READING Avian Influenza Study in Bangladesh

Overview

• A study was conducted in Bangladesh to understand the infection patterns of Highly Pathogenic Avian Influenza (HPAI) H5N1 and Low Pathogenic Avian Influenza (LPAI) H9N2 viruses.
• Cross-sectional studies were conducted on 144 backyard, 106 broiler, and 113 layer chicken farms in 2016 and 2017.
• No sampled birds had H5 virus via RT-PCR, though H5 antibodies were found in unvaccinated birds across all farming systems.

Antibody Prevalence

• Backyard ducks had a higher H5 antibody prevalence at 14.2% (95% CI: 10.0%-19.8%) compared to backyard chickens at 4.2% (95% CI: 2.8%-6.1%).
• Antibody prevalence for H5 virus was lower in broiler chickens 1.5% (95% CI: 0.9%-2.5%), versus layer chickens 7.8% (95% CI: 6.1%-9.8%).
• H9 viruses were found via RT-PCR in 0.5% (95% CI: 0.2%-1.3%) and 0.6% (95% CI: 0.3%-1.5%) of broilers and layers, and in 0.2% (95% CI: 0.0%-1.2%) of backyard chickens.
• H9 antibody prevalence was similar in backyard chickens and ducks: 16.0% (95% CI: 13.2%-19.2%) and 15.7% (95% CI: 11.3%-21.4%), and higher than layers: 5.8% (95% CI: 4.3%-7.6%), and broilers: 1.5% (95% CI: 0.9%-2.5%).
• Over a production cycle, H5 and H9 antibody prevalence increased with age in backyard and layer chickens.
• Multiple ducks usually tested H5 antibody positive within a flock, while only individual birds developed H5 antibodies in backyard chickens, broilers, and layers.
• The findings highlight low virus circulation in healthy chickens across production systems in Bangladesh, unlike high virus circulation in live bird markets.
• Data can inform risk-based surveillance methods and mathematical models for HPAI infection dynamics in poultry.

Introduction Context

• HPAI H5N1 virus is considered endemic in several countries including Bangladesh, causing sporadic human cases from exposure to infected poultry.
• Co-circulation of LPAI H9N2 in H5N1 areas may lead to reassortants that can effectively spread among humans.
• H5N1 has severely affected poultry in infected countries, resulting in over 2.7 million culled poultry in Bangladesh between 2007 and 2019.

Study Objectives and Hypothesis

• A key objective was to quantify H5 and H9 virus circulation in backyard chicken farms and commercial broiler and layer chicken farms in two areas of Bangladesh.
• It was hypothesized that poultry species and husbandry systems affect avian influenza virus transmission, requiring an improved understanding of bird and flock-level infection patterns.

Materials and Methods

• The targeted populations were in the Chittagong and Cox's Bazar districts
• Study design involved two cross-sectional studies
• Backyard chicken farms were visited from February to April 2016
• Commercial broiler and layer chicken farms were visited from February to April 2017

Sample Size and Selection

• Sample size calculations assumed that there would be differences in H5 bird- and flock-level antibody prevalence based on species and means of production.
• Subdistricts for sampling were selected based on factors influencing AI transmission like poultry farm density, location, environment, and distance to live bird markets.
• Villages were selected randomly from quartiles based on the number of households or backyard farms per village using Microsoft Excel.
• An attempt was made to sample at least 123 backyard chicken farms and in each village selected 3-4 farms which also raised ducks, selected by counting farms from the village entrance.
• 10-13 broiler and 10-11 layer farmers were sampled in each subdistrict using random number generation in Microsoft Excel.

Bird Selection and Sample Collection

• Backyard chickens and in-contact ducks were sampled by capturing them, while commercial farm chickens were selected across the poultry shed.
• Farmers gave written consent before sampling.
• Blood, cloacal, and oropharyngeal swabs were taken from each bird with data taken, 1-3 ml of blood was extracted and placed into sterile plastic tubes.

Diagnostic and Data Analysis

• Serum tested for Influenza A antibodies using ELISA kits.
• Influenza A-positive samples were tested for H5 and H9 specific antibodies using the haemagglutination inhibition (HI) test.
• Swab samples were pooled and tested for Influenza A virus by real-time RT-PCR.
• Statistical analysis was conducted using Microsoft Excel 2013 and STATA 14.1.

Outcomes and Implications

• Apparent and true virus prevalence calculated for Influenza A, H5 and H9
• Prevalence values used to stratify bird-level antibody prevalence by age groups alongside confidence intervals.
• Clustering and low bird-level H5 and H9 antibodies can be used to inform better surveillance tailored to different chicken production systems.