Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Clinical Chemistry and Laboratory Medicine (CCLM)

Published in Association with the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM)

Editor-in-Chief: Plebani, Mario

Ed. by Gillery, Philippe / Lackner, Karl J. / Lippi, Giuseppe / Melichar, Bohuslav / Payne, Deborah A. / Schlattmann, Peter / Tate, Jillian R.

12 Issues per year


IMPACT FACTOR 2016: 3.432

CiteScore 2016: 2.21

SCImago Journal Rank (SJR) 2016: 1.000
Source Normalized Impact per Paper (SNIP) 2016: 1.112

Online
ISSN
1437-4331
See all formats and pricing
More options …
Volume 56, Issue 1

Issues

Performance analysis of automated evaluation of Crithidia luciliae-based indirect immunofluorescence tests in a routine setting – strengths and weaknesses

Wymke Hormann / Melanie Hahn / Stefan Gerlach / Nicola Hochstrate / Kai Affeldt / Joyce Giesen
  • Central Diagnostic Laboratory, Maastricht University Medical Center, Maastricht, The Netherlands
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Kai Fechner / Jan G.M.C. Damoiseaux
  • Corresponding author
  • Central Diagnostic Laboratory, Maastricht University Medical Center, Maastricht, The Netherlands
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
Published Online: 2017-06-23 | DOI: https://doi.org/10.1515/cclm-2017-0326

Abstract

Background:

Antibodies directed against dsDNA are a highly specific diagnostic marker for the presence of systemic lupus erythematosus and of particular importance in its diagnosis. To assess anti-dsDNA antibodies, the Crithidia luciliae-based indirect immunofluorescence test (CLIFT) is one of the assays considered to be the best choice. To overcome the drawback of subjective result interpretation that inheres indirect immunofluorescence assays in general, automated systems have been introduced into the market during the last years. Among these systems is the EUROPattern Suite, an advanced automated fluorescence microscope equipped with different software packages, capable of automated pattern interpretation and result suggestion for ANA, ANCA and CLIFT analysis.

Methods:

We analyzed the performance of the EUROPattern Suite with its automated fluorescence interpretation for CLIFT in a routine setting, reflecting the everyday life of a diagnostic laboratory. Three hundred and twelve consecutive samples were collected, sent to the Central Diagnostic Laboratory of the Maastricht University Medical Centre with a request for anti-dsDNA analysis over a period of 7 months.

Results:

Agreement between EUROPattern assay analysis and the visual read was 93.3%. Sensitivity and specificity were 94.1% and 93.2%, respectively. The EUROPattern Suite performed reliably and greatly supported result interpretation.

Conclusions:

Automated image acquisition is readily performed and automated image classification gives a reliable recommendation for assay evaluation to the operator. The EUROPattern Suite optimizes workflow and contributes to standardization between different operators or laboratories.

Keywords: autoantibodies; automation; Crithidia luciliae; dsDNA; indirect immunofluorescence; systemic lupus erythematosus (SLE)

Introduction

Laboratory diagnostics is of great importance for the diagnosis of systemic lupus erythematosus (SLE). Besides hematologic parameters like the sedimentation rate, leukopenia, or lymphopenia, autoantibodies against components of the cell nuclei (anti-nuclear antibodies, ANA) are the parameter of interest. Among ANA, antibodies against double-stranded DNA (anti-dsDNA) are one of the most reliable serological markers for SLE. Accordingly, they were listed as an important immunological feature for SLE diagnosis in the revised classification criteria by the American College of Rheumatology in 1982, and in the updated version from 1997 [1], [2]. In the most recent revision and validation of the criteria by the Systemic Lupus International Collaborating Clinics (SLICC) [3] anti-dsDNA antibodies were attributed an even more prominent role in SLE diagnosis. Also, anti-dsDNA antibody levels may reflect disease activity and kidney involvement; titer increases can precede a flareup of disease [4], [5].

The prevalence of anti-dsDNA antibodies in SLE is reported to range from 30% to 98% [6], [7], [8]. One cause of this variability is the use of different test systems for antibody detection, including radio immunoassay (RIA), Crithidia luciliae indirect immunofluorescence tests (CLIFT), enzyme-linked immunosorbent assays (ELISA) and other solid phase immunoassays. Each assay is supposed to bind a different subset of anti-dsDNA antibodies, e.g. of high or low avidity [6], [9]. The Farr assay, a certain form of RIA, is often considered as the gold standard for anti-dsDNA antibody detection [10], [11], [12]. However, as many clinics and laboratories avoid handling radioactive material, CLIFT and ELISA are today’s preferred methods [13]. ELISA is easy-to-use and cost effective, and therefore commonly used as screening assay. However, ELISA are often less disease specific as they also detect anti-dsDNA antibodies of low avidity, which are less relevant for SLE diagnostics, resulting in a higher number of false positive results. This is also reflected in the SLICC classification criteria which prescribe a clinical relevant antibody level threshold of >2 times the reference range if measured by ELISA [3]. Furthermore, ELISA are prone to a high performance variability between test kits and manufacturers, depending on the antigenic substrate used [14], [15], [16]. Even though great improvements have been made in recent years [9], [16], [17], confirmation of positive ELISA results with a second, more specific assay, like CLIFT or Farr assay, is advised [18]. The indirect immunofluorescence (IIF) test for anti-dsDNA antibodies is commonly based on C. luciliae as the substrate, a flagellate with a large mitochondrion, called a kinetoplast, containing a complex network of dsDNA, which is almost free of proteins and histones. This increases the assay’s specificity for SLE, as it is selective to high avide anti-dsDNA antibodies, which are most disease relevant [14], [19]. More sensitive CLIFT assays also detecting antibodies of low avidity can be achieved using modified buffer compositions.

A general drawback of the IIF test lies in the visual and subjective evaluation of test results which leads to high intra- and interlaboratory variability [15], [20], [21]. To reduce such variances and to optimize workflow, automated systems have been developed, which assist the operator in the handling of slides and result evaluation [22], [23], [24], [25], [26], [27], [28], [29], [30]. The first computer-aided immunofluorescence microscopy (CAIFM) systems were introduced into the field of ANA diagnostics for IIF on HEp-2 cells, which is regarded as the gold standard [31], [32]. As intended, automation could contribute to standardization of assay evaluation and fluorescence interpretation [33], [34], [35], [36], [37]. Especially software performed discrimination of positive and negative results greatly matches with visual evaluations by the operator. Several commercial platforms are on the market, among them the EUROPattern Suite encompassing the automated EUROPattern microscope and different software modules [27], [30]. The system offers fully automated image acquisition for a high number of different IIF substrates and is capable of automated IIF (pattern) classification for ANA (HEp-2 and HEp-20-10), ANCA (ethanol- and formalin-fixed), and CLIFT assays. Results will be suggested to the operator and verified by him.

Only few CAIFM systems are capable of automated CLIFT analyses [35], [38], [39], [40], [41]. Data on these systems are mainly focused on either their development or their performance characteristic. Only little is published on their performance in routine diagnostics in a clinical laboratory so far. We aimed to test the EUROPattern Suite performance in the routine setting of a clinical laboratory.

Materials and methods

Serum samples

Serum samples were collected at the Central Diagnostic Laboratory of the Maastricht University Medical Centre over a period of 7 months. All samples requested for anti-dsDNA analyses were considered, 332 serum samples in total. Fifteen samples had to be excluded due to insufficient material and five samples were assessed to be not evaluable by the visual reader leaving 312 samples in the final analysis. Collected serum samples included specimens sent to the laboratory for either diagnostic or follow-up purposes. The 312 serum samples were assigned to 281 patients, of which 256 patients provided one sample, 19 patients two samples and six patients three samples. Analysis was performed blinded. Matching with clinical data, being either SLE or non-SLE, was performed afterwards and only in case of positive results. This study was performed in accordance with the 1997 Declaration of Helsinki of the World Medical Association. For the analyses on patient material serum was obtained for diagnostic or follow-up purposes. As rest-serum was used in a semi-anonymous way, ethical approval was not necessary according to the Dutch guidelines.

Anti-dsDNA antibody detection

Three different CLIFT and evaluation procedures were applied for anti-dsDNA detection: (1) Routine analysis of anti-dsDNA antibodies at the Central Diagnostic Laboratory of Maastricht University Medical Centre was performed using the Fluorescent nDNA Test System (ImmunoConcepts, Sacramento, CA, USA) according to the manufacturer’s instructions [42]. Slides were prepared manually and evaluation was performed visually by trained experts utilizing a fluorescent microscope (Axioskop, Carl Zeiss Microscopy GmbH, Jena, Germany) with LED light source. (2) dsDNA EUROPattern test kit, (Euroimmun AG, Lübeck, Germany) and (3) dsDNA (sensitive) EUROPattern test kit (Euroimmun AG) were conducted at the Central Diagnostic Laboratory of Maastricht University Medical Centre according to the manufacturer’s instructions. Each dsDNA EUROPattern slide contains 10 reaction areas, each providing one biochip (2×2 mm fragment of coated cover slip glued into the reaction fields), coated with the protist. Slides were automatically incubated and washed using the IF Sprinter (Euroimmun AG, Lübeck, Germany). For initial screening, samples were applied at a dilution of 1:10. Fluorescein isothiocyanate (FITC)-labeled goat anti-human IgG was used for green fluorescent staining. Antiserum was supplied with Evans Blue used for red fluorescent counterstaining of the cells. Images of each incubated biochip were automatically acquired by the EUROPattern microscope, evaluation of the images was performed either visually by two trained experts or automatically by the EUROPattern software. Samples reactive at screening dilution were further analyzed at dilutions of 1:100 and 1:1000. Anti-dsDNA antibody titers ≥1:10 were considered positive. CLIFT 3 was only performed in case of discrepant results between CLIFT 1 and CLIFT 2.

EUROPattern Suite

The EUROPattern Suite (EUROIMMUN AG, Lübeck, Germany) comprises an automated fluorescence microscope and associated software modules for automated image acquisition and assay evaluation (for details see [40]). A focused image of each biochip is automatically taken by the EUROPattern microscope. A novel software feature induces acquisition of a second image of the same biochip with reduced exposure time in the case of overexposed images (low-light image). Automated titer estimation was performed by the software based on the original image data. During the classification process confidence values are calculated. They are a measure of how certain the system is about the proposed result (for details see [40]).

Statistical analysis

For statistical comparison the Mann-Whitney rank sum test was employed using SigmaPlot 13 (Systat Software, San Jose, USA). Values of p<0.05 were considered significant, whereas values of p<0.01 and p<0.001 were defined as very significant and highly significant, respectively.

Results

Appraisal of dsDNA EUROPattern test kit in the routine environment of the MUMC

Of 312 serum samples, 304 had a clear positive or negative result in the routine CLIFT 1 of the Central Diagnostic Laboratory of the Maastricht University Medical Centre. In 290 cases concordant results (35 positive, 255 negative) were obtained with the EUROPattern CLIFT 2 and subsequent visual evaluation by one expert (Table 1). Of the 14 discrepant samples, one was tested negative in the routine laboratory but positive in CLIFT 2 (diagnosis SLE); 13 were positive in routine testing but negative in CLIFT 2 – of which 12 samples belonged to patients who were finally diagnosed with SLE. These 12 samples were further analyzed with the more sensitive EUROPattern CLIFT 3. To avoid bias due to knowledge of former results, analysis was performed blinded with additional positive and negative samples. Reevaluation with CLIFT 3 revealed 11 positive results. With an overall agreement of 95.4% between CLIFT 1 and 2, and 99.0% between CLIFT 1 and 2/3, respectively, substantial equivalence of the assays can be assumed. The next step was now to assess the performance of the automated assay evaluation by EUROPattern.

Table 1:

Comparison of results by routine CLIFT 1 and CLIFT 2 (visual evaluation).

Automated CLIFT evaluation compared to visual interpretation

Out of 312 samples tested with CLIFT 2, 32 samples were found positive by visual read-out as well as automated EUROPattern classification (Table 2). Thirty-one samples were identified to be derived from patients diagnosed with SLE; one sample with low anti-dsDNA titer originated from a patient with primary anti-phospholipid syndrome. Concordant negative classifications of samples by the expert and EUROPattern were found in 259 cases, while discrepant results were reported for 21 samples. Thus, visual and automated classification showed an overall concordance of 93.3%. Concerning the estimated titers visual reading by one expert and EUROPattern classification agreed in 81.3% of the cases within the range of ±1 dilution step. EUROPattern determined a two steps higher titer for the remaining 18.8% of samples (Figure 1). However, these values vary with inter-reader differences, as agreement in ±1 titer step between two visual reads was also just 81.3%.

Table 2:

Comparison of results by software classification and visual evaluation of all 312 included serum samples.

Titer deviation between visual read and EUROPattern classification. Agreement is shown in %. Titer estimation was based on all three dilutions (1:10, 1:100 and 1:1000).
Figure 1:

Titer deviation between visual read and EUROPattern classification.

Agreement is shown in %. Titer estimation was based on all three dilutions (1:10, 1:100 and 1:1000).

In total, EUROPattern gave apparently false positive results in 19 cases and false negative results for two samples with respect to visual interpretation as reference standard (Table 3), yielding a relative sensitivity of 94.1% and a relative specificity of 93.2%.

Table 3:

Detailed results of samples classified false negative or false positive.

More detailed analysis of the discrepant samples revealed that the two being evaluated negative by EUROPattern had a low titer of only 1:10 as determined visually (and this was confirmed by a re-evaluation of the images by a second visual reader) (Table 3; upper part). Confirmed SLE diagnosis of the respective patients supported the correct evaluation by the experts in these cases. Titers given by EUROPattern in the remaining 19 discrepant cases ranged from 1:32 up to 1:1000 (Table 3). In three patients, the software-generated positive results were confirmed by positive routine testing and final SLE diagnosis; in one sample, the routine test result as well as the diagnosis were doubtful. In the other 15 samples, routine test results agreed with the negative classifications by the expert, clinical diagnosis excluded SLE, or only revealed cutaneous LE, in 14 of the respective patients.

Fluorescent images concerning these 15 patients showed either a strong, granular background staining of the cell body or a comma-like accentuation of the kinetoplast. Also, certain samples revealed a very bright background staining of the cells which perturbed the EUROPattern classification software at the default setting. In these cases a low light image, as depicted in Figure 2, was automatically acquired. Therefore, overexposure was not responsible for any misclassification, as all additional low light images were easily evaluable and could be classified as any image with default settings.

Exemplary images of CLIFT with a serum sample that causes bright background staining. The same biochip is shown without (A) and with (B) enabled low light function.
Figure 2:

Exemplary images of CLIFT with a serum sample that causes bright background staining.

The same biochip is shown without (A) and with (B) enabled low light function.

Looking at the confidence values revealed that the median confidence value of results matching with the visual read-out was considerably higher than the corresponding value for deviating results. Figure 3 shows the fluorescence images for whom positive and negative classification, respectively, revealed the highest confidence values. The distribution of confidence values for concordantly (true) positive and negative as well as discrepant (false positive and false negative) results are given in Figure 4. However, only the difference between the medians of concordantly positive and false positive results was (highly) significant (p<0.001).

Images of samples that were classified true positive (A) and true negative (B) with the highest confidence values of 0.90 and 1.00, respectively. Samples shown were diluted 1:10 in PBS-T.
Figure 3:

Images of samples that were classified true positive (A) and true negative (B) with the highest confidence values of 0.90 and 1.00, respectively.

Samples shown were diluted 1:10 in PBS-T.

Distribution of confidence values for true positive (TP), false positive (FP), true negative (TN) and false negative (FN) results. For statistical comparison the Mann-Whitney rank sum test was employed, non-significant (n.s.) p>0.05; ***p<0.001.
Figure 4:

Distribution of confidence values for true positive (TP), false positive (FP), true negative (TN) and false negative (FN) results.

For statistical comparison the Mann-Whitney rank sum test was employed, non-significant (n.s.) p>0.05; ***p<0.001.

Discussion

Antibodies directed against dsDNA are highly specific diagnostic markers for SLE [1], [2], [8]. A well-established assay with high disease specificity is the CLIFT [34], [43], [44], [45], [46], [47]. The sensitivity of CLIFT for SLE is reported to be approximately 30%–60% [14], [15], [48]; prevalence may vary with the used test system and the manufacturer. This may explain the differences between EUROPattern CLIFT 2 results and the results yielded by routine testing (CLIFT 1). Depending on, e.g. sample buffer composition, binding capacities for antibodies may vary. The hypothesis is that special buffer formulations allow binding of only the clinically relevant antibodies of high avidity making an assay less sensitive but more specific. However, as the clinical data in our study were only accessed in the case of any positive test result, no sensitivity and specificity data for the used CLIFT assays can be calculated. But further analysis of the 12 samples which showed discrepant results between routine CLIFT 1 and CLIFT 2 with the CLIFT 3 revealed that 11 samples reacted positive with the more sensitive test kit; highlighting the importance to know the specifications of an assay in use and to assess the preferred weighting of sensitivity and specificity [10].

Reliable automation and standardization are the future for clinical application of indirect immunofluorescence microscopy to avoid subjective interpretation of immunofluorescence results and to optimize workflow [33], [36], [49], [50]. The EUROPattern Suite is an advanced system for automated image acquisition and classification in indirect immunofluorescence diagnostics including CLIFT. Overall concordance, sensitivity and specificity of the system with respect to classical visual evaluation determined in this study are comparable to those reported in the literature for similar systems [35], [41].

Comparing the results of EUROPattern’s automated CLIFT evaluation with visual reading of the acquired fluorescence images by trained and experienced lab technicians, a relative sensitivity of 94.1% was found. In total numbers, only two samples were missed by the software which were classified positive with a titer of 1:10 by visual interpretation. These data are in line with an earlier study demonstrating the very high accuracy of the EUROPattern Suite [40]. Discrepant positive classification in 19 cases results in a relative specificity of 93.2%. Those deviating classifications of single samples can either be explained by a bright and granular background staining of the whole Crithidia’s cell body or a comma-like accentuation of the kinetoplast, which is judged negative by visual reading but positive by the software. Intensive fluorescence of the basal bodies within the analyzed cells, earlier described to cause misclassification, does not seem to influence automated classification in this study, showing that this issue was successfully corrected [40]. As the confidence values calculated by the software are significantly higher for samples which are concordantly positive compared to those that are false positive with respect to visually obtained results, further optimization of the classification process might be realized in the near future. Still, automated classification by the EUROPattern Suite is already very accurate and reliable. Of all negative classifications, 99.2% were correct compared to visual evaluation. Additionally, it is important to stress that in routine operation of the EUROPattern Suite, software results have to be validated by the operator before being included in the final result report. Negative samples can be verified batch-wise, but it is easily possible to further evaluate single samples by removing the check mark for negative. Samples will then be displayed along with the positive sera and verification is performed one by one. All images generated during sample analysis will be integrated into the patient’s data history, making them readily accessible in any case that might demand reevaluation of the patient’s diagnosis.

Additionally to positive-negative discrimination, titer estimation is an essential part of serological diagnostics to evaluate antibody levels in patient samples. Although automated titer calculation cannot be taken for granted in all CAIFM systems for CLIFT analysis [35], [38], [39], [41], this is an integral feature of the EUROPattern classification software module [40]. Automated titer estimation worked very well compared to visual evaluations with an agreement of 81.3% (expert 1) and 100.0% (expert 2) within the range of ±1 titer step, respectively.

Conclusions

Modern health care demands both, cost effectiveness and high quality results from laboratory diagnostics. Therefore, current methodological improvements in clinical laboratories focus on the standardization of diagnostics and workflow optimization. We show that computer-aided immunofluorescence microscopy with an advanced system like the EUROPattern Suite is able to perform reliable assay analysis and can be of great help in routine laboratories. Automated image acquisition is readily performed and automated image classification gives a reliable result proposal to the human operator. Adjustment of results is only seldom necessary, comprising almost exclusively false positive results, which are verified by the operator one after another, and thus can be easily revised.

Acknowledgments

We thank Tom Avery, Kathleen Mallet, Veerle Mengelers, and Angelina Misiou for their participation in the analyses of the IIF assays and Johanna Fraune for her assistance in the final figure design and for careful and critical reading of the manuscript.

References

  • 1.

    Tan EM, Cohen AS, Fries JF, Masi AT, Mcshane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7. PubMedCrossrefGoogle Scholar

  • 2.

    Hochberg MC. Updating the American college of rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997;40:1725. CrossrefPubMedGoogle Scholar

  • 3.

    Petri M, Orbai AM, Alarcõn GS, Gordon C, Merrill JT, Fortin PR, et al. Derivation and validation of the systemic lupus international collaborating clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum 2012;64:2677–86. PubMedWeb of ScienceCrossrefGoogle Scholar

  • 4.

    Ho A, Magder LS, Barr SG, Petri M. Decreases in anti-double-stranded DNA levels are associated with concurrent flares in patients with systemic lupus erythematosus. Arthritis Rheum 2001;44:2342–9. PubMedCrossrefGoogle Scholar

  • 5.

    Mastroianni-Kirsztajn G, Hornig N, Schlumberger W. Autoantibodies in renal diseases - clinical significance and recent developments in serological detection. Front Immunol 2015;6:221. PubMedWeb of ScienceGoogle Scholar

  • 6.

    Egner W. The use of laboratory tests in the diagnosis of SLE. J Clin Pathol 2000;53:424–32. CrossrefGoogle Scholar

  • 7.

    Cozzani E, Drosera M, Gasparini G, Parodi A. Serology of lupus erythematosus: correlation between immunopathological features and clinical aspects. Autoimmune Dis 2014;2014:e321359. Google Scholar

  • 8.

    Kavanaugh AF, Solomon DH. Guidelines for immunologic laboratory testing in the rheumatic diseases: Anti-DNA antibody tests. Arthritis Rheum 2002;47:546–55. CrossrefPubMedGoogle Scholar

  • 9.

    Antico A, Platzgummer S, Bassetti D, Bizzaro N, Tozzoli R, Villalta D. Diagnosing systemic lupus erythematosus: new-generation immunoassays for measurement of anti-dsDNA antibodies are an effective alternative to the Farr technique and the Crithidia luciliae immunofluorescence test. Lupus 2010;19:906–12. Web of SciencePubMedCrossrefGoogle Scholar

  • 10.

    Rouquette AM, Desgruelles C. Detection of antibodies to dsDNA: an overview of laboratory assays. Lupus 2006;15:403–7. PubMedCrossrefGoogle Scholar

  • 11.

    Hughes R, Ui-Hassan S. Anti-dsDNA antibodies: their role in the detection and monitoring of SLE. Clin Lab Int 2006;7:12–7. Google Scholar

  • 12.

    Agmon-Levin N, Damoiseaux J, Kallenberg C, Sack U, Witte T, Herold M, et al. International recommendations for the assessment of autoantibodies to cellular antigens referred to as anti-nuclear antibodies. Ann Rheum Dis 2014;73:17–23. Web of ScienceCrossrefPubMedGoogle Scholar

  • 13.

    Damoiseaux J, Agmon-Levin N, Van Blerk M, Chopyak V, Eriksson C, Heijnen I, et al. From ANA-screening to antigen-specificity: an EASI-survey on the daily practice in European countries. Clin Exp Rheumatol 2014;32:539–46. PubMedGoogle Scholar

  • 14.

    Haugbro K, Nossent JC, Winkler T, Figenschau Y, Rekvig OP. Anti-dsDNA antibodies and disease classification in antinuclear antibody positive patients: the role of analytical diversity. Ann Rheum Dis 2004;63:386–94. PubMedCrossrefGoogle Scholar

  • 15.

    Chiaro TR, Davis KW, Wilson A, Suh-Lailam B, Tebo AE. Significant differences in the analytic concordance between anti-dsDNA IgG antibody assays for the diagnosis of systemic lupus erythematosus-Implications for inter-laboratory testing. Clin Chim Acta 2011;412:1076–80. Web of ScienceCrossrefGoogle Scholar

  • 16.

    Biesen R, Dähnrich C, Rosemann A, Barkhudarova F, Rose T, Jakob O, et al. Anti-dsDNA-NcX ELISA: dsDNA-loaded nucleosomes improve diagnosis and monitoring of disease activity in systemic lupus erythematosus. Arthritis Res Ther BioMed Central Ltd 2011;13:R26. CrossrefGoogle Scholar

  • 17.

    Dieker J, Schlumberger W, McHugh N, Hamann P, Vlag J van der, Berden JH. Reactivity in ELISA with DNA-loaded nucleosomes in patients with proliferative lupus nephritis. Mol Immunol 2015;68:20–4. CrossrefPubMedWeb of ScienceGoogle Scholar

  • 18.

    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 rhuematic diseases. Am J Clin Pathol 2002;117:316–24. CrossrefGoogle Scholar

  • 19.

    Aarden L, de Groot E, Feltkamp T. Crithidia luciliae, a simple substrate for the determination of anti-dsDNA with the immunofluorescence technique. Ann N Y Acad Sci 1975;254:505–15. PubMedCrossrefGoogle Scholar

  • 20.

    Bizzaro 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 Methods 1998;219:99–107. PubMedCrossrefGoogle Scholar

  • 21.

    Rigon A, Soda P, Zennaro D, Iannello G, Afeltra A. Indirect immunofluorescence in autoimmune diseases: assessment of digital images for diagnostic purpose. Cytom Part B (Clinical Cytom) 2007;72B:472–7. CrossrefGoogle Scholar

  • 22.

    Egerer K, Roggenbuck D, Hiemann R, Weyer M-G, Büttner T, Radau B, et al. Automated evaluation of autoantibodies on human epithelial-2 cells as an approach to standardize cell-based immunofluorescence tests. Arthritis Res Ther 2010;12:R40. Web of ScienceCrossrefPubMedGoogle Scholar

  • 23.

    Perner P, Perner H, Müller B. Mining knowledge for HEp-2 cell image classification. Artif Intell Med 2002;26:161–73. PubMedCrossrefGoogle Scholar

  • 24.

    Hiemann R, Hilger N, Michel J, Nitschke J, Böhm A, Anderer U, et al. Automatic analysis of immunofluorescence patterns of HEp-2 cells. Ann N Y Acad Sci 2007;1109:358–71. PubMedCrossrefGoogle Scholar

  • 25.

    Knütter I, Hiemann R, Brumma T, Buttner T, Grossmann K, Cusini M, et al. Automated interpretation of ANCA patterns - a new approach in the serology of ANCA-associated vasculitis. Arthritis Res Ther 2012;14:R271. PubMedWeb of ScienceCrossrefGoogle Scholar

  • 26.

    Voigt J, Krause C, Rohwäder E, Saschenbrecker S, Hahn M, Danckwardt M, et al. Automated indirect immunofluorescence evaluation of antinuclear autoantibodies on HEp-2 cells. Clin Dev Immunol 2012;2012:651068. Web of ScienceGoogle Scholar

  • 27.

    Krause C, Ens K, Fechner K, Voigt J, Fraune J, Rohwa E. EUROPattern Suite technology for computer-aided immunofluorescence microscopy in autoantibody diagnostics. Lupus 2015;24:516–29. Web of SciencePubMedCrossrefGoogle Scholar

  • 28.

    Tonti S, Di Cataldo S, Bottino A, Ficarra E. An automated approach to the segmentation of HEp-2 cells for the indirect immunofluorescence ANA test. Comput Med Imaging Graph 2015;40:62–9. CrossrefPubMedWeb of ScienceGoogle Scholar

  • 29.

    Di Cataldo S, Tonti S, Bottino A, Ficarra E. ANAlyte: A modular image analysis tool for ANA testing with indirect immunofluorescence. Comput Methods Programs Biomed 2016;128:86–99. Web of ScienceCrossrefPubMedGoogle Scholar

  • 30.

    Fraune J, Gerlach S, Rentzsch K, Teegen B, Lederer S, Affeldt K, et al. Multiparametric serological testing in autoimmune encephalitis using computer-aided immunofluorescence microscopy (CAIFM). Autoimmun Rev 2016;15:937–42. PubMedWeb of ScienceCrossrefGoogle Scholar

  • 31.

    Meroni PL, Schur PH. ANA screening: an old test with new recommendations. Ann Rheum Dis. 2010;69:1420–2. CrossrefPubMedWeb of ScienceGoogle Scholar

  • 32.

    Rohwäder E, Locke M, Fraune J, Fechner K. Diagnostic profile on the IFA 40: HEp-20-10-an immunofluorescence test for reliable antinuclear antibody screening. Expert Rev Mol Diagn 2015;15:451–62. CrossrefWeb of SciencePubMedGoogle Scholar

  • 33.

    Tozzoli R, Antico A, Porcelli B, Bassetti D. Automation in indirect immunofluorescence testing: A new step in the evolution of the autoimmunology laboratory. Auto Immun Highlights 2012;3:59–65. CrossrefPubMedGoogle Scholar

  • 34.

    Bizzaro N, Antico A, Platzgummer S, Tonutti E, Bassetti D, Pesente F, et al. Automated antinuclear immunofluorescence antibody screening: a comparative study of six computer-aided diagnostic systems. Autoimmun Rev 2014;13:292–8. Web of ScienceCrossrefPubMedGoogle Scholar

  • 35.

    Melegari A, Bonaguri C, Russo A, Luisita B, Trenti T, Lippi G. A comparative study on the reliability of an automated system for the evaluation of cell-based indirect immunofluorescence. Autoimmun Rev 2012;11:713–6. CrossrefWeb of SciencePubMedGoogle Scholar

  • 36.

    Bonroy C, Verfaillie C, Smith V, Persijn L, De Witte E, De Keyser F, et al. Automated indirect immunofluorescence antinuclear antibody analysis is a standardized alternative for visual microscope interpretation. Clin Chem Lab Med 2013;51:1771–9. PubMedWeb of ScienceGoogle Scholar

  • 37.

    Hayashi N, Saegusa J, Uto K, Oyabu C, Saito T, Sato I, et al. Evaluation of a computer-aided microscope system and its anti-nuclear antibody test kit for indirect immunofluorescence assay. Rinsho Byori 2016;64:142–1451. PubMedGoogle Scholar

  • 38.

    Soda P, Onofri L, Iannello G. A decision support system for Crithidia Luciliae image classification. Artif Intell Med 2011;51:67–74. Web of SciencePubMedCrossrefGoogle Scholar

  • 39.

    Buzzulini F, Rigon A, Soda P, Onofri L, Infantino M, Arcarese L, et al. The classification of Crithidia luciliae immunofluorescence test (CLIFT) using a novel automated system. Arthritis Res Ther 2014;16:R71. PubMedWeb of ScienceCrossrefGoogle Scholar

  • 40.

    Gerlach S, Affeldt K, Pototzki L, Krause C, Voigt J, Fraune J, et al. Automated evaluation of crithidia luciliae based indirect immunofluorescence tests: A novel application of the EUROPattern-suite technology. J Immunol Res 2015;2015:742402. PubMedWeb of ScienceGoogle Scholar

  • 41.

    Lakos G, Gonzalez M, Flaherty D, Bentow C, Ibarra C, Stimson D, et al. Detection of anti-dsDNA antibodies by computer-aided automated immunofluorescence analysis. J Immunol Methods 2016;433:17–22. CrossrefWeb of SciencePubMedGoogle Scholar

  • 42.

    Avery TY, van de Cruys M, Austen J, Stals F, Damoiseaux JGMC. Anti-nuclear antibodies in daily clinical practice: prevalence in primary, secondary, and tertiary care. J Immunol Res 2014;2014:401739. Web of SciencePubMedGoogle Scholar

  • 43.

    Bossuyt X, Cooreman S, De Baere H, Verschueren P, Westhovens R, Blockmans D, et al. Detection of antinuclear antibodies by automated indirect immunofluorescence analysis. Clin Chim Acta 2013;415:101–6. CrossrefWeb of SciencePubMedGoogle Scholar

  • 44.

    Damoiseaux J, Mallet K, Vaessen M, Austen J, Cohen Tervaert JW. Automatic reading of ANCA-slides: evaluation of the AKLIDES system. Clin Dev Immunol 2012;2012:762874. PubMedWeb of ScienceGoogle Scholar

  • 45.

    Buchner C, Bryant C, Eslami A, Lakos G. Anti-nuclear antibody screening using HEp-2 cells. J Vis Exp 2014;88:e51211. Web of ScienceGoogle Scholar

  • 46.

    Tozzoli R, D’Aurizio F, Villalta D, Bizzaro N. Automation, consolidation, and integration in autoimmune diagnostics. Auto Immun Highlights 2015;6:1–6. PubMedCrossrefGoogle Scholar

  • 47.

    Csernok E, Damoiseaux J, Rasmussen N, Hellmich B, van Paassen P, Vermeersch P, et al. Evaluation of automated multi-parametric indirect immunofluorescence assays to detect anti-neutrophil cytoplasmic antibodies (ANCA) in granulomatosis with polyangiitis (GPA) and microscopic polyangiitis (MPA). Autoimmun Rev 2016;15:736–41. Web of ScienceCrossrefPubMedGoogle Scholar

  • 48.

    Wichainun R, Kasitanon N, Wangkaew S, Hongsongkiat S, Sukitawut W. Sensitivity and specificity of ANA and anti-dsDNA in the diagnosis of systemic lupus erythematosus: A comparison using control sera obtained from healthy individuals and patients with multiple medical problems. Asian Pacific J Allergy Immunol 2013;31:292–8. Google Scholar

  • 49.

    Lutteri L, Dierge L, Pesser M, Watrin P, Cavalier E. A paperless autoimmunity laboratory: myth or reality? Ann Biol Clin 2016;74:477–89. Web of ScienceGoogle Scholar

  • 50.

    Maenhout TM, Bonroy C, Verfaillie C, Stove V, Devreese K. Automated indirect immunofluorescence microscopy enables the implementation of a quantitative internal quality control system for anti-nuclear antibody (ANA) analysis. Clin Chem Lab Med 2014;52:989–98. Web of SciencePubMedGoogle Scholar

About the article

Corresponding author: Jan G.M.C. Damoiseaux, PhD, Central Diagnostic Laboratory, Maastricht University Medical Centre, P. Debyelaan 25, 6229 HX Maastricht, The Netherlands, Phone: +31 43 3876655


Received: 2017-04-15

Accepted: 2017-05-08

Published Online: 2017-06-23

Published in Print: 2017-11-27


Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

Research funding: None declared.

Employment or leadership: W.H., M.H., S.G., K.A., N.H., and K.F. are employees of EUROIMMUN medizinische Labordiagnostika AG, Lübeck, Germany.

Honorarium: None declared.

Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.


Citation Information: Clinical Chemistry and Laboratory Medicine (CCLM), Volume 56, Issue 1, Pages 86–93, ISSN (Online) 1437-4331, ISSN (Print) 1434-6621, DOI: https://doi.org/10.1515/cclm-2017-0326.

Export Citation

©2018 Walter de Gruyter GmbH, Berlin/Boston. Copyright Clearance Center

Comments (0)

Please log in or register to comment.
Log in