Mohammad Mahmudul Hassan
Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin University, Geelong, Australia; Faculty of Veterinary Medicine, Chittagong Veterinary and Animal Sciences University, Chittagong, Bangladesh
Md. Ahasanul Hoque
Faculty of Veterinary Medicine, Chittagong Veterinary and Animal Sciences University, Chittagong, Bangladesh
Nitish Chandra Debnath
Faculty of Veterinary Medicine, Chittagong Veterinary and Animal Sciences University, Chittagong, Bangladesh, 3 FAO, Dhaka, Bangladesh
Mat Yamage
FAO, Dhaka, Bangladesh
Marcel Klaassen
Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin University, Geelong, Australia
AIV, Antibodies, Domestic birds, Resident wild birds, Migratory birds, Spillover, Spill back
Across Bangladesh
Pest Management
Poultry, Wild Birds
We sampled a wide range of wild and domestic birds from a variety of locations across Bangladesh, between May 2012 and December 2015. Although birds were sampled throughout the year, most of them were sampled during the months November through March, when also most outbreaks of HPAI occur (Biswas et al. 2014). Domestic birds were mostly sampled randomly and systematically and ranged from birds kept in commercial poultry sheds, i.e. layer and broiler (i.e. meat) chickens and chickens, ducks, quail and pigeons from live bird markets (LBMs), to birds kept on private properties in a household setting (i.e. backyard or household chickens, ducks, quail and pigeons; scientific names, order and subfamily of all bird species are provided in Supplementary Table S1), to free-ranging or range ducks, which are left unattended for most of their life after being released in wetlands at 4 weeks of age and are rounded-up for sale approximately 48 weeks later. Commercial farm chickens were sampled from 32 randomly selected farms where we targeted five samples from each farm (resulting in n = 159 sampled birds). Domestic or household pigeons (n = 13) and household chickens (n = 111) were randomly sampled, where we targeted one sample from each household farm and household ducks (n = 1232) were randomly sampled, where we targeted five samples from each household farm. Range ducks were sampled from 15 randomly selected flocks where five samples were targeted from each flock (n = 72) at two major wetlands, Hakaluki and Tanguar Hoar, in the vicinity of the city of Sylhet in north-eastern Bangladesh. Chickens (n = 27), ducks (n = 26), pigeons (n = 22), quail (n = 51) and spotted doves (n = 22) were also sampled in 20 randomly selected LBMs of Chittagong Metropolitan Area and its adjacent subdistricts of Anwara and Patiya and Sylhet and Sunamganj, Gazipur and Dhaka Metropolitan Areas, where we targeted one bird/shop. Resident wild birds (n = 1662), which reproduce in and are resident to Bangladesh, were sampled conveniently throughout the year, especially at roosting sites in the vicinity of LBMs, around farms in the area of Chittagong, Dhaka and Sylhet and in the wetlands of Hakaluki and Tanguar Hoar. We conveniently sampled migratory wild birds (n = 188), which visit wetlands in Bangladesh during the winter season (November to March) only, at the Hakaluki and Tanguar Hoar wetlands. The ultimate sample sizes varied due to mixed success in catching wild and migratory birds and convincing the general public and salesman to allow us sample their birds. All wild resident and migratory birds were caught using mist nets. Cloacal and oro-pharyngeal swabs along with blood samples were collected from each bird except for birds from LBMs where we sampled cloacal swabs only. Swabs were taken from birds by inserting swab sticks (until faecal contamination) into the vent for cloacal swabs and oro-pharyngeal airway and wall of oro-pharynx for oropharyngeal swabs. Each of the cloacal and oro-pharyngeal swab samples was placed separately into a vial containing 1 ml of sterile viral transport media (Druce et al. 2012). Samples collected in the Chittagong area were stored in an insulated container with ice packs until transfer to - 80C in the laboratory at Chittagong Veterinary and Animal Sciences University (CVASU), within 2–4 h of collection. Samples collected in the areas of Dhaka and Sylhet were immediately stored in liquid nitrogen after collection. Whole blood samples for AIV antibody prevalence analyses (0.5–3 ml, in all cases <1% of body weight) were drawn aseptically from wing veins or jugular veins and then immediately transferred to individual sterile tubes. Blood samples were subsequently allowed to clot at ambient temperature, kept refrigerated overnight, followed by centrifugation at 10,000 rpm for 30 min at 4C to separate serum. Serum was then transferred into cryovials and preserved at -20C (Basler et al. 1999). Serum samples were evaluated by competitive enzymelinked immunosorbent assay (c-ELISA) (Hoque 2011). Swabs were tested for AIV RNA using RT-PCR directed at the matrix (M) gene in an ABI Fast Real-Time PCR Machine ABI 7500 (AAHL 2014; Heine et al. 2007). For the latter, we used an Invitrogen reaction kit (Superscript iii platinon One-step Quantitative RT-PCR system—Cut No. 11732-088) and a fast cycling programme for the ABI 7500 (fast mode; thermal profile 50C for 5 min, hold 95C for 2 min, hold 40 cycles of: -95C, 2 s; 60C, 30 s). We plotted the sampling locations on a map of Bangladesh using the spatial analyst tool of ArcGIS (ArcMap, version 10.2, Environmental Systems Research Institute, Redlands, California, USA). We used generalized linear models to analyse binomial variation in both sero- and viral prevalence across species. In the basic model, we only included species as a random factor. In a series of subsequent models of increasing complexity, we included up to two fixed factors and their interaction, with one fixed factor distinguishing between Anseriformes and non-Anseriformes and the other fixed factor distinguishing between domestic and wild birds. Using a similar procedure but for wild birds only, we also evaluated possible differences between resident and migratory birds after also distinguishing between Anseriformes and non-Anseriformes. Only species for which a minimum of ten samples was available were used in the analyses.
EcoHealth 14, 490–500, 2017
Journal