Abstract

BACKGROUND: Effective strategies to combat the spread of influenza viruses rely on accurate diagnosis of the infection, which is greatly affected by the sensitivity of the assays used. OBJECTIVES: To determine if the sensitivity of assays used to detect influenza varies with the viral strains. STUDY DESIGN: This prospective study used 142 respiratory specimens submitted to the clinical virology laboratory during the 2003/2004 influenza season, when the H3N2 Fujian strain was first detected in our community, to assess sensitivities of rapid immunoassay (RIA), shell-vial (SV), hemadsorption (HA)-, and fluorescent antibody (FA)-based conventional tissue culture (TC). Results were compared with similar evaluations performed in 2002/2003 or earlier seasons. RESULTS: Sensitivities of RIA, SV, HA-TC, and FA-TC were 44, 70, 82, and 100%, respectively for influenza virus circulating in 2003/2004. The comparison with results obtained in previous influenza seasons showed a significantly decreased sensitivity of HA-TC (p=0.02) and lower sensitivity of RIA but without reaching statistical significance (44% versus 72%, p=0.2). CONCLUSIONS: This report underscores the importance of determining the accuracy and sensitivity of diagnostic tests against emerging influenza variant strains.