This is a cross-sectional pilot study designed to collect air samples from high-risk and low-risk environments in urban and periurban settings of Bangladesh, mainly the Dhaka metropolitan area and Mirzapur Upazilla (subdistrict) of Tangail District. Environmental systematic and grid sampling will be followed according to approaches described by Keith [35]. This is suitable for finding hotspots for target organisms/genes over a specific period of time, as samples are taken at regularly spaced intervals. Data collection started in January 2018, and the laboratory analysis is currently ongoing. The live bird markets and commercial poultry farms are considered to be high-risk areas, while the urban residential areas and the periurban households are low-risk areas (Figure 1).

Mapping with GIS software (ArcGIS, Esri, Redlands, California) will plot concentrations of resistant genes and antimicrobial-resistant bacteria in environmental compartments (high- and low-risk areas) in each location using GPS coordinates. Temporal variation will be observed by comparing the magnitude of resistant bacteria and genes in the dry season with those in the wet season.

Air samples from commercial poultry farms, live bird markets, rural households, and urban residential areas will be collected. All active sampling will be performed using the same Surface Air System Sampler (SAS Super 180 Microbial Air Sampler, Bioscience International, Rockville; Figure 2), with a flow rate of 180 L/min. The following media will be used for collection of samples: standard plate count (SPC) agar (Oxoid, Hampshire, United Kingdom) for aerobic plate count, MacConkey agar (BD Difco, Becton Dickinson, New Jersey) supplemented with cefotaxime (1 mg/L) and MacConkey agar supplemented with meropenem (0.5 mg/L) to obtain gram-negative resistant organisms, Mannitol Salt agar (MSA) (Oxoid) supplemented with oxacillin (8 mg/L), and Slanetz and Bartley (SB) agar (Oxoid) medium supplemented with vancomycin (6 mg/L) to obtain gram-positive resistant organisms. During sampling, the aspirating head will be removed from the air sampler. An identified, closed, and prepared plate will be inserted, and the plate lid will be removed. The aspirating head will be replaced again to cover the plate. The required volume and air flow rate will be adjusted, and the air sampler will be started. Subsequently, the airflow will be directed into the agar surface of the plate and at the end of a cycle, the aspirating head will be removed. The plate will be closed with the lid and removed from the air sampler. An identical method will be followed for different types of agar plates in each location, although the air volume will be different (ie, 1000 L for MacConkey, MSA, and SB agar and 100 L for SPC agar). After sampling, the plates will be carried in a carrying case and transported to the laboratory.

The outcome variables that will be assessed are as follows:

The rise and spread of superbugs have become a key public health and planetary health concern worldwide. Infections caused by multidrug-resistant bacteria are linked to greater mortality rates than antimicrobial-susceptible bacteria [45]. For therapeutic purposes, medically important antimicrobials are used extensively in agricultural and farming industries for disease prevention (eg, prophylaxis and metaphylaxis), treatments, and growth promotions. More than two-thirds of antimicrobials are consumed in the livestock sector each year. It is projected that by 2030, there may be a massive increase (up to 67%) in global antimicrobial use in food-producing animals, especially in Brazil, Russia, India, China, and South Africa (BRICS) [46]. The antimicrobial resistance problem is severe, not only in developed countries but also in nonindustrialized countries. Scarcity of antimicrobial usage policies and substandard hygiene situations are the precipitating factors that drive antimicrobial resistance issue to be more challenging [47].

The burden of the antimicrobial resistance rate depends on the population of a country and its environment. Bangladesh is a highly populated country; therefore, the rate of antimicrobial resistance is extensive due to rapid spread of antimicrobial-resistant organisms. Antimicrobial resistance genes are conferring their resistance value to a wider community including both animals and humans, through close interactions with the environment and the wastes that are disposed in the environment, directly affecting the food chain [48]. Due to the extensive use of antibiotics as therapeutic and prophylactic in farming, bacterial resistance may be developed by either chromosomal gene mutation or gene acquisition through different mechanisms like transformation, transduction, or conjugation [49]. Continuous accumulation of multiple mutations have caused the genome to become resistant to antibiotics. To screen the antimicrobial resistance genes, whole genome sequencing is an important tool for understanding the antimicrobial resistance mechanisms. Whole genome sequencing helps evaluate the number of mutations and particular mechanisms of the mutated gene that drive antimicrobial resistance to develop new drugs (antibiotic) and diagnose the disease state and treatment process [47]. Primary biological aerosol particles such as bacteria, viruses, pollens, and mold spores are important components of airborne particulate matter (PM), and abundant pathogenic bacteria have been identified from PM2.5 and PM10 samples [50]. However, these compounds from pollution and other sources are not prioritized during analysis. Our study will also focus on the resistant bacteria as an integral part of air pollution.

Intrinsic genetic determinants of resistance factors are harbored by bacteria with macromolecules in the environment. Robust evidence suggests that such macromolecules developing “environmental resistomes” are a source from which clinically relevant bacteria acquire antibiotic resistance genes [51]. Poultry, the most common rural domestic species, is considered a major driver for selection of antimicrobial resistance from the environment to human/animal due to fecal shedding of resistant bacteria. As poultry feces are used as fertilizers and household members share sleeping spaces with poultry animals, antibiotic use in poultry in the last 6 months has been reported by more than 50% poultry owners [37]. Commercial poultry farm areas harbor the antimicrobial-resistant isolates in the air, and humans are easily exposed to these antimicrobial-resistant isolates during inhalation of dust and particles. Consequently, antimicrobial-resistant organisms can build up a strong biofilm in the intestinal tract of humans. Through excretion of stool and other body fluids, these can be spread in the environment. Urban live bird markets are the largest site of bird slaughtering, with no proper waste management setting, and the huge amounts of wastes are directly passed into environment through dust and direct wash out [14].

The importance of environmental compartments as the transmission hub of antimicrobial resistance is well established. However, airborne resistomes and their transmission pathway are poorly studied. Without containment of environmental reservoirs, antimicrobial resistance prevention policy will fail [52]. The occurrence of pathogenic resistant organisms and resistance genes in atmospheric air of different locations will guide researchers and policy makers to adopt new strategies for the containment of this alarming issue. However, further large-scale studies need to be carried out to link the clinically important organisms and airborne resistomes. To achieve this, One Health surveillance needs to be carried out in humans (both healthy and clinical patients), animals, and environments at the national level.

Regarding the strengths, our study will focus on the air resistomes, which is a less explored environmental dimension of antimicrobial resistance transmission dynamics. To our knowledge, this is the first study to explore the presence of superbugs in the air in commonly exposed areas in Bangladesh. The limitation exists in the small sample size and lack of baseline data from the atmospheric environment of Bangladesh. Therefore, the findings may not be generalizable for all areas in the country.