Wild Birds as Sentinels

 Canada’s Inter-Agency Wild Bird Influenza Survey has been testing wild birds for the AI virus since 2005

The 2015 Avian Influenza outbreak was the largest animal health event in U.S. history, affecting 48 million commercial birds at 223 farms in 15 states over six months. North of the border the outbreak affected 245,600 birds across Canada, at 11 farms in BC, three in Ontario.

But it was minus forty degrees when the AI virus first appeared in poultry flocks in the Midwest U.S. How did the wild birds interact with the poultry in that extreme cold? Are the wild birds really to blame?

While the experts still shake their heads about the reasons why the outbreak got out of control or even got started, Jane Parmley, Epidemiologist with the Canadian Wildlife Health Cooperative (CWHC), continues to investigate the role of wild birds in the spread of the Avian Influenza (AI) virus.  

Parmley has been part of Canada’s Inter-Agency Wild Bird Influenza Survey coordinated by the CWHC since 2005. From 2005 through 2014, over 50,000 wild birds have been tested for the AI virus in Canada. The first screening determines if the birds carry AI of any type. Positive birds are then further tested for H5 or H7 specifically; further positive tests then lead to investigation of the origin and pathogenicity of the AI virus.

Does the detection of low pathogenic AI in wild birds indicate a risk to domestic poultry? “We tend to blame wild birds when we don’t have an easy explanation,” Parmley told delegates at a Poultry Industry Council Health Day in Stratford, Ont., but would early detection of the AI virus in wild birds could provide a sentinel to poultry producers?

It’s a global story: the high pathogenic H5N8 strain was originally identified in South Korea in 2014, showing similarities to a virus detected in 2014 in China, eventually reaching birds in Russia, North America, Europe and Japan.  There were three HPAI virus strains seen in North America in 2014/15: H5N8 Eurasian lineage, H5N1 and H5N2. It is believed that the North American viruses came across the Pacific because of their closer similarity to the Asian strain than the European strain, and the timing of arrival made more sense, said Parmley. So far the virus is not considered zoonotic, but that could shift quickly.

Wild birds that are considered as natural reservoirs of low pathogenicity strains include waterfowl (ducks, geese and swans), and shorebirds (such as waders and gulls). There are four flyways across North America – the Pacific, Central, Mississippi and Atlantic – connecting wintering and breeding grounds in every part of the continent, from Alaska and Greenland through to Mexico and the Caribbean.

Over 75 per cent of Canadian wild bird species spend at least half of the year outside of Canada; representatives of all of these species are in all the flyways. Because the greatest number and variety of viruses have been seen in waterfowl and shorebirds and their large population, these birds have been the focus of live bird surveillance; most of the live birds sampled have been ducks.

On average, 16 per cent of live birds and one per cent of dead birds have so far tested positive for the low pathogenic North American viruses during the survey period. So far in 2015, 1134 live birds have been tested across Canada, with 93 positive for AI and three for H7 viruses that are not HPAI. In Ontario alone, 624 live birds have yielded 86 positives with no H5 or H7 so far. Also in 2015, 1576 dead birds have been tested across Canada, with 16 positive, four H5 and one H7 (not HPAI); in Ontario alone 266 have been tested with two positive and no H5 or H7. The updated results are available on the CWHC website at http://www.cwhc-rcsf.ca/data_products_aiv.php

But Parmley says that the surveillance effort has varied over the past ten years for several reasons. Survey objectives have changed and available resources have changed; sample sizes are low compared to real populations, and samples are taken haphazardly across the country, often piggybacking on bird banding procedures that may not necessarily be anywhere near poultry farms. We also need to be careful when extrapolating past results over a wide geographic region to the viruses of today.

Moving forward there are still many questions. Will virus based warnings even work? Can wild birds be sentinels? Ultimately we’re trying to develop an early warning system to predict risk and see how the virus is evolving, said Parmley, something she called “incredibly hard to do.” To better protect poultry farms, Parmley says this effort will take more resources – people, time and money.

When is the best time of year to test? How early do we disseminate findings to react in a timely manner? Does wild bird monitoring detect the risk sooner than monitoring the poultry population itself? Can we verify the signals and the risk associated with those signals to avoid an unnecessary response? Is there adequate infrastructure and political will to design and implement a sustainable system?

While wild birds are acknowledged as a reservoir for the AI virus, the relationship between hosts and virus remains diverse and complex. We need to further consider the epidemiological, climatological and agricultural differences across such a vast area, from the Arctic to the Caribbean, as well as at the interface between wild and farmed birds. So far we suspect that the virus can move through migratory birds, but it can also move through trade in poultry and poultry products as well as other human activities.

Nationally, for 2015/16, the goal of testing 1500 dead wild birds has already been exceeded. In Ontario the plan is to test 1500 live wild birds. One live trap site has been set within the control zone of the central Ontario AI outbreak in the spring of 2015.

Despite the challenges, so far Parmley reports a greatly improved understanding of the ecology and epidemiology of AI.  We now have a better understanding of the role of waterfowl as a source and vector, the activity of the virus itself within the host, and a better grasp of how the virus is shed.

Beyond 2015, Parmley suggests a thorough process review to identify gaps, identify locations and populations that may be more vulnerable to infection, to help target both resources and surveillance.  A national strategy would clarify roles, responsibilities and performance expectations.

“The virus keeps changing and so we need to keep learning,” said Parmley. The question that remains from the 2014/15 outbreaks in North America is how the virus got into poultry? “We can’t look at wild birds and domestic poultry separately – it is the points and places where these populations intersect where we need to focus our attention if we hope to prevent and prepare for the next outbreak.”