Article Text

Download PDFPDF

How to use… antinuclear antibodies in paediatric rheumatic diseases
  1. Lehn K Weaver,
  2. Edward M Behrens
  1. Division of Pediatric Rheumatology, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
  1. Correspondence to Dr Edward M Behrens, Department of Pediatric Rheumatology, The Children's Hospital of Philadelphia, 34th and Civic Center Blvd., Philadelphia, PA 19104, USA; behrens{at}email.chop.edu

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

The test for antinuclear antibodies (ANA) is frequently obtained to identify patients at higher risk for a rheumatic condition prior to subspecialty referral. However, a positive ANA occurs in a variety of clinical settings and does not automatically indicate underlying pathology. Therefore, careful consideration is needed to determine if the results will alter diagnostic, treatment, or medical decision making before the ANA test is sent.

Systemic lupus erythematosus (SLE) is the prototypic autoimmune disease related to the presence of ANAs. The clinical manifestations of SLE are diverse, as this disease affects every organ in various combinations and severities unique to each individual lupus patient. Therefore, presentations can range from subtle skin and laboratory abnormalities to life-threatening end-organ damage. Other rheumatic conditions associated with a positive ANA include juvenile idiopathic arthritis, juvenile dermatomyositis, mixed connective tissue disease, localised scleroderma, systemic sclerosis, Sjögren syndrome and drug-induced lupus.1

This article presents the history and current use of the ANA test in paediatric practice related to rheumatic diseases, with an emphasis on three specific clinical scenarios where this test may be useful. We also review the relevant data pertaining to why the ANA test is not a good test to screen for rheumatic conditions. As there is no test capable of discriminating between a patient with and without a rheumatic disease, we present some features of the most common rheumatic diseases to help general paediatricians recognise specific patterns of symptoms that warrant further evaluation and/or referral to a paediatric rheumatologist (box 1).

Box 1

Specific signs and symptoms of common paediatric rheumatic diseases20,22

Systemic lupus erythematosus (SLE) (based on the 1997 American College of Rheumatology Classification of SLE)

  • Oral ulcers (often painless and on the hard palate).

  • Malar rash (spares nasolabial folds).

  • Photosensitivity (unusual persistent skin reaction to sunlight).

  • Discoid rash (inflammatory, scarring rash).

  • Arthritis (often symmetric and affects large and small joints).

  • Serositis (pleuritis or pericarditis).

  • Seizures or psychosis.

  • Lupus nephritis (3+ proteinuria, >0.5 g/dl protein in 24 h urine collection, or cellular casts).

  • Cytopenias (haemolytic anaemia with reticulocytosis, leucopoenia <4000/mm3 or lymphopenia <1500/mm3 on two or more occasions, or thrombocytopenia <100 000/mm3 on one occasion).

  • Autoantibody positive (anti-dsDNA, anti-Smith, or antiphospholipid antibodies).

  • Antinuclear antibodies positive.

Juvenile idiopathic arthritis

  • Objective clinical findings of arthritis in one or more joints of more than 6 weeks duration after other causes (infectious, oncologic, etc) are excluded.

  • Prominent morning stiffness improved with activity.

  • Painful joints that improve with movement.

  • Pain not out of proportion to clinical findings.

  • Long-term changes from arthritis include leg-length discrepancies and muscle atrophy around affected joint(s).

Raynaud's syndrome

  • Attacks of digital pallor (white), followed by cyanosis (blue) and subsequent hyperaemia (red) of the fingers or toes stimulated by cold, excitement or vibration.23

  • Raynaud's syndrome occurs as a secondary feature of multiple rheumatic diseases, and can be the presenting symptom in systemic sclerosis, mixed connective tissue disease and SLE.24

  • Digital ulcers and abnormal nailfold capillaroscopy are predictive of systemic progression in patients with Raynaud's syndrome.25 ,26

Physiological background

What are ANAs and how do they develop?

The field of ANAs originated with the discovery of the lupus erythematosus (LE) cell test in 1942 by Hargraves et al.2 In the LE cell test, in vitro manipulation of whole blood damages cell membranes, and subsequently exposes a cell's nuclear material to the extracellular environment.3 Serum components bind to the exposed nuclear material, which is subsequently engulfed by surrounding phagocytic cells, which can be identified as LE cells on peripheral smears.3 Further studies identified ANAs as the necessary component within the serum for LE cell formation.3

ANAs are directed against large nuclear complexes containing both protein and nucleic acid components.1 When associated with autoimmune disease, ANAs are often high-titre, high-affinity IgG antibodies produced by autoreactive B cells requiring T cell help and driven by host autoantigens.1 In autoimmune disease, autoreactive T and B cells have escaped the multiple checkpoints of immune tolerance leading to autoantibody production, immune complex formation and systemic inflammation.4 Despite these pathophysiologic connections, the presence of ANAs does not always indicate immune complex-mediated pathology. In non-autoimmune settings, ANAs are most likely low-titre antibodies produced by activation of a polyclonal B cell population that has not escaped immune tolerance. The production of these non-specific ANAs may be stimulated by normal homeostasis in ANA-positive healthy individuals, or by systemic inflammation during infection, cancer, or environmental exposures.5

Technological background

How are ANAs measured?

Indirect immunofluorescence using whole cell preparations is the most widely used test to screen for ANAs (figure 1). To perform this assay, a patient's serum is diluted by a factor of 40 and placed on a slide containing a monolayer of cells. After a brief incubation with the diluted serum, the cells are washed and labelled with fluorescent-tagged, secondary, antihuman immunoglobulin. Only a patient's serum containing ANAs will bind to the cells and acquire the fluorescent label. If nuclear fluorescence is detected, the ANA is considered positive and the fluorescent pattern is described. If positive, the sample will be further diluted until the highest serum dilution factor allowing for nuclear fluorescence is detected. This dilution factor is recorded as the ANA titre.5

Figure 1

The antinuclear antibodies (ANA) test. (A) The flow diagram demonstrates the process of ANA detection using the indirect immunofluorescence technique, as described in the text. (B) A single cell is shown with the nucleus highlighted in the centre of the cell. The black antibodies represent ANAs from a patient's serum, which bind to nuclear components within the nucleus of the cell. Following a wash step, fluorescently labelled secondary antibodies (depicted in green) are added to the cell, and detect the presence of ANAs. Only patients’ sera containing ANAs will be labelled with secondary fluorescently labelled antibodies, and demonstrate nuclear fluorescence when visualised under a microscope.

Inconsistencies between the results from different laboratories and published articles regarding the prevalence of a positive ANA in specific populations occur for multiple reasons. These reasons include the subjective interpretation of a positive result requiring observation of cells under a microscope, differences in ‘brightness’ of the fluorescently labelled secondary antibodies, variation of the ANA titre considered as a positive result, and use of different cell substrates for ANA detection.6 For example, previous studies used rat liver cells to detect ANA, which rarely detected ANA in healthy individuals. However, current practices use the HEp-2 human cell line, which is much more sensitive for detecting ANA and produces a positive test result in up to 32% of healthy patients at a titre of 1:40, 13% at a titre of 1:80, 5% at a titre of 1:160 and 3% at a titre of 1:320.7 These striking variations between available testing methods continue to be problematic today with the introduction of ELISA-based ANA detection, which have poor sensitivity when compared with the gold standard method of detecting ANA with indirect immunofluorescence.6 These inconsistencies make the clinical utility of the already confusing ANA test even more difficult to determine.

Indications and limitations

In a child with non-specific symptoms of rheumatic disease, should the ANA test be used to screen for autoimmunity?

The decision to order a test in clinical practice must begin with an understanding of how the results of the test will be used to inform clinical decision making. If an ANA is used to screen for rheumatic disease, but is inherently a poor screening test, then the results of the test will not be useful and the test should not be sent. Therefore, careful evaluation of the ANA as a screening tool for rheumatic diseases is warranted.

For diseases that are rare, such as rheumatic diseases, a test must be both highly sensitive and specific to be useful as a screen for disease. If a test is not sensitive, then it will produce a high false negative rate missing a disease when the disease is present. If a test is highly sensitive, but is not specific, it will generate a high false positive rate indicating disease in patients without underlying disease. Both these scenarios are true for the ANA test, as evidenced by the poor specificity and/or sensitivity of the ANA test in various rheumatic diseases (table 1).

Table 1

Rheumatic diseases associated with a positive antinuclear antibodies6

A screening test must also have a high positive predictive value to be useful. The positive predictive value is the proportion of patients who have a positive test that truly have the disease in question. Unfortunately, the positive predictive value of the ANA test when used to screen for a rheumatic disease is only 11%.6 This means that the proportion of patients with a positive ANA that truly have a rheumatic disease is low. Therefore, patients with a positive ANA cannot automatically be considered to have an underlying rheumatic disease.

Further evidence of the ANA test's inability to screen for rheumatic diseases comes from retrospective studies of patients referred to rheumatology clinics with a positive ANA. Perilloux et al8 reviewed the charts of 245 patients referred to a paediatric rheumatology clinic with a positive ANA, and found that 45% of the ANA-positive patients did not have an underlying rheumatic disease. McGhee et al9 reviewed the charts of 110 patients referred to a paediatric rheumatology clinic for positive ANA of which 72% did not have a rheumatic disease. The majority of the patients in this study had musculoskeletal pain syndromes including hypermobility syndrome, Osgood–Schlatter syndrome, patellofemoral syndrome and other mechanical musculoskeletal disorders.9 McGhee et al also confirmed the previous connection of a positive ANA test in children with acute infections, as 13% of the patients in the study had an identifiable acute infection correlated with the positive ANA test.9–11 These studies indicate that the ANA test continues to be a poor screen for rheumatic disease even in a population biased toward a high baseline concern for rheumatic disease.

The ANA titre provides a quantitative analysis of the concentration of ANAs in a patient's serum. Vaile et al12 demonstrated that the presence of a high-titre ANA (>1:640) does not discriminate between patients with and without a connective tissue disease at presentation. However, it is unclear if patients with high-titre ANA tests are at increased risk of developing rheumatic disease over time.

Musculoskeletal pain is a common complaint in paediatric practices, as a recent report describes greater than 50% of 18-year-olds have experienced musculoskeletal pain, and 30% of children reported having chronic pain lasting more than 6 months.13 It is important to differentiate between inflammatory and non-inflammatory causes for a patient's joint symptoms, as many rheumatic diseases can present with musculoskeletal pain. Cabral et al14 determined that a positive ANA test is not correlated with the development of a rheumatic disease in patients presenting with non-rheumatic musculoskeletal complaints by prospectively tracking ANA-positive patients over time. Of the patients with non-rheumatic musculoskeletal pain, 68% of these subjects were found to be ANA negative at presentation, and 10% of patients were determined to be ANA negative on repeat testing.14 Of the 22% of patients with a persistently positive ANA test, no patient developed overt inflammatory or autoimmune disease during a mean follow-up of 61 months.14 The low rate of progression to frank autoimmune disease was confirmed by Deane et al15 who showed that only one out of 31 ANA positive patients without autoimmune disease at presentation developed a non-rheumatic autoimmune disease during a mean follow-up of 37 months. This indicates that patients with non-rheumatic musculoskeletal pain can have persistently positive ANA tests with a low rate of progression to autoimmune disease.

These studies indicate that the ANA test should not be used unless there are specific signs of rheumatic disease identified on physical exam.

In a child who does not have specific signs of rheumatic disease, does a positive ANA indicate the presence of a rheumatic disease?

Many patients referred to a rheumatology clinic for a positive ANA do not have an underlying rheumatic disease. A variety of non-rheumatic conditions have been associated with a positive ANA including acute infection, drug and environmental exposures, malignancy and non-rheumatic autoimmune diseases, such as autoimmune thyroiditis, hepatitis, idiopathic thrombocytopenic purpura and myasthenia gravis.1 ,5 ,9–11 ,16 In addition, a positive ANA test occurs in between 3% and 32% of healthy individuals.5 ,7 ,16–18 Therefore, interpretation of a positive ANA test must be guided by clinical impression after a careful history and examination is performed.

Based on these studies, many different individuals can have a positive ANA test without a rheumatic disease or underlying pathology. This makes interpreting a positive ANA difficult. If a patient has a positive ANA without clinical features of autoimmune disease, then the ANA test should not have been sent. In this scenario, it is most likely a benign ANA, and reassurance should be provided to the family. In patients with a low-titre positive ANA less than 1:640, there is no reason to recheck the ANA, as the ANA can remain positive for years with a low rate of progression to autoimmune disease.15 However, if new or concerning symptoms of a rheumatic disease develop in a patient with a previously positive ANA, those patients should have further evaluation, as a positive ANA test may precede the development of clinically observable autoimmune disease.19 In the case of a high-titre ANA of 1:640 or greater, our practice is to reassess the child after 6 months to ensure additional features of rheumatic disease have not developed.

In a child with specific clinical signs of a rheumatic disease, does a positive ANA confirm the diagnosis?

The results of an ANA test may influence clinical decision making in the diagnosis of SLE or lupus-like conditions, such as mixed connective tissue disease, Sjögren syndrome, or scleroderma. To aid in the classification of SLE, modified American College of Rheumatology criteria were created in 1997 (table 1). The presence of four or more criteria can increase the specificity of a diagnosis of SLE.20 Therefore, in a patient with two specific classification criteria for SLE, sending an ANA test may prompt further referral or diagnostic testing for SLE if positive. By contrast, the negative predictive value of an ANA is near 100% indicating that a similar patient with a negative ANA should prompt further workup or evaluation for aetiologies other than SLE as a cause of the patient's symptoms.18

The ANA is not useful in the diagnosis of other rheumatic diseases, but is helpful in predicting the risk of developing uveitis in patients with juvenile idiopathic arthritis, and may predict the lack of systemic progression in Raynaud's syndrome if negative, as described in box 2. In conclusion, a positive ANA in a patient with specific clinical signs of SLE or a lupus-like disease does not confirm the diagnosis, but may aid in determining if further evaluation or referral to a paediatric rheumatologist is necessary.

Box 2

Clinically relevant uses of the antinuclear antibodies (ANA) test

  1. The ANA test should be sent in patients with two or more specific criteria for systemic lupus erythematosus or a lupus-like condition, such as mixed connective tissue disease, Sjögren syndrome, or scleroderma, as described in the main text of this article.

  2. Patients diagnosed with juvenile idiopathic arthritis should have an ANA test performed to predict their risk of developing uveitis.

    1. All patients with juvenile idiopathic arthritis (JIA) are at risk for developing uveitis, and require regular ophthalmologic evaluations, as juvenile idiopathic arthritis (JIA)-associated uveitis is frequently asymptomatic and can lead to band keratopathy, cataracts, glaucoma and severe vision loss.22

    2. If a JIA patient has a positive ANA, they require more frequent ophthalmologic evaluations than JIA patients who are ANA negative.27

  3. The ANA test may be helpful in predicting the patients with Raynaud's syndrome who are at low risk for developing progressive disease.

    1. Most patients with Raynaud's syndrome who are ANA negative will not develop progressive rheumatic disease, as its negative predictive value is 93%.26

    2. A positive ANA in patients with Raynaud's syndrome does not always portend progressive disease, as its positive predictive value is 30%.26

Does a negative ANA test rule out rheumatic disease?

One of the defining features of SLE is a positive ANA with a sensitivity and negative predictive value approaching 100%. In other words, almost all SLE patients are ANA positive, and a negative ANA test accurately predicts that a patient does not have SLE. This is not the case when you extend the function of the ANA test to predict if a patient does not have any rheumatic disease when the ANA test is negative. Because rheumatic diseases are rare, the negative predictive value of the ANA test would have to approach 100% to be useful at predicting that a patient does not have a rheumatic disease when the ANA is negative, as in the case of SLE. Since this is not the case, the usefulness of the negative predictive value of the ANA test for all rheumatic diseases is nil. Therefore, a negative ANA test does not indicate the absence of a rheumatic disease.

Topics for further research

Biomarkers are molecular tests that can be used for purposes of diagnosis, treatment and/or prognosis.21 As detailed above, the ANA is a poor biomarker for guiding a clinician to or from a diagnosis of a rheumatic disease. Discovery of new biomarkers or combinations of currently available biomarkers that are better able to predict the presence or absence of a rheumatic disease are needed. However, as rheumatic diseases are both rare and diverse in aetiology, a single biomarker with the ability to predict the presence or absence of a rheumatic disease is unlikely to exist. By contrast, multiple biomarkers are in preclinical development that may aid in the diagnosis of individual rheumatic diseases.21 These tests will need to be used in the correct clinical scenario to be useful, indicating that a careful history and physical exam will continue to be necessary for the appropriate identification of rheumatic disease for many years to come.

Search Strategy

A structured PubMed literature search for original articles and abstracts was performed focusing on childhood rheumatic diseases without restriction of the publication date. The search terms used in various combinations were ‘childhood’, ‘juvenile’, ‘systemic lupus erythematosus’, ‘Raynaud's syndrome’, ‘Sjögren syndrome’, ‘mixed connective tissue disease’, ‘dermatomyositis’, ‘systemic sclerosis’, ‘scleroderma’, ‘juvenile idiopathic arthritis’, ‘juvenile chronic arthritis’, ‘antiphospholipid antibody syndrome’, ‘ANA’ and ‘antinuclear antibodies’. The references lists of the identified papers were also reviewed to identify papers relevant to the topics covered in this article.

Clinical bottom line

  • Information gathered from the use of the ANA test in over half a century of research and clinical medicine has reduced the utility of this test to specific clinical scenarios most often encountered by rheumatologists.

  • The ANA test cannot be appropriately used as a screening test for rheumatic disease because of its poor sensitivity, specificity and positive predictive value.

  • The ANA test should only be performed by indirect immunofluorescence, and NOT by ELISA.

  • There is no single test that can accurately predict whether a child does or does not have an underlying rheumatic condition.

  • Without accurate diagnostic biomarkers, identification of rheumatic disease is dependent on pattern recognition using the careful history and physical examination skills of a trained clinician.

Quiz – test your knowledge. Answers are on page 75

  1. Which of the following statements best depicts the usefulness of the ANA test.

    1. The ANA test is nearly 100% sensitive in patients with systemic lupus erythematosus making it a good screening test for all rheumatic diseases.

    2. A child presenting with two swollen joints for over 6 weeks is not likely to have a rheumatic disease if they are ANA negative.

    3. A positive ANA test highlights underlying pathophysiology as it represents active autoimmune disease.

    4. The ANA test is neither sensitive nor specific for rheumatic diseases, making it an ineffective screening test to discriminate between a patient with and without a rheumatic disease.

    5. The ANA test is a precise test with little variation between laboratories making it easy to compare the prevalence of disease between different studies.

  2. Which statement best describes appropriate counselling for a patient with a positive ANA?

    1. If a patient does not have features of an autoimmune disease, a positive ANA is likely a benign ANA that can occur in healthy individuals.

    2. Positive ANA tests are not found in healthy individuals and warrants referral to a rheumatology clinic.

    3. Patients with a positive ANA test are more likely to have a rheumatic disease than not.

    4. A positive ANA test accurately predicts the presence of systemic lupus erythematosus.

    5. Family members of a patient with a positive ANA are at increased risk for developing an autoimmune disease.

  3. A positive ANA test has been described in patients with:

    1. Acute infections.

    2. Systemic lupus erythematosus.

    3. Healthy individuals.

    4. Malignancy.

    5. All of the above.

  4. In which of the following clinical settings should an ANA test be sent?

    1. A 9-year-old female whose mother has systemic lupus erythematosus who presents with 3 months of leg pain.

    2. An 8-year-old male being treated with 4 weeks of antibiotics for Lyme arthritis with persistent joint swelling after 2 weeks of therapy.

    3. A 13-year-old female with a photosensitive facial rash and thrombocytopenia.

    4. A 12-year-old female with a previously positive ANA who comes in with 3 months of right foot allodynia.

    5. None of the above.

  5. The following is false regarding antinuclear antibodies (ANA).

    1. ANAs are directed against nuclear complexes containing both protein and nucleic acid components.

    2. The presence of ANAs indicates the presence of autoimmune disease.

    3. The presence of lupus erythematosus cells in peripheral smears is dependent on the presence of ANAs coating nuclear material that is subsequently eaten by surrounding phagocytes.

    4. The higher the ANA titre, the less likely it is to occur in a healthy individual.

    5. A negative ANA test does not rule out rheumatic disease.

  6. In which of the following clinical scenarios is the ANA test not useful?

    1. A 12-year-old female with sun photosensitivity, thrombocytopenia and painless palatal ulcer.

    2. An 8-year-old male whose fingers turn white in the cold, and subsequently blue then red upon rewarming.

    3. A 13-year-old female with fatigue, arthralgias and abdominal pain for 3 months that prevents her from going to school.

    4. A 2-year-old female recently diagnosed with juvenile idiopathic arthritis.

    5. A 15-year-old male with arthritis who has proteinuria and cellular casts on urine evaluation.

Answers to the quiz on page 69

1. D

2. A

3. E

4. C

5. B

6. C

References

Footnotes

  • Contributors Both Lehn K Weaver and Edward M Behrens contributed to the planning, writing and editing of the work described in this article.

  • Funding Lehn K Weaver was supported by the NIH T-32 grant (HD043021). Edward M Behrens was supported by an Arthritis Foundation Innovative Research Grant, the Howard Hughes Medical Institute Early Career Physician Scientist Award, and an NIH/NIAID grant (K08AI079396).

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.