Multiple Test Methods for the Detection of Novel H1N1
Abstract & Commentary
By Ellen Jo Baron, PhD, D(ABMM)
Professor of Pathology, Associate Director, Clinical Microbiology Laboratory, Interim Director, Clinical Virology Laboratory, Stanford University Medical Center
Dr. Baron reports no financial relationships relevant to this field of study.
Synopsis: At least two current commercially available and widely used rapid antigen tests performed poorly on respiratory samples with low sensitivity for detecting both the regularly circulating influenza A and B strains this season and they failed to detect the novel influenza A H1N1 (swine-origin) influenza in the majority of specimens tested. The sensitivity of direct fluorescent antibody stain assays seems to vary among laboratories. A new multiplex bead based assay (Luminex technology) was the best performing among all test methods.
Source: Ginocchio CG, 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 Virology 2009;45:191-195.
This article originally appeared in the August 2009 issue of Infectious Disease Alert.
The energetic Christine Ginocchio (Director of Clinical and Molecular Microbiology Laboratories for North Shore-Long Island Jewish Healthcare System) began receiving an unusual gift beginning on April 24, 2009. A group of local New York schoolboys had just returned from Cancun, Mexico, with influenza-like illness rampant among them, which spread quickly throughout the community. News was hitting the airwaves about a new "swine flu" outbreak in Mexico. Dr. Ginocchio's laboratory received more than 6,000 respiratory samples to test within five weeks. She was able to use this treasure trove of patient material to run assays using several methods to determine how the methods compared for detection of the novel virus. Almost before June began, and in plenty of time for other laboratories to benefit from the New York experience, the study had been written up and published. The speed of this accomplishment mirrors that of the local administration's response to her plea for more laboratory capacity. In a special session at the American Society for Microbiology General Meeting in Philadelphia, called to discuss the novel H1N1 influenza, Dr. Ginocchio told the stunned audience that her facility was able to convert two administrative rooms into a functioning virology laboratory complete with two biosafety level 2 cabinets over one weekend — surely a record for creating a laboratory from scratch, and a tribute to the dedication and leadership of her administration. The laboratory at Stanford, after already having discontinued use of rapid antigen tests due to both poor sensitivity and poor specificity, started processing direct fluorescent antibody tests around the clock in response to demand, going from a high of 50 the week before April 24th to 250 samples during the first week of the outbreak in our local area. These numbers, however, pale in comparison with those submitted to the North Shore-Long Island Jewish microbiology/virology laboratory.
The Luminex platform utilizes tiny beads to which either antigens, antibodies, or oligonucleotides that bind a bit of DNA can be attached. The beads come in 100 subtly different fluorescent colors so that it would be possible to create a multiplex assay that could detect up to 100 different analytes in one suspension. When the ligand of interest has bound to the beads, the beads are fed in a stream similar to that of a cell-sorting device where laser lights can interrogate the bead's fluorescence to determine the specific ligand, as well as how many beads actually have bound their targets. The results are a semi-quantitative determination of what targets were in the sample. The Luminex Diagnostics Company in Toronto, Canada, has developed a multiplex assay for respiratory viruses, called xTAG Respiratory Virus Panel (RVP). A total of 12 different respiratory viruses are detected in the system, which requires sample extraction (usually on the NucliSens EasyMag system, available from bioMerieux), amplification by PCR of multiple virus targets, attachment of the amplified targets to specific beads with matching sequences, and reading the information from the beads in a flow-cytometer-like reader instrument. Turnaround time is usually around 24 hours due to the multiple steps required and the need to batch samples for testing. A previous publication on the Luminex platform1 showed that the system was 10- to 100-fold less sensitive than individual real-time PCR tests for each virus individually but that, in their study of previously positive clinical samples, there were virtually no false-negative results. With a small number of prospective samples (18), Ginocchio et al were surprised to find 22% of samples had two viruses detected by the PCR method, whereas cultures and DFA methods usually report only one virus.
Commentary
Ginocchio et al performed more than 2,700 xTAG RVP tests in comparison with rapid antigen (two different manufacturers), DFA, and culture. The largest percentage (40.8%) of the samples was positive for the variant swine-origin H1N1. The next closest prevalent virus was rhinovirus, at 16.6%. Once the new H1N1 enters the population, the other influenza A viruses are quickly supplanted. This also is being seen at our own facility, where around 99% of influenza A detected are the new flu. One would suspect that viruses seen now in the New York area also will be primarily the novel H1N1, but this study was conducted during the initial stages when other viruses were still circulating. As most laboratories have quickly learned, the rapid antigen tests perform poorly with this influenza A. Before this year, we probably did not realize how insensitive the rapid antigen tests were for the other influenza A strains. The sensitivity of the rapid antigen tests for detecting novel variant H1N1 (adding results from both rapid test kits) was only 23.8%. This is not a test that any physician should continue to request.
In Ginocchio's laboratory, the DFA performed marginally better, with a sensitivity of 50%. This is quite different from the Stanford laboratory's results, at least for respiratory viruses overall. A publication done by one of our pediatric infectious diseases fellows in 2003 compared DFA with culture2 and found the sensitivity of DFA (not limited to influenza A) to be 84% in children in whom the test may have greater sensitivity than in adults. That was before we also tested for metapneumovirus by DFA. A more recent publication from the Stanford Virology laboratory3 comparing DFA to another multiplex molecular respiratory panel (NanoChip400 from Nanogen, since withdrawn from the market) showed the sensitivity of DFA (for influenza A, influenza B, respiratory syncytial virus, and parainfluenzas 1, 2, and 3) to be 94%. Of 122 positive samples tested, the DFA only missed one influenza A (the age of the subjects was not stated). This study was conducted before the novel H1N1 variant had surfaced. However, anecdotally, this season, whenever cultures or PCR tests (performed at the Santa Clara County laboratory) were compared with the DFA, there have been only rare discrepancies as well. The take-home message about DFA testing for respiratory viruses is that the experience of the clinical laboratory scientists, their care in specimen handling and testing, and the procedures used can lead to large differences in comparability or reliability of results. The DFA test is very dependent on technical expertise and, thus, laboratory performance can vary.
Fortunately, in the near future, many laboratories will have multiplex molecular respiratory virus panel tests that will yield reliable results, as did the RVP in the New York study. On a subset of 288 samples tested by all four methods, the RVP had a sensitivity of 97.8% for detection of novel H1N1 swine-origin influenza, with a positive predictive value of 100% and negative predictive value of 97.3%.1 The viral culture sensitivity was 88.9%.
Although clinicians will suffer a longer wait for the results of the RVP than for either rapid antigen or DFA results, they will be likely to get the best answer. Depending on the quality of their local laboratory, DFA testing, especially given a more rapid time to results, even if testing is batched and performed twice a day, may be a second option.
Mahony et al have just published a cost analysis of xTAG RVP vs. DFA alone and vs. DFA and rapid culture method (shell vial) to determine the best strategy for testing samples from children in their four Canadian hospitals when the next influenza season hits.4 The cost analysis included all hospitalization, antibiotic, laboratory testing, and isolation costs. Surprisingly, the costs per case were quite similar regardless of which testing strategy was used, ranging from $3,914 to $3,623. They found that the most cost-effective plan would be to use the RVP alone when the prevalence of infection was 11% and DFA alone when the prevalence was < 11%. If the RVP alone was employed to replace their current protocol of DFA followed by shell vial culture, which requires 48 hours for final results, Mahony et al estimated that the four hospitals would save more than $500,000 in one year.4 Their turnaround time for DFA alone (only performing at 72% sensitivity in their system) was four hours, and the xTAG RVP turnaround time would be 24 hours. In fact, if they were able to enhance the sensitivity of the DFA in their analysis model, then the outcomes would have slightly favored the DFA test. And since most of the cost savings found with the RVP-alone strategy resulted from decreased hospital length of stay, based on 24-hour results in comparison with their current protocol, which required 48 hours for the culture results, the use of a highly reliable DFA would result in even greater savings.
There are other benefits to multiplex PCR tests, however. As the number of laboratory technical experts in conventional virology dwindles (already beginning to happen), the skills required to perform DFA will also disappear. Automated testing platforms offer the same or better accuracy with less subjectivity. The platforms can also detect more virus types than DFA panels, and certain respiratory viruses, such as adenovirus and rhinoviruses, which are notoriously difficult to detect in either DFA or culture, depending on the virus, can be effectively targeted in molecular assays. And, in a very short time, the automation will increase and the turnaround time will decrease. Perhaps within a year or two, manual virology testing for respiratory viruses will be another technology of the past.
References
- Brunstein J, Thomas E. Direct screening of clinical specimens for multiple respiratory pathogens using the Genaco Respiratory Panels 1 and 2. Diagn Molec Pathol. 2006;15:169-73.
- Shetty AK, et al. Conventional viral cultures are not necessary if direct fluorescent antibody stains are performed for diagnosis of community-acquired respiratory virus infections in hospitalized children. Pediatr Infect Dis J. 2003;229:789-794.
- Takahashi H, et al. Evaluation of the NanoChip 400 system for detection of influenza A and B, respiratory syncytial, and parainfluenza viruses. J Clin Microbiol. 2003;46:1724-1727.
- Mahony JB, et al. Cost analysis of multiplex PCR testing for diagnosing respiratory virus infections. J Clin Microbiol. 2009 [Epub ahead of print].
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content