| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 University Clinic Magdeburg, Institute of Medical Microbiology, Magdeburg, Germany
2 Institute of Virology, Hannover Medical School, Hannover, Germany
Correspondence
B. Ghebremedhin
beniam.ghebremedhin{at}med.ovgu.de
Received June 22, 2008
Accepted November 14, 2008
Nowadays, influenza antigen detection test kits are used most frequently to detect influenza A or B virus to establish the diagnosis of influenza rapidly and initiate appropriate therapy. This study was conducted to evaluate the performance of the actim Influenza A&B test (Medix Biochemica). Overall, 473 respiratory specimens were analysed in the actim Influenza A&B test and the results were compared with those from an RT-PCR assay; 461 of these samples originated from paediatric patients aged 7 weeks to 6.5 years either with influenza-related symptoms or from the intensive care unit, and 12 samples originated from adults with underlying lung or haematological diseases. Diagnosis of influenza A or B virus could be established using the actim Influenza A&B test (9/473 samples for influenza A virus and 6/473 for influenza B virus). RT-PCR revealed 23 patients with influenza virus (13/473 for influenza A virus and 10/473 for influenza B virus). The sensitivity and specificity of the actim Influenza A&B test were 65 and 100 % compared with the RT-PCR assay. However, 32 external quality assessment samples containing seven different strains of influenza A subtypes H1N1 and H3N2 and the avian H5N1 were detected correctly by the actim Influenza A&B test. No cross-reactivity to a range of bacterial, fungal and other viral pathogens was observed. In conclusion, the actim Influenza A&B test is reliable for positive results due to its high specificity. Nevertheless, negative results from this test need to be confirmed by a more sensitive assay because of the low sensitivity observed with diagnostic samples.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
The rapid detection of influenza viruses is important to apply preventative strategies, initiate antiviral therapy, detect avian influenza viruses and exclude influenza-like bacterial and viral diseases (e.g. anthrax, smallpox). Moreover, it is important to determine not only whether influenza virus is present, but also which type of influenza virus is present (Noyola et al., 2000a; Chan et al., 2002; Effler et al., 2002; Poehling et al., 2002).
Although the gold standard reference method for the diagnosis of influenza is isolation of the virus, commercially available rapid antigen detection tests have the advantage of providing results much more quickly, i.e. within approximately 30 min. These tests, also referred to as near-patient or point-of-care tests, are mostly immunoassays that detect influenza virus antigen, although one test detects viral neuraminidase activity (Noyola et al., 2000b; Uyeki, 2003). They detect and distinguish between influenza A and B viruses, detect but do not distinguish between influenza A and B viruses or detect influenza A virus only. The tests vary in complexity, type of respiratory specimen acceptable for testing and the time needed to obtain results (Seno et al., 1990; Rodriguez et al., 2002; Uyeki, 2003; Hurt et al., 2007).
The actim Influenza A&B test (Medix Biochemica) is a point-of-care test utilizing influenza-specific mAbs for rapid diagnosis and differentiation of influenza A and B viruses. The aim of this study was to conduct a prospective evaluation of this test for qualitative, simultaneous and differential detection of influenza A and B antigens in different respiratory specimens. For this purpose, the actim Influenza A&B test was compared with an established in-house RT-PCR assay (Schweiger et al., 2000) for sensitivity, specificity and the time needed to obtain results.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
). A panel of other micro-organisms, including 11 bacteria (Citrobacter freundii, Enterobacter cloacae, Escherichia coli, Haemophilus sp., Klebsiella pneumoniae, Legionella pneumophila, Mycoplasma pneumoniae, Pseudomonas aeruginosa, Staphylococcus aureus, Staphylococcus epidermidis and Streptococcus pneumoniae), four Candida spp. (C. albicans, C. glabrata, C. krusei and C. tropicalis) and ten viruses (adenovirus, human bocavirus, human cytomegalovirus, enterovirus, Epstein–Barr virus, herpes simplex virus, human metapneumovirus, respiratory syncytial virus, rhinovirus and varicella-zoster virus) were used to test for cross-reactivity in the actim Influenza A&B test. Bacteria and yeasts were cultured and resuspended in Dulbecco's PBS at a concentration of 107–108 c.f.u. ml–1 or, in the case of M. pneumoniae, at a concentration of 107–108 colour-changing units ml–1. For testing cross-reactivity to viruses other than influenza, patient specimens positive for the respective viruses were used. The actim Influenza A&B test was carried out with the investigator blinded to the nature of the pathogen.
|
Additionally, 12 respiratory specimens (stored at –20 °C) collected from adults with underlying lung or haematological diseases who had tested positive for influenza virus were included in the panel of 473 samples.
Ethical approval was obtained from the institutional review boards for this study.
actim Influenza A&B test. The actim Influenza A&B test was carried out according to the manufacturer's instructions. The test was performed with the investigator blinded to the results of the RT-PCR. For liquid specimens, 300 µl sample was dispensed into the test tube together with three drops of extraction buffer, and nasopharyngeal swabs were inserted into test tubes containing six drops of NaCl buffer and three drops of extraction buffer, mixed and incubated for 5 min at room temperature to allow virus particles in the specimens to be disrupted, exposing internal viral nucleoproteins. The appearance of a blue–black control line on the actim test strip confirmed that the test had been performed properly and that the results were therefore valid.
BinaxNOW Influenza A & B test. The BinaxNOW Influenza A & B test (Binax) is an immunochromatographic membrane assay that uses highly sensitive mAbs to detect influenza type A and B nucleoprotein antigens in nasopharyngeal specimens. The test was carried out according to the manufacturer's instructions.
RNA extraction and cDNA synthesis. Total viral RNA was extracted using a commercial kit (QIAamp Viral RNA kit; Qiagen) according to the manufacturer's instructions. Viral RNA was extracted from 140 µl of each sample. The RNA was eluted from columns with 50 µl RNase-free water (Qiagen) and used immediately in the following experiments or stored at –80 °C. cDNA synthesis was carried out at 37 °C for 1 h using 10 µl eluted RNA, 100 U murine leukaemia virus reverse transcriptase (Gibco/Invitrogen), 10 mM dithiothreitol, 20 U RNasin (Promega) and 0.25 µM random hexamer primers (Gibco/Invitrogen).
FastCycler PCR. PCR was carried out in a 20 µl volume containing 2.5 µl cDNA, 2x master mix (FastCycler 9800, 2x GeneAmp Fast PCR Master Mix; Applied Biosystems) and 0.25 µM each primer. Primer sets specific for influenza A and B virus (custom synthesized by MWG-Biotech) were used to detect and differentiate the two influenza viruses as described by Schweiger et al. (2000). PCR conditions were initial denaturation at 95 °C for 10 s, 40 cycles of 95 °C for 2 s and annealing and extension at 67 °C for 15 s, and final extension at 72 °C for 7 min. Amplicons were detected by gel electrophoresis.
Statistical analysis. Specimens positive for influenza A or B virus in the RT-PCR were regarded as true positives. The sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values of the actim Influenza A&B test results compared with those of the RT-PCR assay were calculated using two-by-two contingency tables.
|
RESULTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
Analytical sensitivity of the actim Influenza A&B test in quality assessment influenza samples
A total of 32 quality assessment influenza A and B samples of different designation (INSTAND), including avian influenza virus, were evaluated by the actim Influenza A&B test and RT-PCR. All of these samples tested positive by both methods (Table 1
), revealing a sensitivity and specificity of 100 % for the actim Influenza A&B test.
Comparison of the actim Influenza A&B test and RT-PCR assay
Among the 473 specimens, nine (1.9 %) were antigen-positive for influenza A virus in the actim Influenza A&B test and 13 (2.8 %) were positive in the RT-PCR assay. Six specimens (1.3 %) were positive for influenza B virus by the actim Influenza A&B test and ten (2.1 %) in the RT-PCR assay (Table 2). All influenza samples that were detected by the actim Influenza A&B test were also positive in the RT-PCR assay.
|
The overall sensitivity of the actim Influenza A&B test compared with the RT-PCR assay was 65.2 % (69.2 % for influenza A virus and 60 % for influenza B virus) with a specificity of 100 %. The actim Influenza A&B test revealed a PPV of 100 % (95 % CI 78.2–100 %) for both influenza virus types, whereas the NPV was approximately 98 % (95 % CI 96.5–99.2 %).
From the panel of 473 samples, 149 paediatric respiratory samples were also analysed using a BinaxNOW Influenza A & B kit. The sensitivity of the BinaxNOW Influenza A & B kit [4/7 RT-PCR-positive samples (57 %) for influenza A virus and 3/7 RT-PCR-positive samples (43 %) for influenza B virus] was lower than the sensitivity of the actim Influenza A&B test (71 % for influenza A virus and 57 % for influenza B virus when compared for the 149 samples). The BinaxNOW Influenza A & B test kit revealed a specificity of 100 % for both influenza A and B viruses compared with the RT-PCR assay. When compared with the actim Influenza A&B test as the reference, the BinaxNOW Influenza A & B test yielded a sensitivity of 78 %. RT-PCR-positive samples were confirmed as described above.
Clinical symptoms of influenza A and B virus-positive cases
The majority of the influenza virus-positive samples were detected between February and March 2005 and between January and April 2008. Among the patients with influenza, the youngest was 7 weeks old and the eldest 61 years old (with influenza A and B virus, respectively; Table 3). The clinical presentation of the patients with influenza virus detection as well as the type of specimen in which influenza virus was detected are shown in Table 3. No dual infections with influenza A and B viruses were detected.
|
|
DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
For example, Herrmann et al. (2001) compared an optical immunoassay for rapid detection of influenza A and B virus with RT-PCR and immunofluorescence on nasopharyngeal aspirates and swabs. The sensitivity and specificity of the optical immunoassay in nasopharyngeal aspirates were 40.4 and 94.3 %, respectively, compared with the RT-PCR. In general, a greater yield of positive samples can be obtained with PCR techniques than with rapid antigen detection tests. This has been observed particularly when the sample is taken 3–4 days after the onset of symptoms, when the viral load drops sharply, as does the infecting capacity (Cherian et al., 1994).
The Directigen FluA+B antigen test (BD Diagnostic Systems) was evaluated by Chan et al. (2002) in a study design similar to our present work. Most of the patients studied were children. Compared with an RT-PCR assay, the sensitivity, specificity, and PPV and NPV of the Directigen FluA+B test for influenza virus A were 96, 99.6, 96 and 99.6 %, and for influenza virus B were 87.5, 96.8, 80 and 98 %, respectively. In a French study of the QuickVue Influenza antigen test (Quidel), these values were 87.5, 100, 100 and 89.5 %, respectively (Arsene et al., 2004). These results seem to suggest a better performance of the Directigen FluA+B and QuickVue Influenza antigen tests than of the actim Influenza A&B test in the present study. However, Ruest et al. (2003) compared a number of antigen tests with an RT-PCR assay with different results. A poor specificity (35–58 %) and a poor PPV (41–60 %) were reported for the compared antigen tests, with the Directigen FluA+B test having a higher sensitivity than the QuickVue Influenza antigen test but being associated with a larger number of invalid results. For comparison, the actim Influenza A&B test had a sensitivity, specificity, and PPV and NPV of 65, 100, 100 and 98 %, respectively, and no invalid results were observed in the present study. Cross-reactivity of the actim Influenza A&B test with other bacteria, fungi and viruses that may be encountered in the respiratory tract was not observed. Moreover, false-positive results were not obtained using the actim Influenza A&B test in the panel of 473 patient specimens tested (Table 2). Therefore, the specificity and robustness of the actim Influenza A&B test are probably superior to those of the above-described rapid antigen detection kits.
As the sensitivity of the actim Influenza A&B assay in diagnostic samples seemed to be lower than that of the above-mentioned antigen detection kits, we analysed a subgroup of 149 respiratory samples in a direct comparison of the BinaxNow Influenza A & B and actim Influenza A&B tests. This revealed that the BinaxNow Influenza A & B test did not yield better results than the actim Influenza A&B test. Moreover, the actim Influenza A&B assay achieved a sensitivity of 100 % in comparison with RT-PCR and the expected results in a panel of quality assessment samples (including influenza A H1N1, H3N2 and H5N1) obtained from INSTAND (Table 1).
In summary, the studies cited above, together with our results, suggest that the reported sensitivities of influenza rapid antigen tests may vary highly in relation to the panel of diagnostic specimens tested. If a panel contains diagnostic specimens with low virus concentrations (perhaps because these were taken more than 2 days after the onset of symptoms), the sensitivities of antigen assays drop in comparison with that of RT-PCR, which can still detect influenza RNA due to its low detection limit. Thus RT-PCR offers an alternative method of diagnosing influenza virus infections. It has high sensitivity and specificity (Herrmann et al., 2001), but requires a high level of skill and a complex laboratory infrastructure, and takes several hours to perform, as was also shown in our study (150–170 min). Therefore, RT-PCR does not fill the niche of tests that are rapid and easy to perform with a low level of expertise.
The low influenza activity in the Magdeburg area during the respective seasons between 2004 and 2008 was disadvantageous for this study, resulting in a limited number of positive specimens being obtained. Thus studying a larger number of positive specimens would make evaluation and determination of sensitivity and specificity more reliable. However, during times of low influenza incidence, the NPV of a rapid antigen detection test is higher and the PPV is lower than in times of high influenza activity (WHO, 2005). Therefore, the high PPV of the actim assay observed in this study indicates that positive results do not need to be confirmed by RT-PCR and should be reported to the treating physician immediately in times of both low and high influenza incidence. However, negative actim Influenza A&B results, as well as negative results of other rapid antigen assays, should be confirmed by RT-PCR. This is even more important during the influenza season, because of decreasing NPV with increasing influenza virus incidence.
In conclusion, the actim Influenza A&B test is a rapid in vitro assay that detects and differentiates influenza A and B virus antigens (nucleoprotein) extracted from respiratory specimens. Because the actim Influenza A&B test is easy to perform and has a specificity and PPV of 100 % for each virus, this test is appropriate for acute cases as a first-line emergency diagnostic method. A negative result warrants confirmation with a more sensitive method.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
|
References |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
