Epidemiology of Newcastle disease in village chickens in Ethiopia : risk factors, molecular characterization and role of poultry markets

Abstract

Village chickens contribute considerably to the economy and to the nutritional requirements and livelihood of many rural farmers in developing countries across the globe. The spread of highly pathogenic avian influenza H5N1 into Africa during 2005/6 drew attention to the neglect of avian disease surveillance and research in countries such as Ethiopia, in which predominantly village chickens are reared. Several infectious and non-infectious diseases have limited the productivity of village chickens in Ethiopia, among which Newcastle disease (ND), caused by avian paramyxovirus serotype 1 (APMV-1), is the most important. Newcastle disease virus (NDV) causes subclinical to severe disease depending on the virus strain. To better understand the epidemiology of the disease, a study was performed in the mid-Rift Valley area of Oromia region, Ethiopia, to estimate seroprevalence and incidence of NDV exposure, identify risk factors, evaluate market trade movements and characterize circulating NDV strains. Repeated serological surveys in live bird markets revealed that village chickens were concurrently seropositive for several important infectious diseases, particularly during the wet season. The seroprevalence of ND, Pasteurella multocida infection, Mycoplasma gallisepticum infection and infectious bursal disease virus infection were 5.9%, 66.2%, 57.7% and 91.9%, respectively, during the dry season, and 6.0%, 63.4%, 78.7% and 96.3%, respectively, during the wet season. This underlines the need for a holistic approach to control of infectious disease in village chickens, and further studies are warranted to better understand the circulating strains, their interactions and their economic effect on village poultry production. A cross-sectional study using a multistage random sampling design with repeated sampling periods was done in households, along with a structured questionnaire. The prevalence of household flocks with at least one seropositive chicken was higher during the dry season (27.4%) than during the wet season (17.4%) (P = 0.003) while the proportion of flocks in which viral genome was detected was 24.2% and 14.2 %, respectively. The prevalence of NDV genome detection in individual birds at markets varied from 4.9 % to 38.2, depending on the period of sampling and the reverse transcriptase polymerase chain reaction (RT-PCR) technique employed. Multilevel mixed-effect logistic regression models were used to identify risk factors for NDV seropositivity and for incidence of NDV exposure. Reduced frequency of cleaning of poultry waste, larger flock size and use of an open water source (pond or river) for poultry were associated with increased risk of NDV exposure or seropositivity, while maintaining a closed flock and the use of a grain supplement was associated with lower odds of seropositivity or a lower risk of NDV exposure. Molecular characterization and phylogenetic analysis, based on complete F and HN gene sequencing, was done on NDV isolates obtained at markets and villages. The circulating viruses had amino acid motifs characteristic of virulent strains, indicating endemic circulation of virulent virus in village chickens which poses a threat to improvement of village chicken production and emerging small-scale commercial poultry production. The strains clustered in genotype VI, branching with viruses from subgenotype VIb that commonly affect pigeons, although clustering apart on pairwise distance analysis. The apparent poor biosecurity in village chickens and history of isolation of pigeon variant viruses from domestic chickens in Ethiopia suggest that pigeons could play a role in the epidemiology of ND in village chickens. Further surveillance and virus characterization is required to shed more light on this. Bayesian methods were used to evaluate the performance of two commercial enzyme-linked immunosorbent assay (ELISA) kits (a blocking and an indirect ELISA) and haemagglutination inhibition (HI), in the absence of a gold standard, for their ability to detect antibodies to NDV in chicken serum from villages and live bird markets. The blocking ELISA had the highest sensitivity (Se) of 96.3% (95% posterior credible interval (PCI): 88.1; 99.8%), and specificity (Sp) of 98.9% (95% PCI: 97.8; 99.9%), while the HI had Se of 81.6% (95% PCI: 71.8, 91.9%), and Sp of 96.1% (95% PCI: 95.1; 96.6%). The indirect ELISA also had high Se (95.2%; 95% PCI: 88.5; 99.0%) but had very low Sp (8.9%; 95% PCI: 6.4, 11.8%). There is therefore a need for evaluation of commercial kits before their wider use in village chickens under field conditions. Market trade movement patterns for live chickens were described, using social network analysis, for two different periods during the year 2010, representing high (period one) and low (period two) seasons for poultry trade. The study revealed that the networks exhibited scale-free characteristics with weak connectivity of the markets and low density of the networks. The density for the two periods was not difference (P = 0.29), although a somewhat higher number of markets and links were observed during period one than period two. The low density of the networks indicates that in the event of infectious disease outbreaks in surroundings of the respective markets, the risk of its spread to many others would likely be fairly low. Nevertheless, the close similarity of NDV isolates from distant markets in the study area suggests that markets could play a role in the spread of infectious poultry diseases. A few markets were more central in the networks, in terms of their betweenness and out-degree; these markets could be considered for targeted surveillance, while those markets with high in-degree, mainly situated in the larger urban centres, can be considered for surveillance that involves regular poultry traders.