The evolution of rapid influenza testing and the introduction of digital immunoassays
Influenza viruses are responsible for winter epidemics of respiratory illness among all age groups. About 5% to 20% of the population is infected each year, resulting in several hundred thousand hospitalizations. It is well known that a number of patient populations are at particularly high risk for complications from influenza infection. Specifically, these include young children, pregnant women, older adults, patients in nursing homes or long-term care facilities, and individuals with certain underlying medical conditions such as chronic lung disease, heart disease, asthma or compromised immune system function.
On average, approximately 36,000 individuals die annually in the United States due to influenza and complications from influenza infection, and this occurs most commonly in those aged 65 years or older. The medical and indirect costs attributed to influenza infections in the US exceed $10 billion annually.
There are a number of methods available to potentially identify those infected with influenza, the timeliness and accuracy of which influence treatment decisions and possibly further transmission of the disease. These modalities include clinical diagnosis, rapid influenza diagnostic tests (RIDTs), direct immunofluorescence, conventional and rapid cell culture, and nucleic acid detection methods. Unfortunately, making a diagnosis of influenza infection based solely on clinical signs and symptoms has, to date, proven to be not very accurate. Studies have shown that the only independent predictor of influenza infection is high fever; symptoms such as headache, pharyngitis, myalgia or gastrointestinal disturbances cannot independently predict influenza infection. Therefore, a laboratory test is needed to confirm infection.
The ‘ideal’ influenza diagnostic test
The “ideal” influenza diagnostic test would combine a high degree of sensitivity and specificity of detection with a rapid, easy-to-use test platform. This would allow for better management of patients at the point of care in terms of antibiotic or antiviral use and whether additional diagnostic testing is necessary. The ability to differentiate between viral and bacterial infections and treat them appropriately may reduce selection for antibiotic-resistant bacteria. Having a rapid and accurate test result for influenza detection also is known to influence how long a patient stays in the hospital or ED. Certainly, we would like to be able to quickly identify potential outbreaks so that infection control or public health measures could be instituted to prevent additional cases.
A different kind of RIDT
RIDTs based on immunochromatographic analysis of respiratory specimens typically can be performed within about 15 to 30 minutes, and many commercially available test formats require no specific virology or laboratory expertise. However, many factors can influence the performance characteristics of RIDTs. Technically, false-negatives may occur due to poor analytical sensitivity. The assay’s limit of detection may vary depending on the type or strain of virus. Factors inherent to the host such as age, disease severity and time to presentation and testing also affect test performance. The type and quality of respiratory specimen influences test accuracy, and infections that are primarily of the lower respiratory tract are less likely to be identified by sampling through the nasal cavity.
RIDTs have been used extensively for many years in a variety of settings, including physicians’ offices, urgent care centers and small laboratories where more complex viral diagnostic capabilities may not be available.During periods of higher prevalence of influenza infection in the community, positive results by these rapid methods generally correlate well with actual influenza virus infection. However, the historical performance of most RIDTs has been hampered by a significant lack of analytical and clinical sensitivity and low negative predictive values compared with more complex and technically demanding types of assays such as cell culture and nucleic acid detection methods. These discrepant findings are particularly apparent for novel or pandemic influenza strains.
If a RIDT had diagnostic accuracy approaching or equivalent to the more sensitive analytic methods, it could serve as a stand-alone test for most patients that present with influenza-like illness. To that end, two lateral-flow immunoassays have been designed and developed to employ an instrument-based digital scan of the test strip to enhance the sensitivity and specificity of detection of influenza virus antigens in patient respiratory specimens. One employs europium-based immunofluorescence technology; the other relies on a reflectance-based measurement to evaluate the line signal intensities on the assay test strip. These digital immunoassay (DIA) test platforms eliminate the need for an operator to visualize and interpret test results, a task that is often subjective and variable. In a number of clinical studies, the sensitivities for detection of influenza A and B viruses achieved using the DIAs approached those of viral culture and/or nucleic acid detection. In addition, when other commercially available RIDTs were studied side by side, both digitally read assays displayed significantly better sensitivities for detecting influenza viruses. With respect to novel and variant influenza A strains, the DIAs displayed improved antigen detection capabilities with differences in lower limits of detection often several fold better.
Improving patient outcomes
The FDA has recently proposed up-classifying RIDTs from Class I to Class II devices subject to special controls and performance standards. The intent of the reclassification is to assure the reliability and accuracy of RIDTs, reduce the likelihood of false-negative results, aid clinicians in making appropriate treatment decisions and enable effective infection control and public health response during influenza outbreaks. This new rule would “1) identify the minimum acceptable performance criteria; 2) identify the appropriate comparator for establishing performance of new assays; and 3) call for mandatory annual analytical reactivity testing of contemporary influenza strains, including testing of newly emerging strains that pose a danger of public health emergency.” The proposed minimum performance criteria are quite stringent and many RIDTs currently do not meet these standards. For the most part, the DIAs appear to consistently meet the criteria and may be most useful when nasal wash/aspirate or nasopharyngeal swabs from symptomatic patients are used for testing.
Use of RIDTs in a variety of settings has been shown to improve patient outcomes. The newer DIAs are clearly a significant improvement over previous, commercially available immunochromatographic antigen detection assays. The DIAs have demonstrated enhanced sensitivity of detection for seasonal, novel and variant influenza strains and appear to meet the criteria for FDA up-classification. There are numerous studies demonstrating the benefits of rapid diagnosis of influenza, both in directing the appropriate use of antivirals and in the reduction of unnecessary diagnostic tests. The greatest cost benefit for rapid influenza virus testing is achieved when unnecessary antibiotics are not prescribed for patients with positive test results. The improved performance afforded by digital detection of influenza infection should expand opportunities for appropriate patient management and more propitious outcomes moving forward.
Blaschke AJ. J Pediatric Infect Dis Soc. 2014;3:112-118.
Dunn J. Diagn Microbiol Infect Dis. 2014;79:10-13.
Harper SA. Clin Infect Dis. 2009;48:1003-1032.
Hassan F. J Clin Microbiol. 2014;52:906-910.
Heinonen S. Eur J Clin Microbiol Infect Dis. 2012;31:1569-1574.
For more information:
Jim Dunn, PhD, (D)ABMM, is the director of medical microbiology and virology at Texas Children’s Hospital and an associate professor of pathology and immunology at Baylor College of Medicine. He can be reached at email@example.com.
Disclosure: Dunn reports no relevant financial disclosures.