Postepy Hig Med Dosw. (online), 2012; 66: 752-757
Original Article
Full Text PDF  

Accuracy of rapid influenza diagnostic test and immunofluorescence assay compared to real time RT-PCR in children with influenza A(H1N1)pdm09 infection
Wiarygodność szybkiego testu diagnostycznego i metody immunofluorescencji bezpośredniej w odniesieniu do real-time RT-PCR u dzieci z grypą wywołaną wirusem typu A (H1N1)pdm09
Aneta Nitsch-Osuch1  ABCDEFG, Agnieszka Woźniak-Kosek2  ABDE, Lidia Bernadeta Brydak2,3  ADEF
1Department of Family Medicine, Warsaw Medical University, Poland
2Department of Influenza Research. National Influenza Center, National Institute of Public Health – National Institute of Hygiene, Warsaw, Poland
3Department of Immunology, Faculty of Biology, University of Szczecin, Poland
Corresponding author
Aneta Nitsch-Osuch, MD, PhD, Department of Family Medicine, Warsaw Medical University, Banacha 1a, blok F, 02-097 Warsaw; e-mail: anitsch@amwaw.edu.pl

Authors' Contribution:
A - Study Design, B - Data Collection, C - Statistical Analysis, D - Data Interpretation, E - Manuscript Preparation, F - Literature Search, G - Funds Collection

Source of support
Source of support: scientific grant from the Department of Family Medicine, Warsaw Medical University, 1MH/2010

Received:  2012.03.06
Accepted:  2012.07.14
Published:  2012.10.22

Streszczenie
Wprowadzenie: Grypa stanowi u dzieci istotny, ale niedoceniany i nieoszacowany problem kliniczny i epidemiolo­giczny. Celem pracy było porównanie wiarygodności szybkiego testu diagnostycznego w kierunku grypy BD Directigen EZ Flu A+B® z metodą immunofluorescencji bezpośredniej w odniesieniu do metody real time RT-PCR u dzieci zgłaszających się do ambulatorium z objawami grypopodobnymi.
Materiał/Metody:
Do badania włączono 150 dzieci, kryteriami włączenia do badania były: wiek poniżej 59 miesię­cy życia, występowanie objawów grypopodobnych, zgodnych z definicją CDC: gorączka >=37,8°C, kaszel, ból i/lub zaczerwienienie gardła, utrzymywanie się objawów <=96 godzin. U pacjentów pobrano wymaz z nosa i z gardła i poddano badaniu z użyciem RIDT i/lub DFA, a wyniki zwe­ryfikowano metodą real time RT-PCR.
Wyniki:
Metoda RIDT charakteryzowała się w odniesieniu do wykrywania zakażeń wywołanych wiru­sem grypy A (H1N1)pdm09: czułością 62,2% (95% CI 46,5-76,2%), swoistością 97,1% (95% CI 91,8-99,4%), PPV 90,3% (95% CI 74,3-98%), NPV 85,5% (95% CI 78,1-91,5%), podczas gdy metoda DFA charakteryzowała się: czułością 60% (95% CI 51,9-63,2%), swoistością 96% (95% CI 88,7-98,8%), PPV 93,1% (95% CI 80,5-98%), NPV 72,7% 995% CI 67,2-74,9%). Analiza regresji logistycznej wykazała, że szansa na uzyskanie prawdziwie dodatniego wyniku RIDT była dwukrotnie wyższa, gdy badanie wykonywano w ciągu 48 godzin od pojawienia się objawów chorobowych (OR 0,40 vs OR 0,22).
Wnioski:
Wiarygodność RIDT i DFA u dzieci jest porównywalna, czułość obu testów jest umiarkowana, a swoistość bardzo wysoka. Oba testy mogą być używane jako metody skriningowe w diagno­styce grypy u dzieci z objawami grypopodobnymi
Słowa kluczowe: grypa • szybkie test diagnostyczny • immunofluorescencja bezpośrednia • dzieci


Summary
Introduction: The influenza burden among children is underestimated. The aim of our study was to estimate the accuracy of the rapid influenza detection test (RIDT) BD Directigen™ EZ Flu A+B® and direct immunofluorescence assay (DFA) used among children with influenza-like illness (ILI) consul­ted in the ambulatory care clinic.
Material/Methods: A total of 150 patients were enrolled in the study. Inclusion criteria were: age less than 59 mon­ths, presentation of ILI according to the CDC (Centers for Disease Control and Prevention) defi­nition (fever >37.8°C, cough and/or sore throat in the absence of another known cause of illness), duration of symptoms shorter than 96 hours. Two nasal swabs and one pharyngeal swab were ob­tained from patients and tested by RIDT, DFA and real time RT-PCR as the reference method.
Results: For influenza A (H1N1)pdm09 virus sensitivity of RIDT was 62.2% (95% CI 46.5-76.2%), specificity 97.1% (95% CI 91.8-99.4%), PPV 90.3% (95% CI 74.3-98%), NPV 85.7% (95% CI 78.1-91.5%), for DFA sensitivity was 60% (95% CI 51.9-63.2%), specificity 96% (95% CI 88.7-98.8%), PPV 93.1% (95% CI 80.5-98%), NPV 72.7% (95% CI 67.2-74.9%). Analysis of logistic regression revealed that the chance of receiving a true positive result of RIDT was twi­ce as high when the test was conducted during the first 48 hours of symptoms (OR 0.40 vs OR 0.22).
Conclusions: The accuracy of RIDT is comparable with DFA and both methods are very specific but modera­tely sensitive in diagnosis of influenza in young children. Both methods may be recommended for screening for influenza among children.
Key words: influenza • rapid influenza diagnostic test • direct immunofluorescence assay • children




Abbreviations:
DFA - direct immunofluorescence assay; LR- - negative likelihood ratio; LR+ - positive likelihood ratio; NPV - negative predictive value; PPV - positive predictive value; real time RT-PCR - real time reverse transcriptase polymerase chain reaction; RIDT - rapid influenza diagnostic test.
Background
Influenza virus is a highly contagious agent that affects in­dividuals of all ages and plays an important worldwide role by causing epidemics and pandemics. Since the signs and symptoms of influenza are similar to other respiratory vi­ral infections, the differential diagnosis of influenza may be difficult, even impossible, when based solely on clini­cal symptoms. Laboratory diagnosis of influenza is criti­cal not only for its treatment and surveillance, but also for its prevention and control [27,32].
Laboratory testing for influenza has historically been of questionable value for the management of patients with influenza, because of limited test sensitivity, long turna­round times, and a lack of effective therapy [21,24]. The development of more rapid and accurate tests for the de­tection of influenza now enables the laboratory to provi­de a prompt, definitive diagnosis, which allows clinicians to initiate modern antiviral therapy within 48 hours from the disease onset, limit the injudicious use of antibacterial agents, implement appropriate infection control measures, decrease the duration of hospital isolation, reduce ancillary testing, and decrease health costs [9,10,11,13,17,19,25,26].
Laboratory diagnosis of influenza can be accomplished by detection of the virus or the patient's immune response to the virus. Diagnostic approaches for identification of the virus include viral isolation, detection of viral antigens by immunospecific assays (immunofluorescence microsco­py), rapid influenza detection test (RIDT - based on opti­cal immunoassay or immunoenzymatic assay), and detec­tion of viral nucleic acid by use of amplification testing (RT-PCR, real time RT-PCR). Antibody detection is usu­ally accomplished by virus neutralization (virus NT) and hemagglutinin inhibition (HI) tests conducted to moni­tor seroconversion to a specific virus strain or to determi­ne the immune status, for example, after vaccination [21].
Diagnosis of influenza by immunoenzymatic assay has led to the development of easy-to-use, self-contained diagno­stic kits that can provide results within 15 minutes after specimen collection. The World Health Organization has issued recommendations for the use of such kits [33]. In contrast to traditional laboratory methods, such as virus isolation from cell culture or real-time PCR, RIDTs are widely used in a routine clinical setting. However, RIDTs may have low to moderate sensitivity and their results sho­uld be interpreted very carefully by practitioners [9,11,33].
Another rapid detection method is direct immunofluorescen­ce assay (DFA), which can generate results within 1-4 hours but requires specialized and experienced staff and specialized equipment. Generally, the sensitivity of DFA is higher than that of the rapid antigen test. Sensitivity of DFA is also higher among pediatric patients, especially those admitted with high fever and spreading high amounts of viruses [3,14,16,19,23].
The aim of our study was to estimate the accuracy of two rapid influenza diagnostic methods, RIDT and DFA, com­pared to real time RT-PCR in children younger than 59 months with acute symptoms of respiratory tract infection consulted in an ambulatory care setting. We chose this age group of patients because influenza is an epidemiological, clinical and economic problem for these patients, their ca­regivers and health care services. Depending on age, at­tack rates of influenza may be 1.5-3 times higher than for adults and are estimated to be between 10 and 40% each year [28]. Influenza in the pediatric population leads to a si­gnificant increase in primary care visits, emergency depart­ment visits and hospitalizations due to complications [28].
According to our current knowledge, our study is the first one in Poland exploring this problem.
Material and Methods
A total of 150 patients (70 boys and 80 girls) aged 0-59 months (mean age 34.3 months, SD 14.3 months) were inc­luded in the study. All patients came from ambulatory care settings and met symptoms of influenza-like illness (ILI) according to the CDC definition (fever >38.7°C plus cough and/or sore throat in the absence of another known cause of illness) lasting less than 4 days (96 hours). Exclusion criteria for the study were age older than 59 months, du­ration of symptoms longer than 96 hours, current antibio­tic therapy or antiviral therapy.
The observation was performed in two consecutive influ­enza seasons: 2009/2010 and 2010/2011. Nasopharyngeal swabs were taken from all patients by previously trained staff using special artificial sticks (ending with viscose). The specimens were tested for influenza with both the ra­pid influenza diagnostic test (RIDT) BD Directigen EZ Flu A+B® (Becton, Dickinson and Company, Sparks, MD) (150 patients) and direct immunofluorescence assay IMAGENTM Influenza Virus A and B (Oxoid) (95 pa­tients), strictly according to the manufacturer's instruc­tions. Results of RIDT were available during 15 minutes in the office. DIA and real time RT-PCR were performed by qualified personnel at the National Influenza Center in Warsaw (Poland). Isolation of viral RNA was conducted with Maxwell® 16 Viral Total Nucleic Acid Purification Kit (Promesa Corp., USA). One-step RT-PCR was con­ducted with Transcriptor One-step RT-PCR Kit (Roche Diagnostics, Switzerland). Real time RT-PCR was conduc­ted with Invitrogen SuperScriptTM III Platinum® One-Step Quantitative RT-PCR System (Invitrogen, USA) using the CDC Realtime RTPCR (rRTPCR) Protocol for Detection and Characterization of Swine Influenza (version 2009) (http://www.who.int/csr/resources/publications/swineflu/ CDCrealtimeRTPCRprotocol_20090428.pdf).
The sensitivity, specificity, positive predictive value, ne­gative predictive value, positive likelihood ratio, negative likelihood ratio of rapid influenza detection test and di­rect immunofluorescent assay compared to real time re­verse transcriptase polymerase chain reaction (real time RT-PCR) were calculated, by using standard formulae (table 1) and the medical statistical calculator available at www.medcal3000. 95% confidence intervals were calcula­ted. The analysis of logistic regression was used to descri­be the impact of duration of symptoms on the sensitivity of RIDT (OR and 95% CI were calculated). The statisti­cal program CRAN-R was used. The approval of the Local Ethical Committee of Warsaw Medical University was ob­tained for the study (no. KB 165/2011).
Table 1. Formulae used for calculation of accuracy of RIDT and DFA

Results
For RIDT there were obtained 28 true positive, 102 true negative, 3 false positive and 17 false negative results. For DFA there were obtained 27 true positive, 48 true negati­ve, 2 false positive and 18 false negative results.
The accuracy of RIDT and DFA compared to real time RT-PCR is presented in table 2.
Table 2. Comparison of the accuracy of rapid influenza detection test (RIDT) and direct immunofluorescent assay (DFA)

The positive and negative predictive values were calculated for estimated prevalence of influenza at the level of 20%.
The chance of obtaining a true positive result of RIDT or DFA was twice as high when the duration of symptoms was shorter than 48 hours and, generally, the chance of ob­taining a true positive result of RIDT decreased with lon­ger duration of symptoms (table 3).
Table 3. Duration of symptoms as the predictor of true positive RIDT result

Discussion
Our results indicate a comparable and moderate accura­cy of both rapid influenza diagnostic methods, RIDT and DFA, in the detection of infections caused by A (H1N1)pdm09 virus in a homogeneous group of children aged 0-59 months consulted in primary care settings with symptoms of influenza-like illness.
Values estimated in our study, describing specificity and sensitivity of RIDT BD Directigen EZ Flu A+B®, are lo­wer than these indicated by the manufacturer and calcula­ted for a viral culture as a gold standard [4]. This may be a result of different sample types taken in our study (pha­ryngeal and nasal swabs), while the manufacturer tested nasopharyngeal swabs or pharyngeal swabs. Our observa­tion proves that the accuracy of RIDT should be evaluated in everyday practice because many factors may influence its results, including for example age of patients and the subtype of virus circulating in the population.
One of these possible factors influencing the sensitivity of RIDT and DFA is the duration of symptoms of ILI. In our study we found that the chance of obtaining a true positi­ve result of RIDT was twice as low when the symptoms lasted longer than 48 hours and the logistic regression analysis indicated that longer duration of symptoms was negatively associated with RIDT performance. Our obse­rvation is consistent with the results obtained by Hawkes et al. [18]. Gordon et al. also observed that the sensitivity was significantly lower when the samples for RIDT were collected on the day of symptom onset compared to sam­ples collected one or two days following symptom onset (41.7% vs 72.1%) [16]. Time-dependent sensitivity of a rapid antigen test in patients with A (H1N1)pdm09 in­fluenza was also described by Lee et al. [22] - the most appropriate time frame for sample collection for the de­tection of influenza virus in RIDT was 24-96 hours (sen­sitivity 75-79.9%).
We found that the DB Directigen Ez A+B Flu test perfor­med moderately well, with an intermediate sensitivity and a high specificity in comparison to the CDC real time RT-PCR assay for influenza A (H1N1)pdm09. Several stu­dies have been published on the performance of the BD Directigen EZ A+B Flu test in detecting influenza A (H1N1)pdm09, with reported sensitivities ranging from 20.6% to 70.4% [1,5,6,10,18]. Our sensitivity estimate of 62% falls within the estimate of previously reported sensitivities and is very similar to the estimate of sensitivity of 70.4% from the Blazguez et al. study [5], but higher than reported by Karre et al. (48.7%) [20] and Vasoo et al. (46.7%) [32]. Possible reasons for the higher sensitivity of RIDT in our study may include the fact that we tested young children aged less than 59 months. It has been proven that the sen­sitivity of the RIDT increases with younger age of patients and it is believed to be a result of more intensive and lon­ger lasting replication and shedding (transmission) of in­fluenza virus in children compared to adults. Hawkes et al. found that the sensitivity of RIDT for detection of in­fluenza A (H1N1)pdm09 virus was higher in a pediatric population than in a previously reported mixed adult pa­tient population [18]. Gao et al. reported that the sensiti­vity of RIDT was higher among children younger than 2 years (85.7%) than patients aged 2-39 years (33.3%) [15]. A similar observation was reported by Landry et al. for se­asonal influenza (the sensitivity of BD Directigen EZ Flu A+B - 75%) [21].
Another reason for higher sensitivity of RIDT reported in our study could be that we used fresh samples for testing while some researchers used frozen samples; studies using frozen specimens or specimens delayed in testing reported worse accuracy of RIDT [17,29].
The BD Directigen EZ Flu A+B® in our study revealed a moderate positive predictive value (68.6%), and high ne­gative predictive value (94.1%). These results would be more typical for prevalence of influenza estimated as low (<10%) while we conducted the study during the autumn and winter season (between 1 November and 31 March) when the prevalence of the disease is expected to be hi­gher, especially as the 2009/2010 season was a pandemic season with predominance of type (H1N1)pdm09 influen­za virus. Moderate positive predictive value of the RIDS may be explained by the changing prevalence of influenza in different months, even different weeks of the season, as described by some authors [18,27].
Positive and negative likelihood ratios are independent values from the prevalence of the diseases and in our stu­dy we observed a high positive and low negative likeliho­od ratio (respectively), which is similar to data reported by other investigators: Poehling et al. found a positive and negative likelihood ratio of respectively 126 and 0.18 [27].
The second rapid method of diagnosis of influenza is a di­rect immunofluorescent assay. The sensitivity of DFA for the pandemic virus A (H1N1)pdm09 ranged from 32.35-38.7% to 65% [18,22,23]. The sensitivity of DFA in our study was and our result is consistent with the results of other researchers. Sandora found the sensitivity of DFA in detection of influenza A (H1N1)pdm09 infection as high as 53%, specificity 99% [30]. Bakerman et al. reported the sensitivity of DFA of 65% (specificity 99.6%) [3].
In our study the accuracy of RIDT was comparable with the accuracy of DFA, and we conclude that in young chil­dren aged less than 5 years, consulted with symptoms of ILI in primary care settings, the RIDT may be a sound tool in the diagnosis of influenza and the method of choice for screening for influenza infections and there is no need to conduct immunofluorescence assays as they are more time consuming, more expensive, and require access to a well­-equipped laboratory and highly trained laboratory per­sonnel. RIDT should be especially used in resource-limi­ted settings where equipment, reagents and laboratory staff are not always available.
In our study both methods "missed" one-third of infected individuals. Clinicians must be aware of the limitations of RIDT and the results should be interpreted cautiously. In an epidemic or pandemic season false negative results oc­cur more often than false positive results. The physician should consider sending respiratory specimens of patients with negative results received in RIDT for influenza testing by RT-PCR or viral culture or DIA, especially when com­munity influenza activity is high, and particularly when the patient is in the risk group of severe course of disease and complications [9,11,12,33]. For children at high risk with influenza-like illness during high-prevalence periods of influenza, empiric initiation of antiviral therapy should be considered for patients with negative RIDT results [33].
Both methods - the rapid influenza detection test and di­rect immunofluorescent assay - allow one to differentiate between influenza type A and type B infection, but they do not provide information concerning the subtypes of the vi­rus [2,3]. The gold standard for diagnosis of influenza and for evaluation of diagnostic tests is still viral culture, but molecular biology methods (PCR and RT-PCR) are also considered as gold standard, especially for clinical diagno­sis [9,11,33]. Potentially, the most important aspect of this rapid test is that it can provide timely, accurate and useful information for clinicians. The information can be provi­ded in real time when the diagnostic, isolation and thera­peutic questions need to be addressed and solved, not ho­urs later. Laboratory methods based on molecular biology should be recommended to all patients with a severe cour­se of the disease suspected for influenza, even when RIDT provided a negative result. Laboratory-based surveillance is also needed in order to obtain information on currently circulating influenza viruses, and the emergence and pre­valence of antiviral resistant strains [33].
One advantage of our study is the homogeneous group of patients (children younger than 59 months) while most stu­dies dedicated to evaluation of rapid influenza tests have been conducted on adult, mixed adult and pediatric patients or among patients aged 0-18 years [2,7,8,24,31].
Another advantage of our study is that it was conducted in out-patient settings while most other studies were per­formed in hospital settings or emergency units [24,30,31].
There are several limitations of our study. The sample size for calculation of the accuracy of RIDT and DFA was rela­tively small, although limited to a very age-related homo­geneous group of patients. Another limitation is that DFA was not conducted for each sample (samples were collec­ted in the primary care setting, in the evening hours and at weekends too; laboratory staff and equipment required for DFA testing were not available every time).
Conclusions
The rapid influenza detection test BD Directigen EZ Flu A+B® is characterized by moderate accuracy which is com­parable to the direct immunofluorescence assay accura­cy. RIDT may be used as a screening test in a differential diagnosis of influenza caused by type A virus in children younger than 59 months consulted in the primary care set­ting with symptoms of ILI.
REFERENCES
[1] Al Johani S.M., Al Balawi M., Al Alwan B., Al Hefdhi R., Hajeer A.: Validity of two rapid point of care influenza tests and direct fluorescence assay in comparison of real time PCR for swine of origin influenza virus. J. Infect. Public Health, 2011; 4: 7-11
[PubMed]  
[2] Angoulvant F., Belletre X., Houhou N., Dexpert J.B., Morin L., Siriez J.Y., Soole F., de Lauzanne A., Cohen R., Brun-Vezinet F., Alberti C., Mercier J.C.: Sensitivity and specificity of a rapid influenza diagnostic test in children and clinical utility during influenza A (H1N1) 2009 outbreak. Emerg. Med. J., 2011; 28: 924-926
[PubMed]  
[3] Bakerman P., Balasuriya L., Fried O., Tellez D., Garcia-Filion P., Dalton H.: Direct fluorescent-antibody testing followed by culture for diagnosis of 2009 H1N1 influenza A. J. Clin. Microbiol., 2011; 49: 3673-3674
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[4] Becton Dickinson and Company. 2008. Directigen EZ Flu A+B package insert. Becton Dickinson and Company, Sparks, MD (product characteristics)
[5] Blázguez D., Diaz J., Cruz J., Folgueira D., Acosta J., Marin M.: Rapid influenza diagnostic tests for detection of novel influenza A (H1N1) virus in children. Enferm. Infecc. Microbiol. Clin., 2010; 28: 698-700
[PubMed]  
[6] Chan K.H., Maldeis N., Pope W., Yup A., Ozinskas A., Gill J., Seto W.H., Shortridge K.F., Peiris J.S.: Evaluation of the Directigen FluA+B test for rapid diagnosis of influenza virus type A and B infections. J. Clin. Microbiol., 2002; 40: 1675-1680
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[7] Cheng P.K., Wong K.K., Mak G.C., Wong A.H., Ng A.Y., Chow S.Y., Lam R.K., Lau C.S., Ng K.C., Lim W.: Performance of laboratory diagnostics for the detection of influenza A (H1N1)v virus as correlated with the time after symptom onset and viral load. J. Clin. Virol., 2010; 47: 182-185
[PubMed]  
[8] Cruz A.T., Demmler-Harrison G.J., Caviness A.C., Buffone G.J., Revell P.A.: Performance of a rapid influenza test in children during the H1N1 2009 influenza A outbreak. Pediatrics, 2010; 125: e645-e650
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[9] Dale S.E.: The role of rapid antigen testing for influenza in the era of molecular diagnostics. Mol. Diagn. Ther., 2010; 14: 205-214
[PubMed]  
[10] Esposito S., Marchisio P., Morelli P., Crovari P., Principi N.: Effect of a rapid influenza diagnosis. Arch. Dis. Child., 2003; 88: 525-526
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[11] Evaluation of rapid influenza diagnostic tests for detection of novel influenza A (H1N1) virus - United States, 2009. Morb. Mortal. Wkly. Rep., 2009; 58: 826-829
[PubMed]  [Full Text HTML]  
[12] Faix D.J., Sherman S.S., Waterman S.H.: Rapid-test sensitivity for novel swine-origin influenza A (H1N1) virus in humans. N. Engl. J. Med. 2009; 361: 728-729
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[13] Falsey A.R, Murata Y., Walsh E.E.: Impact of rapid diagnosis on management of adults hospitalized with influenza. Arch. Intern. Med., 2007; 167: 354-360
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[14] Ganzenmueller T., Kluba J., Hilfrich B., Puppe W., Verhagen W., Heim A., Schulz T., Henke-Gendo C.: Comparison of the performance of direct fluorescent antibody staining, a point-of-care rapid antigen test and virus isolation with that of RT-PCR for the detection of novel 2009 influenza A (H1N1) virus in respiratory specimens. J. Med. Microbiol., 2010; 59: 713-717
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[15] Gao F., Loring C., Laviolette M., Bolton D., Daly E.R., Bean C.: Detection of 2009 pandemic influenza A (H1N1) virus infection in different age groups by using rapid influenza diagnostic tests. Influenza Other Respir. Viruses, 2012; 6: e30-e34
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[16] Gordon A., Videa E., Saborio S., López R., Kuan G., Balmaseda A., Harris E.: Diagnostic accuracy of a rapid influenza test for pandemic influenza A H1N1. PLoS One, 2010; 5: e10364
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[17] Harnden A., Brueggemann A., Shepperd S., White J., Hayward A.C., Zambon M., Crook D., Mant D.: Near patient testing for influenza in children in primary care: comparison with laboratory test. Br. Med. J., 2003; 326: 480
[PubMed]  [Full Text HTML]  
[18] Hawkes M., Richardson S.E., Ipp M., Schuh S., Adachi D., Tran D.: Sensitivity of rapid influenza diagnostic testing for swine-origin 2009 a (H1N1) influenza virus in children. Pediatrics, 2010; 125: e639-e644
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[19] Heinonen S., Silvennoinen H., Lehtinen P., Vainionpää R., Heikkinen T.: Feasibility of diagnosing influenza within 24 hours of symptom onset in children 1-3 years of age. Eur. J. Clin. Microbiol. Infect. Dis., 2011; 30: 387-392
[PubMed]  
[20] Karre T., Maguire H.F., Butcher D., Graepler A., Weed D., Wilson M.L.: Comparison of Becton Dickinson Directigen EZ Flu A+B test against the CDC real-time PCR assay for detection of 2009 pandemic influenza A/H1N1 virus. J. Clin. Microbiol., 2010; 48: 343- 344
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[21] Landry M.L.: Diagnostic tests for influenza infection. Curr. Opin. Pediatr., 2011; 23: 91-97
[PubMed]  
[22] Lee H.M., Park H.K., Hwang H.S., Chun M.Y., Pai H.J., Oh S.H., Kim D.A.: Diagnostic value of the rapid influenza antigen test for novel influenza A (H1N1). Scand. J. Infect. Dis., 2011; 43: 43-46
[PubMed]  
[23] Lee J.H., Shin S.R., Cho J.H.: Evaluation of direct immmunofluorescence test with PCR for detection of novel influenza A (H1N1) virus during 2009 pandemic. Yonsei Med. J., 2011; 52: 680-682
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[24] Louie J.K., Guevara H., Boston E., Dahlke M., Nevarez M., Kong T., Schechter R., Glaser C.A., Schnurr D.P.: Rapid influenza antigen test for diagnosis of pandemic (H1N1) 2009. Emerg. Infect. Dis., 2010; 16: 824-826
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[25] McGeer A.J.: Diagnostic testing or empirical therapy for patients hospitalized with suspected influenza: what to do? Clin. Infect. Dis., 2009; 48 (Suppl. 1): S14-S19
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[26] Noyola D.E., Demmler G.J.: Effect of rapid diagnosis on management of influenza A infections. Pediatr. Infect. Dis. J., 2000; 19: 303-307
[PubMed]  
[27] Poehling K.A., Griffin M.R., Dittus R.S., Tang Y.W., Holland K., Li H., Edwards K.M.: Bedside diagnosis of influenzavirus infections in hospitalized children. Pediatrics, 2002: 110: 83-88
[PubMed]  
[28] Principi N., Esposito S., Gasparini R., Marchisio P., Crovari P.; Flu-Flu Study Group: Burden of influenza in healthy children and their households. Arch. Dis. Child., 2004; 89: 1002-1007
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[29] Quach C., Newby D., Daoust G., Rubin E., McDonald J.: QuickVue influenza test for rapid detection of influenza A and B viruses in a pediatric population. Clin. Diagn. Lab. Immunol., 2002; 9: 925-926
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[30] Sandora T.J., Smole S.C., Lee G.M., Chung S., Williams L., McAdam A.J.: Test characteristics of commercial influenza assays for detecting pandemic influenza A (H1N1) in children. Pediatr. Infect. Dis. J., 2010; 29: 261-262
[PubMed]  
[31] Stein J., Louie J., Flanders S., Maselli J., Hacker J.K., Drew W.L., Gonzales R.: Performance characteristics of clinical diagnosis, a clinical decision rule, and a rapid influenza test in the detection of influenza infection in a community sample of adults. Ann. Emerg. Med., 2005; 46: 412-419
[PubMed]  
[32] Vasoo S., Stevens J., Singh K.: Rapid antigen tests for diagnosis of pandemic (Swine) influenza A/H1N1. Clin. Infect. Dis., 2009; 49: 1090-1093
[PubMed]  [Full Text HTML]  [Full Text PDF]  
[33] WHO recommendations on the use of rapid testing for influenza diagnosis (28.08.2012)
http://www.who.int/influenza/resources/documents/RapidTestInfluenza_WebVersion.pdf
The authors have no potential conflicts of interest to declare.