Antimicrobial Resistance Modeling

Antimicrobial resistance is a growing concern in connection with treatment of infections and the World Health Organization (WHO) now recognizes it as being amongst the ten biggest threats to the global health. Antimicrobial resistance bacteria can cause treatment extension or failure due to resistance against antimicrobial drugs. The spread of antimicrobial resistance in livestock constitutes a risk for animal welfare and increases the risk of transferring antimicrobial resistant bacteria from animals to humans.

The overall aim of the Ph.D. thesis is to gain more knowledge on the influence of antimicrobial use and management factors on antimicrobial resistance in Danish slaughter pigs. This is achieved by analyzing and modeling the spread of antimicrobial resistance genes in fecal samples from Danish slaughter pigs in relation to antimicrobial exposure and management factors. The data on antimicrobial resistance originates from a study, where faeces from Danish pigs were sampled at the time of slaughter. The collected fecal samples were analyzed by qPCR quantifying the amount of antimicrobial resistance genes.

An algorithm was develop for estimating the antimicrobial exposure of the sampled pigs. In Denmark, pigs are categorized into three phases based on their weight: piglets, weaners and finishers. The algorithm estimates the average antimicrobial exposure for a Danish pig in each rearing period. This is done by tracing the location(s) based on movements of the pigs registered in the Pig Movement Database and subsequently estimating the antimicrobial exposure derived from the national VetStat-register, which contains information on purchased veterinary drugs.

Mixed effect models were applied to data to examine the relationship between antimicrobial exposure, management factors and antimicrobial resistance in the pig production. Tetracyclines are one of the most used antimicrobial classes in the pig production. The total resistance level against tetracycline, calculated as a sum of the genes coding for resistance against tetracycline and the level of individual resistance genes coding for resistance against different antimicrobial classes were modeled. The measured individual resistance genes, which were observed in at least 50% of the samples were modeled and the results were compared to metagenomic data of an antimicrobial resistance context study.

The key findings are that management factors in the form of movement patterns and ownership of the farms are crucial for the resistance levels of Danish slaughter pigs. In addition, some observed complex antimicrobial exposure and resistance patterns described by the models, might be explained by co-occurrence of antimicrobial resistance genes i.e. antimicrobial resistance genes which occur together in a genomic context. The modeling suggested that the occurrence of the resistance gene tet(X) is affected only by the antimicrobial exposure of macrolide and lincosamide classes, which is in contrast to the commonly accepted hypothesis that it provides resistance against tetracyclines. This result might be a consequence of observed co-occurrence with erm(F), that provides resistance against several antimicrobial classes including macrolides and lincosamides.

The results could contribute to qualified discussions on treatment strategies and targeted interventions in Danish pig production. Reductions in usage of certain antimicrobial classes do not necessarily yield a lower abundance of resistance genes for these classes and the genomic context should be considered in assessments. Furthermore, there should be a focus on more systematic data collection in future studies and the surveillance of the development in antimicrobial resistant bacteria should continue with high quality.