Diagnostic capacity of rapid influenza antigen test: Reappraisal with experience from the 2009 H1N1 pandemic

Abstract

The rapid influenza antigen test (RIAT) has been questioned because of its poor sensitivity. Clinicians are confused as to what diagnostic help it may provide. RIAT was reappraised by other laboratory confirmatory tests for its diagnostic capacity. Records of RIAT, RT-PCR and virus culture, performed for upper respiratory tract samples during the period from July 2009 to January 2010, were reviewed. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of RIAT were evaluated with RT-PCR and virus culture as reference. With either positive RT-PCR or positive virus culture as confirmation of presence of the virus, the overall sensitivity was 44.2% (235/532) and the overall PPV was 91.1% (235/258). With both negative RT-PCR and negative virus culture as confirmation of absence of the virus, the overall specificity was 98.2% (239/264) and the overall NPV was 92.6% (239/258). The PPV reached 96.2% during peak prevalence of infection and the NPV increased to 91.7% with low prevalence. The sensitivity for seasonal H3N2 was 56% (56/100), significantly better than the 39.6%, (156/394) for 2009 pandemic H1N1. Although RIAT positivity correlated to the viral load in samples, a substantial amount of negative RIAT samples had high viral load, with 16.8% (260/1548) of Ct value less than 36 and 8.8% (136/1548) of Ct value less than 31. An algorithm is derived for the fast and inexpensive point-of-care laboratory test RIAT for appropriate application in clinical diagnosis of influenza virus infection. In peak seasons, positive RIAT confirms the diagnosis, with PPV over 96%. In low seasons, negative RIAT sufficiently excludes the diagnosis, with NPV over 91%. The sensitivity of RIAT may vary with different species of the influenza virus. Negative RIAT is not necessarily equal to low viral load in the upper respiratory tract or low infectivity of the patient.