• Users Online: 2156
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2015  |  Volume : 17  |  Issue : 1  |  Page : 7-11

Evaluation of ELISA and indirect immunofluorescence in the diagnosis of autoimmune diseases and their interpretation in the clinical situation

Department of Microbiology, PSG Institute of Medical Sciences and Research, Coimbatore, Tamil Nadu, India

Date of Web Publication16-Jun-2015

Correspondence Address:
Jacinth Angel
Department of Microbiology, PSG Institute of Medical Sciences and Research, Coimbatore, Tamil Nadu
Login to access the Email id

Source of Support: PSG Institute of Medical Sciences and Research, Conflict of Interest: None

DOI: 10.4103/0972-1282.158776

Rights and Permissions

Introduction: Systemic autoimmune diseases include conditions where the immune system fails to recognize self antigens leading to production of "auto antibodies" and subsequent damage to several organs and tissue systems, e.g., Systemic lupus erythematosus, Scleroderma or Systemic sclerosis, Dermatomyosits, Polyarteritis nodosa and Mixed connective tissue disease. Detection of antinuclear antibodies (ANA) has been found to be the single most important criteria for the diagnosis of connective tissue disorders. Materials and Methods: The commonly used methods to detect ANA like ELISA and indirect immunofluorescence (IFA) were evaluated in this study. Results: ELISA showed a sensitivity, specificity, positive predictive value, negative predictive value, false positive rate and false negative rate of 93%, 54%, 60%, 92%, 45% and 7% respectively when compared to IFA. Conclusions: ELISA is non-specific and less useful as a diagnostic test whereas ANA IFA is a more accurate test and the gold standard.

Keywords: Antinuclear antibodies - ELISA and indirect immunofluorescence, detection of antinuclear antibodies, connective tissue diseases diagnosis

How to cite this article:
Angel J, Thomas M, Appalaraju B. Evaluation of ELISA and indirect immunofluorescence in the diagnosis of autoimmune diseases and their interpretation in the clinical situation. J Acad Clin Microbiol 2015;17:7-11

How to cite this URL:
Angel J, Thomas M, Appalaraju B. Evaluation of ELISA and indirect immunofluorescence in the diagnosis of autoimmune diseases and their interpretation in the clinical situation. J Acad Clin Microbiol [serial online] 2015 [cited 2023 Sep 25];17:7-11. Available from: https://www.jacmjournal.org/text.asp?2015/17/1/7/158776

  Introduction Top

Connective tissue disease (CTD) is manifested as a systemic autoimmune response wherein auto antibodies are produced against nuclear self antigens like Smith, Ro 52, SSA, SSB, Centromere, Scl 70 etc., leading to host tissue destruction and organ malfunction. Hence the hallmark of diagnosis of CTD is the detection of the antinuclear antibodies (ANA). The most commonly used method to detect ANA is ELISA, though indirect immunofluorescence assay (IFA) is the gold standard. [1],[2] ANA profile is a more specific test which detects specific nuclear and cytoplasmic autoantibodies in each disease but is expensive. The emphasis on lab diagnosis of autoimmune diseases is important as the major presenting feature is pyrexia of unknown origin [3] and the course of CTDs being chronic with vague clinical symptoms which mostly overlap and make clinical diagnosis very difficult.

  Materials and Methods Top

The study was carried out between the period of June 2010 to September 2011 in the Department of Microbiology, PSG Institute of Medical Sciences and Research, Coimbatore. Serum samples (1-2 ml) were collected from 442 consecutive patients in whom connective tissue disorder was clinically suspected:

Inclusion criteria

Patients with history and clinical examination findings suggestive of CTDs.

Exclusion criteria

  1. Established Rheumatoid arthritis patients.
  2. Patients with CTD who have co existing infectious diseases or carcinomas.

A total of 40 serum samples from blood bank were tested as negative controls for both ANA ELISA and IFA.

All the serum samples were subjected to:

  1. Antinuclear antibodies IFA test at a dilution of 1:40 using Immco Immuglo kits (Manufactured by Immco Diagnostics, supplied by Anand Diagnostics, Bangalore) and ANA IFA was read using the Leica Immunofluorescence Microscope using the blue filter (470-550nm). Positives were seen as apple green fluorescing patterns against a dark background. Homogenous and peripheral pattern were interpreted as Systemic lupus erythematosus (SLE), speckled pattern as Mixed connective tissue disease (MCTD) or SLE or Sjogren's syndrome or Polymyositis or Scleroderma, nucleolar pattern was interpreted as Polymyositis or Scleroderma and centromere pattern as CREST. [13]
  2. Antinuclear antibodies ELISA using Esculisa kit. ELISA was read using the BioRad ELISA reader and values more than one were considered as positive. The positive and negative controls were regularly plotted on the Levey-Jennings (LJ) (LJ quality control chart - Westgard) for quality assurance.
  3. Antinuclear antibodies profile test was done using Euroimmun kit (Manufactured by Euroimmun, Germany) in some patients who were negative by ELISA and IFA but had strong clinical features suggestive of CTD and in some patients who had speckled pattern in IFA with vague clinical signs and symptoms.

Clinical details of the patients were obtained from the patient case files from the medical records department. CTD was diagnosed when there was positivity either by ANA, and/or anti-cytoplasmic antibodies by IFA or by ANA profile.

Ethical clearance to perform the study was obtained from the Institutional Ethical Clearance Committee.

  Results Top

During the period of June 2010 to September 2011, CTD was clinically suspected in 442 patients out of whom 74% (327) were hospitalised and 26% (115) were outpatients, attending the departments of Medicine and Dermatology. Serum samples were collected from all the patients suspected to have CTD.

Of the 442 samples, ANA were positive in 289 samples by the ELISA test and in 185 samples by the IFA test. ELISA and IFA tests were negative for ANA in all the 40 control sera tested.

The sensitivity, specificity, positive predictive value, negative predictive value, false positive rate and false negative rate for ANA ELISA taking IFA as gold standard was 93%, 54%, 60%, 92%, 45% and 7% respectively. ELISA showed high optical density (OD) values from 3 to 8 in 105 samples which were positive by IFA. All the values on LJ chart were within 2 SD in all the test runs.

A total of 193 out of 442 cases were confirmed to have CTD. This was based on antinuclear and anti-cytoplasmic antibodies positivity by IFA and ANA profile positivity in 174, 11 and 8 patients respectively. ELISA was positive in 172 of the total 185 IFA positive samples. The prevalence of CTD was found to be 44% among all suspected cases. Among the 193 patients 170 (88%) were females and 23 (12%) were males. The predominant age group was 21-30 years in 29% of SLE patients and 44% of MCTD patients. The most prevalent CTD was SLE (37%) followed by MCTD (18%). The predominant pattern in SLE was homogenous and in MCTD and Scleroderma the pattern was speckled. In 8 patients in whom CTD was clinically suspected, both ANA IFA and ELISA were negative. Among these three were diagnosed as Sjogren's syndrome, three as SLE and two as MCTD based on ANA profile [Table 1].
Table 1: ANA pattern and other characteristics of autoimmune diseases

Click here to view

Out of the 193 CTD patients, 48 (25%) could not be classified into any particular type of disease. IFA was done in all of them and 47 of them were positive and showed speckled pattern. One sample which was negative for ANA by both IFA and ELISA was positive for Ro 52 antibody by ANA profile.

The cost of ANA ELISA was 81/well as compared with 64/well for ANA IFA excluding costs for controls and overhead expenses [Table 2].
Table 2: IFA pattern in number and (%) in various diseases

Click here to view

  Discussion Top

In this study, out of 442 patients suspected of having CTD, 193 were diagnosed based on a combination of tests like ELISA, IFA, ANA profile and clinical features. The most common CTD was found to be SLE [Table 3].
Table 3: A 2 by 2 table showing ELISA positives against IFA positives shows how the sensitivity and specifi city were arrived at

Click here to view

ELISA was the commonly done test for detection of ANA and it continues to be so, [12] because ELISA is highly sensitive, equipments are easily available and performance of the test is easy, and not expensive. Sensitivity of ELISA in our study was 93% as is seen in other studies also (91%). [4],[5] A high false positive rate (43%), which is reported in other studies also (13-40%) [14] makes the ELISA test very non-specific. Many physicians are unaware of the high false positivity rate of ELISA and request for the same as the first line test. In this study the cut off OD value for ELISA was taken as 1, as suggested in the manufacturer's instructions, whereas in other studies when they considered 2 as the cut off value, the specificity increased and the sensitivity reduced. [12] Tozzoli et al. [6] in their guidelines for ANA testing suggest that ELISA should be used for screening only when it shows good clinical and analytical correlation with IFA, which is not so in our study. Greidinge and Hoffman [7] suggest that at least 3% of the healthy individuals show a false positive result when testing for ANA, because they have antinuclear antibodies but do not suffer from any CTD. In our study 40 normal sera tested as controls were negative by all the tests done. The specificity of ELISA test though is a cause for concern in our study (54%) as well as in other studies (86%, [4] 60% [5] ). ELISA can be used only as a screening test and further testing by IFA is definitely needed for diagnosis. Tozzoli et al. found that ANA IFA is easy to perform, cheaper than ELISA or Profile and gives a more specific diagnosis though it requires more expensive equipment and expertise to read the results. [6] The advantage and importance of IFA is that, it shows different types of antinuclear and anti-cytoplasmic staining patterns which gives more information on the auto antibodies produced against specific antigens seen in each disease, thus leading to a diagnosis of the same, making it a more specific and a gold standard test. [7],[8]

Among the 193 patients, females were the commonest affected in all the CTDs which is similar to the observations by Gunnarsson et al., [18] Paul et al., [19] and Gaubitz. [20] The age group between 21 and 30 years were commonly affected in SLE and 41-50 years in scleroderma which correlated with other studies. [20] The maximum number of patients with MCTD and Sjogren's were also in the third decade in our study but Gunnarsson et al. [18] and Gaubitz [20] report the predominant age group affected in MCTD and Sjogren's to be the fourth decade.

The most common CTD in our study was SLE (38%). Malaviya et al. [15] report a prevalence of 14-60/100,000, and studies from the west report a prevalence of 124/100,000. [16] However, a fair number of cases of SLE are encountered in any large hospital in India. [17] Our study showed a good correlation with other studies where majority of our patients showed homogenous pattern [Figure 2] (31%) followed by peripheral pattern (17%) and a combination of both patterns (6%). [13] White and Robbins have also observed homogenous pattern in 55% of their SLE patients. [12] We observed speckled pattern in few (18%) patients which is lower than White and Robbins (45%). Egner [16] in his article reports elevated ribosomal P titers in SLE. In our study 17% showed anti-cytoplasmic pattern which is due to antibodies produced against ribosomal P protein. Hence in our study IFA appears to be a very specific and feasible diagnostic test in SLE. But in some patients who show only speckled pattern it is difficult to arrive at a specific diagnosis as speckled pattern indicates more than one disease as described earlier. Therefore IFA should be correlated with clinical features for a diagnosis [Figure 3].

Systemic sclerosis was the next commonly occurring CTD (12%). The common IFA patterns seen in our study was speckled in 12 (50%) of the patients and centromere in nine (38%). Roberts-Thomson et al. have also reported the association of centromere pattern with CREST and speckled pattern with scleroderma. [9] Hence IFA, is again very useful in patients with scleroderma as centromere pattern is diagnostic of CREST and a speckled pattern with a good clinical correlation can pin point to scleroderma.

Polymyositis was seen in 3% of the patients. In all 6 cases nucleolar pattern coupled with coarse cytoplasmic staining was present as in other studies. [10],[11] Greidinge and Hoffman also suggest that when polymyositis is being suspected, attention should be paid to the presence of the anti-cytoplasmic antibodies [Figure 1]. [7]
Figure 1: Anticytoplasmic with nucleolar pattern

Click here to view
Figure 2: Homogenous pattern

Click here to view
Figure 3: Speckled pattern

Click here to view

In Sjogren's syndrome and MCTD, an ANA profile was needed for a diagnosis as both presented with overlapping signs and symptoms and in both speckled pattern was seen.

The serum of 25% patients who showed only speckled pattern, and did not have any specific clinical feature could not be subjected to ANA profile test due to unforeseen reasons, leading to an incomplete diagnosis.

Thus, IFA is more specific than ELISA, cheaper, easy to perform and useful in diseases like scleroderma, SLE and polymyositis. Though the diagnosis becomes difficult in case of a speckled pattern and diseases like Sjogren's syndrome and MCTD, it is important to emphasize that as per the ANA testing guidelines, ANA profile should be used only for confirmation in patients in whom screening by IFA is positive and in whom IFA is negative but show strong clinical features. [6]

  Conclusion Top

ELISA is a non-specific test in the diagnosis of autoimmune diseases. A false positive rate of 54% will give a misleading diagnosis to the physician which may actually be an infectious disease. Precious time will be lost for the patient. IFA on the other hand is more accurate than ELISA, and useful in diseases like Scleroderma, SLE and polymyositis. Therefore we recommend that IFA should be used as a diagnostic test and ELISA as a screening test. ANA Profile can be used in patients who are negative by IFA but are clinically suspected to have CTDs as well as in patients who have Speckled ANA pattern by IFA which is difficult to pin point a specific diagnosis.

  References Top

Homburger HA, Cahen YD, Griffiths J, Jacob GL. Detection of antinuclear antibodies: Comparative evaluation of enzyme immunoassay and indirect immunofluorescence methods. Arch Pathol Lab Med 1998;122:993-9.  Back to cited text no. 1
Satoh M, Vázquez-Del Mercado M, Chan EK. Clinical interpretation of antinuclear antibody tests in systemic rheumatic diseases. Mod Rheumatol 2009;19:219-28.  Back to cited text no. 2
Abdelbaky MS, Mansour HE, Ibrahim SI, Hassan IA. Prevalence of connective tissue diseases in egyptian patients presenting with Fever of unknown origin. Clin Med Insights Arthritis Musculoskelet Disord 2011;4:33-41.  Back to cited text no. 3
Divate S, Hardikar P, Bichile LS, Rajadhyaksha A. Clinical utility of screening for antinuclear antibodies by enzyme immunoassay - A preliminary study. J Assoc Physicians India 2004;52:290-3.  Back to cited text no. 4
Gniewek RA, Stites DP, McHugh TM, Hilton JF, Nakagawa M. Comparison of antinuclear antibody testing methods: Immunofluorescence assay versus enzyme immunoassay. Clin Diagn Lab Immunol 1997;4:185-8.  Back to cited text no. 5
Tozzoli R, Bizzaro N, Tonutti E, Villalta D, Bassetti D, Manoni F, et al. Guidelines for the laboratory use of autoantibody tests in the diagnosis and monitoring of autoimmune rheumatic diseases. Am J Clin Pathol 2002;117:316-24.  Back to cited text no. 6
Greidinge EL, Hoffman RW. Antinuclear antibody testing: Methods, indications, and interpretation. Lab Med 2003;34:2.  Back to cited text no. 7
Martinez-Lavin M, Vaughan JH, Tan EM. Autoantibodies and the spectrum of Sjögren′s syndrome. Ann Intern Med 1979;91:185-90.  Back to cited text no. 8
Roberts-Thomson PJ, Nikoloutsopoulos T, Cox S, Walker JG, Gordon TP. Antinuclear antibody testing in a regional immunopathology laboratory. Immunol Cell Biol 2003;81:409-12.  Back to cited text no. 9
Longo L, Fauci S, Dennis L, Hauser L, Jameson J. Harrison′s Principles of Internal Medicine. Ch. 319, 323. McGraw Hill: SLE; 2011. p. II.  Back to cited text no. 10
Parslow G, Stites DP, Terr AI, Imboden JB. Medical Immunology. 10 th ed., Ch. 31. McGraw Hill; 2003.  Back to cited text no. 11
White RH, Robbins DL. Clinical significance and interpretation of antinuclear antibodies. West J Med 1987;147:210-3.  Back to cited text no. 12
Egner W. The use of laboratory tests in the diagnosis of SLE. J Clin Pathol 2000;53:424-32.  Back to cited text no. 13
Check W. Making sense of the ANA hodgepodge. CAP Today 2009; September.  Back to cited text no. 14
Malaviya AN, Chandrasekaran AN, Kumar A, Shamar PN. Systemic lupus erythematosus in India. Lupus 1997; 6:690-700.  Back to cited text no. 15
Uramoto KM, Michet CJ Jr, Thumboo J, Sunku J, O′Fallon WM, Gabriel SE. Trends in the incidence and mortality of systemic lupus erythematosus, 1950-1992. Arthritis Rheum 1999;42:46-50.  Back to cited text no. 16
Kumar A. Indian guidelines on the management of SLE. J Indian Rheumatol Assoc 2002;10:80-96.  Back to cited text no. 17
Gunnarsson R, Molberg O, Gilboe IM, Gran JT; PAHNOR Study Group. The prevalence and incidence of mixed connective tissue disease: A national multicentre survey of Norwegian patients. Ann Rheum Dis 2011;70:1047-51.  Back to cited text no. 18
Paul BJ, Fassaludeen M, Nandakumar, Razia MV. Clinical profile of systemic lupus erythematosus in northern Kerala. J Rheumatol Assoc 2003;11:94-7.  Back to cited text no. 19
Gaubitz M. Epidemiology of connective tissue disorders. Rheumatology (Oxford) 2006;45 Suppl 3:iii3-4.  Back to cited text no. 20


  [Figure 1], [Figure 2], [Figure 3]

  [Table 1], [Table 2], [Table 3]

This article has been cited by
Chinmayi Pramod Bokey, Anil T. Deshmukh, Ramawatar Soni, Ashish A. Tayde
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Materials and Me...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded15    
    Comments [Add]    
    Cited by others 1    

Recommend this journal