Nephrol Dial Transplant (2019) 1–3

doi: 10.1093/ndt/gfz128

Diagnosing renal involvement in connective tissue disease:

interpretation of anti-nuclear autoantibody tests

NDT
Downloaded from https://academic.oup.com/ndt/advance-article/doi/10.1093/ndt/gfz128/5531267 by guest on 20 October 2020
Ailish Nimmo1, Charu Chopra2 and Robert W. Hunter1,3

DIGEST
1
Department of Renal Medicine, Royal Infirmary of Edinburgh, Edinburgh, UK, 2Department of Clinical Immunology, Royal Infirmary of
Edinburgh, Edinburgh, UK and 3Department of Cardiovascular Science, Queen’s Medical Research Institute, University of Edinburgh,
Edinburgh, UK

Correspondence and offprint requests to: Ailish Nimmo; E-mail: ailish.nimmo@googlemail.com

INTRODUCTION salt extraction, giving the name ENAs, but are now commonly
Connective tissue diseases (CTDs) are characterized by im- measured using a solid-phase enzyme-linked immunosorbent
mune system dysregulation, circulating autoantibodies and assay technique. These include Sjögren’s syndrome–related an-
multisystem disease. Diagnosis is made using both clinical and tigen A (SSA; also known as Ro), Sjögren’s syndrome–related
serological features. Renal involvement—manifesting as renal antigen B (SSB; also known as La), Smith (Sm), topoisomerase
failure or urine dipstick abnormalities—may be the presenting 1, ribonucleic acid (RNA) polymerase III, Jo-1 and U1 ribonu-
feature of CTD. Renal and general physicians need to have a cleic protein (U1RNP). The antigen panel used for an ENA
good understanding of how serological testing can help in the screen varies between laboratories. For specific clinical ques-
diagnosis of CTD. tions or in the case of diagnostic uncertainty, it can be helpful to
Here we consider how anti-nuclear antibody (ANA) and ex- discuss extended screening with a local immunologist.
tractable nuclear antigen (ENA) testing can aid the diagnosis of
CTD in patients with renal disease. We review how these assays Principles of test interpretation
are performed and the principles of serological testing. We With any test, it is important to understand the sensitivity
summarize the clinical features, renal pathology and diagnostic and specificity and positive and negative predictive values (PPV
performance of ANA and ENA testing in a range of CTDs. (We and NPV, respectively). Sensitivity and specificity are often gen-
do not discuss lupus nephritis, which is comprehensively cov- erated from case–control studies and, depending on the clinical
ered elsewhere in the literature.) characteristics and disease prevalence, it is possible to over-
estimate the PPV [2]. Furthermore, these are not necessarily in-
How are ANA and ENA assays performed? tuitive concepts, and clinicians often overestimate the chance of
ANAs are directed towards components of the cell nucleus. disease with a positive test result [3].
They are detected by indirect immunofluorescence using the The diseases we discuss have a low prevalence and the PPV
human epidermoid carcinoma (HEp-2) cell line. HEp-2 cells, of ANA is poor if applied to the general population. Testing
with large nuclei containing many autoantigens, are well suited only adds value in the presence of a reasonably high pre-test
for the detection of autoantibodies. Patient serum is incubated probability, and the results need to be carefully interpreted with
with the cells before a fluorescent antibody to human immuno- knowledge of the clinical context.
globulin is added. Samples are tested first at a screening dilution In the remainder of this article we consider individual
(e.g. 1:80) and, if positive, are sequentially diluted until fewer CTDs. We summarize the salient clinical features (i.e. features
than half the cells are positively stained. This dilution is reported that might prompt autoantibody testing) and then discuss the
to reflect the strength of ANA binding present in the serum. diagnostic performance of ENA tests in that disease.
Weakly positive ANA titres are common and are often not
clinically significant [1]. In addition to the titre, most laborato- Sjögren’s syndrome
ries report the staining pattern of the antibody, for example, Sjögren’s syndrome is characterized by B-cell activation
homogeneous, speckled, centromere or nucleolar. This reflects and lymphocytic infiltration of exocrine glands, causing ker-
different nuclear proteins to which the antibody has bound atoconjunctivitis sicca and xerostomia. Renal involvement
(Table 1). occurs in up to 20% of patients and can predate the clinical
Specific nuclear antigens are associated with different CTDs. diagnosis in 30%. In 75% of cases there is an acute or
Such antigens were historically removed from the nucleus using chronic tubulointerstitial nephritis with a CD4þ T-cell
C The Author(s) 2019. Published by Oxford University Press on behalf of ERA-EDTA.
V
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 1
Table 1. Prevalence and associations of antinuclear autoantibodies in CTDs

CTD Autoantibody Prevalence in Sensitivity Specificity Staining pattern Antibody association

disease (%) (%) (%)
Sjögren’s Anti-SSA 50–70 8–70 87 Speckled nuclear Neonatal heart block
syndrome Anti-SSB 25–40 16–40 94 Speckled nuclear
Limited cutaneous Anti-centromere 30 31 97 Centromere Pulmonary hypertension, peripheral
scleroderma vascular disease
Anti-Th/To 5 Nucleolar Pulmonary fibrosis
Anti-U1RNP 5 Speckled nuclear Mixed connective tissue disorder
Diffuse cutaneous Anti-topoisomerase 1 30 26 99.5 Various Pulmonary fibrosis, cardiac
scleroderma (anti-Scl-70) involvement
Anti-RNA polymerase III 15 12 96 Nucleolar Greatest association with sclero-

Downloaded from https://academic.oup.com/ndt/advance-article/doi/10.1093/ndt/gfz128/5531267 by guest on 20 October 2020

derma renal crisis (in 52% of
patients)
Anti-U3RNP (fibrillarin) 5 Nucleolar More common in African Americans
and males; pulmonary hyperten-
sion and skeletal muscle
involvement
Mixed connective Anti-U1RNP 100 Speckled nuclear Usually required for diagnosis
tissue disorder
Polymyositis Anti-Jo-1 20 Cytoplasmic Polymyositis > dermatomyositis
Dermatomyositis Anti-synthetase syndrome: myositis,
idiopathic interstitial lung disease,
arthritis and Raynaud’s
phenomenon

infiltrate. There may be also tubular defects, typically distal hyperplasia and fibrinoid necrosis that progresses to a prolifera-
hypokalaemic renal tubular acidosis. Proximal tubular dys- tive arteriolopathy with characteristic onion-skin appearance in
function can occur, but rarely as full-blown Fanconi syn- arcuate and interlobular arteries.
drome. The remaining 25% of patients have glomerular Patients with SSc may develop other renal lesions, including
disease, most commonly immune complex–mediated mem- penicillamine-related membranous nephropathy, scleroderma-
branoproliferative glomerulonephritis (MPGN) [4]. associated vasculopathy and myeloperoxidase-positive
The most common ENAs are anti-SSA (in 50–70% of vasculitis [6].
patients) and anti-SSB (in 25–40%). While it is possible to have The predominant ENAs are anti-centromere and anti-Th/
anti-SSA but negative anti-SSB, the converse is rare. Their pres- To in lcSSc and anti-topoisomerase 1 and anti-RNA polymer-
ence correlates to diagnosis at a younger age, longer disease du- ase III in dcSSc. Anti-U1RNP, U3RNP and PM-Scl are found in
ration and more severe exocrine gland involvement. Anti-SSA SSc overlap syndromes. Anti-centromere is specific (97%) for
immunoglobulin G can cross the placenta and cause neonatal SSc when tested in patients with other CTDs and has a PPV of
heart block. While anti-SSA can also be found in other CTDs 89.5%. Their presence is associated with peripheral arterial dis-
such as systemic lupus erythematosus (SLE), anti-SSB is more ease and pulmonary arterial hypertension. Anti-Th/To has
specific for Sjögren’s syndrome (Table 1) [5]. been less widely studied and the sensitivity and specificity are
not accurately known, but they are present in 5% of patients
Systemic sclerosis and associated with pulmonary hypertension and fibrosis [7].
Anti-topoisomerase 1 is highly specific (99.5%), with an excel-
Systemic sclerosis (SSc) is characterized by an overproduc-
lent PPV of 98% but a lower sensitivity (26%). It is associated
tion of extracellular matrix proteins and collagen resulting in
with cardiac involvement and pulmonary fibrosis. Anti-RNA
tissue fibrosis. Vascular proliferation leads to an obliterative
polymerase III has the strongest association with scleroderma
vasculopathy. Features include Raynaud’s phenomenon, pul-
renal crisis, found in 50% of patients developing this
monary fibrosis, pulmonary hypertension, oesophageal dysmo-
complication [8].
tility and scleroderma renal crisis. It is classified as diffuse
cutaneous SSc (dcSSc; skin involvement proximal to the elbows
and knees) or limited cutaneous SSc [lcSSc; distal skin involve- Polymyositis and dermatomyositis
ment, also known as CREST (calcinosis, Raynaud’s phenome- These are inflammatory disorders of skeletal muscle, pre-
non, esophageal dysmotility, sclerodactyly, and telangiectasia) senting with symmetrical proximal muscle weakness.
syndrome]. Dermatomyositis occurs as a result of a complement-mediated
Scleroderma renal crisis develops in 5–10% of patients with microangiopathy and can be associated with underlying malig-
dcSSc. This is defined by new-onset accelerated hypertension nancy or an overlap with mixed connective tissue disorder.
and rapidly progressive renal impairment (with microangio- Polymyositis is characterized by an inflammatory infiltrate of
pathic haemolytic anaemia in 50% of patients). Renal biopsy CD8þ T cells in muscle fibres [9]. Renal manifestations are
shows a thrombotic microangiopathy with mucinoid most commonly acute kidney injury secondary to

2 A. Nimmo et al.
rhabdomyolysis and myoglobinuria. Chronic glomerulonephri- FUNDING
tis can occur. MPGN is most commonly described in polymyo- RWH is funded by a Clinical Research Career Development
sitis and membranous nephropathy in dermatomyositis [10]. Fellowship from the Wellcome Trust (award number 209562/
Anti-Jo-1 is an antibody against histidyl-tRNA synthetase Z/17/Z).
and is found in 20% of patients [9]. It can also be found in
patients with interstitial lung disease without myositis. CONFLICT OF INTEREST STATEMENT
Mixed connective tissue disorder None declared. The results presented in this article have not
been published previously in whole or part, including in ab-
Mixed connective tissue disorder has overlapping features of
stract format.
SLE, SSc and dermatomyositis. Anti-U1RNP is suggestive of
the diagnosis. Renal involvement occurs less frequently than in REFERENCES

Downloaded from https://academic.oup.com/ndt/advance-article/doi/10.1093/ndt/gfz128/5531267 by guest on 20 October 2020

other disorders. The most common lesion is membranous
nephropathy [7]. As mixed CTD can evolve into SLE or SSc, 1. Pisetsky DS. Antinuclear antibody testing—misunderstood or misbegotten?
Nat Rev Rheumatol 2017; 13: 495–502
patients can develop lupus nephritis or scleroderma-related 2. Loong T. Understanding sensitivity and specificity with the right side of the
nephropathy. brain. BMJ 2003; 327: 716–719
3. Whiting PF, Davenport C, Jameson C et al. How well do health professio-
nals interpret diagnostic information? A systematic review. BMJ Open 2015;
CONCLUSIONS 5: e008155
Renal involvement is common in CTDs and causes a range of 4. Evans R, Zdebik A, Ciurtin C et al. Renal involvement in primary Sjogren’s
glomerular, tubulointerstitial and vascular pathologies. syndrome. Rheumatology (Oxford) 2015; 54: 1541–1548
5. Fayyaz A, Kurien BT, Scofield RH. Autoantibodies in Sjogren’s syndrome.
Autoantibody testing can play an important diagnostic role in Rheum Dis Clin North Am 2016; 42: 419–434
patients with multisystem symptoms with renal impairment or 6. Shanmugam VK, Steen VD. Renal manifestations in scleroderma: evidence
dipstick abnormalities. However, their effective clinical utility for subclinical renal disease as a marker of vasculopathy. Int J Rheumatol
depends on a reasonably high working pre-test probability of 2010; 2010: 538589
disease, as serological tests are not generally well-performing 7. Koenig M, Dieude M, Senecal J. Predictive value of antinuclear autoantibod-
ies: the lessons of the systemic sclerosis autoantibodies. Autoimmun Rev
screening tools in wider populations. 2008; 7: 588–593
Close liaison with the immunology laboratory can be helpful 8. Nguyen B, Assassi S, Arnett FC et al. Association of RNA polymerase III
in evaluating patients whose clinical picture and serological test antibodies with scleroderma renal crisis. J Rheumatol 2010; 37: 1068, author
results are discordant. In some cases, testing for extended auto- reply 1069
antibody panels is helpful in reaching a firmer diagnosis. In 9. Dalakas MC, Hohlfeld R. Polymyositis and dermatomyositis. Lancet 2003;
362: 971–982
other cases, alternative diagnoses can lead to unexpected sero- 10. Yen T, Lai P, Chen C et al. Renal involvement in patients with polymyositis
logical test results. and dermatomyositis. Int J Clin Pract 2005; 59: 188–193
We summarize the diagnostic performance of ANA and
ENA tests in Table 1. Received: 9.4.2019; Editorial decision: 28.5.2019

Antinuclear antibodies in renal disease 3