A systematic review and meta-analysis of antibiotic resistance of foodborne pathogenic bacteria

Abstract

Antimicrobial drugs are used to treat bacterial pathogens that cause infections in humans and animals. Despite their importance, antimicrobial drugs exhibit inefficiency in treating infections if used irrationally without adherence to standard guidelines. Currently there is a lack of review literatures concerning antimicrobial resistance status in the southern sub Saharan African countries, hence the study is designed for and provides valuable insights into the status and comparison of antimicrobial resistance among foodborne bacteria in Zambia relative to other regions of the world, using systematic literature review and meta-analysis. For meta-analysis of bacterial and AMR prevalence and, generation of forest plots, functions from R packages were used and meta-regression analysis using the random effect model with the R functions "escalc" and "rma" from R "metafor" package was used to determine sample size on bacterial prevalence. A total of 434 articles were identified and downloaded after a systematic research. The study has implicated that the most common foodborne bacteria in the last five years in Zambia are salmonella spp., E. coli., and L. monocytogens. Based on the random effect model, the prevalence of bacterial pathogens across all studies in food samples was observed to be 11% and in human samples was 14%. The study found a significant increase in antimicrobial resistance (AMR) burden among foodborne pathogens in Zambia compared to other regions of the world over the past five years. This rise is attributed to the bacteria's ability to develop resistance mechanisms and easily spread between humans, animals, and the environment. Ineffective surveillance, inadequate management by stakeholders, and public unawareness have further exacerbated the problem, requiring effective policy implementations in the health sector.

Keywords: Antimicrobial drugs; Antimicrobial resistance; Foodborne bacteria; Meta-analysis.

Fig. 1

Diagram illustrating the selection process of research articles using PRISMA

Fig. 2

A funnel plot for food (a) and human (b) samples

Fig. 3

Proportion of publications according to type of sample (Food or Human), countries and year of publication. The proportion of publications are estimated out of the total number of publications from respective sample types (Food: n= 20; Human: n= 11)

Fig. 4

Prevalence of foodborne bacteria across all studies. A: Prevalence of bacteria in food samples across all the studies (n: total number of publications in food samples; Total: sample size from the publications where prevalence data was extracted). B: Prevalence of bacteria in human samples across all the studies (n: total number of publications in human samples; Total: sample size from the publications where prevalence data was extracted)

Fig. 5

Prevalence of bacteria according to countries across all the studies. A. Prevalence of bacteria in food samples according to countries across all the studies (n: total number of publications in food samples; Total: sample size from the publications where prevalence data was extracted). B. Prevalence of bacteria in human samples according to countries across all the studies (n: total number of publications in human samples; Total: sample size from the publications where prevalence data was extracted)

Fig. 6

Regression plots showing impact of sample size bacterial proportion in all bacterial irrespective of countries in food samples

Fig. 7

Non-significant (p ≥ 0.05) differences of AMR prevalence of pathogens to different antibiotics from pooled studies in food samples

Fig. 8

Significant differences (p ≤ 0.05) of AMR prevalence of pathogens to different antibiotics from pooled studies in food samples

Fig. 9

The AMR prevalence of pathogens to different antibiotics from pooled studies in food samples

Fig. 10

The AMR prevalence of pathogens to different antibiotics from pooled studies in human samples with no significant differences in heterogeneity

Fig. 11

The AMR prevalence of pathogens to different antibiotics from pooled studies in human samples with significant differences in heterogeneity