Article in press - uncorrected proof Clin Chem Lab Med 2010;48(1):xxx-xxx 2010 by Walter de Gruyter • Berlin • New York. DOI 10.1515/CCLM.2010.022
Short Communication
Diagnostic performance of rapid influenza antigen assays in patients infected with the new influenza A (H1N1) virus
Ileana Herzum*, Thomas Lutz, Friederike Koch, Roland Geisel and Andreas Gehrt Medical Laboratories Du¨sseldorf, Du¨sseldorf, Germany
Abstract Background: The new swine influenza (H1N1) requires fast and accurate diagnosis. Currently, there are few reports about the diagnostic performance of influenza antigen tests with regard to the H1N1 virus infection. We evaluated the reactivity of eight commercially available rapid antigen tests in samples from confirmed infected patients. Methods: From 23 patients tested positive by polymerase chain reaction (PCR), parallel testing with the antigen tests QuickVue Influenza AqB (Quidel Corporation), Binax NOW Flu A and B (Binax), Clearview Exact Influenza A and B (Inverness medical), TRUFLU (Meridian Bioscience), Directigen EZ Flu AqB (Becton Dickinson), Influenza A and B (Diagnostik Nord), Xpect Flu A and B (remel) and Influenza Antigen Quick (Generic Assays) was performed. Results: Only few antigen tests showed positive reactivity in six of the 23 samples. The virus load of the reactive samples as indicated by the PCR cycle threshold was high. Negative results of rapid influenza tests do not rule out infection with the new influenza A (H1N1) virus. Their analytical sensitivity with reference to the virus load is low and not reproducible. Conclusions: As reliable laboratory results during a pandemic are essential, the diagnostic use of rapid influenza tests seems to be questionable. Clin Chem Lab Med 2010;48 Keywords: H1N1 influenza; PCR; rapid antigen test; virus load.
The rapid spread of the new swine influenza (H1N1) challenged clinicians and laboratories in setting appropriate diagnostic approaches. The Robert Koch Institute (RKI), the German institute of the health department concerned with disease prevention, recommended the use of the real time polymerase chain reaction (RT-PCR) only because of its *Corresponding author: Ileana Herzum, MD, Medical Laboratories Du¨sseldorf, Germany, 40477 Du¨sseldorf, Germany Phone: q40-211-4978189, Fax: 40-211-4930612, E-mail:
[email protected] Received September 30, 2009; accepted November 10, 2009
reliability as a confirmation assay in patients with influenzalike illness (ILI). However, the German National Association of Statutory Health Insurance Physicians (KBV) allows the use of rapid point of care antigen tests in medical facilities, as well as PCR in patients with suspected new influenza according to the case definition in use. At present, there are few reports about the diagnostic performance of influenza antigen tests with respect to the H1N1 virus infection. We evaluated the reactivity of eight commercially available rapid antigen tests in samples from confirmed infected patients. From each patient, two dry nasopharyngeal swabs were sampled according to the preanalytic recommendation of the RKI for PCR testing within 24 h (1). All samples were stored at 48C and were not frozen. Between 22 June and 22 July 2009, 23 patients were confirmed as being H1N1 positive. The positive PCR results were reconfirmed by the RKI using a different PCR for the N1 gene of the virus.
The RT-PCR The RT-PCR is based on reverse transcription
and amplification of certain specific regions of the virus genome. Fluorescent marked oligonucleotides bind to the amplified DNA and the fluorescence intensity is detected in real time during amplification so that qualitative and quantitative detection occurs without reopening the sample. The greater the virus load of the sample, the fewer the number of amplification cycles that are needed before a fluorescent signal is detected. Isolation of viral RNA was performed with the COBAS Ampliprep device (Roche Diagnostics, Mannheim, Germany) and the Total Nucleic Aid Isolation (TNAI) reagent (Roche Diagnostics, Mannheim, Germany). We used the QuantiTect Multiplex RT–PCR NR (Qiagen, Hilden, Germany) containing HotStar Taq DNA polymerase and buffer. The primer sequences (TIB MOLBIOL, Berlin, Germany) were as follows: H1SWS: 59-CATTTgAAAggTTTgAgATATTCCC H1SWAs: 59-ATgCTgCCgTTACACCTTTgT H1SWP: 59-6FAM-ACAAgTTCATggCCCAATCATgACTCg-BBQ During reverse transcription at 508C the polymerase is inactive and becomes activated at 958C in the activation step. The amplification step takes 45 cycles, the process ends with the cooling step at 408C (2).
2010/531
Article in press - uncorrected proof 2 Herzum et al.: Diagnostic performance of rapid influenza antigen assays in patients infected with the new H1N1 virus
Rapid influenza antigen assays We used the patients intact second swab for parallel testing within 48 h with the antigen tests QuickVue Influenza AqB (Quidel Corporation, San Diego, CA, USA), Binax NOW Flu A and B (Inverness Medical Corporation, Bedford, UK), Clearview Exact Influenza A and B (Inverness Medical Corporation, Bedford, UK), TRUFLU (Meridian Bioscience, Cincinnati, OH, USA), Directigen EZ Flu AqB (Becton Dickinson, Franklin Lakes, NJ, USA), Influenza A and B (Diagnostik Nord, Schwerin), Xpect Flu A and B (Thermo Fisher Scientific, Remel Products, Lenexa) and Influenza Antigen Quick
(Generic Assays, Dahlewitz, Germany). These are immunochromatographic membrane assays containing an eluent solution as the fluid phase and test cards or dipsticks as the solid phase (Table 1). The antigen load of each swab was transferred into 0.5 mL normal saline solution by gentle rotation and squeezing against the tube (1). For each assay, 50 mL of the antigen solution was processed according to the manufacturer’s instructions. Only a few antigen tests showed positive reactivity in six of the 23 samples, despite appropriate reactivity of the control bands (Table 2). Sensitivity of the tests was 26%. Reac-
Table 1 Assay overview. Assay
Saline medium acceptable (manufacturer’s recommendations)
Solid phase
Incubation, min
QuickVue Influenza AqB Binax NOW Flu A and B Clearview Exact Influenza A and B TRUFLU Directigen EZ Flu AqB Influenza A and B Xpect Flu A and B Influenza Antigen Quick
Yes Yes Not mentioned Yes Yes Not mentioned Yes Yes
Strip Card Strip Strip Card Card Card Strip
10 15 15 15 15 10 15 10
Article in press - uncorrected proof Herzum et al.: Diagnostic performance of rapid influenza antigen assays in patients infected with the new H1N1 virus 3
Figure 1 Reactive assays.
tive bands in more than one antigen test were determined for four of the six samples (Figure 1). The virus load of the reactive samples, as indicated by the PCR cycle threshold, was high with detection of the florescent signal after an average of 22.31"2.16 and 30.21"4.76 cycles in the strongly and weakly reactive samples, respectively (Table 2). In the samples that tested negative by the antigen tests, PCR virus detection occurred after 30.51"3.48 cycles. The range between the lowest and highest virus load was 20.78–40.85 cycles (Figure 2). Our findings are in agreement with data from the few previously published evaluations of rapid antigen tests. The Centers for Disease Control (CDC) and Prevention reported an overall sensitivity of 40%–69%, depending on the virus load of the respiratory samples (3). Studies with other assays found sensitivities between 18% and 51% (4, 5). When cultured virus strains with high concentrations were used as the antigen solution, the detection rate of the assays increased up to 100%, as reported in another study (6). In our study, dilution with saline solution was kept to a minimum (500 mL), but it is possible that antigen recovery may be higher when using one dry swab for each assay in order to improve the reactivity. Also, there may have been some antigen loss due to storage, when compared with immediate processing of collected samples. Differences in the PCR
Figure 2 PCR reaction slopes of samples with high (A) and low (B) virus load.
Article in press - uncorrected proof 4 Herzum et al.: Diagnostic performance of rapid influenza antigen assays in patients infected with the new H1N1 virus
cycle threshold between the samples with reactive vs. nonreactive rapid tests were within an exponential range, rather than a linear range. Therefore, it appears that improvement in the test sensitivities by adaptation of the assay technique, rather than of handling, would be worthwhile since rapid and reliable testing for H1N1 should be available for regions without routine access to PCR methods. From our experiences, The German Society for Virology, the German Society for Hygiene and Microbiology and the German Association against Viral Diseases discourage the use of the rapid tests (7). We conclude that negative results of rapid influenza tests do not rule out infection with the new influenza A (H1N1) virus. Their analytical sensitivity with reference to the virus load is low and inconstantly reproducible. As reliable laboratory results during the pandemic are essential, the use of rapid influenza tests for diagnosis seems to be questionable.
Acknowledgements We thank the manufacturers for providing the test kits for the rapid tests.
Conflict of interest statement Authors’ conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article.
Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared.
References 1. ProbenentnahmeArzt.pdf (application/pdf-Objekt). Available from: http: / / www.rki.de / cln _ 091 / nn _ 200120 / DE / Content / InfAZ / I / Influenza/IPV/ProbenentnahmeArzt,templateIdsraw,propertys publicationFile.pdf/ProbenentnahmeArzt.pdf. 2. Panning M, Eickmann M, Landt O, Monazahian M, Olschla¨ S, Baumgarte S, et al. Detection of influenza A(H1N1)v virus by real-time RT-PCR. Euro Surveill 2009;36:19329. 3. Evaluation of rapid influenza diagnostic tests for detection of novel influenza A (H1N1) virus – United States, 2009. MMWR Morb Mortal Wkly Rep 2009;58:826–9. 4. Ginocchio CC, Zhang F, Manji R, Arora S, Bornfreund M, Falk L, et al. Evaluation of multiple test methods for the detection of the novel 2009 influenza A (H1N1) during the New York City outbreak. J Clin Virol 2009;45:191–5. 5. Faix DJ, Sherman SS, Waterman SH. Rapid-test sensitivity for novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med 2009;361:728–9. 6. Hurt AC, Baas C, Deng Y, Roberts S, Kelso A, Barr IG. Performance of influenza rapid point-of-care tests in the detection of swine lineage A (H1N1) influenza viruses. Influenza Other Respi Viruses 2009;3:171–6. 7. RKI Influenza. Available from: http://www.rki.de/cln_160/ nn_200120/DE/Content/InfAZ/I/Influenza/IPV/Stellungnahme__ Diagnostik.html.
