Detection of influenza B in clinical specimens: comparison of high throughput RT-PCR and culture confirmation

Influenza virus is one of the major causes of worldwide respiratory tract infections during the winter season. Here we describe a high throughput (HTP) protocol for rapid diagnosis of influenza B that combines automated viral RNA extraction with detection and quantification by TaqMan-based PCR. Using this methodology, we tested 4176 nasal swabs collected from children enrolled in a European influenza vaccine trial during the winter of 2000 to compare our HTP PCR method to culture confirmation for detection of influenza B. Among these, 37 were positive by culture and 169 were positive by PCR irrespective of virus copy number. However, when specimens with fewer than 20 copies of the viral genome were disregarded, a good correlation between two methods was observed. At this cut-off, 34 specimens were positive and 4106 were negative by both methods. Statistical analysis of the data using culture confirmation as the standard indicated that the sensitivity of HTP RT-PCR for influenza B was 92% (95% CI, 0.93-1), and the specificity was 99% (95% CI, 0.99-1). In summary, HTP RT-PCR was proved to be more rapid and sensitive than culture confirmation, and it correlated significantly with culture confirmation for specimens containing more than 20 copies of the viral genome. (C) 2004 Elsevier B.V. All rights reserved.