Rapid H1N1 Test Relatively Insensitive
Preliminary data from the CDC indicate that rapid influenza diagnostic tests (RIDTs) for the novel influenza A (H1N1) virus are relatively insensitive—especially when viral titers are low in specimens from nasopharyngeal and oropharyngeal swabs. Therefore a negative RIDT for the novel H1N1 virus does not rule out the diagnosis of swine flu, particularly in the context of consistent clinical symptoms. The CDC recommends that, if laboratory confirmation is required, further testing should be performed by using the real-time reverse transcription-polymerase chain reaction (rRT-PCR) assay or viral culture.
The CDC determined the following sensitivities of 3 commercially available RIDTs by using respiratory specimens that were positive for the novel H1N1 virus with rRT-PCR. However, the CDC advises against discriminating among the 3 tests' sensitivities, because of the relatively low number of specimens analyzed.
|
Test |
High Novel H1N1 Titer |
Low Novel H1N1 Titer |
All Novel |
|
89% |
28% |
40% | |
|
89% |
38%* |
49% | |
|
100% |
61% |
69% |
* n = 34.
The sensitivities of the RIDTs declined in proportion to declining titers of novel H1N1. Factors that affect influenza titers in specimens include the timing of specimen collection during illness, the age of the patient, the type of specimen collected, and the transportation and storage of the specimen.
The overall sensitivities of the RIDTs for seasonal H1N1 (n = 5) or seasonal H3N2 (n = 15) were observed to be 60%-80% and 80%-83%, respectively. (It is important to note that RIDTs do not distinguish among subtypes of influenza A virus.) The CDC advises that the results of RIDTs should be interpreted in the context of known circulating viral strains and provides this link for guidance.
N. B.—The sensitivity and specificity of the rRT-PCR assay, when compared with the reference standard of viral culture, are 99.3% and 92.3%, respectively.
08/13/09 update: RIDT data from the Naval Health Research Center indicate respective sensitivities of 51% for the detection of novel H1N1 (n = 39 patients), 63% for seasonal H1N1 (n = 19), and 31% for H3N2 (n = 19). Specificities for all influenza A virus types were 99%. Values are for the QuickVue Influenza A+B test; the reference standard is rRT-PCR.
