Chapter 45 Avian Influenza H5N1 Virus
Epidemiology in Wild Birds, Zoo Outbreaks, and Zoo Vaccination Policy
Influenza A viruses are a genus of highly variable, negative-strand RNA viruses within the family Orthomyxoviridae. Influenza A viruses infect birds and a number of mammals, including humans, but exhibit greatest diversity among avian taxa, particularly Anseriformes (ducks, geese, swans, and allies) and Charadriiformes (waders, gulls, auks, and allies), which are widely believed to constitute the natural reservoir.25 The genus is characterized using the antigenic characteristic of two surface glycoproteins, hemagglutinin (16 subtypes) and neuraminidase (9 subtypes). Some strains of the H5 and H7 subtypes are capable of causing high mortality in domestic poultry (up to 100%), and are termed highly pathogenic avian influenza (HPAI) viruses, with others causing low mortality to poultry being termed low- pathogenic avian influenza (LPAI) viruses.26 HPAI virus strains are believed to arise through mutation and selection of an LPAI progenitor virus, following introduction into domestic birds.2 In the last half-century, there have been 26 recorded epizootics of HPAI virus, the largest of which is the current epizootic of HPAI H5N1 virus that emerged in 1996.2
Current Highly Pathogenic Avian Influenza H5N1 Virus Epizootic
In 1996, an HPAI virus of the H5N1 subtype was isolated from an outbreak affecting domestic geese in Guangdong Province, China. A year later, a related virus emerged in Hong Kong that affected poultry, and led to the first human clinical respiratory cases, with hospitalization of 18 patients, of whom 6 died. Subsequent isolates in neighboring territories from 1998 to 2002 suggested that the virus continued to circulate in China, undergoing a number of genetic reassortments.23 In late 2003 and early 2004, eight countries in East and Southeast Asia reported outbreaks for the first time,1 with the virus establishing itself in some areas, particularly those integrating rice cultivation with free grazing of domestic ducks.13 By 2005, multiple sublineages of HPAI H5N1 virus had become established among domestic poultry in geographic subregions of Asia, indicating long-term endemicity and spatial isolation.7 Phylogenetic classification of these viruses, using a unified nomenclature based on H5N1 hemagglutination (HA) sequences from the goose Guangdong lineage, identified 10 major clades (designated 0 to 9) and numerous subclades.28 In spring 2005, an outbreak affecting wild migratory waterfowl in Qinghai Province, China,6 marked the onset of a range expansion that saw outbreaks in wild and domestic birds recorded over an area extending progressively westward through Central Asia to Europe to the Middle East and Africa.
By the end of 2009, 62 countries or territories had recorded outbreaks of HPAI H5N1,26 with 468 human cases and 282 deaths reported across 15 countries.29 Cases have also been recorded in a number of mammalian species, including canids, felids, viverids, mustelids, lagomorphs, suids, and primates. Globally, the number of outbreaks or cases in poultry and humans peaked annually in the January through March period each year, but the size of these peaks has declined annually.9 Although variation in the intensity of national surveillance and frequency of reporting inhibit firm conclusions, effective control measures appear to have led to a steady reduction in the numbers of countries affected, with the virus now largely confined to endemic regions in Northeast Africa and South and Southeast Asia.
The ongoing HPAI H5N1 epizootic has been unusual in the extent to which wild birds have been affected. Prior to this, the only records of HPAI in wild birds were the isolation of HPAI H5N3 virus following the death of 1300 common terns (Sterna hirundo) in South Africa in 1961 and a case of H7 infection in a saker falcon (Falco cherrug) in Italy at the time of an HPAI H7N1 virus outbreak affecting poultry. The first wild bird cases of HPAI caused by the H5N1 subtype were detected in Hong Kong in December 2002 in wild and ornamental birds at four sites. Initially, further (sporadic) cases in wild birds occurred in the vicinity of infected poultry and were likely the result of local spillover from poultry. The possibility that wild birds might be capable of long-distance transmission of virus arose with the mortality of over 6000 wild birds during the outbreak at Qinghai Lake in April 2005.6 Although the 2005 mortalities at Qinghai remain the largest reported in wild birds, further outbreaks involving tens or hundreds of birds of more than 60 species occurred at sites in Europe and Central Asia in 2006 and 2007. The regularity of wild bird outbreaks of Qinghai-like HPAI H5N1virus (clade 2.2) has declined since 2007. However, evidence is emerging that another strain of HPAI H5N1 virus, clade 2.3.2, may have established itself in wild birds, with isolates from Hong Kong in 2007 and 2008, Japan in 2008, Russia in 2009 and Mongolia in 2009 and 2010.
Considerations in Wild Birds
Ecology of Low and Highly Pathogenic Avian Influenza
The comparative lack of precedent for HPAI virus infections among wild birds prior to 2002 limits our ability to predict how HPAI H5N1 viruses will behave in wild populations. Although extrapolation based on the epidemiology of LPAI viruses may be helpful, inferences should not be overinterpreted, because differences in pathogenesis may significantly affect the dynamics of HPAI virus transmission in wild populations. Recognizing these shortcomings, a discussion of HPAI with reference to the ecology of LPAI may still be instructive in predicting the behavior of the virus.
Influenza A viruses are able to persist for a prolonged period in an aquatic environment, which together with ecologic factors, such as feeding behavior and sociality, may explain the prominence of water birds in the epidemiology of LPAI viruses in the wild.17 LPAI viruses follow a fecal-oral transmission cycle, with viral replication and shedding occurring within the intestinal mucosa. By contrast, HPAI viruses have a wider tissue affinity, with greater respiratory involvement, although the significance of this on the ecology of the virus in wild birds is unknown.22
Infections with LPAI viruses are traditionally believed to incur minimal cost on wild bird hosts, in most cases remaining entirely subclinical. However, more recently, this view has been challenged, and LPAI infections may have more subtle affects on behavior and ecology, such as those on migration, feeding rate, and body weight.24
Immune status plays an important role in the cycling of LPAI virus in host populations, with prevalence rates highest among naïve juvenile birds, resulting in seasonal peaks of infection in the postbreeding period.16 The role of prior exposure to influenza A viruses in the survival and viral shedding patterns of birds infected with HPAI viruses is poorly known, although experimental infection of mute swans has implied that naturally acquired avian influenza–specific antibodies protect swans from clinical HPAI H5N1 virus infection, although viral shedding still occurs.
Source of Highly Pathogenic Avian Influenza H5N1 Virus Infection
The vast majority of isolates of HPAI H5N1 virus from wild birds have been collected from incapacitated or dead birds,2 but this is of limited value in understanding how the virus perpetuates in diverse avian communities. Of greater importance is the identity of species that may withstand, shed, and potentially disseminate the virus over moderate or longer distances. Species variation in survival and shedding of HPAI virus has been demonstrated experimentally, yet few isolates have been obtained from live, apparently healthy birds in the wild. The failure to identify wild asymptomatic carriers of HPAI viruses should not be taken as evidence that such birds do not exist, because the costs and logistics of obtaining statistically robust sample sizes are prohibitive. However, considering the many tens of thousands of wild, healthy birds that have been tested in endemic areas,16 we may conclude that at most, asymptomatic carriers are extremely rare within the wild bird population. This should be contrasted with the observation that predominant strains of LPAI virus may be detected from multiple individuals within a migratory flyway in any year,15 emphasizing the differences in the epidemiology of HPAI with respect to LPAI in wild birds.
There has been intense debate over the relative importance of wild birds and domestic fowl in the dissemination of HPAI H5N1 virus.10,12 The situation is complicated by the presence of backyard or extensive husbandry systems employed in many areas, and the complexity of supply chains that serve the poultry industry. This extensive overlap of domestic and wildlife sectors, coupled with inherent difficulties in wild bird surveillance, make it impossible to draw firm conclusions about the source of outbreaks in most cases. However, the occurrence of outbreaks in remote areas such as Mongolia or the Tuva Republic, where poultry are effectively absent, suggests that wild birds are able to carry virus, at least over moderate distances, and seed new outbreaks. The frequency at which this occurs is unknown, and the continued absence of HPAI of the H5N1 subtype in some areas that receive large populations of migrants from endemic regions, notably Australasia, suggests that dissemination of virus by wild birds, at least in some taxonomic groups, is very rare indeed.