Current practices in antinuclear antibody testing ...

Article in press - uncorrected proof Clin Chem Lab Med 2009;47(1):102–108 2009 by Walter de Gruyter • Berlin • New York. DOI 10.1515/CCLM.2009.021

2007/346

Current practices in antinuclear antibody testing: results from the Belgian External Quality Assessment Scheme

Marjan Van Blerk1,*, Christel Van Campenhout1, Xavier Bossuyt2, Jean Duchateau2, Rene´ Humbel2, Genevie`ve Servais2, Jean-Paul Tomasi2, Adelin Albert3, Wim Coucke1 and Jean-Claude Libeer1 1

Department of Clinical Biology, Scientific Institute of Public Health, Brussels, Belgium 2 EQA Advisory Board, Belgium 3 Department of Medical Statistics, University of Lie`ge, Lie`ge, Belgium

Abstract Background: This study aimed to assess the state-ofthe-art of antinuclear antibody (ANA) testing as practiced in the Belgian and Luxembourg laboratories, using the results obtained in the Belgian National External Quality Assessment Scheme from 2000 to 2005. Methods: During this period, nine samples with different specificities were sent for analysis. Participants were surveyed for methodology used and were asked to report staining pattern and titer of ANAs. In 2002, an attempt was made to improve the comparability of quantitative ANA results by the provision of a commercial reference material and to relate observed differences to methodology. Results: With one exception, all participants employed a microscope-based indirect immunofluorescence assay with human epithelial cell line 2 cells. Most laboratories were accurate in describing the pattern. The percentage of unacceptable answers was greater for samples with borderline levels of antibody and for samples showing a cytoplasmic pattern. An improvement in the detection of anticentromere antibodies was observed. For all samples, a wide range of titers was reported. The provision of the secondary reference preparation led to improved inter-laboratory concordance. Comparison of methodology variables revealed a correlation between unstandardized titers and the power of the lamp of the microscope and the use of a dark room. Conclusions: The EQAS results presented in this work provide valuable insights into the state of the art of ANA testing as practiced in the Belgian and Luxembourg Laboratories and illustrate the important value of a national EQAS for ANA testing as a tool to *Corresponding author: Marjan Van Blerk, Wetenschappelijk Instituut Volksgezondheid, Klinische Biologie, J. Wytsmanstraat 14, 1050 Brussels, Belgium Phone: q32-2-6425383, Fax: q32-2-6425645, E-mail: [email protected] Received July 11, 2008; accepted October 13, 2008; previously published online December 10, 2008

improve performance and interlaboratory comparability of laboratory results. Clin Chem Lab Med 2009;47:102–8. Keywords: antinuclear antibodies; external quality assessment; indirect immunofluorescence; quality control; standardization.

Introduction Autoantibodies to nuclear antigens (antinuclear antibodies, ANAs) are useful as diagnostic markers for a variety of autoimmune diseases (1–6). Tests for ANAs are widely used in clinical laboratories and many different reagents and techniques are available. The most common method is a microscope-based indirect immunofluorescence assay (IIF-ANA) with cultured human epidermoid carcinoma (human epithelial cell line 2, HEp-2) cells as substrate for antibody binding (3, 5, 7). The precision and accuracy of the technique depends on the assay configuration, the quality control procedures, and the experience of the reader (4, 5, 7). A mandatory External Quality Assessment Scheme (EQAS) for ANA testing was introduced in Belgium in 1991 and has been in use since then. The program is organized by the Department of Clinical Biology of the Belgian Scientific Institute of Public Health and includes the qualitative and quantitative identification of antibodies to nuclear antigens, ANA, DNA, and extractable nuclear antigens (ENAs). This study aims to provide an assessment of the state-of-the-art of ANA testing as practiced in the Belgian and Luxembourg laboratories, using the results obtained in this EQAS from 2000 to 2005. In addition, the paper discusses an attempt made in 2002 to improve the comparability of quantitative results in ANA testing by the provision of a commercial reference material and describes an effort to relate observed differences to methodology.

Materials and methods Sent-out specimens Table 1 lists the exercise code of the nine samples under study and tabulates the clinical diagnosis, the ANA staining pattern and the presence of antibodies to DNA and nuclear and cytoplasmic antigens. Figure 1 (A-I) illustrates the IIFANA patterns on HEp-2 cells (Medica, Carlsbad, CA, USA). Sample 03S1 (Figure 1G) from a patient with the CREST (Calcinosis, Raynaud’s phenomenon, Esophageal dysmotility, Sclerodactyly, Telangiectasias) subset of progressive sys-

Brought to you by | ULB - Bibliotheque des Sciences Humaines (Universite Luxembourg Bruxelles) Authenticated Download Date | 2/6/15 1:28 PM

Article in press - uncorrected proof Van Blerk et al.: Belgian EQAS for ANA testing 103

Table 1 Summary of exercise code, clinical diagnosis, ANA staining pattern, and presence of antibodies to DNA and nuclear and cytoplasmic antigens of the samples included in the study. Year

Sample

Clinical diagnosis

ANA pattern

DNA

Antigen

2000

00S1

Renal insufficiency/ no autoimmune disease SLE SCLE Systemic sclerosis MCTD CREST Polymyositis Primary biliary cirrhosis Healthy volunteer

Nuclear matrix

–

–

Homogeneous Fine speckled Homogeneousqnucleolar Coarse speckled Centromere Granular cytoplasma Mitochondrial-like Negative

q – – – – – – –

RNP SS-A/Ro Scl-70 RNP CENP-B Jo-1 M2 –

2001 2002 2003 2004 2005

00S2 01S1 01S2 02S1 03S1 04S1 04S2 05S1

SLE, systemic lupus erythematosus; SCLE, subacute cutaneous lupus erythematosus; MCTD, mixed connective tissue disease.

Figure 1 IIF-ANA patterns on HEp-2 cells (Medica, Carlsbad, CA, USA) of the samples included in the study. (A) Nuclear matrix, 00S1. (B) Homogeneous pattern, 00S2. (C) Fine speckled pattern, 01S1. (D) Homogeneousqnucleolar pattern, 01S2. (E, F) Coarse speckled pattern, 02S1 and commercial reference material. (G) Centromere pattern, 03S1. (H) Granular cytoplasmic pattern, 04S1. (I) Mitochondrial-like pattern, 04S2. temic sclerosis had been sent earlier to the participating laboratories (surveys 1992, 1996, 1999).

Sample preparation and distribution Sample material was obtained after informed consent and tested negative for hepatitis B surface antigen and antibodies to hepatitis C virus and human immunodeficiency viruses 1 and 2. The test material was studied extensively by five expert laboratories using different techniques and only sent

to the laboratories after overall agreement of the expert group’s results. The specimens were preserved with 0.1% sodium azide, divided aseptically into plastic tubes, packaged in accordance with postal regulations and distributed by overnight mail. The laboratories were surveyed for methodology used and were asked to report staining pattern and titer of ANAs. They were encouraged to process the samples according to their usual procedures. Completed reports were to be postmarked within 2 weeks of the initial shipping date.


Article in press - uncorrected proof 104

Van Blerk et al.: Belgian EQAS for ANA testing

Data analysis and reporting of results to participants

The data returned by the laboratories were entered into a centralized and secured computer database at the Department of Clinical Biology of the Belgian Scientific Institute of Public Health. The laboratories were coded in a random manner and a preliminary report was sent to the participants within 1–2 weeks after responses had been received. Acceptable responses were determined from reference laboratory results. There was uniform agreement on strict confidentiality of results between the scheme organizer and participants. The results of each survey were discussed within the advisory board. A global report including the results of all laboratories and the comments of the advisory board was sent to the participants prior to the following survey.

Statistical analysis Results were expressed as median and range for quantitative variables and as frequencies and percentages for qualitative variables. Mean values were compared by one-way analysis of variance (ANOVA) followed by multiple comparisons according to Tukey, or by the Mann-Whitney U-test (ns2) or Kruskal-Wallis (n)2) non-parametric test when normality assumptions could not be fulfilled. Relations between continuous variables were assessed by Pearson’s correlation coefficient and between discontinuous variables by Spearman’s rank correlation coefficient. Proportions were compared by the classical x2-test for contingency tables. Results were considered to be statistically significant at the 5% critical level (p-0.05).

Results Attempt to reduce the variation in ANA results by use of a commercial reference preparation Test material In 2002, a commercial reference material (semi-quantitative titratable ANA control, Immuno Concepts, Sacramento, CA, USA, Figure 1F) standardized against the World Health Organization standard for ANA-homogeneous pattern 66/233, was provided to the laboratories together with a sample from a patient with mixed connective tissue disease (MCTD, sample 02S1, Figure 1E) showing the same coarse speckled pattern as the reference material. The participants were asked to assay both samples together, to report their results in titers, and to convert the titer obtained for sample 02S1 to international units (IUs) by dividing the titer obtained on specimen 02S1 by the titer obtained on the commercial reference material and multiplying the quotient by 25 (the number of IU/mL in the commercial reference material). Questionnaire The participants were also requested to fill out a questionnaire covering the following topics: brand of microscope, type, power and lifespan of the lamp, magnification for reading slides, use of a dark room, screening dilution for serum testing, source and type of substrate, conjugate, and mounting medium, use of Evans blue.

Participation Participation in the scheme is mandatory for Belgian laboratories. The number of laboratories enrolled in the program dropped from 192 in 2000 to 157 in 2005, due to the fact that several laboratories merged with another laboratory and no longer performed ANA testing. IIF-ANA methods of the participating laboratories In 2001, one participant switched to enzyme-linked immunosorbent assay (ELISA) to screen for ANA wQUANTA Lite ANA ELISA (Inova Diagnostics, San Diego, CA, USA)x. All other laboratories employed IIFANA on HEp-2 cells or used the HEp-2000 substrate, consisting of HEp-2 cells transfected with SS-A/Ro cDNA. Nearly all laboratories utilized commercially available kits most often from Immuno Concepts (Sacramento, CA, USA; HEp-2, 0.8%; HEp-2000, 32.8%), Inova Diagnostics (27.5%), Euroimmun (Lu¨beck, Germany; 15.3%), Alphadia (Waver, Belgium; 10.7%), or Kallestad-Sanofi Diagnostics Pasteur (Marnes-La-Coquette, France; 6.9%) (data from 2005).

Table 2 Summary of percentages of acceptable responses and unacceptable answers for the samples included in the study. Sample

na

Acceptable results

Unacceptable answers

00S1

192

85.9%

00S2

186

94.1%

01S1

180

92.8%

01S2 02S1

178 177

97.8% 93.1%

03S1 04S1

174 157

97.6% 81.4%

04S2

155

87.3%

05S1

151

99.3%

Negative (7.8%) Centromere, nuclear dots, spindle fiber, nucleolar (6.3%) Negative (0.5%) Speckled, nuclear dots, centriole, mitochondrial, granular cytoplasma (5.4%) Negative (2.8%) Homogeneous, spindle fiber, PCNA, positive (4.4%) Nucleolar, mitochondrial, mixed (2.2%) Homogeneous, nuclear dots, nuclear matrix, cytoplasmic, positive (6.9%) Speckled (2.4%) Negative (6.4%) Speckled, homogeneous, nucleolar, lysosomal, ribosomal (12.2%) Negative (9.1%) Speckled, nuclear membrane, vimentin, positive (3.6%) Speckled (0.7%)

a

Number of participants. PCNA, proliferating cell nuclear antigen.



At the beginning of the program, 4.3% of the participants were still using rodent substrate, but this number diminished quickly, until only one laboratory continued to use rat liver cells up to 2002. Antinuclear staining on immunofluorescence Staining pattern Table 2 shows the percentage of acceptable responses for the samples included in this study. Results in the last column of the Table indicate the unacceptable answers and the percentage of false-negative results. The percentage of unacceptable answers ranged from 0.7% to 18.6% and was greater for sample 00S1 with borderline level of antibody and for the samples showing a cytoplasmic pattern: 18.6% and 12.7% of the participants failed to report the cytoplasmic staining for sample 04S1 and 04S2, respectively. No differences between kit manufacturers were found in the reporting of patterns (data not shown). During the program, an improvement in the detection of anticentromere antibodies (ACAs) was observed: 83.0% of the participants reported the centromere pattern for sample 03S1 in 1992, whilst 97.6% of the laboratories did so in 2003 (Table 3). The one laboratory that switched to ELISA to screen for ANAs (positive/negative) reported correctly on all samples. ANA titer The median ANA titer of the participants and the percentage of results within one and two twofold dilutions of the median ANA titer are summarized in Table 4. Usually, 10% of the IIF-ANA results were outside the acceptable limits wmedian titer"two twofold dilutions (3)x. There were wide ranges of end point titers for all samples (e.g., Figure 2) and all substrates (data not shown). Table 3 Improvement in the frequency of detecting anticentromere antibodies (sample 03S1). Year

Participants reporting centromere pattern, %

1992 1996 1999 2003

83.0 93.5 94.4 97.6

Table 4 Summary of the median titer of the participants and the percentages of results within one and two two-fold dilutions of the median titer for the samples included in the study. Sample

Median titer

%, median"1 titer

%, median"2 titers

00S1 00S2 01S1 01S2 02S1 03S1 04S1 04S2

160 1280 320 640 1280 1280 320 1280

88.8 72.3 75.6 82.0 78.3 73.9 59.2 85.5

99.4 92.1 93.6 97.5 94.3 91.7 91.3 93.4

Attempt to reduce the variation in ANA results by use of a commercial reference preparation ANA results Figure 2 shows the distribution of results for the raw, unstandardized titers and the standardized results. The percentage of results within one two-fold dilution of the median ANA titer of the participants increased substantially after standardization. Only 78.3% of the laboratories were within one twofold dilution of the median result before standardization, but 89.0% were within this range after standardization. Questionnaire Most participants replied to the questionnaire on the methodology used (ns152, 88.3%). Six different brands of microscope with several models were used: Leitz, Wetzlar, Germany; (69.0%), Zeiss, Go¨ttingen, Germany; (11.4%), Leica, Wetzlar, Germany; (7.6%), Olympus, Hamburg, Germany; (5.1%), Jena, Go¨ttingen, Germany; (4.4%), and Nikon, Amstelveen, the Netherlands; (2.5%). Almost all laboratories utilized mercury pressure lamps. Of these laboratories, 92.3% used a 50 watt lamp and 5.6% a 100 watt lamp. One participant employed a xenon lamp and three other participants a halogen lamp. The number of hours the bulb was lit at the moment of reading expressed towards the working life of the lamp ranged from 0% to 99% (mean"standard deviation: 49.5%"27.9%). The majority of participants used a =40- or =50-lens (20.9% and 69.3%, respectively). A dark room was used by 86% of the laboratories. The screening dilution varied from 1: 20 to 1: 100. Almost all laboratories (96.8%) used an initial screening dilution of 1:40 (29.5%) or 1:80 (67.3%). Substrate, conjugate, and mounting medium were almost always (95.1%) purchased from the same manufacturer wImmuno Concepts (29.4%), Inova (27.6%), Alphadia (17.2%), Euroimmun (16.6%), Kallestad (4.3%), Medica (2.5%), or other (2.5%)x. Approximately 80% of the participants employed Evans blue to counterstain the HEp-2 and HEp-2000 cells. The titers obtained by the participants employing a 50 watt lamp were significantly lower than the titers obtained by the participants employing a 100 watt lamp (median titer of 1:1280 vs. 1:2560, ps0.047). Significantly lower titers were also obtained by the participants not using a dark room (median titer of 1:640 vs. 1:1280, ps0.044). For the standardized results, by contrast, these differences were not observed. All other methodology variables studied showed no influence on titers or standardized results.

Discussion The Belgian EQAS for ANA testing was introduced in 1991. Participation in the scheme is mandatory for Belgian laboratories. The goal of the program is to improve laboratory performance by analyzing results, detecting deficiencies, evaluating methods, and disseminating results and comments on problems that are encountered during the surveys. This paper reports the results from the period 2000–2005.




Figure 2 Frequency distribution of ANAs expressed in dilution titer (upper panel) and after conversion to IU/mL by comparison with secondary reference material (lower panel).

Most laboratories were accurate in detecting moderate or strongly positive ANAs and in describing the pattern. The percentage of unacceptable answers was greater for samples with borderline levels of antibody and for samples showing a cytoplasmic pattern. The latter were included in the program to emphasize the fact that cytoplasmic antibody patterns can be detected on the HEp-2 and HEp-2000 substrate and should also be considered, as these cytoplasmic antibodies are also associated with autoimmune diseases. The correct interpretation of the centromere pattern is important as ACAs are of high diagnostic value for the CREST syndrome (3, 8). An improvement in the frequency of detecting these autoantibodies was observed over the course of the program. The reasons for the improvement are probably two-fold: on the one hand, all laboratories gained experience (9) in recognizing the centromere pattern, and on the other hand, the laboratories abandoned the use of rodent tissues where mitotic figures are infrequent (3, 10, 11). In this study, all substrates were found overall to be acceptable and to result in uniform specificity and sensitivity, but this was probably due to the fact that only very easily interpreted samples were used. In all specimens and for all IIF-ANA methods, there were wide variations in the degree of positivity of ANA measurement, as reflected by the broad range of ANA titers. This lack of concordance between laboratories has been demonstrated earlier (10, 12–19)

and is not at all surprising given the subjective interpretation of results being dependent on the experience and knowledge of the operator and the variations in IIF-ANA methodologies (7, 20, 21), including substrate and fixative variations, differences in conjugate and in microscope optics. To reduce the rather large between-laboratory variability in ANA titers, a commercial secondary reference material standardized against the World Health Organization standard for ANA-homogeneous pattern 66/233 was provided to the laboratories together with a questionnaire covering the following subjects: brand of microscope and type, power and lifespan of the lamp, magnification for reading slides, use of a dark room, screening dilution for serum testing, source and type of substrate, conjugate, and mounting medium, use of Evans blue. The provision of the secondary reference preparation led to improved inter-laboratory concordance with smaller dispersion of the results. The effectiveness of the reference material in improving comparability of results was probably due to the fact that the antibodies in the test serum and the secondary reference preparation were similar wanti-RNP, (13)x. The improved precision in ANA testing by use of a common standard serum has been shown previously (14–17). Comparison of methodology variables revealed a correlation between unstandardized titers and the power of the lamp of the microscope and the use of a dark room. In contrast, results did not relate to methodology after standard-



ization. These findings show the effect of the elimination of bias in ANA test results obtained by use of a common standard serum in each laboratory and illustrate that titered positive and negative controls should be routinely used in laboratories performing IIF-ANA testing to avoid intra- and inter-laboratory variability (21, 22). ANA screening systems that employ enzyme immunoassay (EIA) technology and that use antigens extracted from HEp-2 cells and/or specific purified or recombinant antigens as substrates (EIA-ANA) have recently become commercially available. These EIAANAs are less time-consuming than IIF-ANAs, require less knowledge and are suitable for automation. However, large differences in sensitivity and response to disease-specific ANAs have been reported, depending on the antigens used (6, 23–38). Discrepant negative EIA-ANA results may be due to the fact that relevant antigens are either lacking or are present in too low concentration to be detected (6, 24, 30, 33–35, 37). Thus, if EIA screening for ANAs is chosen as routine, the limitations of the EIA need to be known by the clinicians ordering these serology tests (3, 6, 24, 37, 38). In 2001, one of the participating laboratories started using the QUANTA Lite ANA ELISA, a hybrid assay that uses purified HEp-2 cell nucleolar and nuclear extracts, spiked with highly purified antigens: dsDNA, histones, Sm, RNP, centromere, Jo-1, SS-A/ Ro60, SS-B/La, PCNA, Scl-70, mitochondria M-2, and ribosomal-P protein (35). As could be expected from the design of the kit, the laboratory detected the autoantibodies in all samples tested. In conclusion, the EQAS results presented in this work provide valuable insights into the state-of-theart of ANA testing as practiced in the Belgian and Luxembourg laboratories and illustrate the important value of a national EQAS for ANA testing as a tool to improve performance and inter-laboratory comparability of laboratory results.

References 1. Nakamura RM, Tan EM. Update on autoantibodies to intracellular antigens in systemic rheumatic diseases. Clin Lab Med 1992;12:1–23. 2. Von Mu¨hlen CA, Tan EM. Autoantibodies in the diagnosis of systemic rheumatic diseases. Semin Arthritis Rheum 1995;24:323–58. 3. Kavanaugh A, Tomar R, Reveille J, Solomon DH, Homburger HA. Guidelines for clinical use of the antinuclear antibody test and tests for specific autoantibodies to nuclear antigens. Arch Pathol Lab Med 2000;124:71–81. 4. Egner W. The use of laboratory tests in the diagnosis of SLE. J Clin Pathol 2000;53:424–32. 5. Griesmacher A, Peichl P. Autoantibodies associated with rheumatic diseases. Clin Chem Lab Med 2001;39:189– 208. 6. Wiik AS. Anti-nuclear autoantibodies: clinical utility for diagnosis, prognosis, monitoring, and planning of treatment strategy in systemic immunoinflammatory diseases. Scand J Rheumatol 2005;34:260–8. 7. Nossent H, Rekvig OP. Antinuclear antibody screening in this new millennium: farewell to the microscope? Scand J Rheumatol 2001;30:123–6.

8. Moroi Y, Peebles C, Fritzler MJ, Steigerwald J, Tan EM. Autoantibody to centromere (kinetochore) in scleroderma sera. Proc Natl Acad Sci USA 1980;77:1627–31. 9. Tholen D, Lawson NS, Cohen T, Gilmore B. Proficiency test performance and experience with College of American Pathologists’ programs. Arch Pathol Lab Med 1995;119:307–11. 10. Roberts-Thomson PJ, McEvoy R, Gale R, Jovanovich S, Bradley J. Quality assurance of immunodiagnostic tests in Australasia. Pathology 1991;23:125–9. 11. Nakamura RM, Bylund DJ, Tan EM. Current status of available standards for quality improvement of assays for detection of autoantibodies to nuclear and intracellular antigens. J Clin Lab Anal 1994;8:360–8. 12. Taylor RN, Fulford KM. Assessment of laboratory improvement by the Center for Disease Control Diagnostic Immunology Proficiency Testing Program. J Clin Microbiol 1981;13:356–68. 13. Nakamura RM, Rippey JH. Quality assurance and proficiency testing for autoantibodies to nuclear antigen. Arch Pathol Lab Med 1985;109:109–14. 14. Feltkamp TE, Klein F, Janssens MB. Standardisation of the quantitative determination of antinuclear antibodies (ANAs) with a homogeneous pattern. Ann Rheum Dis 1988;47:906–9. 15. Roberts-Thomson PJ, McEvoy R, Gale R, Jovanovich S, Bradley J. Quality assurance of immunodiagnostic tests in Australasia: II. Five year review. Asian Pac J Allergy Immunol 1993;11:29–37. 16. Tan EM, Feltkamp TE, Smolen JS, Butcher B, Dawkins R, Fritzler MJ, et al. Range of antinuclear antibodies in ‘healthy’ individuals. Arthritis Rheum 1997;40:1601–11. 17. Hollingsworth PN, Bonifacio E, Dawkins RL. Use of a standard curve improves precision and concordance of antinuclear antibody measurement. J Clin Lab Immunol 1987;22:197–200. 18. Bizarro N, Tozzoli R, Tonutti E, Piazza A, Manoni F, Ghirardello A, et al. Variability between methods to determine ANA, anti-dsDNA and anti-ENA autoantibodies: a collaborative study with the biomedical industry. J Immunol Meth 1998;219:99–107. 19. Pham BN, Albare`de S, Guyard A, Burg E, Maisonneuve P. Impact of external quality assessment on antinuclear antibody detection performance. Lupus 2005;14:113–9. 20. Humbel RL. Detection of antinuclear antibodies by immunofluorescence. In: Van Venrooij WJ, Maini RN, editors. Manual of biological markers of disease. Deventer: Kluwer, 1993;A2:1–16. 21. Feltkamp TE. Antinuclear antibody determination in a routine laboratory. Ann Rheum Dis 1996;55:723–7. 22. Molden DP, Nakamura RM, Tan EM. Standardization of the immunofluorescence test for autoantibody to nuclear antigens (ANA): use of reference sera of defined antibody specificity. Am J Clin Pathol 1984;82:57–66. 23. Jaskowski TD, Schroder C, Martins TB, Mouritsen CL, Litwin CM, Hill HR. Screening for antinuclear antibodies by enzyme immunoassay. Am J Clin Pathol 1996;105: 468–73. 24. Emlen W, O’Neill L. Clinical significance of antinuclear antibodies. Comparison of detection with immunofluorescence and enzyme-linked immunosorbent assays. Arthritis Rheum 1997;40:1612–8. 25. Gniewek RA, Sandbulte C, Fox PC. Comparison of antinuclear antibody testing methods by ROC analysis with reference to disease diagnosis. Clin Chem 1997;43: 1987–9. 26. 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. 27. Olaussen E, Rekvig OP. Screening tests for antinuclear antibodies (ANA): selective use of central nuclear anti-



28.

29.

30.

31.

32.

33.


gens as a rational basis for screening by ELISA. J Autoimmun 1999;13:95–102. Tan EM, Smolen JS, McDougal JS, Butcher BT, Conn D, Dawkins R, et al. A critical evaluation of enzyme immunoassays for detection of antinuclear autoantibodies of defined specificities. 1. Precision, sensitivity, and specificity. Arthritis Rheum 1999;42:455–64. Reisner BS, DiBlasi J, Goel N. Comparison of an enzyme immunoassay to an indirect fluorescent immunoassay for the detection of antinuclear antibodies. Am J Clin Pathol 1999;111:503–6. Bossuyt X. Evaluation of two automated enzyme immunoassays for detection of antinuclear antibodies. Clin Chem Lab Med 2000;38:1033–7. Kern P, Kron M, Hiesche K. Measurement of antinuclear antibodies: assessment of different test systems. Clin Diagn Lab Immunol 2000;7:72–8. Ulvestad E, Kanestrøm A, Madland TM, Thomassen E, Haga HJ, Vollset SE. Evaluation of diagnostic tests for antinuclear antibodies in rheumatological practice. Scand J Immunol 2000;52:309–15. Bayer PM, Fabian B, Hu¨bl W. Immunofluorescence assays (IFA) and enzyme-linked immunosorbent assays (ELISA) in autoimmune disease diagnostics – technique, benefits, limitations and applications. Scand J Clin Lab Invest 2001;61(Suppl 235):68–76.

34. Hayashi N, Kawamoto T, Mukai M, Morinobu A, Koshiba M, Kondo S, et al. Detection of antinuclear antibodies by use of an enzyme immunoassay with nuclear HEp-2 cell extract and recombinant antigens: comparison with immunofluorescence assay in 307 patients. Clin Chem 2001;47:1649–59. 35. Ulvestad E. Performance characteristics and clinical utility of a hybrid ELISA for detection of ANA. APMIS 2001; 109:217–22. 36. Bernardini S, Infantino M, Bellincampi L, Nuccetelli M, Afeltra A, Iori R, et al. Screening of antinuclear antibodies: comparison between enzyme immunoassay based on nuclear homogenates, purified or recombinant antigens and immunofluorescence assay. Clin Chem Lab Med 2004;42:1155–60. 37. Fenger M, Wiik A, Høier-Madsen M, Lykkegaard JJ, Rozenfeld T, Hansen MS, et al. Detection of antinuclear antibodies by solid-phase immunoassays and immunofluorescence analysis. Clin Chem 2004;50:2141–7. 38. Tonutti E, Bassetti D, Piazza A, Visentini D, Poletto M, Bassetto F, et al. Diagnostic accuracy of Elisa methods as an alternative screening test to indirect immunofluorescence for the detection of antinuclear antibodies. Evaluation of five commercial kits. Autoimmunity 2004; 37:171–6.
