Comparison of sensitivities of virus isolation, antigen detection, and nucleic acid amplification for detection of equine influenza virus.

Four seronegative foals aged 6 to 7 months were exposed to an aerosol of influenza strain A/Equi/2/Kildare/89 at 10(6) 50% egg infective doses (EID(50))/ml. Nasopharyngeal swabs were collected for 10 consecutive days after challenge. Virus isolation was performed in embryonated eggs, and the EID(50) was determined for all positive samples. The 50% tissue culture infective dose was determined using Madin-Darby canine kidney (MDCK) cells. Samples were also tested by an in vitro enzyme immunoassay test, Directigen Flu A, and by reverse transcription-PCR (RT-PCR) using nested primers from the nucleoprotein gene and a single set of primers from the matrix gene. RT-PCR using the matrix primers and virus isolation in embryonated eggs proved to be the most sensitive methods for the detection of virus. The Directigen Flu A test was the least sensitive method. The inclusion of 2% fetal calf serum in the viral transport medium inhibited the growth of virus from undiluted samples in MDCK cells but was essential for the maintenance of the virus titer in samples subjected to repeated freeze-thaw cycles.