Rational Use of Blood Tests in the Evaluation of Rheumatic Diseases

Abstract

Many blood tests used in the evaluation of rheumatic diseases are non-specific and should not be ordered and interpreted in isolation. False positive results can lead to inappropriate therapies with serious adverse events while false negatives may steer the provider to incorrect diagnoses. Relying on blood tests alone without considering the clinical context should be avoided.

Introduction

Blood tests are more reliable when the pre-test likelihood of the rheumatic disease (RD) under consideration is high.¹ Both false positive (See Table 1) and false negative results are common. The approach of using “rheumatology panels” containing multiple serologies cannot substitute for good history taking and physical examination. Some RDs such as osteoarthritis and soft tissue conditions are diagnosed without specific blood tests. Definitive “screening tests” to rule out RD do not exist. As an example, many patients with life-threatening systemic vasculitis have negative serologies. In this article, we review the general principles behind the commonly used rheumatology blood tests and delineate how the non-rheumatologist can use them appropriately.

Table 1.

Examples of clinical circumstances in which serologies can be falsely positive

(ESRD= End Stage Renal Disease)

ESR Pregnancy Obesity Anemia ESRD

CRP Infections Obesity Malignancies

RF Smoking Infections Elderly patients

ANA 10% of normal population Elderly patients Autoimmune thyroiditis Multiple sclerosis Hepatitis C

ANCA Tuberculosis Inflammatory bowel disease Medications

CK Post-strenuous exercise African American race Higher muscle mass

Erythrocyte Sedimentation Rate

Erythrocyte sedimentation rate (ESR) measures the distance, in millimeters, that red blood cells fall inside a tube over an hour. Many factors such as age, co-morbidities, infections, obesity, pregnancy, and the presence of anemia or polycythemia can affect this measurement. It can not be assumed that a high ESR indicates RD, or that a normal ESR makes it unlikely. Rheumatologists frequently treat patients with inflammatory RD who repeatedly have a normal ESR. It is prudent to repeat an otherwise unexplained high ESR to rule out lab error.

Relatively new onset of symmetric shoulder and hip stiffness with fatigue and a high ESR may suggest a diagnosis of polymyalgia rheumatica (PMR). A good response to prednisone is often confirmatory. However, a normal ESR may be found in 20% of patients with PMR.² Similarly, giant cell arteritis (GCA) can occur with a normal ESR in 5 to 10% of patients.² In addition, other factors such as age, headache description, presence of optic neuritis, and jaw claudication should be included in the diagnostic process. Therefore, in an elderly patient with high pre-test likelihood, a normal ESR does not exclude GCA. Although an ESR may be useful to monitor disease course, the clinical impression usually overrides the lab result. A persistent and significant ESR elevation should alert the provider to look for additional causes.

C Reactive Protein

C Reactive Protein (CRP) is a slightly more reliable marker of inflammation than ESR, but is more expensive.¹ Unlike the ESR it does not rise with anemia. CRP is also nonspecific and can be elevated with cigarette smoking, diabetes, infections, and obesity. Vasculitis may cause a significant elevation of CRP (e.g. >5 mg/dL), sometimes as high as seen with infections. A mild to moderate elevation can be useful supportive evidence, when clinical parameters suggests a RD. It is common for RDs, such as rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PsA), to be diagnosed even when the CRP is normal. Many patients with systemic lupus erythematosus (SLE) have minimal or no elevation in CRP even during intense disease activity.³ Although the reasons for this are not fully established, hypotheses include genetics, antibody binding of CRP, and CRP consumption. Significant CRP elevations in SLE should raise concerns for infection or serositis.³

Rheumatoid Factor

Rheumatoid factors (RF) are antibodies directed against the Fc component of the immunoglobulin molecule. Increased polyclonal immunoglobulin production in diverse non-rheumatic diseases can cause a positive RF. This phenomenon can be seen in primary biliary cirrhosis, sarcoidosis, malignancies, and infections such as hepatitis C, tuberculosis, infectious endocarditis and connective tissue diseases (CTD) such as Sjogren’s syndrome. An increased prevalence of RF is also found in smokers and in the healthy elderly population.⁴

Various techniques such as agglutination of sheep red blood cells, latex agglutination, enzyme linked immunosorbent assay (ELISA), and nephelometry are used to measure RF. Inadequate standardization of these techniques causes variability across laboratories. The sensitivity of RF for RA ranges from 50% to 85%. Approximately 70% of RA patients exhibit a positive RF. RF may be negative in early RA and remain negative in 15% of patients. If the clinical evaluation suggests RA, as in a patient with symmetric polyarthritis of more than six weeks duration, a positive RF can help secure the diagnosis. While a RF titer is not useful to monitor disease activity, ESR and CRP levels may be.⁴ Higher titers of RF are associated with more aggressive arthritis, rheumatoid nodules, and extra-articular manifestations such as lung disease and vasculitis.

Anti-CCP Antibodies

Anti-cyclic citrullinated peptide (CCP) antibodies, commercially introduced in 2003, have specificity as high as 98%, for RA.⁵ Still, the test is not 100% specific and with widespread use clinical circumstances where CCP can be false positive have been identified.⁶ In a patient with undifferentiated arthritis, the presence of CCP can predict the development of RA. CCP are also associated with more aggressive diseases and are useful to distinguish RA from mimickers such as hepatitis C arthropathy, PsA, gout and pseudogout. CCP negative RA has been characterized similar to RF negative RA, and in both circumstances the term “sero-negative RA” is used. When the clinical picture suggests RA, both RF and anti-CCP are ordered. In this setting RA may be diagnosed when either tests are positive, and at times when both are negative.

Antinuclear Antibodies

Antinuclear antibodies (ANA) should not be used as general screening tests and should not be ordered when the pre-test likelihood of SLE is low.^{7, 8, 9} A positive ANA can be seen in 10% or more of the normal population, in individuals with a family history of RD, in as many as 75% of healthy elderly population and with non-rheumatic conditions such as hepatitis C, autoimmune thyroiditis, multiple sclerosis and autoimmune hepatitis. A positive ANA reflects nonspecific autoimmunity in these conditions and most of these patients do not develop SLE.

Several methods are used to detect ANA. Most commercial laboratories use the automated ELISA since it is faster, cheaper and less labor intensive than the immunofluorescence (IF) method in which each slide is manually examined. In the ELISA method, results are described in international units per ml whereas in IF, titers with dilutions (e.g. 1:320) are reported. However, ELISA results can be misleading as both false positives and false negatives are more commonly seen than with IF.¹⁰ The American College of Rheumatology (ACR) issued a policy statement in 2009 advising the preferred use of IF.¹⁰ To detect IF, permeabilized cells fixed to a slide are incubated with a patient’s serum. A fluoresceinated second antibody is added which binds to the ANA already affixed to nuclei. The sample is serially diluted and the greatest dilution at which the fluorescence persists is reported as the titer. Some labs stop at the 1:1280 dilutions, while others report up to 1:5120. Variable thresholds, 1:40 at some labs and higher dilutions at others, are used to report ANA positivity. Only a titer of 1:160 or higher is considered to be positive in the ANA laboratory at the University of Missouri. However, lower titers are reported and could be meaningful in the context of symptoms, especially in children. Reporting a titer 1:40 as positive can lead to a very high frequency of false positive ANA results, noted in up to 32% of samples.¹ In general, ANA titers are not useful to monitor autoimmune disease activity.

“Multiplex assay” also known as “bead technology” has recently gained popularity since it reduces the manpower needs through automation and allows simultaneous IF testing of a much larger number of specimens.

It is very rare for patients with SLE to have negative ANA by IF particularly with the current use of specialized cell lines. The increased sensitivity of these cell lines can generate false positive results much more commonly than false negatives. ANA may be negative in other autoimmune diseases such as polymyositis and vasculitis thus making this test of limited diagnostic utility in these diseases. A positive ANA may result from antibodies directed against hundreds of nuclear antigens. However, antibodies against certain “Extractable Nuclear Antigens (ENA)” such as SS-A, SS-B, Smith, double stranded DNA (ds-DNA), and ribonucleoprotein (RNP) are clinically more relevant and are reported in the “ANA panel.” (See Table 2). It is reasonable to order the complete panel when the pre-test clinical suspicion for CTD is high. Antibodies in the ANA panel are checked by ELISA or by the more labor intensive immunodiffusion technique. When evaluating a patient with significant complaints of dry eyes or dry mouth, ordering anti-SSA and anti-SSB may be useful to diagnose Sjogren’s syndrome (SS) (See Table 2). Many patients with SS do not have these antibodies. SS-A and SS-B antigens are also known by Ro and La respectively. Pregnant women with anti-SS-A are at increased risk of delivering babies with neonatal lupus or congenital heart block. Positive anti-SS-A with a negative ANA has been reported.

Table 2.

Serologies and rheumatic disease associations

Serology	Rheumatic Disease
Anti-Sm	SLE
Anti-SSA	SLE, Sjogren’s Syndrome
Anti-SSB	Sjogren’s Syndrome, SLE
Anti-RNP	MCTD, SLE
Anti-Jo-1, Anti-SRP & Anti-ku	Myositis
Anti-histone	Drug-Induced SLE
Anti-Centromere	Limited Scleroderma
Anti-topoisomerase (Scl-70)	Scleroderma

Anti-Sm=anti-Smith

Anti-SSA=Sjogren’s Syndrome A

Anti-SSB= Sjogren’s Syndrome B

Anti-RNP=anti-ribonucleoprotein

Anti-SRP=anti-signal recognition particle

SLE=systemic lupus erythematous

MCTD=mixed connective tissue disease

A positive ANA is seen in all patients with drug induced lupus. In addition, 100% of patients with mixed connective tissue disease (MCTD) will have a positive, often high titer, anti-RNP. Anti-ds-DNA antibodies are frequently noted in SLE patients with severe renal and central nervous system involvement. Although inconsistent, a high or rising anti ds-DNA titer may predict an SLE renal flare. Anti-Sm is considered specific for the diagnosis of SLE but lacks sensitivity.

ANA patterns were described with the expectation they may be useful to predict a specific CTD. Homogeneous and speckled patterns tend to be non-specific. Some association of the nucleolar pattern with diffuse scleroderma and the rim pattern with SLE has been described. The clinical utility of patterns has diminished owing to the subjective nature of interpretation and to the lack of specificity. One important exception is the anti-centromere pattern which is strongly associated with limited scleroderma.

Anti-neutrophil Cytoplasmic Antibodies

Anti-neutrophil cytoplasmic antibodies (ANCA) are directed against antigens in the cytoplasm of neutrophils. “c-ANCA” and “p-ANCA” refer to the microscope appearance of IF patterns, cytoplasmic and perinuclear respectively. (See Figure 1) An ANCA may be non-specific since they can target numerous antigens. In vasculitis those directed against serum proteinase 3 (PR-3) and myeloperoxidase (MPO) contribute to the pathogenesis and are considered clinically relevant.¹¹ Experts recommend that ANCA testing should be done sequentially with immunofluorescence testing done first. If this is positive, then an ELISA for antibodies against PR-3 and MPO is performed for confirmation.¹² Some laboratories have stopped offering the first step and offer only the second step; this is controversial.

Figure 1.

Anti-Neutrophil Cytoplasmic Antibodies (ANCA)

ANCA may display a cytoplasmic staining pattern (C-ANCA, left) or a perinuclear staining pattern (P-ANCA, right).

Courtesy: Inova Diagnostics, Inc. San Diego, CA

The presence of c-ANCA is sensitive for the diagnosis of Wegener’s granulomatosis (WG). A Positive c-ANCA is found in 90% of systemic WG, 75% of limited WG (no renal disease), and 50% of microscopic polyangiitis (MPA) patients. The presence of c-ANCA is 98% specific in these conditions. It is usually due to antibodies against PR3, though they can be directed against other neutrophil cytoplasmic enzymes. Rising c-ANCA titers may indicate a disease flare but this is not always consistent.

The p-ANCA pattern results from staining around the nucleus, and can be seen in MPA and pauci-immune glomerulonephritis (GN) as well as a wide variety of non-rheumatic conditions such as infections, inflammatory bowel disease, other autoimmune diseases and malignancies.¹¹ Antibodies causing the p-ANCA pattern can be directed against numerous antigens such as lactoferrin, elastase, MPO and others. However, only those against MPO appear to be clinically relevant, especially in MPA and pauci-immune GN. Due to this lack of specificity, p-ANCA is a less reliable predictor of vasculitis than c-ANCA.

ANCA testing can be a useful accessory tool when a patient is being evaluated for possible vasculitis. It is important to remember that up to 10% of WG and 50% Churg-Strauss syndrome patients do not have a positive ANCA. Most patients with polyarteritis nodosa, a medium vessel vasculitis, have a negative ANCA. Obtaining a tissue biopsy, if feasible, is often the best way to confirm vasculitis. Concerns about the inappropriate use of ANCA for diagnostic and treatment decisions, have lead experts to propose gating policies restricting the widespread ordering of ANCA.¹² Of interest, there is no good evidence that ANCA titer elevations or continued ANCA positivity predict recurrence of vasculitis in an allograft, and therefore should not be the sole basis for rejection of candidates waiting for renal transplant.¹³

Uric Acid

The best way to confirm gout is by joint aspiration with the presence of synovial fluid uric acid crystals.¹⁴ In gout, the onset of joint inflammation is rapid and the pain is often more than in other arthritic conditions. Eliciting a history about rapidity of onset and intensity of pain should supplement lab testing. Although a high uric acid is a risk factor for gout, only 22% of patients with uric acid levels greater than 9.0 mg/dL develop gout within a period of five years (annual incidence of 4.5%). For patients with serum uric acid levels between 7 and 8.9 mg/dL, the annual incidence is only 0.5%. Many patients with acute gout exhibit serum uric acid levels in the normal range. Therefore, hyperuricemia does not automatically confirm gout and a normal serum uric acid by itself does not rule out the diagnosis. In the appropriate clinical setting the sensitivity and specificity of serum uric acid for the diagnosis of acute gout may approach 92% and 91% respectively.¹⁵ When joint aspiration is not feasible, checking serum uric acid during an acute attack can be useful since a higher uric acid level increases the probability of gout. In patients with recurrent gout attacks, treatment attempts are geared to reduce the uric acid level below 5–6 mg/dl.

Human Leukocyte Antigen B27

There is wide ethnic variation in the prevalence of Human Leukocyte Antigen (HLA-B27) and its association with AS. Seven to ten percent of the American Caucasian population is HLA-B27 positive. Some Native American and Eskimo populations have prevalence upwards of 50%. The majority of these HLA-B27 positive individuals do not develop AS. HLA-B27 should not be routinely ordered in all patients with back pain. However, if the history suggests inflammatory back pain (stiffness), and there are supportive clinical features such as the presence of uveitis or a family history of AS, checking an HLA-B27 may help confirm the diagnosis of spondyloarthritis.¹⁶ Generally, inflammatory back pain occurs more commonly in patients younger than 40 and the onset is insidious. The pain worsens with rest and improves with activity. The classic description includes night time back pain which worsens in the early morning hours, and is relieved with movement.¹⁶ Among African American patients with AS, only 50% may be HLA-B27 positive whereas in Caucasians with AS HLA-B27 may be seen in up to 80% or more.

Creatinine Kinase and Myositis Serologies

Baseline creatinine kinase (CK) levels can be higher in patients with significant muscle mass and in African-American patients. Levels may also rise with physical exertion. Prior to prescribing statins, a baseline CK is important. During treatment, various complaints such as myalgia, weakness and fatigue may prompt the provider to check a CK. Modest elevations may be confusing unless a baseline comparison is available. It is not uncommon for myalgias and CK elevations to persist for several months after statins are switched or discontinued. Rarely, statins induce autoimmune myositis manifesting as muscle weakness that may persist more than six months after statin discontinuation.¹⁷ Persistent CK elevations, especially in the presence of muscle weakness should warrant a rheumatology consultation. A detailed discussion of the numerous antibodies used by experts in neuromuscular diseases and their association with various myositis syndromes is beyond the scope of this paper. However, some examples are shown in Table 2. These antibodies lack sensitivity since each is present only in a minority of patients with different myositis syndromes. Anti-Jo-1 antibodies, checked most frequently, are present in approximately 20% to 30% of adult-onset polymyositis patients and in greater than 65% of patients with myositis and interstitial lung disease.¹⁸ The myositis serologies are of supportive value to the clinical impression, CK level, electromyography findings and muscle biopsy. The biopsy remains the diagnostic gold standard.

Anti-Phospholipid Antibodies

In patients with recurrent miscarriages, thromboembolic events, unexplained premature myocardial infarction or stroke, the anti-phospholipid (APL) syndrome should be considered. It is important to order the three different serologies including the Lupus anti-coagulant (LAC), anti-cardiolipin antibodies IgG, IgM, IgA and anti-beta2-glycoprotein-1 antibodies. The positivity of any one of these is sufficient to entertain the diagnosis. Assessing for the persistence of positive serologies by rechecking at three months is required to confirm the diagnosis.¹⁹ Frequently, the beta-2-microglobulin test is ordered by mistake since the name sounds similar to beta-2-glycoprotein-1 antibodies.

Miscellaneous Tests

Ferritin, iron studies, thyroid stimulating hormone, parathyroid hormone, calcium, magnesium and phosphorous are ordered to assess for endocrine and metabolic disorders especially when evaluating for pseudogout (Calcium Pyrophosphate Deposition Disease). Mild elevations of ferritin can be a non-specific marker of inflammation. However, very high ferritin levels e.g. >1000 ng/ml can be seen in adult onset Still’s disease. Cytopenias affecting any cell line may accompany SLE, but lymphopenia is most common. When RD manifestations are due to immune-complex deposition in the organs, a decline in complements C3 and C4 levels can be a useful indicator of an SLE flare or can help confirm a new SLE diagnosis. Cryoglobulins, immunoglobulins which precipitate from blood at cold temperature and dissolve on rewarming, can occur with viral infections (e.g. hepatitis C), leukemia’s and lymphomas in addition to RD. When vasculitis is considered in the differential, as in a patient with neuropathy or glomerulonephritis, cryoglobulins should be measured. A tissue biopsy is most reliable in confirming cryoglobulinemic vasculitis. Lack of attention to special collection requirements unfortunately can lead to incorrect results.²⁰

Conclusion

Blood tests, if ordered in the appropriate clinical context and if interpreted with their limitations in mind, can be useful accessory tools in the diagnosis and management of rheumatic diseases.^{5, 9} Comprehensive history taking and physical examination should guide testing.

Biography

Chokkalingam Siva, MD, MS, is Assistant Professor of Clinical Medicine; Emily C. Larson, DO, is a Rheumatology Fellow, and Mark Barnett, MD, is an Family Medicine Resident. All are at the University of Missouri School of Medicine, Division of Immunology and Rheumatology.

Contact: sivac@health.missouri.edu