Prevalence of Influenza A viruses in wild migratory birds in Alaska: patterns of variation in detection at a crossroads of intercontinental flyways

Abstract

Background: The global spread of the highly pathogenic avian influenza H5N1 virus has stimulated interest in a better understanding of the mechanisms of H5N1 dispersal, including the potential role of migratory birds as carriers. Although wild birds have been found dead during H5N1 outbreaks, evidence suggests that others have survived natural infections, and recent studies have shown several species of ducks capable of surviving experimental inoculations of H5N1 and shedding virus. To investigate the possibility of migratory birds as a means of H5N1 dispersal into North America, we monitored for the virus in a surveillance program based on the risk that wild birds may carry the virus from Asia.

Results: Of 16,797 birds sampled in Alaska between May 2006 and March 2007, low pathogenic avian influenza viruses were detected in 1.7% by rRT-PCR but no highly pathogenic viruses were found. Our data suggest that prevalence varied among sampling locations, species (highest in waterfowl, lowest in passerines), ages (juveniles higher than adults), sexes (males higher than females), date (highest in autumn), and analytical technique (rRT-PCR prevalence = 1.7%; virus isolation prevalence = 1.5%).

Conclusion: The prevalence of low pathogenic avian influenza viruses isolated from wild birds depends on biological, temporal, and geographical factors, as well as testing methods. Future studies should control for, or sample across, these sources of variation to allow direct comparison of prevalence rates.

Figure 1

Geographical location of sampling sites in Alaska in 2006 and 2007. Hunter-harvest sampling locations are noted in red. Live bird surveillance sampling locations are marked in green. Habitat classifications based on Bird Conservation Regions [36,37].

Figure 2

Predicted probability of prevalence based on rRT-PCR results by logistic regression. Lines cover the range of dates sampled for each age and sex class. Grey (dashed) lines represent the 95% confidence intervals from the logistic regression. Lines are not extrapolated beyond the range of data and therefore represent dates of sampling. A. Northern Pintails. B. American Green-winged Teal. C. Mallards.