Rapid-test sensitivity for novel swine-origin pandemic influenza A

• Published in: Journal of the Formosan Medical Association 2012-08, Vol.111 (8), p.427-430
• Main Authors: Tai, Chun-Fu, Lu, Chun-Yi, Shao, Pei-Lan, Lee, Ping-Ing, Chang, Luan-Yin, Huang, Li-Min
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Age; Antigens, Viral - isolation & purification; Child; Children; Data processing; Female; Fever; Humans; Influenza; Influenza A; Influenza A Virus, H1N1 Subtype - immunology; Influenza A Virus, H1N1 Subtype - isolation & purification; Influenza, Human - diagnosis; Influenza, Human - epidemiology; Influenza, Human - virology; Internal Medicine; Logistic Models; Male; Middle Aged; novel pandemic influenza A; Outbreaks; Pandemics; Polymerase chain reaction; Prospective Studies; rapid antigen test; Reagent Kits, Diagnostic; Real-Time Polymerase Chain Reaction; Regression analysis; Reverse Transcriptase Polymerase Chain Reaction; Sensitivity; Sensitivity and Specificity; Taiwan - epidemiology; Viral Load; Young Adult
• ISSN: 0929-6646
• DOI: 10.1016/j.jfma.2011.06.018

Background/Purpose Rapid diagnosis of influenza was important in the global pandemic influenza A/H1N1 outbreak. The QuickVue Influenza A+B test is one of the most common tests for rapid diagnosis of influenza. We evaluated the sensitivity and specificity of the test in children. Methods We collected data from 970 patients with influenza-like illness who received rapid influenza antigen tests using the QuickVue Influenza A+B test as well as viral isolation. We compared the results with that of viral isolation and reverse-transcriptase polymerase chain reaction (RT-PCR) assays. Results Based on viral culture, the QuickVue Influenza A+B test had an overall sensitivity of 0.82 (419/513) and specificity of 0.99 (70/71), with a positive predictive value (PPV) of 0.74 (419/563) and a negative predictive value (NPV) of 0.77 (313/407). The sensitivity of QuickVue was significantly higher in specimens with high viral loads. If the viral loads were less than 106 , the sensitivity of QuickVue was 0.62, while the sensitivity of QuickVue was 0.88 if the viral loads were higher than 106 ( p < 0.001). Logistic regression analysis showed that higher viral loads correlated with positive QuickVue results ( p < 0.001). On the first day of fever, the sensitivity of QuickVue was only 0.67; on the second day, the sensitivity was 0.86; on the third day, the sensitivity was 0.98, and on the fourth day, the sensitivity was 0.90. The sensitivity is significantly higher on days 2−3 in comparison with the first day of fever ( p < 0.05). We calculated the correlation of viral load and fever days, and the result showed higher mean viral load on the second and third days of fever. Age did not affect the sensitivity. Conclusion In children, the sensitivity of QuickVue Influenza A+B test was 0.82. In addition, the sensitivity was significantly elevated in the higher viral load group and on the third day of fever.