An evaluation of avian influenza diagnostic methods with domestic duck specimens.

Monitoring of poultry, including domestic ducks, for avian influenza (AI) virus has increased considerably in recent years. However, the current methods validated for the diagnosis and detection of AI virus infection in chickens and turkeys have not been evaluated for performance with samples collected from domestic ducks. In order to ensure that methods for the detection of AI virus or AI virus antibody will perform acceptably well with these specimens, samples collected from domestic ducks experimentally infected with a U.S. origin low pathogenicity AI virus, A/Avian/NY/31588-3/00 (H5N2), were evaluated. Oropharyngeal (OP) and cloacal swabs were collected at 1, 2, 3, 4, 5, 7, 10, 14, and 21 days postinoculation (PI) for virus detection by virus isolation, which was considered the reference method, and real-time RT-PCR. In addition, two commercial antigen immunoassays were used to test swab material collected 2-7 days PI. Virus isolation and real-time RT-PCR performed similarly; however, the antigen immunoassays only detected virus during the peak of shed, 2-4 days PI, and both kits detected virus in fewer than half of the samples that were positive by virus isolation. Cloacal swabs yielded more positives than OP swabs with all virus detection tests. To evaluate AI virus antibody detection serum was collected from the ducks at 7, 14, and 21 days PI and was tested by agar gel immunodiffusion (AGID) assay, a commercial blocking enzyme-linked immunosorbent assay (ELISA), and homologous hemagglutination inhibition (HI) assay, which was used as the reference method. Results for the ELISA and HI assay were almost identical with serum collected at 7 and 14 days PI; however, by 21 days PI 100% of the samples were positive by HI assay and only 65% were positive by ELISA. At all time points AGID detected antibody in substantially fewer samples than either ELISA or HI assay.