Search

Introduction: Escherichia coli, Klebsiella pneumoniae and Enterobacter (EKE) are the leading cause of mortality and morbidity in neonates in Africa. The management of EKE infections remains challenging given the global emergence of carbapenem resistance in Gram-negative bacteria. This study aimed to investigate the source of EKE organisms for neonates in the maternity environment of a national referral hospital in Uganda, by examining the phenotypic and molecular characteristics of isolates from mothers, neonates, and maternity ward.Methods: From August 2015 to August 2016, we conducted a cross-sectional study of pregnant women admitted for elective surgical delivery at Mulago hospital in Kampala, Uganda; we sampled (nose, armpit, groin) 137 pregnant women and their newborns (n=137), as well as health workers (n=67) and inanimate objects (n=70 – beds, ventilator tubes, sinks, toilets, door-handles) in the maternity ward. Samples (swabs) were cultured for growth of EKE bacteria and isolates phenotypically/molecularly investigated for antibiotic sensitivity, β-lactamase and carbapenemase activity. To infer relationships among the EKE isolates, spatial cluster analysis of phenotypic and genotypic susceptibility characteristics was done using the Ridom server.Results: Gram-negative bacteria were isolated from 21 mothers (15%), 15 neonates(11%), 2 health workers (3%), and 13 inanimate objects (19%); a total of 131 Gramnegative isolates were identified of which 104 were EKE bacteria i.e., 23 (22%) E. coli, 50 (48%) K. pneumoniae, and 31 (30%) Enterobacter. Carbapenems were the most effective antibiotics as 89% (93/104) of the isolates were susceptible to meropenem; however, multidrug resistance was prevalent i.e., 61% (63/104). Furthermore, carbapenemase production and carbapenemase gene prevalence were low; 10% (10/104) and 6% (6/104), respectively. Extended spectrum β-lactamase (ESBL) production occurred in 37 (36%) isolates though 61 (59%) carried ESBL-encoding genes, mainly blaCTX-M (93%, 57/61) indicating that blaCTX-M is the ideal gene for tracking ESBL-mediated resistance at Mulago. Additionally, spatial cluster analysis revealed isolates from mothers, new-borns, health workers, and environment with similar phenotypic/genotypic characteristics, suggesting transmission of multidrugresistant EKE to new-borns.Conclusion: Our study shows evidence of transmission of drug resistant EKE bacteria in the maternity ward of Mulago hospital, and the dynamics in the ward are more likely to be responsible for transmission but not individual mother characteristics. The high prevalence of drug resistance genes highlights the need for effective infection prevention/control measures and antimicrobial stewardship programs to reduce spread of drug-resistant bacteria in the hospital and improve patient outcomes.

Introduction: At the peak of Uganda’s first wave of SARS-Cov2 in May 2020, one in three COVID-19 cases was linked to the haulage sector. This triggered a mandatory requirement for a negative PCR test result at all ports of entry and exit (POEs), resulting in significant delays as haulage drivers had to wait for 24-48 hours for results, which severely crippled the regional supply chain. To support public health and economic recovery, we aim to develop and test a mobile phone-based digital contact tracing (DCT) tool that not only augments conventional contact tracing but increases its speed and efficiency. Methods and analysis: To test the DCT tool, we will use a stratified sample of haulage driver journeys, stratified by route type (regional and local journeys). We will include at least 65% of the haulage driver journeys ~ 83,200 on the network through Uganda. This allows us to capture variations in user demographics and socio-economic characteristics that could influence the use and adoption of the DCT tool. The developed DCT tool will include a mobile application and web interface to collate and intelligently process data, whose output will support decision making, resource allocation and feed mathematical models that predict epidemic waves.The main expected result will be an open source tested Digital Contact Tracing (DCT) Tool tailored to haulage use in developing countries. This study will inform the safe deployment of digital contact tracing technologies needed for combatting pandemics in low-income countries. Ethics and dissemination: This work has received ethics approval from School of Public Health Higher Degrees, Research and Ethics Committee at Makerere University and The Uganda National Council for Science and Technology. This work will be disseminated through peer reviewed publications, our websites https://project-thea.org/ and Github for the open source code https://github.com/project-thea/

Additional file 5. Taxonomic association with shedding. The relationship between shedding and the abundance of individual families colored by treatment group is demonstrated.

Additional file 6. Taxonomic association with pathology (microscopic ileum score). The relationship between pathology and the abundance of individual families colored by treatment group is demonstrated.

Additional file 2. Association of histopathology and shedding for each of the treatment groups.

Additional file 3. Taxa and percentage of bacteria present the inoculum used to challenge the pigs.

Bovine Tuberculosis (bTB) caused by Mycobacterium bovis is a livestock disease of global economic and public health importance. There are currently no effective vaccines available for livestock and so control relies on animal level surveillance and pasteurisation of dairy products. A new alternative control approach is to exploit the genetic variability of the host; recent studies have demonstrated that breeding Bos taurus cattle for increased resistance to bTB is feasible. The utility of such an approach is still unknown for the Bos indicus cattle population. This study aims to assess genetic variation in bTB resistance and the underlying genomic architecture in Bos indicus breeds in Cameroon. We conducted a cross-sectional study of slaughter cattle in Cameroon and genotyped a sample of 213 cattle. Their genomic diversity was characterised using PCA, hierarchical clustering and admixture analysis. We assessed genetic variation in bTB resistance using heritability analysis and compared quantitative trait loci. Previous studies had found that breed was an important factor in explaining the epidemiology of bTB, with Fulani cattle appearing to be more susceptible than mixed breeds. However, we show that the apparent phenotypic differences in visual appearance between the breeds was not reflected by clear genomic differences. At the genetic level, cattle belonging to different hierarchical genomic clusters differed in their susceptibility to bTB. There was evidence of a genomic association between M. bovis infection status with specific SNPs. We highlight the need to understand the challenges faced by livestock in specific settings both in terms of pathogens and the environment, in addition to their intended purpose and how they fit into a defined management system. It is only at this point livestock keepers can then make informed breeding choices, not only for resistance to disease but also for increasing production

The use of network analysis to support livestock disease control in low middle-income countries (LMICs) has historically been hampered by the cost of generating empirical data in the absence of animal movement recording schemes. To fill this gap, we have adopted methods which exploit freely available demographic and archived molecular data to generate livestock networks based on phylogeographic and gravity modelling techniques. We compare output from these network methodologies to empirical and randomly generated data. We simulate disease scenarios on the networks to evaluate the potential utility of our methodologies to inform robust livestock disease control strategies. The molecular network was the closest approximation to the empirical network, both in relation to topological and epidemic characteristics. The gravity network tended to overestimated disease epidemics. However, better agreement across all three networks was observed if less specific epidemic characteristics such as the size of outbreak were investigated. Moreover, these methods consistently identified the same important animal movement and trade hotspots as the empirical networks. We therefore consider this proof-of-concept that demographic data such as censuses and archived molecular data could be repurposed to inform livestock disease management in LMICs.