Introduction

Food animals such as cattle, poultry and pigs have been extensively reared worldwide, not only as a source of food but also a source of income. The modes of productions are intensive [1], due to the rapidly increasing demand for livestock products driven by human population growth and urbanization [2, 3]. This has necessitated the uncontrolled use of antimicrobials [4], which has been associated with increase of antimicrobial resistance [5].

The subsequent contamination of soil, sediments, sludge, groundwater, wastewater, tap and surface water, and plants contribute to the emergence and spread of multi-drug (MDR) organisms in environment [6, 7]. Furthermore, unmonitored quantities of waste that contain antimicrobials generated by pharmaceutical manufacturers, hospitals, and livestock producers promote selection of resistomes in the environment, with potential spill over to animals and humans [8, 9]. Some of important organisms that have been found to circulate in different compartments include Campylobacter spp., Salmonella spp., Staphylococcus spp., Enterococcus spp. and ESBL- producing Enterobacteriaceae [10,11,12].

Although AMR is a global threat [13], the situation in Africa is compounded by a number of factors that include lack of access to appropriate antimicrobial therapy, weak of regulation in the use of antimicrobials for human and animal, weak surveillance systems, lack of updated antimicrobial use and treatment guidelines. Others are lack of continuing education on antimicrobial use (AMU) for prescribers, tendency for animal owners to stock drugs and engaging unskilled people to treat animals, high degree of drugs abuse by livestock keepers and unregulated disposal of waste in dumps [14,15,16,17].

Despite these facts, there is paucity of consolidated information on the antimicrobial use (AMU), antimicrobial resistance (AMR), surveillance and stewardship programmes in Africa. The few available reviews on antimicrobial use and resistance are mostly country specific [18, 19]. This study reviewed the use of antimicrobial agents, prevalence of antimicrobial resistance and status of surveillance systems in food-producing animals and the environment in Africa.

Methods

This review was carried out between October 2018 and April 2019. Pub-Med, Google Scholar, Web of Science, Africa Wide Information and African Journal Online databases were searched for information on AMU and AMR covering a period from 2005 to 2018. Full-length research articles and review papers written in English were considered. In addition, publications from Food and Agriculture Organization (FAO), World Health Organization (WHO), International Livestock Research Institute (ILRI), Office International des Epizooties (OIE) websites were also searched and reviewed.

Combinations of search terms used were ‘antimicrobial usage’, ‘antimicrobial use’, ‘antibiotic use’, ‘antimicrobial resistance’, ‘antimicrobial resistant’, ‘food-producing animals’, ‘food animals’, animal husbandry’, ‘animal farming’, ‘domestic animal farming’, ‘farmed animals’, ‘environment’, ‘environmental’, ‘waste water’, treated waste water’, ‘sea water’, ‘river water’, ‘effluent’, ‘irrigation water’, ‘surface water’, ‘soil’ and ‘vegetables’. Others were specific food animal descriptors such as ‘poultry’, ‘chickens’, ‘pigs’, ‘swine’, ‘cattle’, ‘beef cattle’, ‘dairy cattle’, ‘fish’, specific country by name, and the word ‘Africa’. The articles were scrutinized to extract information on the antimicrobial use, prevalence of AMR and availability of a surveillance system.

Results

A total of 200 articles were recovered, of which 170 were original research articles and 30 were reviews, books, reports and perspectives or policy briefs. On further evaluation, five were removed, of which two reported similar data and three were abstracts whose full length papers could not be accessed. Of the 195 articles assessed, 19 were excluded from review because they were not relevant (Fig. 1), therefore only 176 studies were included in this review.

Fig. 1
figure 1

Flow diagram of number of articles obtained and those eliminated

Full size image

Types of study and geographical distribution

The majority of the studies were cross-sectional in design, with only two being retrospective (one presenting data on AMU and the other on AMR patterns). Only three studies quantified the amount of antimicrobial using the defined daily dose per animal (DDDA). Some studies involved both surveillance of antimicrobial use and laboratory analysis of organisms’ (Table 1). Geographically, most (56%) of the studies were from the northern and southern Africa regions (Fig. 2).

Table 1 Different study designs used in data collection on antimicrobial use and resistance
Full size table
Fig. 2
figure 2

Geographical distribution of AMU and AMR studies in animal and environment

Full size image
Fig. 3
figure 3

Complex interactions involved in the spread antimicrobial resistance between sectors

Full size image

Antimicrobial use in animals

As shown in Table 2, antimicrobial agents were administered to different food animal species, mostly in poultry. The percentage of farms using antimicrobial agents ranged from 77.6% in Nigeria to 100% in Tanzania, Cameroon, Zambia, Ghana and Egypt. In total 14 different classes of antimicrobial agents were used, mainly tetracycline, aminoglycosides and penicillin. Antimicrobial agents such as macrolides, which are restricted for use in animal in some developed countries due to their impact in human health, were used in Africa.

Table 2 Percentage of farms using antimicrobials by country, type of animal and class of antimicrobials
Full size table

Antimicrobials use in the environment

Figure 3 below summarizes the complex interactions between human, animal husbandry, veterinary medicine and the environment with potential for occurrence and spread of AMR resistomes. Such complexity complicates the interventions involved in curbing the burden of AMR.

Antimicrobial resistance in food animals

This review highlights different levels of antimicrobial resistance within and between countries. Proportion of MDR strains among E. coli, which is an indicator organism, is shown in Table 3. The multi-drug resistance is referred as non-susceptibility to more than three antimicrobial categories [54]. The prevalence of MDR E.coli ranged from 20% in Nigeria to 100% in South Africa, Zimbabwe and Tunisia, in most cases being above 80%. Of the 14 different classes of antimicrobial agents, higher prevalence of resistance was reported for tetracycline, sulphonamides and penicillin, which are the cheapest and therefore more widely used antibiotics.

Table 3 Proportion of MDR strains among Escherichia coli isolated from food animals
Full size table

Antimicrobial resistance in environment

In this review only a few studies reported AMR in the environment. These studies involved samples collected from domestic and biomedical waste, waste water, river sediments, surface and drinking water, treated waste water, river water and vegetables (Table 4). The prevalence of MDR E. coli in environmental samples ranged from 33.3% in South Africa to 100% in Algeria and South Africa. In total E. coli exhibited resistance to 16 different antimicrobial agents.

Table 4 Percentage of MDR Escherichia coli from environmental samples
Full size table

Genetic relatedness of resistant genes in animals, humans and environment

We identified a few studies that had determined AMR in all three compartments, namely animals, humans and the environment. Although most of the studies involved phenotypic methods, few involved molecular characterizations as well. The most frequently used molecular methods were Polymerase Chain Reaction (PCR), Pulse Gel Field Electrophoresis (PGFE), and Multilocus Sequence Typing (MLST). A summary of AMR genes in the three compartments is shown in Table 5. Some genes such as blaCTX-M, blaTEM, blaSHV, blaOXA, aac(6′)- Ib-cr, tet(A), tet(B), sul1, sul2, and qnr were commonly detected in all of the three compartments, indicating a potential flow of AMR genes across humans, animals and environment [64,65,66,67].

Table 5 A summary of resistant genes detected in humans, animals and environment
Full size table

Surveillance systems

Our review revealed that there were no country with a documented national AMU and AMR surveillance programme specific for either animals or environment. Most countries rely on point-prevalence rather than nation-wide surveillance programmes [19, 77]. Some countries have developed a national antimicrobial plan, as an initial stage towards developing AMU and AMR surveillance system in human, animal and environment. Fifteen countries have approved National action Plan for Antimicrobial resistance (NAP), while eight are waiting for the authority’s approval or they are heading towards finalization [78]. It was noted that there is support on capacity to develop surveillance and monitoring of AMU and AMR in food and agriculture provided by the World Health Organization in collaboration with Food and Agriculture Organization and World Organization for Animal Health (OIE) [79].

Other forms of supports included: (i) value chain analysis in animal health from farm to retail step, focusing on microbiology analysis [80]; (ii) training, technical support, strategic guidance and software tools for Web-based data entry and collaboration for the studies targeting animals, environment and other sectors [81]; (iii) training of laboratory technicians from human, agriculture and veterinary sectors from eight countries on laboratory surveillance and control of major food borne diseases [82]; and (iv) research projects on integrated surveillance of AMR in foodborne bacteria for Chad, Tanzania and Ethiopia [82]. Recently, the Africa Centre for Disease Control has launched a network called Antimicrobial Resistance and Surveillance Network comprising experts from animal, environmental and human health sectors. The network targets to mitigate harm from antimicrobial resistant organisms arising from animals, environment, agriculture and human [83]. According to WHO [82] other initiatives provided to countries in Africa included studies on ESBL-producing E.coli in animals and the environment organized in Egypt, Morocco and Sudan.

Discussion

This review involved studies with different designs in evaluation of AMU and AMR. Such variations, especially in sampling designs, complicate comparison of the reported data, and may impact on the reliability of data [32, 78]. There are also potential challenges such as: financial constraints, skilled human resource, lack of research facilities and lack of awareness on the role of animals and environment in the spread of antimicrobial resistances that differ within and between countries. As a consequence, most of the reported studies are from northern Africa, where the issues of AMU and AMR could be less critical than in other parts of Africa.

The review has revealed that more than 80% of farms use antimicrobial agents in animal production, driven by the increase in demand for food of animal origin and emerging trade opportunities [2, 84]. Some of the antimicrobial agents used such as macrolides that have been restricted for use elsewhere due health risk concerns [33, 35] were commonly used in Africa. We found higher AMR prevalence in tetracycline, aminoglycosides and penicillin, these are the cheapest antibiotics hence widely used. The challenge is that most farmers cannot afford alternative, relatively more expensive drugs. This coupled with low awareness on AMU and AMR is likely to further exacerbate the burden of AMR in the region. This review has revealed high prevalence of AMR, including MDR bacteria in the environment, probably associated with anthropogenic activities, application of manure in farming activities and agricultural wastes from community [61, 69, 85, 86]. There is potential spillage of resistant isolates from sewage, humans, companion and domestic animals and industries to the environment [69, 87].

A few studies compared AMR genes in humans, animals and environment and detected blaCTX-M, blaTEM, blaSHV, blaOXA, aac(6′)- Ib-cr, tet(A), tet(B), sul1, sul2, and qnr in all the three compartments [64,65,66,67] This finding seems to suggest that these MDR pathogens have high propensity to spread widely and cause infections that are difficult treat [18, 68, 88]. Due to limited laboratory capacity in most of the countries the identified AMR genes could just be a tip of iceberg, representing only a fraction of transmitted genes. This calls for more support to these countries to establish systematic AMR surveillance using more advanced techniques such as Whole Genome Sequencing for better understanding the magnitude, spread and evolution of MDR pathogens [67, 89]. Such information is critical in planning effective interventional measures.

The findings indicate that currently most countries in Africa do not have AMU and AMR surveillance systems and are at different stage of developing them. It may takes a long while before implementation. The different approaches used in different countries leads to inconsistency results which are difficult to compare at national, regional and international perspectives [86]. The results reported do not conform to the international reporting systems since there is no appropriate system for data collection, identifications, coordination and reporting [90]. It is important for countries in Africa to adopt internationally recommended surveillance systems that will take into account data collected from animal, agriculture, environment and human sectors. This review suggests that African countries should seek support from global AMR network to be assisted to develop and implement AMR surveillance under one health approach.

It is important to highlight a number of limitations that were uncovered during the review. First of all, information on AMU is based on percentage of farms using antimicrobial agents rather than the Defined Daily Doses Animals as recommended by the World Health Organization. Furthermore, most of the AMR studies were based on phenotypic rather than molecular techniques, thus limiting understanding of transmission dynamics. In addition, lack of standardized laboratory protocol may account for variations in the level of AMR reported within and between countries. More importantly, there is limited One Health approach surveillance of AMR; hence different sectors employ different approaches in monitoring AMU and AMR. Finally, weak enforcement of the available regulations leads to unmonitored production, distribution, handling, storage and sale of veterinary drugs. It is important that these limitations are taken into consideration as countries in the region continue to plan, develop, refine and implement their National One Health AMR plans.

Conclusion

There are very high levels of antimicrobial use and antimicrobial resistance, especially for tetracycline, aminoglycoside and penicillin in animal production systems in Africa. This is likely to escalate the already high prevalence of antimicrobial resistance and multi drug resistance in the continent. This, coupled with weak regulations and antimicrobial resistance surveillance systems in the region is a great concern to the animals, environment and humans as well. It is important that African countries strengthen their respective AMU and AMR regulations and surveillance systems to address the challenges identified in this review.