Between December 2003 and January 2004 highly pathogenic avian influenza (HPAI) H5N1 infections of poultry were declared in China, Japan, South Korea, Laos, Thailand, Cambodia, Vietnam, and Indonesia. In 2004 an outbreak was reported in Malaysia. In 2005 H5N1 outbreaks were recorded in poultry in Russia, Kazakhstan, Mongolia, Romania, Turkey, and Ukraine, and virus was isolated from swans in Croatia. In 2004 HPAI H5N1 virus was isolated from smuggled eagles detected at the Brussels Airport and in 2005 imported caged birds held in quarantine in England. In 2006 HPAI was reported in poultry in Iraq, India, Azerbaijan, Pakistan, Myanmar, Afghanistan, and Israel in Asia; Albania, France, and Sweden in Europe; and Nigeria, Cameroon, and Niger in Africa; as well as in wild birds in some 24 countries across Asia and Europe. In 2003, over 25,000,000 birds were slaughtered because of 241 outbreaks of HPAI caused by virus of H7N7 subtype in the Netherlands. The virus spread into Belgium (eight outbreaks) and Germany (one outbreak). HPAI H5N2 virus was responsible for outbreaks in ostriches in South Africa during 2005. HPAI H7N3 virus was isolated in Pakistan in 2004. Low-pathogenicity avian influenza (LPAI) H5 or H7 viruses were isolated from poultry in Italy (H7N3 2002–2003; H5N2 2005), the Netherlands (H7N3 2002), France (H5N2 2003), Denmark (H5N7 2003), Taiwan (H5N2 2004), and Japan (H5N2 2005). Many isolations of LPAI viruses of other subtypes were reported from domestic and wild birds. Infections with H9N2 subtype viruses have been widespread across Asia during 2002–06.

Between 2002 and 2005, three outbreaks of highly pathogenic avian influenza (HPAI) occurred in the Americas: one outbreak in Chile (H7N3) in 2002, one outbreak in the United States (H5N2) in 2004, and one outbreak in Canada (H7N3) in 2004. The outbreak in Chile was limited to a large broiler breeder operation and a nearby turkey flock and represented the first outbreak of HPAI in that country. The outbreak of HPAI in the United States occurred in Texas and was limited to one premise where chickens were raised for sale in nearby live-bird markets. The outbreak in Canada was the largest of the three HPAI outbreaks, involving 42 premises and approximately 17 million birds in the Fraser Valley, British Columbia. In each of the HPAI outbreaks, the disease was successfully eradicated by depopulation of infected farms. All other reports of infections in poultry and isolations from wild bird species pertained to low pathogenicity avian influenza (LPAI) viruses. Animal Health Officials in Canada reported subtypes H3, H5, and H6 in domestic poultry, and H3, H5, H11, and H13 from imported and/or wild bird species. An LPAI H5N2 virus continues to circulate in Mexico and the Central American countries of Guatemala and El Salvador. Each country reported isolations of H5N2 virus from poultry and the large-scale use of inactivated and recombinant H5 vaccines in their AI control programs. In Colombia, AI was reported for the first time when antibodies to H9N2 were detected in chickens by routine surveillance. Intensive surveillance activities in the United States detected AI virus or specific antibodies to 13 of the 16 hemagglutinin (H1–H13) and all nine neuraminidase subtypes in live-bird markets, small holder farms, and in commercial poultry from 29 states. The largest outbreak of LPAI in the United States occurred in 2002, when 197 farms were depopulated (4.7 million birds) to control an outbreak in Virginia and surrounding states. The outbreak was caused by an LPAI H7N2 virus closely related to an H7N2 virus that has been circulating in the live-bird marketing system in the northeastern United States since 1994.

Numerous lessons have been learned so far in controlling H5N1 avian influenza in Asia. Early detection of incursions of virus prevented establishment of the disease in several countries, notably Japan, South Korea, and Malaysia. In countries where detection of early cases was delayed, infection is endemic and has been for three or more years. Control measures implemented in these countries need to reflect this finding. Vaccination will continue to be one of the key measures used in these endemically infected countries. Used alone, vaccination will not result in elimination of H5N1 viruses from a country, but, if used correctly, it will markedly reduce the prevalence of and susceptibility to infection. Vaccination has already played a valuable role in reducing the adverse effects of H5N1 viruses. Mass culling also reduces the level of infection in infected areas. However, the long-term benefits are limited in endemically infected countries owing to the high probability of reinfection on restocking unless other measures are used in parallel. Full epidemiological studies have not been conducted in many infected countries. Nevertheless, it is recognized that the number of clinical cases does not truly reflect the levels of infection. Domestic ducks and large live poultry markets have played a key role in the persistence of infection, because they can be infected silently. In tackling this disease, countries should adopt integrated control programs using the combination of measures best suited to the local environment. All surveillance data should be shared, both positive and negative, and should include information on cases of infection and disease. Socioeconomic and ecological implications of all control measures should be assessed before implementation, especially the impact on the rural poor.

Outbreaks of H5N1 highly pathogenic avian influenza (HPAI) occurred in various types of domestic poultry in Thailand during 2004–05. H5N1 viruses were also detected in humans and other mammalian species. Infections were mainly detected in backyard chickens and domestic ducks. The geographic distribution of the 2004 outbreaks was widespread throughout Thailand; most outbreaks occurred in the Central Region, the southern part of the Northern Region, and the Eastern Region. In 2005, the H5N1 outbreaks continued and showed a clustered pattern in four provinces in the southern part of the Northern Region and in one province in the Central Region. H5N1 HPAI outbreaks caused serious socioeconomic consequences to the poultry industry, the social community, farmers' livelihood, and human health. After key measures were implemented, the incidence of the outbreaks declined remarkably in 2005.

From November 2003 to June 2004 an epidemic of high pathogenicity avian influenza (HPAI) virus of subtype H7N3 affected the major layer and broiler-breeder raising areas of the country. This was accompanied by an outbreak of low pathogenicity avian influenza (LPAI) virus of type H9N2 in broilers and layers, which continued during 2005. Subsequently, in February 2006 avian influenza virus (AIV) subtype H5N1 was for the first time found in two isolated commercial flocks in this country. The HPAI outbreak of 2003–2004 was eventually overcome by enforcing biosecurity measures, controlling poultry movements, using inactivated vaccines, and introducing a comprehensive AI surveillance network throughout the country. However, similar measures undertaken to control H9N2 outbreaks have not been successful in the affected areas, with continuing increased mortality and heavy production losses in broilers and layers, respectively. A similar strategy has been devised to combat the spread of newly introduced H5N1 HPAIV. The description of these outbreaks and the results of the control strategy are reported here.

Avian influenza (AI) outbreaks were first reported in Thailand in January 2004. In the past 2 yr, AI viruses have caused three epidemic waves. Disease prevention and control in all aspects have been actively carried out. Active and passive surveillance based on clinical observation and laboratory analysis were intensively conducted, as well as monitoring of genetic variation of the viruses. H5N1 viruses isolated from different avian species from different cases and locations were selected. We have sequenced specific genes (HA, NA, M, Ns, and part of PB2 genes) of 58 H5N1 isolates, as well as whole genome sequencing of 21 Thai influenza A (H5N1) viruses isolated during the 2004–2005 outbreak. Cluster analysis study showed that AI isolates were identified as highly pathogenic avian influenza (HPAI) and belonged to genotype Z. The virus had a multiple basic amino acid motif at the cleavage site of HA, deletions in the NA stalk region, a five amino acid deletion in the NS1 gene, and genetic markers for amantadine resistance in the M2 gene. All 58 H5N1 isolates were closely related and grouped into the same cluster, together with isolates from wild birds, cats, tigers, and humans. Phylogenetic analysis also revealed that Thai isolates were in the same cluster as Vietnamese isolates but aligned in a different cluster from Indonesian, Hong Kong, and Chinese viruses. In addition, genetic analysis showed that most avian influenza virus (AIV) isolates from Thailand had no major genetic changes in each gene such as HA (HA cleavage site, receptor binding site, N-link glycosylation site), NA (NA stalk region, oseltamivir resistance marker), M (the amantadine resistance marker, host specificity site), NS (five amino acid deletion site), and PB2 (host specificity site). All Thai poultry isolates contained the amantadine resistance marker while none of them had the oseltamivir resistance marker. To this end, the molecular characterization of H5N1 viruses from Thailand showed that there were no significant point mutations in the critical regions, and there was no evidence of changes in the viruses that indicate they are capable of sustained human-to-human transmission.

Molecular diagnostic tests are commonly used to diagnose avian influenza virus because they are sensitive and can be performed rapidly, with high throughput, and at a moderate cost. Molecular diagnostic tests recently have proven themselves to be invaluable in controlling disease outbreaks around the world. Several different methods, including traditional reverse transcription-polymerase chain reaction (PCR), real-time reverse transcription-polymerase chain reaction, and nucleic acid sequence-based amplification among others, have been described for the diagnosis of avian influenza in poultry with many different variations of primers, probes, enzymes, etc. Few of these tests have been validated, with the understanding that validation should be described as a level of comparison testing to show “fitness for purpose.” None of the molecular diagnostic tests are validated for all species or specimen types that might be presented to a diagnostic laboratory. The sensitivity and specificity for all the molecular tests are governed by three critical control points, including RNA extraction, enzymes used for amplification, and the sequence of primers and probes. The RNA extraction step is of particular concern, since high-quality RNA is needed for any of the molecular tests. Some sample types, including cloacal (fecal) swabs and tissues, are difficult to process, with issues of poor RNA extraction or PCR inhibitors being common. The development of internal controls, robotics, and bead reagents are providing improved performance of existing tests, and new technologies will likely provide better tests for the future. With any molecular test, assay assurance must be performed on an ongoing basis, which includes the use of proficiency panels to measure test performance.

In order to support eradication efforts of avian influenza (AI) infections in poultry, the implementation of “differentiation of infected from vaccinated animals” (DIVA) vaccination strategies has been recommended by international organizations. These systems enable the detection of field exposure in vaccinated flocks, and through this detection, infected flocks may be properly managed, thus interrupting the perpetuation of the infectious cycle. A promising system, based on the detection of antibodies to the nonstructural 1 (NS1) protein of AI, has been deemed a good candidate. However, there are presently no data available, in support of this DIVA system, with regard to the kinetics of antibody production against the NS1 proteins in poultry following infection. The present investigation was undertaken to establish the dynamics of the appearance of anti-NS1 antibodies in a naïve population. Following experimental infection of turkeys, antibodies to a peptide spanning the c-terminal of the NS1 protein were detected by enzyme-linked immunosorbent assay (ELISA) starting between day 3 and day 5 postinfection. In contrast, no antibodies to the NS1 peptide could be detected in chickens over the test period. In addition, the turkeys and chickens reacted differently at a clinical level to the infection by the H9N2 challenge virus. Taken together, these findings indicate that there is a significant difference in the viral replication in turkeys and chickens, resulting in a variation in the production of antibodies to NS1, as detected by the peptide-based ELISA used. This fact must be taken into consideration when using a DIVA system based on the identification of antibodies to the NS1 protein.

Real-time reverse transcriptase–polymerase chain reaction (RRT-PCR) is becoming an established first-line diagnostic assay as well as a precise quantification tool for avian influenza virus detection. However, there remain some limitations. First, we show that the sensitivity of RRT-PCR influenza detection can be 10- to 100-fold inhibited in oropharyngeal and cloacal swabs. Adding 0.5 U of heat-activated Taq DNA polymerase successfully reverses PCR inhibition. Second, an excellent strategy for detecting false negative samples is the coamplification of an internal control from each sample. We developed a universal avian endogenous internal control (bird β-actin) and apply it to influenza A diagnosis. Moreover, this internal control proves useful as a normalizer control for virus quantification, because β-actin gene expression does not change in infected vs. uninfected ducks. A combined panel of wild bird cloacal swabs, wild bird tissue samples, experimental duck swabs, and experimental duck and chicken tissue samples was used to validate the endogenous control. The application of an endogenous internal control proves an excellent strategy both for avoiding false negative diagnostic results and for standardizing virus quantification studies.

Vaccination programs for the control of avian influenza (AI) in birds have restrictions because of some limited efficacy and the difficulty of discriminating between vaccinated and virus-infected poultry. We studied M2e, the highly conserved external domain of the influenza A M2 protein, as a potential differential diagnostic marker for influenza virus infection. The M2 protein is an integral membrane protein, scarcely present on virus particles, but abundantly expressed on virus-infected cells. M2e-specific enzyme-linked immunosorbent assays (ELISAs) for different avian influenza strains were developed by coating the peptides corresponding to the first 18 amino acids, without the first methionine, of the universal human consensus M2e sequence and the specific M2e sequence of two highly pathogenic AI (HPAI) strains, H7N7 and H5N1. Using the M2e ELISAs, M2e-specific antibodies were observed in chickens and ducks experimentally infected with H7 or H5 HPAI, respectively, that correlated well with hemagglutination inhibition (HI) antibodies. Conversely, sera from chicken and ducks inoculated with inactivated AI vaccines were positive for HI test but negative for the M2e ELISAs. Moreover, ducks inoculated with inactivated vaccine and challenged with a HPAI H5N1 seroconverted for antibodies to the M2e peptide, with significantly different levels from those measured between the vaccinated and infected groups. These results indicate the potential benefit of a simple and specific M2e ELISA in the assessment of the efficacy of vaccination as well as for diagnostic and survey applications.

Many different polymerase chain reaction (PCR) protocols have been used for detection and characterization of avian influenza (AI) virus isolates, mainly in research settings. Blind ring trials were conducted to determine the most sensitive and specific AI PCR protocols from a group of six European Union (EU) laboratories. In part 1 of the ring trial the laboratories used their own methods to test a panel of 10 reconstituted anonymized clinical specimens, and the best methods were selected as recommended protocols for part 2, in which 16 RNA specimens were tested. Both panels contained H5, H7, other AI subtypes, and non-AI avian pathogens. Outcomes included verification of 1) generic AI identification by highly sensitive and specific M-gene real-time PCR, and 2) conventional PCRs that were effective for detection and identification of H5 and H7 viruses. The latter included virus pathotyping by amplicon sequencing. The use of recommended protocols resulted in improved results among all six laboratories in part 2, reflecting increased sensitivity and specificity. This included improved H5/H7 identification and pathotyping observed among all laboratories in part 2. Details of these PCR methods are provided. In summary, this study has contributed to the harmonization of AI PCR protocols in EU laboratories and influenced AI laboratory contingency planning following the first European reports of H5N1 highly pathogenic AI during autumn 2005.

The recent spread of highly pathogenic H5N1 avian influenza (AI) has made it important to develop highly sensitive diagnostic systems for the rapid detection of AI genome and the differentiation of H5N1 variants in a high number of samples. In the present paper, we describe a high-throughput procedure that combines automated extraction, amplification, and detection of AI RNA, by an already described TaqMan real-time reverse transcription–polymerase chain reaction (RRT-PCR) assay targeted at the matrix (M) protein gene of AI virus (AIV). The method was tested in cloacal and tracheal swabs, the most common type of samples used in AI surveillance, as well as in tissue and fecal samples. A robotic system (QIAGEN Biosprint 96) extracted RNA and set up reactions for RRT-PCR in a 96-well format. The recovery of the extracted RNA was as efficient as that of a manual RNA extraction kit, and the sensitivity of the detection system was as high as with previously described nonautomated methods. A system with a basic configuration (one extraction robot plus two real-time 96-well thermocyclers) operated by two persons could account for about 360 samples in 5 hr. Further characterization of AI RNA–positive samples with a TaqMan RRT-PCR specific for H5 (also described here) and/or N1 was possible within 2 hr more. As this work shows, the system can analyze up to 1400 samples per working day by using two nucleic acid extraction robots and a 384-well-format thermocycler.

Avian influenza (AI) viruses are a diverse group of viruses that can be divided into 144 subtypes, based on different combinations of the 16 hemagglutinin and nine neuraminidase subtypes, and two pathotypes (low and high pathogenicity [HP]), based on lethality for the major poultry species, the chicken. However, other criteria are important in understanding the complex biology of AI viruses, including host adaptation, transmissibility, infectivity, tissue tropism, and lesion, and disease production. Overall, such pathobiological features vary with host species and virus strain. Experimentally, HPAI viruses typically produce a similar severe, systemic disease with high mortality in chickens and other gallinaceous birds. However, these same viruses usually produce no clinical signs of infection or only mild disease in domestic ducks and wild birds. Over the past decade, the emergent HPAI viruses have shifted to increased virulence for chickens as evident by shorter mean death times and a greater propensity for massive disseminated replication in vascular endothelial cells. Importantly, the Asian H5N1 HPAI viruses have changed from producing inconsistent respiratory infections in 2-wk-old domestic ducks to some strains being highly lethal in ducks with virus in multiple internal organs and brain. However, the high lethality for ducks is inversely related to age, unlike these viruses in gallinaceous poultry, which are highly lethal irrespective of the host age. The most recent Asian H5N1 HPAI viruses have infected some wild birds, producing systemic infections and death. Across all bird species, the ability to produce severe disease and death is associated with high virus replication titers in the host, especially in specific tissues such as brain and heart.

Ducks and other wild aquatic birds are the natural reservoir of type A influenza viruses, which normally are nonpathogenic in these birds. However, the Asian highly pathogenic avian influenza (HPAI) viruses have evolved from producing no disease or mild respiratory infections in ducks to some strains producing severe systemic disease and mortality. To further understand the pathogenicity of these strains in ducks, we studied the gross and histologic lesions and tissue distribution of viral antigen in 2- and 5-wk-old white Pekin ducks infected with different Asian-origin H5N1 AI viruses. Seven of eight 2-wk-old ducks inoculated with A/Egret/HK/757.2/02 developed acute disease, including severe neurological dysfunction and death. However, this virus killed only two of eight 5-wk-old ducks. Two additional viruses, A/Vietnam/1203/04 and A/Crow/Thailand/04, also produced high mortality in 2-wk-old ducks. Microscopic lesions and AI viral antigen were observed most frequently in the nasal cavity, brain, heart, adrenal glands, and pancreas. Another virus, A/Thailand PB/6231/04, killed three of eight 2-wk-old ducks but did not induce neurological signs. Furthermore, older ducks infected with this virus did not present clinical signs or gross lesions, and their tissues showed very few microscopic lesions. All the viruses studied established systemic infections in both younger and older ducks, with viral replication in tissues correlating with the severity of the clinical signs. The differences in mortality induced by HPAI H5N1 viruses in ducks are reflected in the pathological findings and antigen distribution in tissues. However, the observed differences in pathology between ducks infected at different ages is unclear and may be associated with a variety of factors including the virus strain, host immune response, host cell maturation, and capacity to support viral replication.

An avian influenza (AI) isolate can be classified as a high pathogenicity avian influenza (HPAI) virus based upon the results of the standard intravenous pathogenicity index test; molecular classification, which is derived by sequencing the hemagglutinin gene across the site coding for the cleavage site; or a combination. However, discordant results between the molecular classification and virulence for experimentally infected chickens have been observed with several H5 and H7 subtype AI viruses. Because the declaration of HPAI virus results in severe effects on trade for the entire country, the gap between the genetic and phenotypic markers is an important issue, and it requires us to reexamine what should be considered an HPAI virus by the Office International des Épizooties standards. To better understand and assess the true virulence of the virus, potential pathogenicity of H5 and H7 subtype AI virus isolates has been assessed by examining the plaquing efficiency of the virus in chicken embryo fibroblast cells, conducting 14-day-old embryo passage and selection system, and applying in vitro mutagenesis coupled with reverse genetics. The potential value of these complimentary methods in assessing potential pathogenicity of the AI virus is discussed.

To assess the potential of quail as an intermediate host of avian influenza, we tested the influenza A/Mallard/Potsdam/178-4/83 (H2N2) virus to determine whether through adaptation a mallard strain can replicate and transmit in quail, as well as other terrestrial birds. After five serial passages of lung homogenate a virus arose that replicated and transmitted directly to contact cage mates. To test whether adaptation in quail led to interspecies transmission, white leghorn chickens were infected with the wild-type (mall/178) and quail-adapted (qa-mall/178) viruses. The results show that mall/178 H2N2 does not establish an infection in chickens nor does it transmit, while qa-mall/178 H2N2 infects and transmits to contact chickens causing clinical signs like depression and diarrhea. Completed sequences indicate six amino acid changes spanning four genes, PB2, PB1, HA, and NP, suggesting that the internal genes play a role in host adaptation. Further adaptation of qa-mall/178 in white leghorn chickens created a virus that replicated more efficiently in the upper and lower respiratory tract. Sequence analysis of the chicken-adapted virus points to a deletion in the neuraminidase stalk region.

The H5N1 virus currently circulating is continuing to evolve, and it has already resulted in the extension of its host and geographical range. It is likely that H5N1 will become a global problem for the poultry industry. How many of the recent H5N1 changes observed have been induced by changing patterns in poultry raising? A change in attitude on the use of high-quality vaccines is a change that would drastically help in the control of the current epidemic in the poultry industry. This article provides an overview of the changing properties that have been observed during the current H5N1 outbreaks.

Apart from an outbreak in commercial poultry in Chile in 2002, there have been few reports of avian influenza in South America. However, surveillance in free-flying birds has been limited. An avian influenza virus was isolated from a Cinnamon Teal (Anas cyanoptera) in Bolivia in 2001 from samples collected for an avian influenza virus and avian paramyxovirus surveillance study. This isolate was determined to be an H7N3 virus by gene sequencing. Analysis of all eight genes revealed that five genes were most closely related to the H7N3 in Chile in 2002. Two genes were most closely related to North American wild aquatic bird virus lineages and one gene was most closely related to an equine influenza virus from South America.

Anseriformes are the reservoir of low-pathogenicity avian influenza viruses (LPAIV). Studies have shown a high LPAIV prevalence associated with low antibody detection in a wild duck population in northern European countries, whereas in winter areas (Mediterranean basin), low viral detection and high seroprevalence were observed.

In order to gain insight into the role played by both population recruitment and migration on AIV persistence, an epidemiological model was developed. A susceptible, infectious and removed (immune or dead)–individuals model coupling population and infection dynamics was developed to simulate LPAIV circulation in dabbling ducks throughout the entire year. The transmission coefficient (β) was calculated using the original dataset of published works, whereas dabbling duck demographic parameters were obtained from the literature. The estimated host density threshold for virus persistence is 380 susceptible individuals per day whereas the critical community size needed for maintaining the virus throughout the winter has been estimated to be about 1200 individuals. The model showed peaks of viral prevalence after nesting and during the moult period because of population recruitment and high host density, respectively. During the winter and spring periods, the viruses reach the minimal endemic level and local extinction is highly probable because of stochastic phenomena, respectively 80% and 90% of probabilities. The most sensitive parameters of the model are the recruitment rate of young susceptible animals and the duration of virus shedding.

Low-pathogenicity (LPAI) and high-pathogenicity (HPAI) avian influenza viruses are periodically isolated from South African ostriches, but during 2002 the first recorded outbreak of LPAI (H6N2) in South African chickens occurred on commercial farms in the Camperdown area of KwaZulu/Natal (KZN) Province. Sequence analysis of all eight genes were performed and phylogenetic analysis was done based on the hemagglutinin and neuraminidasc sequences. Results from phylogenetic analyses indicated that the H6N2 chicken viruses most likely arose from a reassortment between two South African LPAI ostrich isolates: an H9N2 virus isolated in 1995 and an H6N8 virus isolated in 1998. Two cocirculating sublineages of H6N2 viruses were detected, both sharing a recent common ancestor. One of these sublineages was restricted to the KZN province. The neuraminidase gene contained a 22–amino acid deletion in the NA-stalk region, which is associated with adaptation to growth in chickens, whereas the other group, although lacking the NA-stalk deletion, spread to commercial farms in other provinces. The persistence of particular H6N2 types in some regions for at least 2 yr supports reports from Asia and southern California suggesting that H6N2 viruses can form stable lineages in chickens. It is probable that the ostrich H6N8 and H9N2 progenitors of the chicken H6N2 viruses were introduced to ostriches by wild birds. Ostriches, in which AI infections are often subclinical, may serve as mixing vessels for LPAI strains that occasionally spill over into other poultry.

Although fecal–oral transmission of avian influenza viruses (AIV) via contaminated water represents a recognized mechanism for transmission within wild waterfowl populations, little is known about viral persistence in this medium. In order to provide initial data on persistence of H5 and H7 AIVs in water, we evaluated eight wild-type low-pathogenicity H5 and H7 AIVs isolated from species representing the two major influenza reservoirs (Anseriformes and Charadriiformes). In addition, the persistence of two highly pathogenic avian influenza (HPAI) H5N1 viruses from Asia was examined to provide some insight into the potential for these viruses to be transmitted and maintained in the environments of wild bird populations. Viruses were tested at two temperatures (17 C and 28 C) and three salinity levels (0, 15, and 30 parts per thousand sea salt). The wild-type H5 and H7 AIV persistence data to date indicate the following: 1) that H5 and H7 AIVs can persist for extended periods of time in water, with a duration of infectivity comparable to AIVs of other subtypes; 2) that the persistence of H5 and H7 AIVs is inversely proportional to temperature and salinity of water; and 3) that a significant interaction exists between the effects of temperature and salinity on the persistence of AIV, with the effect of salinity more prominent at lower temperatures. Results from the two HPAI H5N1 viruses from Asia indicate that these viruses did not persist as long as the wild-type AIVs.

Since 2000, hundreds of H9N2 viruses have been isolated from all types of domestic birds. Although H9N2 is a low-pathogenicity virus, disease has been observed in all types of poultry in the field. Clinical signs ranged from very mild disease to high morbidity and mortality when the virus was associated with a secondary pathogen. Because of the wide range of the virus and the great losses it caused, initially a local vaccination program was implemented, but mass vaccination was quickly authorized. A local strain, isolated in 2002 was selected and is currently in use as an inactivated vaccine. An intensive operation is in progress to characterize the isolates.

Several genes (hemagglutinin [HA], neuraminidase, nonstructural protein, nucleoprotein, and matrix) were sequenced, revealing three main groups: the first group included two isolates from 2000, the second group included isolates from 2001 to the beginning of 2003, and the third group included all isolates from 2003 to date. The differences between the second and third groups, in a part of the HA gene, ranged from 3.49% to 6.97% (average 4.57%) of the nucleotides. Similar differences were recorded in the other tested genes. These data could indicate the probable introduction of distinct progenitor viruses into the Israeli poultry population.

Furthermore, sequencing of the HA protein of some Israeli isolates revealed the presence of L²¹⁶ in the binding site; this finding was typical of the H9N2 viruses isolated from humans, which raises the possibility of an influence on host specificity and virulence.

Emergency vaccination for avian influenza (AI) infections caused by viruses of the H5 or H7 subtypes has been used in several instances over the past years. It has been applied primarily in the chicken and turkey industry with the general objective of controlling, and in some instances eradicating, infection. The use of vaccination as a tool to eradicate AI requires the enforcement of a coordinated set of control and monitoring measures. In fact, only certain attempts at eradicating AI with the support of vaccination have been successful, and the outcome of the vaccination campaign has been shown to depend greatly on effective application of the field strategy that complemented the vaccination program. While it is taken for granted that the product and companion diagnostic test are suitable for that given situation, a monitoring system must be in place to promptly identify whether vaccinated birds have been field exposed, and the latter should be dealt with in an appropriate manner, avoiding the spread of infection to other premises. Prophylactic vaccination could also become a tool for AI management in the European Union, provided that its application is based on a systematic assessment of AI risk. The correct use of this tool can be a valuable support for the control of AI in poultry, with the added value of limiting the economic losses to the industry and to the taxpayer. Eventually, this will also reduce human exposure to potentially dangerous viruses.

Daily within-flock mortality data, from a few days before until a few days after onset of increased mortality, from H7N7-infected flocks were analyzed with nonlinear regression for layer (organic and free-range or caged), broiler, and turkey flocks. The following notification thresholds were recommended for the Netherlands: 1) organic layer flocks, broiler flocks, and turkey flocks ≤11 wk of age: ≥0.5% mortality/day for two consecutive days; 2) layer flocks with birds housed in cages: ≥0.25% mortality/day for two consecutive days; 3) turkey flocks ≥16 wk of age: ≥1% mortality/day for two consecutive days. Notification of increased mortality to the veterinary authorities should take place on the second day of increased mortality. Interpretation of mortality thresholds should be on the level of the poultry barn in which clinical problems arise. Because of nonoptimal specificity of proposed thresholds (mortality possibly caused by other diseases), use of PCR-diagnostics (results within 24 hr) without costs to the individual farmer should be promoted to exclude avian influenza in suspect clinical situations in order to minimize negative economic consequence for farmers and stimulate notification by farmers and veterinary practitioners.

In February 2004 a highly pathogenic avian influenza outbreak erupted in the Fraser Valley of British Columbia, Canada. The index farm was a chicken broiler breeder operation comprising two flocks, 24 and 52 wk of age. Birds in the older flock presented with a mild drop in egg production and a small increase in mortality. Pathological specimens taken from the older flock were submitted to the provincial veterinary diagnostic laboratory from which an influenza A virus was isolated. While still under investigation by the provincial veterinary authorities, a spike in mortality was observed in birds belonging to the younger flock. Diagnostic material from both flocks was forwarded to the Canadian Food Inspection Agency's National Centre for Foreign Animal Disease. A low-pathogenicity H7N3 virus was detected in the older flock and a novel highly pathogenic H7N3 virus was found in specimens collected from the younger flock. Despite destruction and disposal of birds on the index farm, the virus spread to adjacent farms. Given the high density of poultry operations in the Fraser Valley and the high level of integration amongst industry support services, a total of approximately 17 million chickens, turkeys, ducks, geese, and speciality birds were put at immediate risk. Despite movement controls the virus spread and established itself in three distinct clusters. To prevent further spread, healthy, marketable birds outside of the surveillance areas were pre-emptively slaughtered. Although highly pathogenic avian influenza is a federal responsibility, the successful control and eradication of this outbreak would not have been possible without the cooperative involvement of federal and provincial diagnostic laboratories. The success of this collaboration was partly responsible for the formation of a national avian influenza laboratory network.

In spring 2004, an outbreak of highly pathogenic avian influenza (HPAI), subtype H7N3, occurred in the Fraser Valley of British Columbia, Canada. The active outbreak lasted more than 90 days; 42 commercial poultry farms were identified as infected premises, and more than 17 million birds were culled. Through the depopulation of HPAI-positive farms and the strategic depopulation of adjacent test-negative farms, a total of 410 commercial poultry farms were emptied.

The goals for the commercial poultry industry were to expedite restocking, reduce nonproductive downtime, negotiate equitable financial compensation, review and restructure emergency disease response plans, and identify and implement mitigation strategies.

After the outbreak, multijurisdictional reviews identified the strengths and weaknesses of the outbreak control strategy. Lessons learned were incorporated into current emergency disease response protocols for both industry and government. The industry-led challenge to initial compensation values, especially for specialty poultry and breeder birds, resulted in a review of the federal Health of Animals Act. The British Columbia poultry industry, in collaboration with the British Columbia Ministry of Agriculture and Lands, developed an Enhanced Biosecurity Initiative that included the identification of mandatory on-farm biosecurity standards for commercial producers, an educational biosecurity self-assessment guide, and provisions for a producer self-quarantine to be enacted upon the first suspicion of disease.

Avian influenza (AI) is an Office International des Epizooties listed disease that has become a disease of great importance both for animal and human health. The increased relevance of AI in animal and human health has highlighted the lack of scientific information on several aspects of the disease, which has hampered the adequate management of some of the recent crises. Millions of animals have died, and there is growing concern over the loss of human lives and over the management of the pandemic potential.

The present article reviews the currently available control methods for notifiable AI infections in poultry. The application of control policies, ranging from stamping out to emergency and prophylactic vaccination, is discussed on the basis of data generated in recent outbreaks and in light of new regulations, also in view of the maintenance of animal welfare.

Poultry veterinarians working for the industry or for the public sector represent the first line of defense against the pandemic threat and for the prevention and control of this infection in poultry and in wild birds.

The highly pathogenic H5N1 avian influenza virus is widespread among domestic ducks throughout Southeast Asia. Many aspects of the poultry industry and social habits hinder the containment and eradication of AI. Vaccination is often put forward as a tool for the control of AI. However, vaccination will only lead to eradication when it reduces the virus spread to such an extent that herd immunity is obtained. To study the effect of a single vaccination dose on the transmission of H5N1 in domestic ducks we performed experiments in which infected and uninfected ducks were housed together and the infection chain was monitored by means of virus isolation and serology. Specifically, Peking ducks were vaccinated with A/Chicken/Mexico/232/94/CPA H5N2 and challenged with A/Chicken/GxLA/1204/04 H5N1 one week after vaccination. In both the control and vaccinated groups all inoculated and contact animals were quickly infected. However, the disease signs and mortality differed between the control and treatment groups. This finding may have important implications for the control of H5N1 by means of vaccination.

The efficacy of an inactivated vaccine containing the Eurasian isolate A/chicken/Italy/22A/98 H5N9 (H5N9-It) was compared with that of the fowlpox-vectored TROVAC™-AIV H5 (rFP-AIV-H5) vaccine against an H5N1 highly pathogenic avian influenza challenge. Five-week-old Muscovy ducks were vaccinated with either H5N9-It (0.5 ml) or rFP-AIV-H5 (5 log₁₀ 50% tissue culture infectious dose (TCID₅₀)/dose), followed by a boost at 7 wk of age with the same vaccine (1.0 ml of H5N9-It or 5 log₁₀ TCID₅₀/dose rFP-AIV-H5), and a challenge at 9 wk of age with 10⁷ egg infectious dose (lethality 50%) of A/crested eagle/Belgium/01/2004 (H5N1). All unvaccinated challenged birds showed severe nervous signs (loss of balance, torticollis) starting 7 days postinfection (dpi). None of the vaccinated ducks showed these nervous signs. Shedding was measured in oropharyngeal and cloacal swabs, sampled from 3 to 19 dpi by titration in chicken embryo fibroblasts and by real-time reverse transcription–polymerase chain reaction. Virus shedding was significantly higher in oropharyngeal compared to cloacal swabs. Both vaccines reduced the percentage of positive swabs and the viral load in the swabs, but the reduction was higher with the H5N9-It vaccine. The inactivated vaccine induced hemagglutination inhibition (HI) titers (5.4 log₂) that were boosted after the second administration (7.5 log₂). rFP-AIV-H5–induced HI titers were lower (3 log₂ only after the second administration), most probably because the fowlpox vector does not replicate in ducks. Altogether, these results indicate that significant protection from clinical signs and reduction in virus shedding may be achieved in ducks with conventional inactivated or fowlpox-vectored vaccine as compared with nonvaccinated challenged control birds.

The objective of this study was to compare the efficacy of two avian influenza (AI) H5-inactivated vaccines containing either an American (A/turkey/Wisconsin/68 H5N9; H5N9-WI) or a Eurasian isolate (A/chicken/Italy/22A/98 H5N9; H5N9-It). Three-week-old specific pathogen-free chickens were vaccinated once and challenged 3 wk later with a H5N1 highly pathogenic AI (HPAI) virus isolated from a chicken in Thailand in 2004. All unvaccinated challenged birds died within 2 days, whereas 90% and 100% of chickens vaccinated with H5N9-WI and H5N9-It, respectively, were protected against morbidity and mortality. Both vaccines prevented cloacal shedding and significantly reduced oral shedding of the challenge HPAI virus. Additional chickens (vaccinated or unvaccinated) were placed in contact with the directly challenged birds 18 hr after challenge. All unvaccinated chickens in contact with unvaccinated challenged birds died within 3 days after contact, whereas unvaccinated chickens in contact with vaccinated challenged birds either showed a significantly delayed mortality or did not become infected. All vaccinated contacts were protected against clinical signs, and most chickens did not shed detectable amount of HPAI virus. Altogether, these data indicate that both vaccines protected very well against morbidity and mortality and reduced or prevented shedding induced by direct or contact exposure to Asian H5N1 HPAI virus.

In 2002, the World Organisation for Animal Health began a review of the chapter on avian influenza by convening a group of experts to revise the most recent scientific literature. The group drafted the initial text that would provide the necessary recommendations on avian influenza control and prevention measures. The main objectives of this draft were to provide clear notification criteria, as well as commodity-specific, risk-based mitigating measures, that would provide safety when trading and encourage transparent reporting.

The Department for Environment, Food and Rural Affairs (Defra) has monitored epidemiologic developments following outbreaks of H5N1 in Asia since the beginning of 2004 and publishes risk assessments as the situation evolves. The U.K. applies safeguard measures that reflect EU rules to enable imports to continue when they present negligible risk. Defra risk assessments (RA) identify possible pathways by which the H5N1 virus may be introduced to the U.K. These assessments provide a basis for identifying appropriate surveillance activities to ensure early detection, should the virus be introduced, and disease control measures to be taken, should the virus be detected in the U.K. Nevertheless, these assessments have highlighted that many fundamental uncertainties still remain. These uncertainties center on the geographic and species distribution of infection outside Asia and the means of dissemination of the virus. However, the evolving developments demonstrated that regulatory decisions had to be made despite these uncertainties. Improvements in our current RA abilities would greatly benefit from systematic studies to provide more information on the species susceptibility, dynamics of infection, pathogenesis, and ecology of the virus along with possible pathways by which the H5N1 virus may be disseminated. Such an approach would assist in reducing uncertainties and ensuring that regulatory risk management measures are regularly reviewed by taking into account the most recent scientific evidence. The likelihood of the persistence of H5N1 outside Asia in the coming years and the effects of control programs in Asia and other affected regions to reduce the prevalence of infection are also important factors.

Avian influenza (AI) is a disease of concern for the poultry industry. In its highly pathogenic form, AI viruses (AIVs) can cause a high morbidity and case fatality rate as well as severe economic consequences. Low pathogenic AIVs (LPAIVs), in contrast, only cause localized infections in the respiratory and gastrointestinal tracts of affected birds. Although there is apparently sufficient scientific evidence documenting the absence of LPAIV in poultry meat, several countries still place restrictions for international trade of poultry meat on LPAIV-infected countries. These restrictions are extremely trade disruptive and entail significant losses to the poultry industry. This article presents a quantitative approach to assess the probability of LPAIV presence in chicken meat and provides a model that can be tailored to reflect the epidemiology of LPAIV and surveillance systems in different countries. Results show that the probability of introducing LPAIV through chicken meat imports is insignificant.

Highly pathogenic avian influenza has not been reported in Nepal to date. Surveillance for the presence of avian influenza viruses was conducted in 16 districts of Nepal from February 2004 to December 2005. Four hundred forty-six serum samples were collected from ducks, chickens, and pigeons and tested for antibodies to all influenza A viruses by competitive enzyme-linked immunosorbent assay (C-ELISA). Any sera positive by C-ELISA were tested for antibodies to H5, H7, and H9 influenza viruses by hemagglutination inhibition tests. One hundred and thirty-five cloacal swabs from healthy ducks and chickens were tested by commercial avian influenza antigen detection kits. A further 13 tissue samples from diseased birds were tested for the presence of virus by virus isolation in eggs, cell culture, and immunohistochemistry. No influenza viruses were detected in any of the tissues or swabs. All serum samples collected before October 2005 were negative for antibodies. The first sera positive for antibodies were collected on October 13, 2005, which were determined to be of the H9N2 subtype. This is the first report of serologic evidence of an avian influenza virus infection in Nepal.

In October 2005, the second Swiss national avian influenza monitoring in wild waterfowl and commercial poultry with free range management started. Cloacal swabs were examined by real-time reverse transcription-polymerase chain reaction for both M gene of influenza A virus and H5 subtype. The monitoring (more than 2000 samples tested) documented the introduction of H5N1 in Swiss wild waterfowl in mid-February 2006. Until the end of March, 29 water bird carcasses were found H5 positive. In the same period, domestic poultry flocks with a permit of free-range management were kept under surveillance, with negative results.

Avian influenza (AI) diagnostic laboratories in Laos and Cambodia have been established recently to conduct AI surveillance and diagnostic tests to coordinate regional efforts for the control of highly pathogenic avian influenza (HPAI) in South East Asia. Two laboratories have been provisioned with equipment, supplies, and reagents for routine diagnostic testing. Laboratory staff has received training to conduct serologic and virologic tests for isolation and identification of AI virus. Development of a disease reporting system and an AI surveillance program is in progress in Laos and Cambodia. There are plans to further upgrade laboratory facilities and to provide more comprehensive and advanced molecular diagnostic tests for control of HPAI in Laos and Cambodia. These two countries are on the frontline in the battle to fight HPAI H5N1 virus and to prevent it from spreading to other regions and mutating to a major human pathogen.

Avian influenza (AI) was diagnosed in May 2002 for the first time in Chile and South America. The epidemic was caused by the highly pathogenic AI (HPAI) virus subtype H7N3 that emerged from a low pathogenic virus. The index farm was a broiler breeder, located in San Antonio, V Region, which at the time was a densely populated poultry area. Stamping of 465,000 breeders, in 27 sheds, was immediately conducted. Surveillance activities detected a second outbreak, 1 wk later, at a turkey breeding farm from the same company. The second farm was located 4 km from the index case. Only 25% of the sheds were infected, and 18,500 turkeys were destroyed. In both outbreaks, surveillance zones and across-country control measures were established: prediagnosis quarantine, depopulation, intensive surveillance, movement control, and increased biosecurity. Other measures included cleaning, disinfection, and controlling the farms with sentinels to detect the potential presence of the virus. Zoning procedures were implemented to allow the international trade of poultry products from unaffected areas. Positive serologic results to H5N2 virus also were detected in other poultry farms, but there was no evidence of clinical signs or virus isolation. Epidemiological investigation and laboratory confirmation determined that positive serology was related to a contaminated imported batch of vaccine against inclusion body hepatitis. All actions taken allowed the control of the epidemic, and within 7 mo, Chile was free of AI. Epidemic and control measures that prevented further spread are described in this article, which illustrates the importance of a combination of control measures during and after an outbreak of AI. This study is a good example of how veterinary services need to respond if their country is affected by HPAI.

The production and supply of reference reagents for the diagnosis of avian influenza (AI) is one of the duties of the World Organization for Animal Health reference laboratories. The lyophilized reagents are routinely shipped on dry ice to both national and international clients. In order to determine the effect of different short-term storage temperatures on the activity of AI reference reagents, vials containing lyophilized avian influenza A antigens and antisera preparations were maintained at 4 C, 22 C, and 37 C over a 21-day period. At days 0, 3, 7, 14, and 21 the reagents were titrated using the hemagglutination test, the hemagglutination inhibition test, or the agar gel immunodiffusion test (AGID). All of the AI antigens that were kept at 4 C and 22 C retained hemagglutinating activity for at least 21 days, but when they were stored at 37 C several lost their hemagglutinating activity. All of the reference antisera tested were still able to inhibit hemagglutination after 21 days, and the antigen used in AGID also gave clear results after 21 days even after incubation at 37 C. Our results therefore indicate that lyophilized avian influenza antigens maintain their hemagglutinating activity at temperatures between 4 C and 22 C for at least 21 days, and both antisera and antigens prepared for AGID remain stable for 21 days between 4 C and 37 C. This information will allow for alternative shipping temperatures than those presently recommended, in addition to the short-term storage of these reagents at nonrefrigerated temperatures.

Highly pathogenic avian influenza (AI) H5N1 viruses have been spreading from Asia since late 2003. Early detection and classification are paramount for control of the disease because these viruses are lethal to birds and have caused fatalities in humans. Here, we described TaqMan reverse transcriptase-polymerase chain reaction assays for rapid detection of all AI viruses (influenza type A) and for identification of H5N1 of the Eurasian lineage. The assays were sensitive and quantitative over a 10⁵–10⁶ linear range, detected all of the tested AI viruses, and enabled differentiation between H5 and H7 subtypes. These tests allow definitive confirmation of an AI virus as H5 within hours, which is crucial for rapid implementation of control measures in the event of an outbreak.

Real time reverse transcriptase (RRT)–polymerase chain reaction (PCR) for the detection of Eurasian H5 avian influenza virus (AIV) isolates was adapted from an existing protocol, optimized, and validated using a number of genetically diverse H5 isolates (n = 51). These included 34 “Asian lineage” H5N1 highly pathogenic avian influenza (HPAI) viruses (2004–2006), plus 12 other H5 isolates from poultry outbreaks and wild birds in the Eastern Hemisphere (1996–2005). All 51 were positive by H5 Eurasian RRT-PCR. Specificity was assessed by testing representative isolates from all other AI virus subtypes (n = 52), non-AI avian pathogens (n = 8), plus a negative population of clinical specimens derived from AI-uninfected wild birds and poultry (n = 604); all were negative by H5 Eurasian RRT-PCR. RNA was directly extracted from suspect HPAI H5N1 clinical specimens (Africa, Asia, and Europe; 2005–2006; n = 58) from dead poultry and wild birds, and 55 recorded as positive by H5 Eurasian RRT-PCR: Fifty-one of these 55 were in agreement with positive AIV isolation in embryonated chickens' eggs. H5 Eurasian RRT-PCR was invaluable in H5 outbreak diagnosis and management by virtue of its rapidity and high degree of sensitivity and specificity. This method provides a platform for automation that can be applied for large-scale intensive investigations, including surveillance.

This work describes the development of a real-time RT-PCR (RRT-PCR) procedure for detection of the N1 gene from avian influenza virus (AIV), based on the use of specific primers and a TaqMan-MGB (minor groove binder) probe. Nucleotide sequences of the neuraminidase type 1 gene from a collection of H5N1 Eurasian strains of AIV were aligned using ClustalW software. Conserved regions were located and used to design specific primers and a TaqMan-MGB probe using Primer Express software. A one-step RRT-PCR method was optimized using RNA from the Turkey 2005 H5N1 strain of AIV and can be completed in about 2 hr once the RNA is extracted from the sample. The specificity of the assay was assessed with non-N1 AIV strains, another related avian virus, and different avian tissue samples from healthy animals. Sensitivity was determined using 10-fold serial dilutions of the H5N1 Turkey 2005 strain and was compared with the generic RRT-PCR detection method, targeted at the matrix protein gene of AIV, commonly used at the Spanish AIV National Reference Laboratory. The N1 detection method proved to be even more sensitive than the generic (matrix-based) method, allowing a very quick confirmation (or discarding) of any Eurasian N1 strain when a positive result was obtained with the matrix RRT-PCR assay. Combined with RRT-PCR tests for general detection of AIV and H5 typing in use at the NRL, the procedure here described allows characterizing of any H5N1 Eurasian AIV strain in a field sample within a working day.

Nota de InvestigacónBulgaria has a surveillance program on avian influenza to keep a close watch on health of poultry, exotic birds, and wild birds. The samples include sera for serologic examination, and carcasses, tissue samples, tracheal and cloacal swabs, and feces for virologic examination. The number of samples depends on the epizootic situation in the country and in neighboring countries. Mmigration and resting and living areas of wild birds also are under consideration. The territory of Bulgaria is divided into 28 regions. For each of these regions, there are exact types and numbers of samples, depending on latitude, water reservoir, farms, and backyards. Since October 2005, surveillance has become harder because of the situation in Romania and Turkey. Since 2006, the number of samples has vastly increased.

Avian influenza is a serious threat to both animal and human health. To enable cutting-edge research in this field, we developed a molecular test on the basis of real-time polymerase chain reaction (real-time PCR), which detects influenza virus RNA. The test enables highly sensitive detection of influenza virus A and B strains, including H5N1, and specific identification of influenza A virus H5 subtypes. The kit was tested against a broad panel of influenza A and B subtypes and other respiratory viruses in collaboration with worldwide authoritative laboratories and shows a very high specificity and sensitivity. An internal control verifies RNA extraction as well as real-time PCR success. With this kit, rapid and reliable detection of influenza A and B viruses and identification of H5 subtypes can be achieved.

Clinical signs, serologic response, viral contents of the trachea and intestine, and histopathological and ultrastructural changes of the tracheal epithelium of Japanese quail experimentally infected with field isolate of H9N2 avian influenza were studied. Vaccinated and unvaccinated quail were inoculated with 10^6.3 50% embryo infectious dose/bird of A/chicken/Iran/SH-110/99 (H9N2) virus via nasal inoculation. Clinical signs such as depression, ruffled feathers, diarrhea, and nasal and eye discharges were observed 6 days postinfection (PI). No mortality was observed; however, there was reduction in feed and water consumption and egg production. However, the serologic response of vaccinated challenged and unvaccinated challenged birds was not significantly different. Unvaccinated challenged quail showed more severe histopathologic reaction in their lungs and trachea. Hyperemia, edema, infiltration of inflammatory cells, and deciliation and sloughing of the tracheal epithelium were observed. Ultrastructural study showed dilatation of endoplasmic reticulum and degeneration of Golgi apparatus and cilia of the tracheal lining cells of respiratory epithelium.

We previously described the use of an established reverse genetics system for the generation of recombinant human influenza A viruses from cloned cDNAs. Here, we have assembled a set of plasmids to allow recovery of the avian H5N1 influenza virus A/Turkey/England/50-92/91 entirely from cDNA. This system enables us to introduce mutations or truncations into the cDNAs to create mutant viruses altered specifically in a chosen gene. These mutant viruses can then be used in future pathogenesis studies in chickens and in studies to understand the host range restrictions of avian influenza viruses in humans.

Two highly pathogenic avian influenza (HPAI) virus clones that met the criteria for high-pathogenicity avian influenza viruses, by possessing a multibasic hemagglutinin (HA) cleavage site, were isolated from an H5N1 outbreak in Norfolk, England, in 1991–92. These two isolates, A/turkey/England/50-92/91 (50-92) and A/turkey/England/87-92/91 (87-92), displayed differences in virulence as determined by intravenous pathogenicity index-3 and -0, respectively. DNA sequencing of these two isolates identified 10 amino acid differences throughout the genome: three in HA and polymerase B2 (PB2) and two in polymerase B1 (PB1) and single mutations in nucleoprotein (NP) and polymerase A (PA). Serial intracerebral passages were performed in 1- or 2-day-old specific pathogen free (SPF) chicks with 87-92. Viruses reisolated from each bird passage displayed increases in intracerebral pathogenicity index values (from 0 to 1.9) and therefore virulence. Reverse transcriptase polymerase chain reaction and DNA sequencing on viruses isolated at each passage displayed nine out of the 10 mutations associated with the higher pathogenic genotype of 50-92, except for the mutation found in NP, which retained the amino acid residue associated with 87-92. Serial passage through 9-day-old SPF embryonated chicken eggs and serial intravenous passage in 6-wk-old birds could not reproduce these results. These results further highlight that nucleotide changes in the genome other than at the HA cleavage site can attenuate the virulence of HPAI viruses.

On October 18, 2004, two crested hawk eagles, Spizaetus nipalensis, smuggled into Europe from Thailand were seized at Brussels International Airport. A highly pathogenic avian influenza virus, denominated A/crested eagle/Belgium/01/2004, was isolated from these birds and antigenically characterized as H5N1. Here we report on the molecular characterization of A/crested eagle/Belgium/01/2004 (H5N1). We completely sequenced all eight genome segments. The hemagglutinin (HA) and neuraminidase (NA) sequences clustered within the Z genotype and were closely related to strains circulating in Thailand during 2004, although some mutations in the HA were evident, notably a unique arginine (R) > lysine (K) replacement in the cleaving site. The HA cleavage site contained six basic amino acids, confirming its high pathogenicity (intravenous pathogenicity index = 2.94). The 20–amino acid deletion in the NA stalk region is consistent with its Thai–Viet origin. We further discuss the assembled genetic information in the light of currently known host adaptation, virulence, and antiviral resistance factors. Using infection experiments, we show that pathogenicity in chickens depends on breed, inoculation route (oculonasal vs. intramuscular), and dose. Additionally, in Muscovy ducks, pathogenicity proved to be age dependent.

Prevalence of avian influenza infection in free-range mule ducks (a cross between Muscovy [Cairina moschata domesticus] and Pekin ducks [Anas platyrhychos domesticus]) is a matter of concern and deserves particular attention. Thus, cloacal swabs were collected blindly from 30 targeted mule flocks at 4, 8, and 12 wk of age between October 2004 and January 2005. They were stored until selection. On the basis of a positive H5 antibody detection at 12 wk of age with the use of four H5 antigens, the samples from eight flocks were selectively analyzed.

Positive samples were first screened with a matrix gene–based real-time reverse transcriptase–polymerase chain reaction assay before virus isolation. Eight avian influenza subtypes (H5N1, H5N2, H5N3, H6N2, H6N8, and H11N9) and three avian paramyxovirus type 1 viruses were isolated. All 11 are characterized as low pathogenicity (LP) and avirulent, respectively, by in vivo tests and, when relevant, nucleotide sequencing of the hemagglutinin (or fusion [F]) protein cleavage site. Regarding H5 isolates, all of their eight genes belong to the avian lineage and some particular genetic traits were determined. H5 genes were fully sequenced and phylogenetically analyzed; they all belong to the Eurasian lineage, lack additional glycosylation sites, and do not cluster, suggesting separate introductions from the wild reservoir. None were grouped with the Asian isolates. The N1 gene (H5N1 isolate) was very close genetically to an Italian LP-H7N1 gene. Antigenic relationships between these H5 isolates and others were assessed comparatively by crossed hemagglutination inhibition tests.

All these data are very useful to control the evolution of H5 viruses at the genetic and antigenic level to better understand the source of new outbreaks (new introductions from wild birds or the result of spread among poultry) and to assess the immunity afforded by available vaccines. These data are useful also to update antigens for avian influenza survey and to choose the most suitable vaccine in the case of preventive vaccination of ducks.

Following the avian influenza (AI) epidemics occurring in different areas of the world, a surveillance program funded by the Italian Ministry of Health was implemented. In the framework of this program, an investigation of wild birds was carried out to assess the circulation of AI viruses in their natural reservoir. More than 3000 samples, mainly cloacal swabs, were collected from migratory wild birds belonging to the orders Anseriformes and Charadriiformes. Samples were screened by means of a real-time reverse transcriptase polymerase chain reaction (RRT-PCR), then processed for attempted virus isolation in embryonated fowl's specific pathogen-free eggs. Approximately 5% of the samples were positive for type A influenza viruses by RRT-PCR, and from 14 of those samples AI viruses were isolated and fully characterized. The isolates, belonging to 8 different avian influenza virus subtype combinations (H10N4, H1N1, H4N6, H7N7, H7N4, H5N1, H5N2, and H5N3), were obtained from migratory Anseriformes.

In 2005 the National Animal Health Monitoring System conducted a survey in 183 live poultry markets throughout the United States. The objectives of this study were to describe characteristics of live poultry markets in the United States and to identify potential risk factors for markets to be repeatedly positive for low-pathogenicity avian influenza virus (LPAIV) H5/H7. A questionnaire was administered to market operators that included questions regarding types of birds and other animals in the market, biosecurity, and cleaning and disinfecting practices. A history of testing for avian influenza from March 2004 through March 2005 was obtained for each market. Cases were defined as markets with at least 2 positive LPAI/H5/H7 test results during the year (separate occasions), and controls were defined as markets that were tested at least twice during the year with all negative results. Markets in the North region (New York, New Jersey, Pennsylvania, New England) were larger than markets in the South (Florida, California, Texas) and were more likely to slaughter birds on-site. Testing for avian influenza virus (AIV) was performed more frequently in the North region than in the South region. Markets in the North region tested positive for H5 or H7 at 14.6% of the testing visits, and no markets in the South region tested positive for H5/H7 at any time during the year. Factors associated with repeated presence of LPAIV H5/H7 included number of times the market was cleaned and disinfected, being open 7 days per week, and trash disposal of dead birds.

The effect of proximity on infected premises was evaluated during the highly pathogenic avian influenza (HPAI) epidemic that struck northern Italy in 1999–2000 by quantifying the spatial and temporal clustering of cases. The epidemic was caused by an H7N1 subtype of type A influenza virus that originated from a low-pathogenic AI virus that spread among poultry farms in northeastern Italy in 1999 and eventually became virulent by mutation. More than 90% of 413 infected premises were located in Lombardy and Veneto regions; of 382 outbreaks, 60% occurred in the Lombardy region and 40% in the Veneto region. Global and local spatial statistics were used to estimate the location and degree of clustering of cases with respect to the population at risk. Outbreaks were spatially clustered primarily in Lombardy, with a large cluster in Brescia province and another in Mantua province, on the border of Veneto. Time series analysis was used to assess the temporal clustering of outbreaks. Temporal aggregation increased during the first 5 wk and decreased thereafter (probably as a result of eradication measures enforced in the Veneto region). Spatio-temporal clustering was assessed considering the Temporal Risk Window (TRW), the time period during which premises remain infectious and infection can spread to neighboring premises. The clustering pattern was similar to the one detected when considering spatial clustering (i.e., the larger clusters were identified in the Brescia and Mantua provinces of Lombardy). These results highlight the role of proximity in the spread of AI virus and, when considering the TRW, indicate the possible direction of virus spread.

The prevalence of influenza A virus infection, and the distribution of different subtypes of the virus, were studied in 604 geese and ducks shot during ordinary hunting 2005. The study was based upon molecular screening of cloacal swabs taken by the hunters. The sampling included the following species: greylag (Anser anser), mallard (Anas platyrhynchos), wigeon (Anas penelope), teal (Anas crecca), goosander (Mergus merganser), tufted duck (Aythya fuligula), common scoter (Melanitta nigra), goldeneye (Bucephala clangula), and red-breasted merganser (Mergus serrator).

The samples found to be positive in the initial pan-influenza A virus reverse transcription-polymerase chain reaction (RT-PCR) were further subtyped by using a specific H5 RT-PCR and full-length RT-PCRs for the hemagglutinin (HA) and neuraminidase genes.

None of the greylag samples (0/185) were positive for influenza A virus, whereas 19.1% of the ducks (80/419) were positive. The prevalences of influenza A virus in the different duck species were as follows: mallard, 20.4% (58/284); wigeon, 12.5% (8/64); teal, 30.9% (13/42); goosander, 0% (0/5); tufted duck, 0% (0/4); common scoter, 14.3% (1/7); goldeneye, 0% (0/11); and red-breasted merganser, 0% (0/2). H5N1 subtype was found in one mallard and H5N2 subtype in another mallard and one teal. Sequencing of the HA gene identified all three viruses as low-pathogenic strains, closely related to low-pathogenic H5 influenza A viruses evidenced in recent years in Sweden and the Netherlands. The other subtypes identified included H1N1, H2, H3N2, H3N8, H6N1, H6N2, H6N8, H8N4, H9N2, H11N9, and H12 in mallards; H3N2, H6N2, H6N8, and H9N2 in teals; and H6N2 in wigeons and common scoter.

In the summer of 2005 a Canadian national surveillance program for influenza A viruses in wild aquatic birds was initiated. The program involved collaboration between federal and provincial levels of government and was coordinated by the Canadian Cooperative Wildlife Health Centre. The surveillance plan targeted young-of-the-year Mallards along with other duck species at six sampling locations along the major migratory flyways across Canada. Beginning in early August, cloacal swabs were taken from 704 ducks on two lakes adjacent to one another near Kamloops, British Columbia. The swabs were screened for the presence of influenza A RNA using a real-time reverse transcription–polymerase chain reaction (RRT-PCR) assay that targets the M1 gene. Swab samples that gave positive results underwent further testing using H5- and H7-specific RRT-PCR assays. One hundred and seventy-four cloacal swab specimens gave positive or suspicious results for the presence of an H5 virus. A portion of these (28/35) were confirmed using an H5-specific conventional reverse transcription–polymerase chain reaction assay and an H5 virus was eventually isolated from 24/127 swab specimens. Neuraminidase typing revealed the presence of H5N2 and H5N9 viruses. In mid-November of 2005 an H5N2 virus was detected in a commercial duck operation in the lower mainland of British Columbia, approximately 120 km from where the H5N2-positive wild ducks were sampled. Molecular genetic analysis of the H5N2 viruses isolated from wild and domestic ducks was carried out to determine their kinship.

The role of migrating birds as potential vectors for avian influenza virus (AIV) was investigated. We captured 543 migrating passerines during their stopover on the island of Helgoland (North Sea) in spring and autumn 2001. These birds were sampled for avian influenza A viruses (AIV), specifically the subtypes H5 and H7. For virus detection, samples were taken from 1) short-distance migrants, such as chaffinches (Fringilla coelebs; n = 131) and song thrushes (Turdus philomelos; n = 169); and 2) long-distance migrants, such as garden warblers (Sylvia borin; n = 142) and common redstarts (Phoenicurus phoenicurus; n = 101). Virus isolation assays failed to identify AIV. Therefore, regarding the actual low number of samples, we speculate that the tested four species of passerines were not infected by AIV, indicating that the passerine species examined in this study may play only a minor role as potential vectors of AIV.

The avian influenza virus (AIV) has eight genomic segments (hemagglutinin [HA], neuraminidase [NA], RNA polymerase subunit A [PA], RNA polymerase subunit B1 [PB1], RNA polymerase subunit B2 [PB2], nucleoprotein [NP], nonstructural gene [NS], and matrix protein [M]). The genetic reassortments, recombinations, and mutations lead to a rapid emergence of novel genotypes of the AIVs during their evolution. These emerging viruses provide a large reservoir for pandemic strains. Here we describe a novel computational strategy for genetic reassortment identification. In contrast to the traditional phylogenetic approaches, our method views the genotypes through the modules in networks. Genetic segments with short phylogenetic distance are grouped into modules. Our method is not limited to the number of sequences. We applied this method in reassortment identification of NP segments in H₅N₁ AIVs. We identified two new potential reassortments for H₅N₁ AIVs beyond the reported genotypes in literature.

There is much debate about the relative roles of poultry movements and wild bird movements in the spread of highly pathogenic avian influenza H5N1. This article looks at the problem from an ornithologic perspective. Outbreaks in wild birds are examined in relation to three scenarios of possible wild bird involvement in virus transmission. These scenarios are examined separately for five phases of the outbreak that began in 1997 and which has recently become more dynamic in terms of virus spread. Most outbreaks in wild birds seem to reflect local acquisition of infection from a contaminated source, followed by rapid death nearby. Outbreaks in Europe in early 2006 indicate that the virus can be spread further by wild birds and thus that they can become infected and travel varying distances before dying, and probably passing the infection to other wild birds before death. There is only limited evidence that some wild birds can carry the virus asymptomatically, and no evidence from wild bird outbreaks that they have done so over long distances on seasonal migration routes. Other potential sources of infection and evidence for asymptomatic infection in wild birds are discussed, and the need for more ornithologic input into epidemiological studies of HPAI H5N1 is highlighted.

Because sequence data on H9 avian influenza virus (AIV) from wild birds are currently limited, we set out to determine the sequence of the hemagglutinin (HA) gene of H9 viruses circulating in North American waterfowl and shorebirds. In this study, we examined the HA gene from H9 AIV isolated from mallards (Anas platyrhynchos) sampled during 1998 and 1999 in Minnesota and ruddy turnstones (Arenaria interpres) sampled during 2003 in Delaware and New Jersey. At these sites, the H9N2 subtype represented 12% and 4% of the avian influenza isolates from mallards in 1998 and 1999, respectively, and 8% of the AIVs isolated from shorebirds between 2000 and 2002. Sequences from these viruses were compared with sequences from H9 AIV isolated from commercial poultry and aquatic birds from North America, Europe, Asia, and the Middle East: four previously reported and three new clades were observed. Sequence data from the HA gene of North American waterfowl and shorebird isolates generated in this study most closely group with the Eurasian H9 viruses in the Y439 clade. In addition, the HA cleavage site (AASNR/G) and receptor binding site was identical to the representative virus of that group (DK/Hong Kong/Y439/97). Viruses in that clade are commonly found in ducks and chickens in Hong Kong and Korea. Positive evolutionary selection (dNonsynonymous > dSynonymous) was observed for the HA gene among the North American waterfowl and shorebird H9N2 viruses, indicating that the H9N2-type viruses are changing in their natural hosts.

During the 1990s, several outbreaks of avian influenza (AI), caused by viruses of the H9N2 subtype, were reported in poultry in various parts of the world. Currently, this infection seems to be endemic in poultry in the Middle and Far East, and the extensive circulation of H9N2 in poultry represents a risk factor for infection of humans, because viruses of this subtype have been sporadically introduced into the human population. Little is known about the gene constellation of the H9N2 viruses that are currently circulating in the Middle East; thus, gene sequences of eight IA viruses of the H9N2 subtype isolated in Jordan in 2003 from poultry were analyzed. The results of this investigation show that all eight Jordanian isolates are closely related to each other and to other H9N2 isolates from the Middle East. Seven of eight genes of the Jordanian strains show a percentage of homology not higher than 95% with the genes of two H9N2 strains, A/HK/1073/99 and A/HK/1074/99, isolated from humans in Hong Kong. The M gene is more closely related to the corresponding gene of the two H9N2 human isolates from Hong Kong (97.7–98.2% homology). This homology suggests that the M gene of the Jordanian and human Hong Kong strains could derive from a common ancestor.

H5N1 avian influenza viruses circulating in early 2004 in eastern Asia appeared to be under strong purifying selection, except for the hemagglutinin (HA) and nonstructural 1 (NS1) genes, where few amino acid positions were found under positive selection pressure. To evaluate whether the widespread circulation of the H5N1 viruses in the following years was accompanied by a change in the evolution of the HA and NS1, phylogenetic and positive selection analyses were performed on 89 HA and 57 NS1 sequences. Results showed that the number of HA positively selected sites decreased compared to 2004; no selection pressure for NS1 was found. These findings suggest a possible change in the adaptation of the H5N1 virus to birds.

After a consultancy mission funded by a nongovernmental organization (NGO), information was collected on the dynamics of avian influenza (AI) infection at the rural level in a Vietnamese province with several ongoing outbreaks. AI outbreaks are frequent at village level due to environmental, ecological, agroecological, physical, social, and cultural factors, the underlying factor being poor hygienic conditions. Viral circulation is facilitated by the interactions of the integrated aquaculture, animal raising, horticulture agroecosystem, which relies in the peculiar integration of aquaculture (ponding), animal activities, and horticulture and by the connections with the live-bird market system. The interactions of these factors determine the complex system in which wild birds interact with domestic birds and in which people are constantly exposed to sources of infection, leading to the association between poverty and AI infection in humans. This experience underlines that despite all efforts by the Vietnamese Government, international institutions, and the NGO sector, awareness of AI at the village level needs to be improved. In turn, the leading institutions and international donors funding projects of technical cooperation aimed at tackling AI in Vietnam should invest in a system based on a deep knowledge of the practical problems of village condition to address AI with an effective approach. On the basis of the data collected during the mission, particularly on rural and semi-intensive poultry rearing systems, proposals that encompass the application of an effective vaccination strategy including backyard flocks coupled with dissemination of relevant information on biosecurity measures have been developed for decision makers.

Meat-type turkey farms in Verona province (Veneto, Italy) have been affected by three low-pathogenicity avian influenza (LPAI) epidemics between 2000 and 2004. Control measures implemented ranged from stamping out to controlled marketing in conjunction with restocking bans and movement restriction on live poultry, vehicles, and personnel. These measures were complemented with two emergency vaccination programs (2000–01, 2002–03) started after the beginning of the epidemics, while 2004 outbreaks occurred in vaccinated farms. The three epidemics differed in the number of outbreaks, duration, and economic impact. The aim of the investigation was to estimate the risk of infection and the effect of vaccination on the LPAI epidemics affecting turkey farms in Verona province. Farm probability to avoid infection (Ps) was calculated by Kaplan–Meier for each epidemic. The vaccination effect was evaluated for the 2000–01 and 2002–03 epidemics considering different risk before or after the emergency vaccination. The epidemics and vaccination entered as predictors in a Cox regression model and hazard ratios (HR) were calculated. Ps values at the end of the epidemics were as follows: 2000–01 = 0.66, 2002–03 = 0.51, and 2004 = 0.91. Vaccination reduced dramatically the risk of infection. The measures implemented had different effects on the three epidemics. The lower probability of being infected during the 2004 epidemic was most likely related to the protection level of the vaccinated farms acquired before the beginning of the epidemic, which was also responsible for the reduced spread of infection.

Avian influenza (AI) is an exotic disease in Argentina. A surveillance program for AI was conducted in backyard poultry during 1998–2005 in two regions: 1) region A, which included the avian population in the provinces that border Brazil, Bolivia, and Paraguay, and 2) region B, which included the rest of the provinces of the country. More than 8000 serum samples were tested for antibodies by enzyme-linked immunosorbent assay and/or agar gel immunodiffusion tests, and more than 18,000 tracheal and cloacal swabs were tested for virus by isolation in embryonated specific-pathogen-free eggs. This study was part of the AI prevention program in Argentina, which includes other avian populations such as commercial poultry and all the controls for importation and exportation of live birds. The results from backyard poultry were negative for AI.

Italian poultry production was affected by several outbreaks of low-pathogenicity avian influenza (LPAI) between 2000 and 2005. Intervention measures (IM), such as stamping out of infected and suspected farms, controlled marketing, restocking bans, movement restriction, and emergency vaccination, were put into force in the most affected areas of Lombardia and Veneto regions. These two regions also showed differences in terms of measures applied and timeliness of application. In this study we describe the epidemics and discuss the effectiveness of the IM put into effect. The regional surveillance systems provided the data on the epidemics and the IM description. The IM effectiveness was compared between the different epidemics and the Lombardia and Veneto regions, considering the number of farms involved, the duration of the epidemics, and the extension of the area affected. With regard to the IM applied, reductions in the number of outbreaks (from 388 in 2002–03 to 15 in 2005), the duration of the outbreaks (from more than 1 yr to ∼1 mo), and the spatial extension of the outbreaks (from 89 to 8 municipalities involved) were observed. The emergency vaccination, depopulation, and pre-emptive slaughtering reduced significantly the spread of the epidemic. Comparing the dynamics of the epidemics, more effective results were observed in the Veneto region, where the IM were applied to a greater extent. Emergency vaccination and depopulation were effective in the eradication of the disease during an epidemic, but vaccination and farm density reduction showed the most effective results in controlling the spread of LPAI.

A low pathogenic avian influenza virus of the H5N2 subtype was isolated for the first time from layer chickens in Japan in 2005. Surveillance in trading restriction zones and epidemiologically related farms revealed 41 seropositive farms, and 16 H5N2 viruses were isolated and characterized from nine of these farms. That these viruses were genetically and antigenically similar to each other suggested that these isolates were derived from a common origin. Complete genomic characterization of all eight gene segments showed that these H5N2 isolates in Japan had high homology to the H5N2 strains prevalent in Central America since 1994. The virus was reisolated from tracheal and cloacal swabs of experimentally inoculated chickens and efficiently transmitted to sentinel chickens in adjacent cages.

The 2004 Asian H5N1 epizootic outbreak indicates the urgent need for vaccines against highly pathogenic avian influenza (HPAI) virus. The manufacture of inactivated whole-virus vaccines from HPAI viruses by traditional methods is not feasible for safety reasons as well as technical issues. The low pathogenic avian influenza A/wild bird feces/CSM2/02 (H5N3) virus was used as a heterologous neuraminidase vaccine, and HPAI A/CK/Korea/ES/03 (H5N1) virus was used as a homologous neuraminidase vaccine. Protection efficacy of both vaccines was evaluated by clinical signs, mortality rates, and virus shedding from oropharynx and cloaca of vaccinated chickens after challenge with HPAI A/CK/Korea/ES/03 (H5N1) virus. One dose of 128 hemagglutinin (HA) homologous H5N1 vaccine induced 100% protection in mortality and prevented viral shedding completely after lethal dose virus challenge, whereas one dose of 64 HA unit of heterologous H5N3 vaccine only induced 50% protection in mortality, and it did not prevent viral shedding. However, two doses at a 3-wk interval of 64 HA unit of heterologous H5N3 vaccine as well as one dose of 1024 HA unit of heterologous H5N3 vaccine induced 100% survival rate and could prevent viral shedding completely. Furthermore, we could differentiate the sera of infected birds from those of vaccinated birds by indirect immunofluorescent antibody test. These results suggest that heterologous neuraminidase H5N3 vaccine could be a useful tool for the control of H5N1 HPAI epidemic in poultry.

The recent outbreaks of avian influenza (AI) worldwide have highlighted the difficulties in controlling this disease. Vaccination has become a recommended tool to support the eradication efforts and to limit the economic losses due to AI. A vaccination system based on the use of a vaccine containing a heterologous neuraminidase to the field virus has been shown to be efficacious in reducing the viral shedding and clinical symptoms and in differentiating vaccinated from infected animals (2). To further develop this so called differentiating infected from vaccinated animal vaccination system, two reassortant avian influenza viruses of the H7N5 subtype have been generated. The aim of this study was to generate a prototype strain with a rare N subtype to avoid interference with the anti-N discriminatory test.

A study was conducted to evaluate efficacy of inactivated, reverse genetics-based H5N3 avian influenza vaccines and the predictive ability of a vaccination/serology model for testing vaccine batches. Serologic response, protection against mortality, and protection against viral shedding from trachea and cloaca were assessed for vaccines prepared varying only in antigen content. When the birds were grouped according to serologic response, a clear association with protection could be seen. In general, for birds possessing a nonmeasurable titer (<10), mortality after challenge was a near certainty. Low titers of 10 to 40 resulted in the prevention of mortality but not viral shedding. Titers greater than 40 prevented mortality and reduced viral shedding. The results suggest that a test model including vaccination of chickens at 3 wk, bleeding at 6 wk, and quantitative assessment of serologic response by using a standardized hemagglutination inhibition assay system can be an excellent predictor of vaccine efficacy.

In Asia, domestic ducks have been shown to play a pivotal role in H5 high-pathogenicity avian influenza virus transmission. We have also observed that the same situation may exist for H5 low-pathogenicity avian influenza (LPAI) virus. No data are available regarding the protection afforded by commercial inactivated vaccines against H5 LPAI virus infection in ducks, and two preliminary experiments using commercial inactivated vaccines gave poor results. Virus-like particles (VLPs) have been shown to be immunogenic in different species. With regard to the influenza model, the matrix (M) protein has been shown to be necessary for the formation of VLPs. In order to attempt to develop a VLP influenza vaccine expressing hemagglutinin and neuraminidase (NA) of interest, we generated a triple recombinant baculovirus (rB) expressing three structural proteins: H5, N3, and M, derived from a recent French LPAI virus strain. Although the three proteins were successfully expressed in rB-infected cells and displayed the expected biological activity, no VLPs were observed. Despite this result, the protection afforded to ducks by rB-infected cell lysates was assessed and was compared with the protection afforded by an inactivated commercial H5N9 vaccine. For this purpose, specific-pathogen-free Muscovy ducks (15 per group) received rB-infected cell lysates (3 wk apart), while a second group received the H5N9 vaccine. Ten days after the boost, a homologous virus challenge was implemented. Both vaccines induced positive hemagglutination inhibition titers and M immune response, whereas lysates of rB-infected cells elicited NA immune response. Tracheal and cloacal sheddings were measured using M-based real-time-reverse transcription–polymerase chain reaction and were compared with the sheddings of vaccinated and unvaccinated infected controls. Lysates of rB-infected cells afforded a significant decrease of cloacal shedding and a delayed peak of tracheal shedding, whereas the inactivated commercial vaccine afforded a significant decrease of tracheal shedding only.

Control of H5/H7 low-pathogenic avian influenza (LPAI) virus circulation is a major issue regarding animal and public health consequences. To improve vaccines and to prevent vaccinated poultry from becoming infected and from shedding wild viruses, we initiated studies targeting prevention of H7 infection through DNA vaccines encoding H7 and M1 viral proteins from an Italian H7N1 LPAI virus isolated from poultry in 1999. More recently, we expressed recombinant H7 and M1 proteins in the baculovirus system to assess whether they might enhance immunity when given as a boost after DNA vaccination. The protection afforded by three vaccine combinations—DNA/DNA, DNA/protein, protein/protein—given 3 wk apart were experimentally compared in 20 specific-pathogen-free chickens per group. Ten days after the boost, chickens were challenged with a homologous (Italian H7N1 LPAI) or heterologous (French H7N1 LPAI isolated from mallards in 2001) virus. Tracheal and cloacal shedding was measured by a matrix gene (M)–based real-time reverse transcription polymerase chain reaction assay and compared with that displayed by unvaccinated infected controls. After the homologous challenge, chickens of every vaccinated group displayed a significant decrease in cloacal shedding, whereas tracheal shedding was not significantly reduced in the protein/protein group. After the heterologous challenge, only the DNA/DNA group showed a nonsignificant decrease in tracheal shedding. According to these two trials, prime–boost DNA/protein vaccination appeared be more advantageous. Further development could be aimed at improving protein expression, shifting subtype (H5), and assessing the interest of proteins as a boost of recombinant vaccines.

Avian influenza represents one of the greatest concerns for public health that has emerged in recent times. Highly pathogenic avian influenza viruses belonging to the H5N1 subtype are endemic in Asia and are spreading in Europe and Africa. Vaccination is now considered a tool to support eradication efforts, provided it is appropriately managed. This study was carried out to establish the degree of clinical protection and reduction of viral shedding induced by a high-specification, commercially available avian influenza vaccine of a different lineage and containing a strain with a heterologous neuraminidase (H5N9 subtype) to the challenge virus isolate A/chicken/Yamaguchi/7/2004 (H5N1 subtype).

A recombinant fowlpox-avian influenza (AI) H5 vaccine (rFP-AIV-H5) expressing the hemagglutinin of the A/turkey/Ireland/1378/83 H5N8 AI isolate has been used in Central America since 1998 to control H5N2 low pathogenicity AI. Previously, this vaccine was shown to induce full protection against a panel of H5 highly pathogenic (HP) AI isolates, including HPAI H5N1. Here, we evaluate the efficacy of rFP-AIV-H5 against escalating doses of HPAI H5N1 A/chicken/SouthKorea/ES/03 isolate and against the HPAI H5N1 A/chicken/Vietnam/0008/2004 isolate. In both studies, 1-day-old specific pathogen-free (SPF) chickens were vaccinated by subcutaneous route with rFP-AIV-H5 and challenged 3 wk later by the oronasal route. In the first study, full protection was observed up to a challenge dose of 6.5 log₁₀ embryo infectious dose (EID₅₀), and the 50% chicken infectious dose was estimated to be 3.1 and 8.5 log₁₀ EID₅₀ in the control and the rFP-AIV-H5-vaccinated group, respectively. A 2–4 log₁₀ and >4 log₁₀ reduction of oral and cloacal shedding was observed in rFP-AIV-H5 vaccinated birds, respectively. The rFP-AIV-H5 vaccine induced hemagglutination inhibition antibodies (5.2 log₂) detectable with homologous H5N8 antigen. In the second study, rFP-AIV-H5-vaccinated chicks were fully protected against morbidity and mortality after challenge with the 2004 Vietnam isolate, whereas unvaccinated chickens died within 2 days of challenge. Shedding in cloacal swabs was detected in all unvaccinated controls but in none of the rFP-AIV-H5-vaccinated chickens. Together, these results confirm the excellent level of protection induced by rFP-AIV-H5 in SPF chickens against two recent Asian HPAI H5N1 isolates.

Outbreaks of highly pathogenic avian influenza (HPAI) (2000–2003) resulted in 50 million EU birds culled or dead. The circulation of H5N1 in Asia could represent the origin of a human pandemic. Questions have been raised to combat the ongoing AI crisis. HPAI H5N1 has spilled over to resident and migratory wild bird populations which could represent a means of the virus reaching the EU, but lack of data make any forecast imprudent. Poultry holdings located close to migratory bird breeding and resting sites are considered at greater risk of exposure and methods to prevent exposure should be implemented. Legal safeguards for importation of poultry commodities currently only apply to HPAI and rely on detection of clinical signs that may not be observable during incubation period. Illegal imports represent an additional risk. Insufficient data on the effectiveness of commodity processing are available and few indications can be deducted. Biosecurity is the primary tool to prevent AI introduction and secondary spread. Massive spread was observed in densely populated poultry areas resulting in vaccination programs. Vaccination should be used to support eradication together with enhanced biosecurity and restriction measures, which shall also be implemented in case of prophylactic vaccination. Animal welfare aspects of AI include use of appropriate culling methods, correct vaccine application, and availability of trained staff. EFSA has recently set up a new scientific work group to further assess the risk of HPAI introduction and spread posed in particular by wild, migratory birds, as well as further follow-up of recent AI developments.

The avian influenza (AI) epidemic is threatening Africa mainly because the flyways of migratory birds link the endemic and newly infected countries with disease-free areas in this continent and because of the risk of introduction through trade. Risk analysis provides a set of tools for supporting decision making by the veterinary services and other stakeholders, resulting in more effective surveillance and emergency preparedness. The risk assessment process could be split into three different steps: 1) risk release through the migratory birds and the official and unofficial poultry-product marketing chains; 2) risk exposure by means of studying interfaces among imported and exposed poultry and among wild and domestic birds; and 3) risk consequences for establishing the probability of AI spreading within the poultry population and the probability of it escaping detection. A conceptual framework is presented based on preliminary data and field missions carried out in Ethiopia. Field surveys and expert opinion will be necessary for the parameterization of the risk model. Spatial analysis will be used to identify high risk of exposure among wild and domestic birds. Risk communication and risk management will be based on the findings from the risk assessment model.

Notifiable avian influenza (NAI) had never been reported in Spain, until July 2006 when a dead Great Crested Grebe (Podiceps cristatus) was found positive to the highly pathogenic H5N1 subtype as part of the active wild bird surveillance plan. The current program of the Spanish Ministry of Agriculture, Fisheries, and Food (MAPA)'s strategic preventive plan against NAI is divided in the following parts: identification of risk areas and risk wild bird species, increased biosecurity measures, early detection of infection with surveillance intensification and development of rapid diagnostic tests, and other policies, which include continuing education and training to ensure early detection of the disease. In 2003 an active surveillance plan was introduced for domestic fowl; the plan was extended to wild birds in 2004. A total of 18,780 samples in poultry and 3687 samples in wild birds had been analyzed through December 2005 to detect the presence and spread of avian influenza subtypes H5 and H7.

In the present work we suggest some contributions to be implemented in MAPA's action plan: 1) the identification of risks because of migratory birds, within the risk assessment of the introduction of NAI virus in Spain and 2) an interactive digital simulator of the disease developed for continuing education and training.