Journal of Clinical Virology 60 (2014) 165–167
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Short Communication
Comparison of two automated immunoassays for the determination of Puumalavirus IgM and IgG Astrid Muyldermans a,∗ , Katrien Lagrou a , Sofie Patteet a , Marjan Van Esbroeck b , Marc Van Ranst a , Veroniek Saegeman a a b
National Reference Center for Hantavirus, Department of Laboratory Medicine, University Hospitals Leuven, Herestraat 49, 3000 Leuven, Belgium Institute of Tropical Medicine Antwerp, Kronenburgstraat 43/3, 2000 Antwerp, Belgium
a r t i c l e
i n f o
Article history: Received 29 December 2013 Received in revised form 14 March 2014 Accepted 15 March 2014 Keywords: Hantaviruses Puumalavirus Nephropathia epidemica ELISA Sensitivity Specificity
a b s t r a c t Background: Puumala virus (PUUV), a member of the genus hantavirus, can cause nephropathia epidemica, a mild form of haemorrhagic fever with renal syndrome. The method of choice for the serodiagnosis of hantavirus infections are enzyme-linked immunosorbent assays (ELISAs). Objectives: Two commercially available PUUV ELISA kits were compared: Hantavirus (Puumala) IgM/IgG ELISA (Progen, Heidelberg, Germany) and PUUMALA IgM and IgG EIA AutoM (Reagena, Toivala, Finland). Study design: The sensitivity of the ELISA kits was evaluated with a panel of 55 serum samples from patients with an acute (n = 27) or past (n = 28) infection based on Progen or Reagena. A panel of 56 serum samples was composed to evaluate the specificity: samples with potentially false positive Progen Puumala IgM results (n = 12), seronegative samples for Puumala IgG/IgM with Progen (n = 20), and potentially cross reacting samples (n = 24). Discrepancies between the two assays were resolved with strip immunoblot. As measure of agreement between Progen and Reagena results, Cohen kappa coefficient was calculated. Results: Reagena showed a higher specificity (IgM 100%, IgG 100%) than Progen Puumala (IgM 73.21%, IgG 100%). However, Reagena showed a slightly lower sensitivity (IgM 96.15%, IgG 97.78%) compared with Progen (IgM 100%, IgG 100%). Substantial agreement with a Cohen kappa of 0.67 and 0.76 was found between the two assays for Puumala IgM and IgG respectively. Conclusions: This study showed a higher specificity of Reagena in comparison to Progen with a lower sensitivity, probably caused by selection bias. In spite of Reagena’s lower sensitivity, no acute infection was missed with this assay. © 2014 Elsevier B.V. All rights reserved.
1. Background Puumala virus (PUUV), a member of the genus hantavirus, can cause nephropathia epidemica (NE), a mild form of haemorrhagic fever with renal syndrome (HFRS). A rapid diagnosis of NE is important to differentiate PUUV from other acute febrile diseases. The method of choice for the serodiagnosis of hantavirus infections are enzyme-linked immunosorbent assays (ELISAs) according to their high sensitivity and the capability of processing large numbers of samples rapidly [1,2]. The ELISAs are predominantly based on the
Abbreviations: CMV, cytomegalovirus; EBV, Epstein Barr virus; ELISA, enzymelinked immunosorbent assay; HFRS, haemorrhagic fever with renal syndrome; IB, immunoblot; NE, nephropathia epidemica; Np, nucleocapsid protein; PUUV, Puumalavirus; RF, rheumatoid factor. ∗ Corresponding author. Tel.: +32 484 14 85 99; fax: +32 16 34 70 10. E-mail address: [email protected] (A. Muyldermans). http://dx.doi.org/10.1016/j.jcv.2014.03.009 1386-6532/© 2014 Elsevier B.V. All rights reserved.
PUUV nucleocapsid protein (Np), which is the major antigen in the acute phase antibody response [2–4]. 2. Objectives The aim of this study was to compare the performance characteristics of two commercially available PUUV ELISA kits: (1) Hantavirus (Puumala) IgM/IgG ELISA (Progen, Heidelberg, Germany); and (2) PUUMALA IgM and IgG EIA AutoM (Reagena, Toivala, Finland). Progen IgM ELISA is based on a direct-coated E. coli-expressed aminoterminal (aa 1–119) PUUV Np. Reagena IgM ELISA is a -capture ELISA with the full-length baculovirusexpressed PUUV Np added in the second step. Both IgG ELISAs are based on a direct-coated PUUV Np. 3. Study design Patients’ serum samples (n = 111) were used that had previously been analysed during routine diagnostic testing at the University
166
A. Muyldermans et al. / Journal of Clinical Virology 60 (2014) 165–167
Table 1 Results of Puumalavirus IgM and IgG detection on 111 serum samples with Progen and Reagena ELISA and Mikrogen immunoblot. Samples (n)
Progen IgM Pos.
Sensitivity Acute NE (27) Past NE (28) Specificity Potentially false positive Progen PUUV IgM (12) Seronegative Progen PUUV (20) Potentially crossreacting (24) Acute EBV (9) Acute CMV (8) Positive RF (4) Haemolytic (3)
IgG Grey
Neg.
Pos.
28
24 28
27
12
2 1
Referencea
Reagena IgM Grey
Neg. 3
Pos.
IgG Grey
25
IgM
Neg.
Pos.
Grey
2 28
16 22
7 1
Neg. 4 5
IgG
Pos.
Neg.
Pos.
26
1 28
22 23
Neg. 5 5
20
12 20
12 20
12 20
12 20
12 20
7 7 4 3
9 8 4 3
9 8 4 3
9 8 4 3
9 8 4 3
9 8 4 3
a The reference result was determined by either concordant test results with both ELISA testkits or the result of a confirmatory immunoblot for discrepant ELISA test results (CMV, cytomegalovirus; EBV, Epstein Barr virus; NE, nephropathia epidemica; PUUV, Puumalavirus; RF, rheumatoid factor).
Hospitals Leuven (Belgium) by Progen IgM/IgG ELISA (n = 82), or at the Institute of Tropical Medicine (Antwerp, Belgium), by Reagena IgM and IgG EIA (n = 5). Potentially crossreacting samples (n = 24) were included likewise (see further). The samples were stored at −20 ◦ C and thawed at room temperature immediately before analysis. To evaluate the sensitivity of the assays, a panel of 55 serum samples was selected from patients with an acute (n = 27) (IgM positive) or a past (n = 28) (exclusively IgG positive) infection as determined with Progen or Reagena ELISA. A second panel, aimed to evaluate the specificity of the assays consisted of 56 serum samples determined with Progen ELISA as potentially false positive (n = 12) (grey zone results of IgM and no seroconversion of IgG in two consecutive samples); samples with negative results with Progen ELISA (n = 20); and potentially crossreacting samples (n = 24) (samples positive for Epstein Barr virus (EBV) IgM (n = 9), cytomegalovirus (CMV) IgM (n = 8), rheumatoid factor (RF) (n = 4) and haemolytic samples (n = 3)). EBV positivity was determined with Enzygnost® Anti-EBV/IgMII (Siemens, Erlangen, Germany) on a Bep® III automate (Siemens) (≥1 Index); CMV positivity with ARCHITECT CMV IgM (Abbott Diagnostics, Lake Forest, IL, USA) on an Architect automate (Abbott) (OD/cut-off > 1); RF positivity with Rheumatoid Factor (Beckman Coulter, Brea, CA, USA) on an IMMAGE® 800 Immunochemistry System (Beckman Coulter) (>40 IU/mL). All 111 samples were tested with Hantavirus (Puumala) IgM/IgG ELISA (Progen) and PUUMALA IgM and IgG EIA AutoM (Reagena) as per the manufacturers’ instructions on a Bep® III automate. Grey zone results were considered positive in the calculation of the sensitivity and specificity. Concordant test results with both the Progen and the Reagena assay were considered as correct. Discrepant results were retested by a confirmatory immunoblot (IB), recomLine HantaPlus IgM or IgG (Mikrogen, Neuried, Germany), of which the results were considered as correct. As measure of agreement between Progen and Reagena results, the Cohen kappa coefficient was calculated categorising according to qualitative results: positive, grey zone, negative. Statistical analyses were performed using Analyse-it Version 2.26 Software (2008, Leeds, United Kingdom). 4. Results IgM ELISA. Discordance between Progen and Reagena IgM ELISA was found in 17 out of 111 analysed samples. Substantial agreement was observed with a Cohen kappa coefficient of 0.67. The problem area as shown in Table 1 is situated in the grey zone results as determined by Progen IgM ELISA (n = 14).
A 100% sensitivity was observed for Progen IgM ELISA and 96.15% for Reagena IgM ELISA. Reagena IgM ELISA (R-value = 0.746; range grey zone results: 0.8–1) generated one discordant result with Progen IgM (Q-value = 2.24; range grey zone results: 1–2) in a patient with a past PUUV infection, correctly detecting IgG. The PUUV specific symptoms were observed in March 2007, sampling was in May 2007. This might show that IgM is picked up for a longer time with Progen IgM ELISA than with Reagena IgM ELISA. A specificity of 73.21% and 100% for the Progen and Reagena IgM ELISA respectively was found. All 12 samples strongly suspected to have a persistent false positive Progen IgM ELISA result, indeed were negative on IB. The three remaining false positive Progen ELISA results were caused by cross-reaction: grey zone results from patients with positive EBV IgM (n = 2) and a positive result from a patient with positive CMV IgM (n = 1). If the grey zone results would be interpreted as negative, no acute infection would have been missed and Progen IgM ELISA specificity would increase to 98.21%. IgG ELISA. From the 111 analysed patient sera, we found discordance between Progen and Reagena IgG ELISA in 14 samples. Most of the discordant results are distributed around the Progen ELISA cut-off for positivity. Substantial agreement with a Cohen kappa coefficient of 0.76 was found. Sensitivity of the Progen and Reagena IgG ELISAs counted 100% and 97.78% respectively. One false negative result was found with Reagena ELISA. Nevertheless, IgM was correctly detected as positive, so no acute infection was missed. A 100% specificity was found with both assays for the IgG determination with the samples included in the panel for specificity determination. 5. Discussion A rapid diagnosis of NE is important to differentiate PUUV from other acute febrile diseases. This study aimed to evaluate the performance characteristics of two commercially available PUUV ELISA assays: Progen IgM/IgG and Reagena IgM/IgG. In spite of the relative lower sensitivity of the Reagena IgM assay (96.15%) compared to Progen IgM (100%), the Reagena assay did not miss any acute infections. Furthermore, selection bias could be responsible for the observed difference between Progen and Reagena; the majority of samples from the sensitivity panel was selected based on Progen ELISA positivity. That could explain the difference with previous studies where it appeared that -capture ELISAs showed a similar or slightly higher sensitivity than directcoated ELISAs [1,5]. Also the use of the full-length Np (whether or
A. Muyldermans et al. / Journal of Clinical Virology 60 (2014) 165–167
not baculovirus-expressed) would offer a higher sensitivity than the E. coli-expressed aminoterminal Np [1]. The Reagena IgM ELISA showed a 100% specificity (n = 56), while the specificity of the Progen IgM ELISA was 73.21%. Previous studies showed a better specificity for Progen IgM ELISA (93–100%) [1,6]. The observed difference could be explained by selection bias: we intentionally compromised the specificity of the Progen IgM ELISA by including samples suspected to be false positive in order to examine the weak point of this assay. This study showed that the relatively low specificity of Progen IgM could be ascribed to the inclusion of a high number of grey zone results which were not confirmed with IB. A potential approach to overcome this problem would be to interpret grey zone results as negative. In that case no acute infection would have been missed and Progen IgM ELISA specificity would increase to 98.21%. Hujakka et al. reported similar findings; if grey-zone results were considered positive, a specificity of 93.33% was observed with the Progen ELISA, versus a specificity of 97% if they were considered negative [6]. However in the study of Hujakka et al. acute infections would be missed in the latter case. The sensitivity of Progen IgG and Reagena IgG was 100% and 97.78% respectively. One false negative Reagena IgG ELISA determination was found. Nevertheless the positive IgM was correctly detected, so no acute infection was missed. Furthermore, six IgG grey zone results were generated by the Reagena assay, while the IgG positive result correctly was picked up with the Progen assay. In five of those, positive IgM was correctly detected with both assays. This probably suggests a lower sensitivity of Reagena ELISA for IgG in acute infections without missing the diagnosis of an acute infection because of the high sensitivity of the Reagena IgM assay. The 100% specificity achieved with both IgG assays corresponds well to data from previous authors (range 98.64–100%) [1,5,7]. In summary this study showed that the Reagena ELISA has a higher specificity than the Progen ELISA with a little lower sensitivity, possibly caused by selection bias. In spite of Reagena’s lower sensitivity, no acute infections were missed.
167
Funding None. Competing interests None declared. Ethical approval Not required. Acknowledgements We thank the laboratory technicians of infectious serology (Laboratory Medicine, University Hospitals Leuven) for their experienced help in performing the analyses. References [1] Sjolander KB, Elgh F, Kallio-Kokko H, Vapalahti O, Hagglund M, Palmcrantz V, et al. Evaluation of serological methods for diagnosis of Puumala hantavirus infection (nephropathia epidemica). J Clin Microbiol 1997;35(12):3264–8. [2] Maes P, Clement J, Gavrilovskaya I, Van Ranst M. Hantaviruses: immunology, treatment, and prevention. Viral Immunol 2004;17(4):481–97. [3] Jonsson CB, Figueiredo LT, Vapalahti O. A global perspective on hantavirus ecology, epidemiology, and disease. Clin Microbiol Rev 2010;23(2):412–41. [4] Vapalahti O, Mustonen J, Lundkvist A, Henttonen H, Plyusnin A, Vaheri A. Hantavirus infections in Europe. Lancet Infect Dis 2003;3:653–61. [5] Kallio-Kokko H, Vapalahti O, Lundkvist A, Vaheri A. Evaluation of Puumala virus IgG and IgM enzyme immunoassays based on recombinant baculovirusexpressed nucleocapsid protein for early nephropathia epidemica diagnosis. Clin Diagn Virol 1998;10(1):83–90. [6] Hujakka H, Koistinen V, Eerikainen P, Kuronen I, Laatikainen A, Kauppinen J, et al. Comparison of a new immunochromatographic rapid test with a commercial EIA for the detection of Puumala virus specific IgM antibodies. J Clin Virol 2001;23(1–2):79–85. [7] Kelo E, Parviainen M, Sirola H, Vaheri A, Vapalahti O, Avsic-Zupanc T, et al. Enzyme immunoassay tests for detection of hantavirus infection and immunity. Eur Infect Dis 2012;6(1):9–13.
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Short Communication
Comparison of two automated immunoassays for the determination of Puumalavirus IgM and IgG Astrid Muyldermans a,∗ , Katrien Lagrou a , Sofie Patteet a , Marjan Van Esbroeck b , Marc Van Ranst a , Veroniek Saegeman a a b
National Reference Center for Hantavirus, Department of Laboratory Medicine, University Hospitals Leuven, Herestraat 49, 3000 Leuven, Belgium Institute of Tropical Medicine Antwerp, Kronenburgstraat 43/3, 2000 Antwerp, Belgium
a r t i c l e
i n f o
Article history: Received 29 December 2013 Received in revised form 14 March 2014 Accepted 15 March 2014 Keywords: Hantaviruses Puumalavirus Nephropathia epidemica ELISA Sensitivity Specificity
a b s t r a c t Background: Puumala virus (PUUV), a member of the genus hantavirus, can cause nephropathia epidemica, a mild form of haemorrhagic fever with renal syndrome. The method of choice for the serodiagnosis of hantavirus infections are enzyme-linked immunosorbent assays (ELISAs). Objectives: Two commercially available PUUV ELISA kits were compared: Hantavirus (Puumala) IgM/IgG ELISA (Progen, Heidelberg, Germany) and PUUMALA IgM and IgG EIA AutoM (Reagena, Toivala, Finland). Study design: The sensitivity of the ELISA kits was evaluated with a panel of 55 serum samples from patients with an acute (n = 27) or past (n = 28) infection based on Progen or Reagena. A panel of 56 serum samples was composed to evaluate the specificity: samples with potentially false positive Progen Puumala IgM results (n = 12), seronegative samples for Puumala IgG/IgM with Progen (n = 20), and potentially cross reacting samples (n = 24). Discrepancies between the two assays were resolved with strip immunoblot. As measure of agreement between Progen and Reagena results, Cohen kappa coefficient was calculated. Results: Reagena showed a higher specificity (IgM 100%, IgG 100%) than Progen Puumala (IgM 73.21%, IgG 100%). However, Reagena showed a slightly lower sensitivity (IgM 96.15%, IgG 97.78%) compared with Progen (IgM 100%, IgG 100%). Substantial agreement with a Cohen kappa of 0.67 and 0.76 was found between the two assays for Puumala IgM and IgG respectively. Conclusions: This study showed a higher specificity of Reagena in comparison to Progen with a lower sensitivity, probably caused by selection bias. In spite of Reagena’s lower sensitivity, no acute infection was missed with this assay. © 2014 Elsevier B.V. All rights reserved.
1. Background Puumala virus (PUUV), a member of the genus hantavirus, can cause nephropathia epidemica (NE), a mild form of haemorrhagic fever with renal syndrome (HFRS). A rapid diagnosis of NE is important to differentiate PUUV from other acute febrile diseases. The method of choice for the serodiagnosis of hantavirus infections are enzyme-linked immunosorbent assays (ELISAs) according to their high sensitivity and the capability of processing large numbers of samples rapidly [1,2]. The ELISAs are predominantly based on the
Abbreviations: CMV, cytomegalovirus; EBV, Epstein Barr virus; ELISA, enzymelinked immunosorbent assay; HFRS, haemorrhagic fever with renal syndrome; IB, immunoblot; NE, nephropathia epidemica; Np, nucleocapsid protein; PUUV, Puumalavirus; RF, rheumatoid factor. ∗ Corresponding author. Tel.: +32 484 14 85 99; fax: +32 16 34 70 10. E-mail address: [email protected] (A. Muyldermans). http://dx.doi.org/10.1016/j.jcv.2014.03.009 1386-6532/© 2014 Elsevier B.V. All rights reserved.
PUUV nucleocapsid protein (Np), which is the major antigen in the acute phase antibody response [2–4]. 2. Objectives The aim of this study was to compare the performance characteristics of two commercially available PUUV ELISA kits: (1) Hantavirus (Puumala) IgM/IgG ELISA (Progen, Heidelberg, Germany); and (2) PUUMALA IgM and IgG EIA AutoM (Reagena, Toivala, Finland). Progen IgM ELISA is based on a direct-coated E. coli-expressed aminoterminal (aa 1–119) PUUV Np. Reagena IgM ELISA is a -capture ELISA with the full-length baculovirusexpressed PUUV Np added in the second step. Both IgG ELISAs are based on a direct-coated PUUV Np. 3. Study design Patients’ serum samples (n = 111) were used that had previously been analysed during routine diagnostic testing at the University
166
A. Muyldermans et al. / Journal of Clinical Virology 60 (2014) 165–167
Table 1 Results of Puumalavirus IgM and IgG detection on 111 serum samples with Progen and Reagena ELISA and Mikrogen immunoblot. Samples (n)
Progen IgM Pos.
Sensitivity Acute NE (27) Past NE (28) Specificity Potentially false positive Progen PUUV IgM (12) Seronegative Progen PUUV (20) Potentially crossreacting (24) Acute EBV (9) Acute CMV (8) Positive RF (4) Haemolytic (3)
IgG Grey
Neg.
Pos.
28
24 28
27
12
2 1
Referencea
Reagena IgM Grey
Neg. 3
Pos.
IgG Grey
25
IgM
Neg.
Pos.
Grey
2 28
16 22
7 1
Neg. 4 5
IgG
Pos.
Neg.
Pos.
26
1 28
22 23
Neg. 5 5
20
12 20
12 20
12 20
12 20
12 20
7 7 4 3
9 8 4 3
9 8 4 3
9 8 4 3
9 8 4 3
9 8 4 3
a The reference result was determined by either concordant test results with both ELISA testkits or the result of a confirmatory immunoblot for discrepant ELISA test results (CMV, cytomegalovirus; EBV, Epstein Barr virus; NE, nephropathia epidemica; PUUV, Puumalavirus; RF, rheumatoid factor).
Hospitals Leuven (Belgium) by Progen IgM/IgG ELISA (n = 82), or at the Institute of Tropical Medicine (Antwerp, Belgium), by Reagena IgM and IgG EIA (n = 5). Potentially crossreacting samples (n = 24) were included likewise (see further). The samples were stored at −20 ◦ C and thawed at room temperature immediately before analysis. To evaluate the sensitivity of the assays, a panel of 55 serum samples was selected from patients with an acute (n = 27) (IgM positive) or a past (n = 28) (exclusively IgG positive) infection as determined with Progen or Reagena ELISA. A second panel, aimed to evaluate the specificity of the assays consisted of 56 serum samples determined with Progen ELISA as potentially false positive (n = 12) (grey zone results of IgM and no seroconversion of IgG in two consecutive samples); samples with negative results with Progen ELISA (n = 20); and potentially crossreacting samples (n = 24) (samples positive for Epstein Barr virus (EBV) IgM (n = 9), cytomegalovirus (CMV) IgM (n = 8), rheumatoid factor (RF) (n = 4) and haemolytic samples (n = 3)). EBV positivity was determined with Enzygnost® Anti-EBV/IgMII (Siemens, Erlangen, Germany) on a Bep® III automate (Siemens) (≥1 Index); CMV positivity with ARCHITECT CMV IgM (Abbott Diagnostics, Lake Forest, IL, USA) on an Architect automate (Abbott) (OD/cut-off > 1); RF positivity with Rheumatoid Factor (Beckman Coulter, Brea, CA, USA) on an IMMAGE® 800 Immunochemistry System (Beckman Coulter) (>40 IU/mL). All 111 samples were tested with Hantavirus (Puumala) IgM/IgG ELISA (Progen) and PUUMALA IgM and IgG EIA AutoM (Reagena) as per the manufacturers’ instructions on a Bep® III automate. Grey zone results were considered positive in the calculation of the sensitivity and specificity. Concordant test results with both the Progen and the Reagena assay were considered as correct. Discrepant results were retested by a confirmatory immunoblot (IB), recomLine HantaPlus IgM or IgG (Mikrogen, Neuried, Germany), of which the results were considered as correct. As measure of agreement between Progen and Reagena results, the Cohen kappa coefficient was calculated categorising according to qualitative results: positive, grey zone, negative. Statistical analyses were performed using Analyse-it Version 2.26 Software (2008, Leeds, United Kingdom). 4. Results IgM ELISA. Discordance between Progen and Reagena IgM ELISA was found in 17 out of 111 analysed samples. Substantial agreement was observed with a Cohen kappa coefficient of 0.67. The problem area as shown in Table 1 is situated in the grey zone results as determined by Progen IgM ELISA (n = 14).
A 100% sensitivity was observed for Progen IgM ELISA and 96.15% for Reagena IgM ELISA. Reagena IgM ELISA (R-value = 0.746; range grey zone results: 0.8–1) generated one discordant result with Progen IgM (Q-value = 2.24; range grey zone results: 1–2) in a patient with a past PUUV infection, correctly detecting IgG. The PUUV specific symptoms were observed in March 2007, sampling was in May 2007. This might show that IgM is picked up for a longer time with Progen IgM ELISA than with Reagena IgM ELISA. A specificity of 73.21% and 100% for the Progen and Reagena IgM ELISA respectively was found. All 12 samples strongly suspected to have a persistent false positive Progen IgM ELISA result, indeed were negative on IB. The three remaining false positive Progen ELISA results were caused by cross-reaction: grey zone results from patients with positive EBV IgM (n = 2) and a positive result from a patient with positive CMV IgM (n = 1). If the grey zone results would be interpreted as negative, no acute infection would have been missed and Progen IgM ELISA specificity would increase to 98.21%. IgG ELISA. From the 111 analysed patient sera, we found discordance between Progen and Reagena IgG ELISA in 14 samples. Most of the discordant results are distributed around the Progen ELISA cut-off for positivity. Substantial agreement with a Cohen kappa coefficient of 0.76 was found. Sensitivity of the Progen and Reagena IgG ELISAs counted 100% and 97.78% respectively. One false negative result was found with Reagena ELISA. Nevertheless, IgM was correctly detected as positive, so no acute infection was missed. A 100% specificity was found with both assays for the IgG determination with the samples included in the panel for specificity determination. 5. Discussion A rapid diagnosis of NE is important to differentiate PUUV from other acute febrile diseases. This study aimed to evaluate the performance characteristics of two commercially available PUUV ELISA assays: Progen IgM/IgG and Reagena IgM/IgG. In spite of the relative lower sensitivity of the Reagena IgM assay (96.15%) compared to Progen IgM (100%), the Reagena assay did not miss any acute infections. Furthermore, selection bias could be responsible for the observed difference between Progen and Reagena; the majority of samples from the sensitivity panel was selected based on Progen ELISA positivity. That could explain the difference with previous studies where it appeared that -capture ELISAs showed a similar or slightly higher sensitivity than directcoated ELISAs [1,5]. Also the use of the full-length Np (whether or
A. Muyldermans et al. / Journal of Clinical Virology 60 (2014) 165–167
not baculovirus-expressed) would offer a higher sensitivity than the E. coli-expressed aminoterminal Np [1]. The Reagena IgM ELISA showed a 100% specificity (n = 56), while the specificity of the Progen IgM ELISA was 73.21%. Previous studies showed a better specificity for Progen IgM ELISA (93–100%) [1,6]. The observed difference could be explained by selection bias: we intentionally compromised the specificity of the Progen IgM ELISA by including samples suspected to be false positive in order to examine the weak point of this assay. This study showed that the relatively low specificity of Progen IgM could be ascribed to the inclusion of a high number of grey zone results which were not confirmed with IB. A potential approach to overcome this problem would be to interpret grey zone results as negative. In that case no acute infection would have been missed and Progen IgM ELISA specificity would increase to 98.21%. Hujakka et al. reported similar findings; if grey-zone results were considered positive, a specificity of 93.33% was observed with the Progen ELISA, versus a specificity of 97% if they were considered negative [6]. However in the study of Hujakka et al. acute infections would be missed in the latter case. The sensitivity of Progen IgG and Reagena IgG was 100% and 97.78% respectively. One false negative Reagena IgG ELISA determination was found. Nevertheless the positive IgM was correctly detected, so no acute infection was missed. Furthermore, six IgG grey zone results were generated by the Reagena assay, while the IgG positive result correctly was picked up with the Progen assay. In five of those, positive IgM was correctly detected with both assays. This probably suggests a lower sensitivity of Reagena ELISA for IgG in acute infections without missing the diagnosis of an acute infection because of the high sensitivity of the Reagena IgM assay. The 100% specificity achieved with both IgG assays corresponds well to data from previous authors (range 98.64–100%) [1,5,7]. In summary this study showed that the Reagena ELISA has a higher specificity than the Progen ELISA with a little lower sensitivity, possibly caused by selection bias. In spite of Reagena’s lower sensitivity, no acute infections were missed.
167
Funding None. Competing interests None declared. Ethical approval Not required. Acknowledgements We thank the laboratory technicians of infectious serology (Laboratory Medicine, University Hospitals Leuven) for their experienced help in performing the analyses. References [1] Sjolander KB, Elgh F, Kallio-Kokko H, Vapalahti O, Hagglund M, Palmcrantz V, et al. Evaluation of serological methods for diagnosis of Puumala hantavirus infection (nephropathia epidemica). J Clin Microbiol 1997;35(12):3264–8. [2] Maes P, Clement J, Gavrilovskaya I, Van Ranst M. Hantaviruses: immunology, treatment, and prevention. Viral Immunol 2004;17(4):481–97. [3] Jonsson CB, Figueiredo LT, Vapalahti O. A global perspective on hantavirus ecology, epidemiology, and disease. Clin Microbiol Rev 2010;23(2):412–41. [4] Vapalahti O, Mustonen J, Lundkvist A, Henttonen H, Plyusnin A, Vaheri A. Hantavirus infections in Europe. Lancet Infect Dis 2003;3:653–61. [5] Kallio-Kokko H, Vapalahti O, Lundkvist A, Vaheri A. Evaluation of Puumala virus IgG and IgM enzyme immunoassays based on recombinant baculovirusexpressed nucleocapsid protein for early nephropathia epidemica diagnosis. Clin Diagn Virol 1998;10(1):83–90. [6] Hujakka H, Koistinen V, Eerikainen P, Kuronen I, Laatikainen A, Kauppinen J, et al. Comparison of a new immunochromatographic rapid test with a commercial EIA for the detection of Puumala virus specific IgM antibodies. J Clin Virol 2001;23(1–2):79–85. [7] Kelo E, Parviainen M, Sirola H, Vaheri A, Vapalahti O, Avsic-Zupanc T, et al. Enzyme immunoassay tests for detection of hantavirus infection and immunity. Eur Infect Dis 2012;6(1):9–13.
