatology. In: Spiegel TM, ed. Practical Rheumatology. New York: Wiley and Sons; 1983:143-168.
3. Hargraves MM, et al. Presentation of two bone marrow elements: The "tart cell" and the "L.E." cell. Proc Staff Meet Mayo Clin 1948;23:25-28.
4. Peter JB, Reyes HR. Use and interpretation of tests in rheumatology, First edition. Santa Monica, CA: Specialty Laboratories, Inc.; 1996:11-12.
5. Wernick R. Avoiding laboratory test misinterpretations in geriatric rheumatology. Geriatrics 1989;44:61-80.
6. Mills JA. Systemic lupus erythematosus: Review article. N Engl J Med 1994;330:1871-1879.
7. von Mühlen CA, Tan EM. Autoantibodies in the diagnosis of systemic rheumatic diseases. Semin Arth Rheum 1995;24:323-358.
8. Reichlin M, Harley JB. Antinuclear antibodies: An overview. In: Wallace DJ, Hahn BH, eds. Dubois’ Lupus Erythematosus. Philadelphia: Williams & Wilkins; 1997:397-405.
9. Peter JB. Use and Interpretation of Tests in Clinical Immunology. Eighth edition. Santa Monica, CA: Specialty Laboratories, Inc.; 1991:224-225.
10. Shart GC, et al. Mixed connective tissue disease. An apparently distinct rheumatic disease syndrome associated with a specific antibody to an extractable nuclear antigen (ENA). Am J Med 1972;52:148-159.
11. Prockop DJ, et al. Collagen in normal and diseased connective tissue. In: McCarty DJ, Koopman WJ, eds. Arthritis and Allied Conditions, Volume I . Philadelphia: Lea and Febiger; 1993:213-262.
12. Isshi K, Hirohata S. Association of anti-ribosomal P protein antibodies with neuropsychiatric systemic lupus erythematosus. Arthritis Rheum 1996;39:1483-1490.
13. Peter JB. Use and interpretation of tests in clinical immunology. Eighth Edition. Santa Monica, California: Specialty Laboratories, Inc., 1991:214.
14. Peter JB. Use and interpretation of tests in clinical immunology. Eighth Edition. Santa Monica, CA: Specialty Laboratories, Inc., 1991:230-231.
15. Gladman DD, Urowitz MB. Systemic lupus erythematosus. In: Primer on the rheumatic diseases, 10th edition. Atlanta: Arthritis Foundation; 1993:106-111.
16. Hannaman RA. Rheumatology. In: MedStudy a Condensed Review of Internal Medicine. 6th Edition. Woodland Park, CO: MedStudy Corp.; 1996:7.
17. Frank MM. Complement in disease: Inherited and acquired complement deficiencies. In: Santer’s Immunologic Diseases, 5th edition, Volume I. New York: Little, Brown & Co.;1995:484-500.
18. Atkinson JP. Complement deficiency: Predisposing factor to autoimmune syndromes. Clin Exp Rheumatol 1989;7:95.
19. Roubey RA. Antiphospholipid antibody syndrome. In: McCarty DJ, Koopman WJ, eds. Arthritis and Allied Conditions, Volume I. Baltimore: Williams & Wilkins; 1997:1393-1406.
20. Lockshin MD. Antiphospholipid antibody syndrome. Rheumatic Dis Clin North America 1994;20:45-59.
21. Harris EN. Antiphospholipid syndrome. In: Primer on the rheumatic diseases. 10th Edition. Atlanta: Arthritis Foundation; 1993:116-117.
22. Harris EN. Diagnosis & management of antiphospholipid syndrome. Hospital Practice 1994;29:65-76.
23. Rao JK, et al. The role of antineutrophil cytoplasmic antibody (C-ANCA) testing in the diagnosis of Wegener’s granulomatosis: A literature review and metaa-analysis. Ann Intern Med 1995;123:925-932.
24. Nolle B, et al. Anticytoplasmic autoantibodies: Their immunodiagnostic value in Wegener’s granulomatosis. Ann Intern Med 1984;111:28-110.
25. St. Claire EW. Vasculitis. In: Weisman MH, Weinblatt ME, eds. Treatment of the Rheumatic diseases: Companion to the Textbook of Rheumatology. Philadelphia: WB Saunders, 1995:137-159.
26. Falk RJ, Jennette SC. Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Engl J Med 1988;318:1651-1657.
27. Jennette JC, Falk RJ. Anti-neutrophil cytoplasmic autoantibody-associated glomerulonephritis and vasculitis. Am J Pathol 1989;135:921-930.
28. Hannaman RA. Rheumatology. In: MedStudy a Condensed Review of Internal Medicine. 6th Edition. Woodland Park, CO: MedStudy Corp.; 1996:13-14.
29. LeRoy EC, Silver RM. Systemic sclerosis and related syndromes. In: Primer on the rheumatic diseases, 10th edition. Atlanta: Arthritis Foundation; 1993:118-120.
30. Metzger AL, Morris RI. Rheumatology Diagnostics Laboratory, Inc. Reference Cards. Santa Monica, CA.
31. Medsger TA, Oddis CV. Inflammatory muscle disease. In: Klippel JH, Dieppe PA, eds. Rheumatology. St. Louis: Mosby; 1994:6.12.1-6.12.14.
32. Moutsopoulos HM, Tzioufas AG. Sjögren’s syndrome. In: Klippel JH, Dieppe PA, eds. Rheumatology. St. Louis: Mosby; 1994:6.27.1-6-27-13.
33. Waaler E. On the occurrence of a factor in human serum activating the specific agglutination of sheep blood corpuscles. Acta Path Microbiol Scand 1940;17:172-188.
34. Rose HM, et al. Differential agglutination of normal and sensitized sheep erythrocytes by sera of patients with rheumatoid arthritis. Proc Soc Exper Biol Med 1948;68:1-6.
35. Lipsky PE. Rheumatoid arthritis. In: Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, Kasper DL. eds. Harrison’s Principles of Internal Medicine. San Francisco: McGraw Hill; 1994:1648-1655.
36. Wigley FM. Rheumatoid Arthritis. In: Barker LR, Burton JR, Zieve PD, eds. Principles of Ambulatory Medicine, 4th edition. Philadelphia: Williams & Wilkins; 1995:943-974.
37. Leavelle DE. Mayo Medical Laboratories Interpretive Handbook. Mayo Medical Laboratories, Rochester, MN; 1988:133-134.
38. Sox HC, Liang MH. The erythrocyte sedimentation rate: Guidelines for rational use. Ann Intern Med 1986;104:515-523.
Phsyician CME Questions
40. The ANA test is most useful to:
a. rule out SLE.
b. monitor the disease activity of SLE.
c. diagnose SLE.
d. differentiate between SLE and RA.
e. subclassify SLE.
41. The best test to confirm Wegener’s granulomatosis is:
c. Westergren ESR.
d. open lung biopsy.
e. nasal passage biopsy.
42. Which one of the following tests is least specific for a connective tissue disease?
d. Anti-Smith (Sm)
43. Which one of the following tests is most sensitive for a rheumatic disease?
a. Rhematoid factor
b. Anti-cardiolipin Abs
c. Westerngren ESR
e. C-reactive protein (CRP)