Bayesian Network analysis of piglet scours

Identifying Associations in Minimum Inhibitory Concentration Values of Escherichia coli Samples Obtained From Weaned Dairy Heifers in California Using Bayesian Network Analysis

Objective

Many antimicrobial resistance (AMR) studies in both human and veterinary medicine use traditional statistical methods that consider one bacteria and one antibiotic match at a time. A more robust analysis of AMR patterns in groups of animals is needed to improve on traditional methods examining antibiotic resistance profiles, the associations between the patterns of resistance or reduced susceptibility for all isolates in an investigation. The use of Bayesian network analysis can identify associations between distributions; this investigation seeks to add to the growing body of AMR pattern research by using Bayesian networks to identify relationships between susceptibility patterns in Escherichia coli (E. coli) isolates obtained from weaned dairy heifers in California.

Methods

A retrospective data analysis was performed using data from rectal swab samples collected from 341 weaned dairy heifers on six farms in California and selectively cultured for E. coli. Antibiotic susceptibility tests for 281 isolates against 15 antibiotics were included. Bayesian networks were used to identify joint patterns of reduced susceptibility, defined as an increasing trend in the minimum inhibitory concentration (MIC) values. The analysis involved learning the network structure, identifying the best fitting graphical mode, and learning the parameters in the final model to quantify joint probabilities.

Results

The graph identified that as susceptibility to one antibiotic decreases, so does susceptibility to other antibiotics in the same or similar class. The following antibiotics were connected in the final graphical model: ampicillin was connected to ceftiofur; spectinomycin was connected with trimethoprim-sulfamethoxazole, and this association was mediated by farm; florfenicol was connected with tetracycline.

Conclusions

Bayesian network analysis can elucidate complex relationships between MIC patterns. MIC values may be associated within and between drug classes, and some associations may be correlated with farm of sample origin. Treating MICs as discretized variables and testing for joint associations in trends may overcome common research problems surrounding the lack of clinical breakpoints.

Comparison of the fecal microbiomes of healthy and diarrheic captive wild boar

Diarrhea caused by Enterotoxigenic Escherichia coli (ETEC) is one of the most common clinical diseases observed in captive wild boars, is usually caused by an imbalance in the gut microbiome, and is responsible for piglets significant mortality. However, little research has been undertaken into the structure and function of the intestinal microbial communities in wild boar with diarrhea influenced by enterotoxigenic E. coli. In this study, fecal samples were collected and 16S-rRNA gene sequencing was used to compare the intestinal microbiome of healthy captive wild boar and wild boar with diarrhea on the same farm. We found that the intestinal microbial diversity of healthy wild boar (HWB) was relatively high, while that of diarrheic wild boar (DWB) was significantly lower. Line Discriminant Analysis Effect Size showed that at the genus level, the abundance of Escherichia-Shigella and Fusobacterium was significantly higher in DWB. Phylogenetic Investigation of Communities by Reconstruction of Unobserved States analysis showed that the expression of genes in pathways including infectious diseases: bacterial, metabolism of amino acids, membrane transport, and signal transduction was significantly higher in DWB. In summary, this study provides a theoretical basis for the design of appropriate means of diarrhea treatment in captive wild boar.

Additive Bayesian networks for antimicrobial resistance and potential risk factors in non-typhoidal Salmonella isolates from layer hens in Uganda

Background

Multi-drug resistant bacteria are seen increasingly and there are gaps in our understanding of the complexity of antimicrobial resistance, partially due to a lack of appropriate statistical tools. This hampers efficient treatment, precludes determining appropriate intervention points and renders prevention very difficult.

Methods

We re-analysed data from a previous study using additive Bayesian networks. The data contained information on resistances against seven antimicrobials and seven potential risk factors from 86 non-typhoidal Salmonella isolates from laying hens in 46 farms in Uganda.

Results

The final graph contained 22 links between risk factors and antimicrobial resistances. Solely ampicillin resistance was linked to the vaccinating person and disposal of dead birds. Systematic associations between ampicillin and sulfamethoxazole/trimethoprim and chloramphenicol, which was also linked to sulfamethoxazole/trimethoprim were detected. Sulfamethoxazole/trimethoprim was also directly linked to ciprofloxacin and trimethoprim. Trimethoprim was linked to sulfonamide and ciprofloxacin, which was also linked to sulfonamide. Tetracycline was solely linked to ciprofloxacin.

Conclusions

Although the results needs to be interpreted with caution due to a small data set, additive Bayesian network analysis allowed a description of a number of associations between the risk factors and antimicrobial resistances investigated.

Progression and risk factors of pododermatitis in part-time group housed rabbit does in Switzerland

In rabbits (Oryctolagus cuniculus L.), pododermatitis is a chronic multifactorial skin disease that appears mainly on the plantar surface of the hind legs. This presumably progressive disease can cause pain leading to poor welfare, yet the progression of this disease has not been thoroughly assessed on the level of individual animals. The aim of this longitudinal study thus was to investigate the possible risk factors and the progression of pododermatitis in group housed breeding does in Switzerland on litter and plastic slats. Three commercial rabbit farms with part-time group housing on litter and plastic slats were visited every four weeks throughout one year. During every visit, the same 201 adult female breeding rabbits (67 does per farm) were evaluated for the presence and severity of pododermatitis. Additionally, the does' age, parity, body weight, reproductive state, hybrid, claw length, cleanliness and moisture of the paws and the temperature and humidity inside the barns were recorded as potential risk factors. The risk factors were analysed through general linear models and additive Bayesian network (ABN) modelling using a directed acyclic graph (DAG) for visualising associations between potential risk factors. The progression of pododermatitis was analysed with a transition matrix. Relative humidity inside the barns, body weight, number of kindlings, age, and claw length were the most important risk factors, all being positively associated with pododermatitis. In contrast to expectations, the cleanliness of the left hind paw was negatively associated with the occurrence of pododermatitis, but the effect was small. In young does, the severity of pododermatitis quickly increased and in some rabbits proceeded to ulcerated spots. It was shown that 60.00%, 14.17% and 3.33% of ulcerated lesions recovered to a state without ulceration within 4, 8 or >12 weeks, respectively.

Modeling Dynamic Human Behavioral Changes in Animal Disease Models: Challenges and Opportunities for Addressing Bias

Over the past several decades, infectious disease modeling has become an essential tool for creating counterfactual scenarios that allow the effectiveness of different disease control policies to be evaluated prior to implementation in the real world. For livestock diseases, these models have become increasingly sophisticated as researchers have gained access to rich national livestock traceability databases, which enables inclusion of explicit spatial and temporal patterns in animal movements through network-based approaches. However, there are still many limitations in how we currently model animal disease dynamics. Critical among these is that many models make the assumption that human behaviors remain constant over time. As many studies have shown, livestock owners change their behaviors around trading, on-farm biosecurity, and disease management in response to complex factors such as increased awareness of disease risks, pressure to conform with social expectations, and the direct imposition of new national animal health regulations; all of which may significantly influence how a disease spreads within and between farms. Failing to account for these dynamics may produce a substantial layer of bias in infectious disease models, yet surprisingly little is currently known about the effects on model inferences. Here, we review the growing evidence on why these assumptions matter. We summarize the current knowledge about farmers' behavioral change in on-farm biosecurity and livestock trading practices and highlight the knowledge gaps that prohibit these behavioral changes from being incorporated into disease modeling frameworks. We suggest this knowledge gap can be filled only by more empirical longitudinal studies on farmers' behavioral change as well as theoretical modeling studies that can help to identify human behavioral changes that are important in disease transmission dynamics. Moreover, we contend it is time to shift our research approach: from modeling a single disease to modeling interactions between multiple diseases and from modeling a single farmer behavior to modeling interdependencies between multiple behaviors. In order to solve these challenges, there is a strong need for interdisciplinary collaboration across a wide range of fields including animal health, epidemiology, sociology, and animal welfare.

Current and future antimicrobial resistance issues for the Australian pig industry

Antimicrobial use and antimicrobial resistance (AMR) in intensive pig production and its potential impacts to human and animal health are very much under the spotlight, both internationally, and within Australia. While the majority of AMR of medical importance is associated with the exclusive use of antimicrobials in humans, resistance in zoonotic foodborne pathogens such as Salmonella and Campylobacter, and livestock commensal bacteria such as Escherichia coli and Enterococcus spp., is under increased scrutiny. This is primarily due to the current reliance on many of the same drug classes as used in human medicine for treatment and control of bacterial diseases of livestock. Furthermore, the development of multidrug resistance in pathogens such as enterotoxigenic E. coli may drive off-label use of critically important drug classes such as 3rd-generation cephalosporins. This could lead to the emergence and amplification of resistance genes of potential public health significance in both pathogens and commensal bacteria. Livestock-associated and community-associated methicillin-resistant Staphylococcus aureus has also recently been detected in Australian pigs as a result of human-to-animal transmission and are a potential public health issue for in-contact piggery workers. Australia is in a unique position compared with many of its international trading partners due to its isolation, ban on importation of livestock and conservative approach to antimicrobial registration, including reservation of the fluoroquinolone class for use in humans and companion animals only. Cross-sectional AMR surveys of pathogens and commensals in healthy pigs have identified only low frequency of resistance to critically important drug classes. Nevertheless, resistance to critically important antimicrobials has emerged and careful antimicrobial stewardship is required to ensure that these low levels do not increase. In this report, we review AMR of significance to the Australian pig industry and identify potential prevention and control measures.