Laboratory testing in the rheumatic diseases: a practical review.

Abstract: Laboratory testing for the rheumatic diseases can allow for rapid diagnosis and appropriate management, while false-positive tests can lead to inappropriate management and unnecessary concern for the patient. An evaluation of laboratory testing for rheumatic illnesses is discussed, including the well-known acute phase proteins, the use of ANA in screening, and the newer antibodies which may potentially allow for an earlier diagnosis. A thorough history and examination are arguably the best screening tests. Clinicians should be judicious in their use of laboratory testing, and should only do so in an attempt to further refine the diagnosis.

**********

Laboratory testing in the rheumatic illnesses can, for the practicing clinician, result in perplexing data. Carefully used, laboratory tests can be very specific and allow rapid diagnosis and appropriate management. False-positive testing may result in inappropriate management and unnecessary concern. In this review, we attempt to circumvent this type of problem. We review current thoughts on well-known acute phase proteins to include the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). We discuss the use of the antinuclear antibody (ANA) in screening and specific subserologies that often may lead to specific and accurate diagnoses. Furthermore, we reflect on newer antibodies (to include anti-cyclic citrullinated peptide [CCP] antibodies), which have the potential of allowing earlier diagnosis and possibly to promote more aggressive management for seemingly more destructive disease. However, it should strongly be argued that the best screening tests always remain the standard tools available to the clinician: namely, an insightful history and an appropriate examination. Testing should be performed to further refine the differential diagnosis and not simply to placate the anxious patient's desire for early diagnosis. Both referring physicians and patients need to understand that while an early, correct diagnosis is sought, this is not always possible in the nonspecific and vague early symptoms so typical of many of the rheumatic diseases. The pressure to provide early diagnosis often leads to indiscriminate and cavalier ordering of these tests.

We will start by considering the more traditional and cheaper "mainstays" of rheumatic tests: the ESR and CRP, which form part of the acute phase response. These remain a cornerstone of early screening and can prove useful in monitoring disease activity.

Acute Phase Proteins

The acute phase response is defined as the pathophysiologic activity accompanying inflammation. Proteins whose plasma concentrations change by 25% during inflammatory states are defined as acute phase proteins. There are a multitude of such proteins, and the change in concentration can either increase (as seen in ceruloplasmin, complement, CRP, ESR, amyloid A, fibrinogen, alpha-1 antitrypsin, haptoglobin, ferritin) or decrease (as seen in albumin, transferrin, transthyretin). Many of the proteins are synthesized in the liver, driven by cytokine levels such as interleukin (IL)-6. The clinical combined effects of the acute phase response consist of a spectrum from fever, increased cortisol release, and fatigue, to anemia, cachexia, amyloidosis, impaired growth, and septic shock. Two of the most widely used acute phase laboratory tests by physicians are the ESR and CRP.

Erythrocyte sedimentation rate

The ESR increases during an acute phase response. The ESR is a measure of the distance that erythrocytes fall through plasma in either a Westergren or Wintrobe tube over a period of 1 hour (measured as mm/h). This is, in large part, influenced by plasma proteins surrounding them, such as fibrinogen, which increase during an acute phase response. An increased fibrinogen level has the effect of dissipating erythrocyte repulsive forces, allowing closer aggregation (ie, rouleaux formation), thus causing erythrocytes to fall a longer distance during a set amount of time. The Wintrobe method uses a shorter tube and cannot quantify values over 50 to 60 mm/h.

Elevations in the ESR can be seen in inflammatory conditions such as infection and in a large number of connective tissue diseases such as systemic lupus erythematosus (SLE), vasculitis, polymyalgia rheumatica, and rheumatoid arthritis. However, the ESR is a very nonspecific marker. Since the ESR is an indirect measurement of acute phase proteins, non-inflammatory conditions (such as change in the morphology and concentration of erythrocytes) or a change in activity or concentrations of other plasma proteins (as in multiple myeloma) can elevate the ESR. The Westergren method uses dilution to attempt to correct for the effects of anemia. The change in ESR often lags behind an abrupt change in inflammatory conditions, sometimes decreasing by only 50% 1 week after resolution of an inflammatory event. ESR also increases with factors such as obesity, age, and female sex. Because of the increase in ESR with age, it is important to adjust for age when interpreting the normal range for the ESR. A general rule of thumb for a normal ESR value in mm/h is as follows: men = age/2; women = (age + 10)/2. (1)

The experienced clinician is obviously aware of the conundrum associated with using the ESR to screen for inflammatory arthritis or vasculitis. It is not unusual to see a very active rheumatoid synovitis in the setting of a patient with a normal ESR. On the other hand, an elevated ESR does not necessarily imply an underlying disease process. A common clinical mistake is to ignore the patient's age. Elderly patients with diffuse musculoskeletal complaints are often screened with an ESR and are inaccurately diagnosed with polymyalgia rheumatica. Patients with polymyalgia rheumatica tend to be white, closer to the age of 60 years, and present with nonspecific complaints to include arthralgias of the shoulder and pelvic girdle. The ESR tends to be dramatically elevated. Intermediate values for the ESR need to be interpreted with caution. Implicit in the diagnosis of polymyalgia rheumatica is a fairly dramatic response to glucocorticoid therapy, usually within a 24- to 48-hour period. Therefore, there should be some hesitation in diagnosing a 75-year-old woman with an ESR of 52 with polymyalgia rheumatica, whose complaints do not particularly fit the above-mentioned description and whose response to glucocorticoid therapy is intermediate, at best.

C-reactive protein

C-reactive protein is synthesized in the liver and increases in the acute phase response. The role of CRP itself is complex and not fully understood. On one hand, it seems to have pro-inflammatory effects, such as activating the classic complement pathway and decreasing IL-6 receptor concentrations in areas where IL-6 has anti-inflammatory effects. (2) On the other hand, it also has anti-inflammatory properties, such as reducing neutrophil adhesion to endothelium. (3) The general rule of thumb for interpreting CRP levels in mg/dL is as follows: <0.2 mg/dL = normal; 0.2 mg/dL - 1.0 mg/dL = indeterminate; and >1 mg/dL = inflammatory. A more accurate rule for calculating normal CRP levels in patients age 25 to 70 is as follows: women = (age/65) + 0.7 mg/dL; men (age/65) + 0.1 mg/dL. (4) In general, values 0.2 to 1.0 mg/dL are nonspecific and can be seen in instances such as cigarette smoking and diabetes mellitus. Elevated levels are caused by the same inflammatory conditions listed with the ESR; however, the rise and fall of the CRP is much more abrupt than the ESR. Early elevation is seen within 4 hours of an event, peaking at 24 to 72 hours. Infection is common with high levels of CRP, with up to 85% incidence of bacterial infection in patients with a CRP greater than 10 mg/dL. (5)

Acute phase proteins such as ESR and CRP are nonspecific and in many cases, nonsensitive. They can be useful to indicate the presence and severity of an inflammatory condition, which in the case of a connective tissue disease may prompt therapy. In some cases, these values may be useful for monitoring therapy to an extent. However, it should be strongly argued that the patient is often an excellent judge of response to therapy, and the dose of glucocorticoids in the setting of polymyalgia should be titrated against the patient's perceived clinical response rather than the acute phase proteins themselves. It should be remembered that normal values are often seen in inflammatory conditions, and never rule out the possibility of connective tissue diseases. In one study, up to 10% of patients with moderate rheumatoid arthritis had a normal ESR. (6) Active SLE is also often seen in the setting of normal ESR and CRP, and it has been suggested that up to 20% of patients with polymyalgia rheumatica can have a normal ESR, though this very concept itself is quite contentious. Other conditions, such as malignancy, are seen with elevated acute phase reactants. There is emerging evidence of the association of CRP with cardiac events, and its clinical usefulness is being determined and debated.

Antinuclear antibodies

Antinuclear antibodies are present in a wide array of autoimmune diseases. When used correctly, they provide valuable diagnostic and prognostic information; however, it must be kept in mind that their presence or absence alone can neither rule in nor rule out any particular disease. The laboratory method of performing ANA testing has varied over the years, from the use of the lupus erythematosus (LE) cell test in the 1940s to the current use of immunofluorescence. Along with the change in laboratory methods, the performance characteristics of the ANA have also changed. While the occurrence of "ANA-negative lupus" has declined significantly with more sensitive laboratory methods, the presence of ANA in healthy patients has risen. It is therefore important to understand the predictive value of the ANA in different autoimmune diseases.

The initial LE test of the 1940s was performed by incubating a bare nucleus with a patient's serum. A polymorphonuclear leukocyte was then added, and if sufficient antibodies in the patient's serum had bound to the nucleus, opsonization occurred. The LE cell is a polymorphonuclear leukocyte containing a phagocytosed nucleus. The main method used now is immunofluorescence. Hep-2 cells (derived from a human epithelial tumor line) are incubated with a patient's serum. After washing, a fluoresceinated antibody is then added, which binds to the patient's antibodies bound to the nucleus. Nuclear fluorescence can be seen through a fluorescence microscope, along with a staining pattern.

A precise characterization of the performance of the ANA is difficult to define. Varying laboratory methods, as well as studies in differing patient subsets, have yielded a wide range of sensitivity and specificity. Solomon et al (7) critically reviewed available published data on ANA to calculate a weighted average of performance characteristics. These results, listed in Table 1, provide the sensitivity, specificity, and likelihood ratios (LR) of the ANA in different disease states. Most notable is the high sensitivity in SLE, in which a negative ANA makes the probability of SLE unlikely (negative LR = 0.11). However, a positive ANA result is less helpful, with a positive predictive value of only 11% (positive LR = 2.2). Given the low specificity, an ANA should only be sent in patients suspected of SLE and should be interpreted in the context of the clinical presentation. One reason for such poor specificity for SLE is that the ANA is commonly seen in other disease states such as other connective tissue diseases and chronic infections (ie, hepatitis C), can be associated with multiple medications, and can be seen in healthy patients. The ANA is found in 5 to 10% of patients without evidence of connective tissue disease, increasing in prevalence with age and in patients of relatives with rheumatic disease. The ANA is also seen in many other autoimmune disease states not associated with a connective tissue disease, such as autoimmune hepatitis, primary autoimmune cholangitis, primary biliary cirrhosis, primary pulmonary hypertension, Graves disease, and Hashimoto thyroiditis.

The low sensitivity and specificity of the ANA in other connective tissue diseases, shown in Table 1, makes the ANA a weak test for ruling in or ruling out polymyositis, dermatomyositis, rheumatoid arthritis, and Sjogren syndrome. Although not useful for ruling in scleroderma, it can be helpful to rule it out, with a sensitivity of 85% (negative LR = 0.27). Of particular note is the marginal likelihood ratio when used to distinguish primary from secondary Raynaud, in which the ANA is not very helpful. In the instance of a patient with Raynaud, an ANA should only be sent if a diagnosis such as scleroderma is suspected, but not as a screening tool.

The staining pattern of an ANA is important and is determined by the target antigen. Table 2 lists the different patterns and their association with different disease states. In general, ANA patterns are not sensitive or specific and have been supplanted by other tests discussed in this paper. Patterns can also be unreliable, as pattern recognition is somewhat operator dependent and can also change at different dilutions.

ANA titers can be important in the interpretation of the ANA test. In one study of 125 individuals with a positive ANA but no other evidence of a connective tissue disease, titers of greater than 1:40 were seen in 32%, greater than 1:80 were seen in 13%, and greater than 1:320 were only seen in 3% of patients. (8) All patients in this study had a negative dsDNA. Although useful in making a diagnosis, the ANA titer is not useful to predict disease activity. It is a common mistake for patients to have ANAs repeated again and again, as if they will become negative with treatment or as if titers will decline. Patients themselves often become somewhat entrapped in "how high is my ANA titer today?" It should always be remembered that a particular laboratory may incur an error. If the diagnosis of SLE is strongly suspected, and an ANA test is negative, we would argue that this test be repeated either at the same laboratory or at some other laboratory. On the other hand, a low-positive titer is simply that and nothing else. It is somewhat inappropriate to screen a 57-year-old woman with arthralgias with an ANA test who simply has generalized osteoarthritis. There should be a high index of suspicion of a connective tissue disease process before using this laboratory test as a screening predictor. Specifically, if the patient is a young woman of childbearing potential who has a rash and joint pain, the ANA test would be quite appropriate. If positive, specific subserologies can then be ordered.

Types of ANAs

Laboratory tests can also assess the presence of antibodies to specific nuclear proteins, which will be discussed below. While specificity for most of these tests is high, sensitivity markedly decreases. In contrast to the ANA, which has a high sensitivity and can be useful for ruling out (but not ruling in) SLE, the tests listed below in Table 3 have a high specificity and are most useful for ruling in (but not ruling out) certain connective tissue diseases.

Anti-ssDNA and anti-dsDNA antibodies

Anti-DNA antibodies are most commonly associated with SLE. Antibodies to single stranded DNA (anti-ssDNA) are very nonspecific, clinically not very useful, and therefore will not be discussed. Antibodies to double stranded DNA (anti-dsDNA), however, are of particular interest. Kavanaugh and Solomon (9) critically reviewed available published data on anti-dsDNA to calculate a weighted average of performance characteristics. Based on this review of anti-dsDNA data measured by three different laboratory methods, a sensitivity of 57.3%, and specificity of 97.4% was calculated (shown in Table 3). The presence of anti-dsDNA is highly suggestive of SLE (positive LR = 16.3); however, its absence does not significantly decrease post test probability of disease (negative LR = 0.49). Anti-dsDNA has been associated with more severe disease, including renal involvement. In general, it is debated whether the levels correlate with disease activity, but it does appear that at least in some patients this is the case. (9-12) Laboratory tests can either be done by radioimmunoassay (Farr method) or indirect immunofluorescence (Crithidia Lucilia). Despite the high specificity for SLE, it should be remembered that although exceedingly rare, anti-dsDNA has been found in other rheumatologic ailments. Occasionally, patients receiving minocycline, etanercept, infliximab, and penicillamine have also been found to have positive serologies.

Anti-Sm, anti-U1-RNP, anti-histone antibodies

The anti-Smith (anti-Sm) antibody binds to a series of nonhistone proteins complexed with small nuclear RNAs, the complex being referred to as small nuclear ribonucleoprotein particles/proteins (snRNP). These snRNPs are involved in the machinery of messenger RNA splicing. Anti-Sm Ab is highly specific and when present indicates SLE. However, it is only seen in roughly 25 to 30% of SLE, and its absence is not helpful in ruling out disease. Whereas anti-dsDNA antibodies can become undetectable with quiescent disease, anti-Sm may remain elevated. Anti-U1-RNP Ab is another type of snRNP antibody; however, it is associated with mixed connective tissue diseases consisting of a combination of SLE, scleroderma, and polymyositis. As the definition of mixed connective tissue disease involves anti-U1-RNP Ab, performance characteristics cannot be calculated. Anti-histone antibodies are found in 95% of drug-induced lupus. However, the presence of anti-histone antibodies does not verify drug-induced lupus; as most patients with the presence of these antibodies do not go on to develop symptoms. For example, anti-histone Abs occur in 80% of patients taking procainamide, but most will not have development of drug-induced lupus.

Anti-Ro/SS-A and anti-La/SS-B antibodies

Anti-Ro/SS-A and anti-La/SS-B antibodies recognize cellular proteins complexed with small nuclear RNAs, and are found mainly in association with Sjogren syndrome and SLE, although they can be found in other conditions such as scleroderma, polymyositis, mixed connective tissue disease, and rheumatoid arthritis. Anti-Ro/SS-A Abs are found in 50% of patients with SLE, being associated with photosensitivity (subacute cutaneous lupus), cutaneous vasculitis, and interstitial lung disease. They are also found in 75% of patients with primary Sjogren syndrome but in only 10 to 15% of those with secondary Sjogren. (13) These antibodies are also associated with neonatal lupus syndrome and congenital heart block. Approximately 80% of babies born with congenital heart block have the presence of anti-Ro/SS-A or anti-La/SS-B in the mother, even in cases without a connective tissue disease. However, of all mothers with the presence of anti-Ro/SS-A or anti-La/SS-B, less than 1% will be associated with neonatal lupus syndrome or congenital heart block. Anti-Ro/SS-A and anti-La/SS-B Abs have also been associated with patients who fit the criteria for SLE but have a negative ANA (ANA-negative lupus); however, as mentioned earlier, this is becoming less common. Anti-Ro/SS-A and anti-La/SS-B antibodies are also seen in 0.1 to 0.5% of healthy adults.

Anti-centromere, anti-Scl-70, anti-U3-RNP antibodies

Antibodies against topoisomerase I, or Scl-70, have been found in roughly 20% of patients with diffuse systemic sclerosis. (14) Although the sensitivity is low, specificity approaches 100%. When present, the diagnosis of scleroderma is almost certain. The presence of anti-Scl-70 Abs are also associated with an increased risk of radiographic pulmonary fibrosis, abnormal pulmonary function tests, and diffuse cutaneous involvement. When compared with controls of primary Raynaud, Scl-70 has a sensitivity of 28%, and specificity of 98% (positive LR = 10, negative LR = 0.7). (14) It is therefore useful to rule in (but not out) scleroderma in a patient with Raynaud. Anti-centromere antibodies are associated with limited cutaneous systemic sclerosis (formerly CREST). Compared with healthy control subjects, the specificity for CREST is very high (99.9%), although the sensitivity is low (65%). (14) When present, they almost certainly rule in disease, although their absence does not make disease significantly less likely. The use of anti-centromere Ab to distinguish CREST from diffuse systemic sclerosis is less useful (positive LR = 3.9, negative LR = 0.5). Anti-centromere Ab can also be useful in ruling out CREST in a patient presenting with Raynaud (negative LR = 0.2 when comparing CREST with primary Raynaud), although not as useful for ruling in scleroderma in this instance (positive LR = 3.5). Antifibrillin nucleolar (Anti-U3-RNP) antibodies are present in approximately 12% of patients with scleroderma. Although their sensitivity is low, their specificity is high (96%), and when present, they can be associated with muscle, small bowel, renal, and cardiac involvement, as well as pulmonary hypertension. (15, 16)

Rheumatoid Arthritis

Rheumatoid factor

Rheumatoid factor (RF) is commonly used in the diagnosis of rheumatoid arthritis. RF is an antibody against the Fc portion of IgG, but in clinical practice is measured as IgM. The role of RF in disease is not clear but may be related to the binding by RF on B-cells to immunoglobulins attached to antigens for antigen presentation, resulting in amplification of the humoral response. The sensitivity of RF for rheumatoid arthritis varies, depending on the patient population. RF is associated with more severe disease with greater extra-articular features. It then would make sense that sensitivity of RF in studies of more severe RA is higher than in those with mild disease. In general, the sensitivity of RF for RA is around 50 to 85%, increasing over time, as some people with rheumatoid arthritis are initially RF negative only to later seroconvert. However, RF alone cannot make a diagnosis of rheumatoid arthritis, as roughly 5% of a young healthy population can be RF positive, and this rate increases with age. Although most associated with rheumatoid arthritis, RF can also be seen in Sjogren syndrome, mixed connective tissue disease, mixed cryoglobulinemia, SLE, and polymyositis, as well as nonrheumatic disorders such as chronic infections, inflammatory disorders, and malignancy.

Only 50% of early rheumatoids demonstrate RF positivity. Furthermore, around 15% of patients with RA never have RF positivity. Therefore, this test should only be ordered where the diagnosis of RA is strongly suspected: namely, in the setting of obvious joint swelling, synovitis, or effusions. Another setting would be the finding of erosive changes radiologically. As previously suggested, screening the elderly somewhat randomly with rheumatoid factors in the setting of minimal joint findings may indeed yield false-positive results. The consequence of these false-positive investigations result in great patient fear and anxiety and lead to expensive, extensive and inappropriate investigations.

Anti-CCP antibodies

Anti-cyclic citrullinated peptide (anti-CCP) antibodies have been around for decades but have recently been gaining popularity. These were first described as markers for RA in 1964. It was because of very exacting technical requirements that anti-CCP antibodies never gained wide-spread acceptance, despite their specificity for RA. Some research suggests that citrullinated extracellular fibrin in synovium is a significant autoantigen involved in the immune response in rheumatoid arthritis. It has similar sensitivity for RA (50 to 85%) but higher specificity (90 to- 95%). (17) Its clinical usefulness is still evolving; however, these antibodies are quite obviously potentially important surrogate markers for diagnosis of RA. They may also be useful in the prognosis of this condition. They are as sensitive and seemingly more specific than IgM RF, particularly in early prognosis, but also in fully established disease. An argument has been made that their usefulness may indeed be in predicting the eventual evolution into RA, particularly when found in the setting of an undifferentiated inflammatory arthropathy, which is not an unusual presentation that faces a clinician. These antibodies have been detected in healthy individuals before the onset of clinical RA and seem to also occur in the setting of erosive disease.

Other Antibodies

Antineutrophil cytoplasmic antibodies

Antineutrophil cytoplasmic antibodies (ANCA) are directed against several neutrophil cytoplasmic components. They are often used in the evaluation of vasculitis, specifically Wegener granulomatosis, microscopic polyarteritis, and Churg-Strauss syndrome. The two main target antigens are proteinase 3 and myeloperoxidase, which are located in the azurophilic granules of neutrophils and the peroxidase-positive lysosomes of monocytes. The two basic staining patterns are cytoplasmic (c-ANCA) and perinuclear (p-ANCA). ANCA-associated conditions can be either c-ANCA or p-ANCA; however, some diseases have a predilection for one pattern. The c-ANCA pattern is highly sensitive for Wegner granulomatosis, seen in more than 90% of active disease. The antigen involved is proteinase 3, and an ELISA assay for this can be done individually. Even though sensitive, the specificity of a c-ANCA is low, in one series roughly 50%. (18) Using ELISA for proteinase 3 gives a higher specificity. The p-ANCA pattern is directed against several proteins including myeloperoxidase and elastase, and is much less sensitive and specific. It is most commonly associated with microscopic polyarteritis (70%) (19) but can also be seen in pauci-immune glomerulonephritis without extrarenal disease (80%) (20) and Wegener. Similarly to proteinase 3, using the ELISA test for myeloperoxidase is more specific for vasculitis. Either ANCA pattern is commonly seen in Churg-Strauss (50%), (21) inflammatory bowel disease, rheumatoid arthritis, other connective tissue diseases, and drug-induced illness (such as with propylthiouracil). Although the pattern is helpful to establish vasculitis, both c-ANCA and p-ANCA can be seen in microscopic polyarteritis and Wegener granulomatosis; therefore, the result alone does not make the distinction. It can, however, be useful in distinguishing microscopic polyarteritis and Wegener granulomatosis from polyarteritis nodosa, which is rarely ANCA-positive.

The usefulness of ANCA can particularly be seen in the clinical situation in which a biopsy can be circumvented. An open lung biopsy is associated with significant morbidity and mortality. If a c-ANCA returns strongly positive, the diagnosis of Wegener granulomatosis would seem to be more solidified and may indeed avoid a costly and dangerous biopsy. Similarly, renal biopsy may be avoided.

Myositis-specific antibodies (anti-Jo 1, anti-Mi2, anti-SRP, anti-MAS)

Although autoantibodies are rarely used in the initial diagnosis of an inflammatory myopathy, once the diagnosis has been established they can give information as to disease manifestation. The most common is anti-synthetase (Jo-1), which can be seen in interstitial lung disease and mechanic's hands. Anti-Mi2 can be seen in dermatomyositis and is correlated with a good prognosis. Anti-SRP is associated with cardiac disease and poor response to medical therapy. Anti-MAS can be seen in alcoholic rhabdomyolysis.

Conclusion

In the above review, we have outlined the dichotomy offered by the laboratory in terms of the usefulness in diagnosis and prognosis of rheumatologic diseases. We clearly believe that these laboratory tests will never circumvent the need for a detail-oriented history and insightful, specific clinical examination. An appropriate differential diagnosis followed by thoughtful screening should prove rewarding both to the clinician and the patient. The scare factor in terms of false-positive diagnoses may be avoided and inappropriate, expensive, subsequent investigations bypassed. On the other hand, appropriate screening with early diagnosis and treatment should indeed prove rewarding.

Anything that is too stupid to be spoken is sung.
--Voltaire

Table 1. Performance characteristics of antinuclear antibodies in
connective tissue diseases

                                                Positive       Negative
                                               likelihood/    likelihood
Disease            Sensitivity  Specificity      ratio           ratio

Systemic lupus         93%          57%           2.2             0.11
  erythematosus
Scleroderma            85%          54%           1.86            0.27
Polymyositis/          61%          63%           1.67            0.61
  dermatomyositis
Rheumatoid             41%          56%           0.93            1.06
  arthritis
Sjogren                48%          52%           0.99            1.01
  syndrome
Secondary              64%          41%           1.08            0.88
  Raynaud

Table 2. Antinuclear antibody patterns and associated diseases (a)

Antinuclear
antibody
pattern         Associated antigen         Associated disease

Homogeneous     DNA histone complex        SLE, drug-induced lupus,
                  (ribonucleoprotein.        other diseases
                  nucleosome)
Peripheral/rim  DNA, nuclear envelope      SLE, autoimmune hepatitis
                  antigens
Speckled        Smith, RNP, SS-A, SS-B,    SLE, MCTD, Sjogren,
                  Scl-70, centromere, RNA  PSS, other diseases
                  polymerase II and III
Nucleolar       Nucleolar RNA, RNA         Diffuse systemic sclerosis,
                  polymerase I               hepatocellular carcinoma
Centromeric     Centromeres                Limited cutaneous systemic
                                             sclerosis

(a) SLE, systemic lupus erythematosus; MCTD, mixed connective tissue
disease; PSS, progressive systemic sclerosis.

Table 3. Performance characteristics of specific antinuclear
antibodies (a)

Antigen             Associated condition                    Sensitivity

anti-dsDNA Ab       Systemic lupus erythematosus            57%
anti-Smith (Sm) Ab  Systemic lupus erythematosus            25-30
anti-Ro/SS-A Ab     Sjogren, SCLE, neonatal lupus syndrome   8-70
anti-La/SS-B Ab     Sjogren, SCLE, neonatal lupus syndrome  16-40
Scl-70              Systemic sclerosis                      20%
Anticentromere      Limited cutaneous systemic sclerosis    65%
Anti-U3-RNP Ab      Scleroderma                             12%

                                   Positive            Negative
Antigen             Specificity  likelihood ratio  likelihood ratio

anti-dsDNA Ab           97%            16.3             0.49
anti-Smith (Sm) Ab      High*           *                  *
anti-Ro/SS-A Ab         87              *                  *
anti-La/SS-B Ab         94              *                  *
Scl-70                  100%          >25                0.8
Anticentromere          99.9%         650                0.4
Anti-U3-RNP Ab          96%             3                0.92

(a) SCLE, subacute cutaneous lupus erythematosus.
*Precise data not available.

Accepted January 21, 2004.

References

1. Sox HC Jr, Liang MH. The erythrocyte sedimentation rate: guidelines for rational use. Ann Intern Med 1986;104:515-523.

2. Jones SA, Novick D, Horiuchi S. et al. C-reactive protein: a physiologic activator of interleukin 6 receptor shedding. J Exp Med 1999;189:599-604.

3. Zouki C, Beauchamp M, Baron C, et al. Prevention of in vitro neutrophil adhesion to endothelial cells through shedding of L-selectin by C-reactive protein and peptides derived from C-reactive protein. J Clin Invest 1997;100:522-529.

4. Wener MH, Daum PR, McQuillan GM. The influence of age, sex, and race on the upper reference limit of serum C-reactive protein concentration. J Rheumatol 2000;27:2351-2359.

5. Morley JJ, Kushner I, et al. Serum C-reactive protein levels in disease. Ann N Y Acad Sci 1982;389:406-418.

6. Amos RS, Crockson RA, Crockson AP, et al. Rheumatoid arthritis: C-reactive protein and erythrocyte sedimentation rate during initial treatment. BMJ 1978;1:1396.

7. Solomon DH, Kavanaugh AJ, Schur PH, et al. Evidenced-based guidelines for the use of immunologic tests: antinuclear antibody testing. Arthritis Rheum 2002;47:434-444.

8. Tan EM, Feltkamp TE, Smolen JS, et al. Range of antinuclear antibodies in "healthy" individuals. Arthritis Rheum 1997;40:1601-1611.

9. Kavanaugh AF, Solomon DH. Guidelines for immunologic laboratory testing in the rheumatic diseases: anti-DNA antibody tests. Arthritis Rheum 2002;47:546-555.

10. Schur PH, Sandson J. Immunologic factors and clinical activity in systemic lupus erythematosus. N Engl J Med 1968;278:533-538.

11. Ter Borg EJ, Horst G, Hummel EJ, et al. Measurement of increases in anti-double-stranded DNA antibody levels as a predictor of disease exacerbation in systemic lupus erythematosus. Arthritis Rheum 1990;33:364.

12. Swaak AJ, Groenwold J, Bronsveld W. Predictive value of complement profiles and anti-dsDNA in systemic lupus erythematosus. Ann Rheum Dis 1986;45:359-366.

13. Harley JB, Alexander EL, Bias WB, et al. Anti-Ro (SS-A) and anti-La (SS-B) in patients with Sjogren's syndrome. Arthritis Rheum 1986;29:196-206.

14. Reveille JD, Solomon DH, American College of Rheumatology Ad Hoc Committee of Immunologic Testing Guidelines. Evidenced-based guidelines for the use of immunologic tests: anticentromere, Scl-70, and nucleolar antibodies. Arthritis Rheum 2003;49:399-412.

15. Okano Y, Steen VD, Medsger TA Jr. Autoantibody to U3 nucleolar ribonucleoprotein (fibrillarin) in patients with systemic sclerosis. Arthritis Rheum 1992;35:95-100.

16. Falkner D, Wilson J, Fertig N, et al. Studies of HLA-DR and DQ alleles in systemic sclerosis patients with autoantibodies to RNA polymerase and U3-RNP (fibrillarin). J Rheumatol 2000;27:1196-1202.

17. Kroot EJ, de Jong BA, van Leeuwen MA, et al. The prognostic value of anti-cyclic citrullinated peptide antibody in patients with recent-onset rheumatoid arthritis. Arthritis Rheum 2000;43:1831-1835.

18. Stone JH, Talor M, Stebbing J, et al. Test characteristics of immunofluorescence and ELISA tests in 856 consecutive patients with possible ANCA-associated conditions. Arthritis Care Res 2000;13:424-434.

19. Guillevin L, Durand-Gasselin B, Cevallos R, et al. Microscopic polyangiitis. Arthritis Rheum 1999;42:421-430.

20. Woodworth TG, Abuelo JG, Austin HA III, et al. Severe glomerulonephritis with late emergence of classic Wegener's granulomatosis: report of 4 cases and review of the literature. Medicine (Baltimore) 1987;66:181-191.

21. Hauschild S, Scmitt WH, Csernok E, et al. ANCA in systemic vasculitides, collagen vascular diseases, rheumatic disorders and inflammatory bowel diseases. Adv Exp Med Biol 1993;336:245-251.

RELATED ARTICLE: Key Points

* While laboratory testing can help with the diagnosis of rheumatic disease, overreliance on a lab result may lead the clinician to an incorrect diagnosis.

* Antinuclear antibodies are present in a number of connective tissue diseases; while nearly all patients with lupus are ANA-positive, the high number of false positives limits its utility as a screening test.

* New tests, such as the cyclic citrullinated protein antibody, may improve early identification of patients with rheumatoid arthritis, particularly those with a prognosis for a more aggressive disease.

Christopher Lee Colglazier, MD, and Paul George Sutej, MD

From the Section on Rheumatology and Clinical Immunology, Department of Internal Medicine, Wake Forest University School of Medicine, Wake Forest, NC.

Reprint requests to Dr. Christopher Morris, 3 Sheridan Square, Kingsport, TN 37660. Email: [e-mail]