Antimicrobial Resistance in Bacteria Isolated from Foods in Cuba

INTRODUCTION Antimicrobial drug resistance constitutes a health risk of increasing concern worldwide. One of the most common avenues for the acquisition of clinically-relevant antimicrobial resistance can be traced back to the food supply, where resistance is acquired through the ingestion of antimicrobial resistant microorganisms pres-ent in food. Antimicrobial resistance constitutes a health risk, leading to production losses and negative consequences for livelihood and food safety. OBJECTIVE Determine whether resistant bacteria are present in foods in Cuba. METHODS A descriptive observational study was conducted in the Microbiology Laboratory of Cuba’s National Institute of Hygiene, Epi-demiology and Microbiology from September 2004 through December 2018. Researchers analyzed 1178 bacterial isolates from food samples. The isolates were identi ﬁ ed as Escherichia coli, Salmonella, Vibrio cholerae and coagulase-positive Staphylococcus . The antimicrobial susceptibility study was performed using the Bauer-Kirby disk diffusion method, following procedures outlined by the Clinical and Laboratory Standards Institute. The data were analyzed using WHO-NET version 5.6. RESULTS Of the total isolates, 62.1% were resistant to at least one antibiotic. Within each group, >50% of isolates showed some type of resistance. E. coli and V. cholerae exceeded 50% resistance to tetracycline and ampicillin, respectively. Staphylococcus showed the highest resistance to penicillin, and Salmonella to tetracycline, nali-dixic acid and ampicillin. The highest percentages of non-susceptible microorganisms were identi ﬁ ed in meats and meat products. CONCLUSIONS These results serve as an alert to the dangers of acquiring antibiotic-resistant bacteria from food and demonstrate the need to establish a surveillance system and


INTRODUCTION
Antimicrobial resistance (AMR) is a health risk worldwide, leading to production losses and negative effects on livelihood, food safety and the economy, [1] including in Cuba.Statistics from the national program for prevention and control of healthcare-associated infections show an increase in resistance to the most commonly used hospital antibiotics in the last few years, as well as longer hospitalizations and higher spending on these infections.[2] The public health sector is acting to promote the rational prescription and use of antimicrobials, and is conducting various susceptibility studies on clinically-obtained isolates.[3] However, there are few reports on antimicrobial-resistant foodborne bacteria.
Quantitatively, foodborne AMR is the most common route for the spread of antibiotic-resistant bacteria.The presence of these microorganisms in the food chain, the environment and water can lead to their appearance in the human intestinal microbiome, turning it into a major reservoir for resistant genes in the body.It also increases the risk of their dissemination among commensal bacteria and pathogens that cause intra-and extraintestinal infections.[4] Among the most clinically important foodborne pathogenic bacteria in AMR are strains of Salmonella and E. coli, which carry extended-spectrum beta lactamases, fl uoroquinoloneresistant Campylobacter and Salmonella, and methicillin-resistant Staphylococcus aureus.[5] However, commensal bacteria also found in foods play a key role in AMR evolution and spread.
They predominate in the environment and show greater genetic diversity and host variety in nature, which makes them a potential indicator for AMR.Thus, studying these agents can provide early warning of emerging AMR.[6] WHO suggests regular, periodic surveillance to address the problem of AMR, with permanent monitoring of changes in its prevalence in humans, animals, foods and the environment.[7] Clearly, it is important to discover foodborne AMR as quickly as possible.This includes studying risks by identifying dangers: antimicrobial-resistant microorganisms, the antimicrobials to which they are resistant, and the food products in which this resistance is found.Cuba has no program dedicated to ongoing surveillance of this problem.For these reasons, this study was performed with the aim of assessing antimicrobial resistance in clinically relevant bacteria isolated from foods in Cuba.

METHODS
A descriptive observational study was conducted from September 2004 through December 2018 on 1178 isolates identifi ed in foods (381 isolates of E. coli, 402 of Salmonella, 113 of V. cholerae and 282 of coagulase-positive Staphylococcus).The isolates were performed at the Provincial Hygiene, Epidemiology and Microbiology Centers in 13 Cuban provinces and in the Microbiology Laboratory of the National Hygiene, Epidemiology and Microbiology Institute (INHEM) in Havana, following current standards in Cuba.[8][9][10][11] The microorganisms were identifi ed in a variety of 146 foods subject to microbiological surveillance in the study of foodborne disease outbreaks and health inspections of foods before sale.These were categorized in 14 groups, according to Cuban microbiological criteria standard NC 585, 2017.[12] The food types were: IMPORTANCE This paper highlights the importance of antimicrobial resistance surveillance in foods commonly consumed in Cuba.
Results were interpreted following the manufacturer's criteria.E. coli ATCC 25922 strains were tested as a negative control, with ESβL Klebsiella pneumoniae ATCC 700603 strains tested as a positive control.
Results were analyzed using a database created in WHONET version 5.6, a WHO digital platform for surveillance of antimicrobial resistance and infection control.[14] The antibiogram interpretation criteria cutoff points were updated according to CLSI standards.Susceptibility was analyzed by isolate source, for which contingency tables were established, and the chi-square test was applied with a signifi cance level of 0.05%.The data were processed using the EPIDAT program (EpiData Association, Denmark) for epidemiological analysis of tabular data, version 3.0 of 2004.[15] Results of the in vitro susceptibility tests were expressed as absolute frequencies and percentages.Isolates with full growth around the antibiotic disk or those in which growth inhibition did not reach the diameter established for the CLSI susceptibility criterion (reduced susceptibility) were considered resistant.Otherwise, they were considered sensitive to the antibiotic.

Ethical considerations
No clinical assays were performed on persons or animals in this study, and the study was authorized by INHEM's scientifi c council.This document contains no company, institution or brand names of foods from which the isolates were obtained.

RESULTS
AMR was analyzed according to the microorganisms retrieved from different food types (Table 1 V. cholerae was isolated in fruits and vegetables, and in fi sh, seafood and fi shery products, which had the highest percentage of resistant isolates at 69.3%.

Original Research Peer Reviewed
Table 2 shows the relation between AMR in Salmonella, E. coli and Staphylococcus and their isolate sources.Salmonella was not associated with any specifi c food type.The highest percentage of resistant isolates was found in meats and meat products.E. coli had a higher proportion of resistant isolates compared to subgroup size in meats and meat products.Additionally, Staphylococcus had a higher proportion of resistant isolates found in meat and dairy products.
Resistance by antibiotic type was low overall, except for tetracycline in E. coli and ampicillin in V. cholerae, for which resistance was over 50% (Table 3).Of the 19 antibiotic agents analyzed (14 for Salmonella and E. coli, 12 for Staphylococcus and 6 for V. cholerae) Salmonella expressed in vitro resistance to 12, and E. coli, to 14. Tetracycline, nalidixic acid and ampicillin showed the highest resistance levels.More than 75% of Staphylococcus isolates were resistant, mainly against penicillin, erythromicin and tetracycline, in decreasing order.V. cholerae was resistant to three antibiotics, namely tetracycline,

DISCUSSION
More than half of the bacterial isolates recovered from foods were resistant to at least one of the drugs tested.The most clinically important isolates were E. coli and Salmonella, since they often cause gastrointestinal disease or extraintestinal infections requiring treatment.The least effective antibiotics administered in vitro were tetracycline, ampicillin, nalidixic acid and penicillin, as also found in international studies.[16][17][18][19][20] For WHO-classifi ed antibiotics, [18] specifi cally those appropriate for only limited use in humans (including ciprofl oxacin, cefotaxime, ceftriaxone and ceftazidime), resistance was low and observed more often in E. coli and Staphylococcus.The international literature reports resistance percentages higher than those in this study.[19][20][21] The foods that most often contained resistant isolates were meats and meat products; for Salmonella, this result is consistent with those of other researchers, which show that these products are among the main sources of resistant bacteria in this genus.[22,23] The 173 Salmonella isolates from meats and meat products were obtained from 31 different foods.Hamburger showed the highest number of resistant isolates.Among fresh meats, resistance was most often found in poultry, where isolates from ground turkey were predominant, followed by those from ground chicken and mechanically deboned meat.These results agree with international reports, which found that in ground meats, the Salmonella detected often presents with high virulence and high levels of AMR.[24,25] Since most poultry meats in Cuba are imported, [26] this could be considered a route for spreading resistance, in addition to antibiotics found in imported meat that are not used in domestic animal production, such as cefotaxime, ceftriaxone and ceftazidime.
Resistant E. coli isolates were most often found in pork, mortadella and smoked pork loin.Three isolates carrying ESβL were found in imported poultry meat and beef, and in domestically produced pork, at a lower percentage than has been reported in other countries.[27,28] Globally, antimicrobial susceptibility of E. coli is studied in different foods depending on geographic region.In the European Union and the United States, emphasis is on meats and antibiotics such as cephalosporins and fl uoroquinolones.[29,30] In Asia and Latin America, there are more studies on ready-to-eat foods.[31,32] This could be due to greater availability of industrially processed ready-to-eat foods in developed countries, while in developing nations there are more prepared foods sold by small-scale manufacturers who generally do not monitor product preparation, potentially allowing bacterial contaminants to survive and multiply.In this study, which analyzed meats and ready-to-eat foods, antibiotic resistance was frequent regardless of food type.
Currently, AMR in commensal bacteria such as E. coli is cause for growing concern because resistant genes can be replaced with bacteria that are pathogenic to humans.The scientifi c literature has demonstrated transfer of multidrug resistance through E. coli plasmids to other enterobacteria such as Salmonella.[33] Most antibiotic-resistant Staphylococcus isolates were identifi ed in meats and meat products such as sausages, ground meats and hamburger.In milk and dairy products, most isolates were found in cheese, mainly artisanal cheeses.This last food group was shown to be associated with resistant isolates.Other countries report varying percentages of AMR to at least one of the antibiotics tested, among which S. aureus was the most prevalent in meats and cheeses.[21,34] It should be noted that foodborne staphylococcal intoxication does not require antibiotic treatment, and there is no evidence that consuming foods contaminated with this bacteria is associated with infection in humans.[35] However, there is now special interest in antimicrobial susceptibility studies because of the possible transfer of resistant genes between microorganisms, and thus from the environment to humans.[7] V. cholerae is a species endemic to aquatic environments, and thus may be an indicator of antibiotic resistance in bacteria found in these ecosystems.In this study, it was mainly found in fi sh, seafood and other fi sh products.Its expressed resistance was low except to ampicillin, to which resistance was seen in >50% of isolates.No resistance was found to ciprofl oxacin, azithromycin or doxycycline, which are often used as fi rst-line treatments for infections of toxigenic agents of this species.For V. cholerae, the international literature reports AMR usually higher than that found in this study.[36,37] The highest percentage of isolates analyzed came from foods inspected at INHEM as part of the institution's responsibilities in sanitary registration including imported products and those domestically produced by various Cuban companies.Foods that do not meet the bacterial limits in the standard [11] are not approved for sale.However, there are currently no trade regulations that address antibacterial resistance, which is why studies focusing on risk are needed to accurately determine the scope of the problem.[38] We observed an unequal distribution in both the number and geographic origin of isolates received from laboratories in other provinces participating in the study, as well as in numbers of isolates of each bacteria type received.There were low percentages of E. coli, Staphylococcus and V. cholerae, which made it impossible to analyze antibiotic resistance for each region of the country.This would be possible if a national antimicrobial resistance surveillance system were established to obtain standardized information that would allow comparisons by region and over time.
One of the study's main limitations was the unequal numbers of bacterial isolates sent from each province.The study was based on the isolates received, which did not allow nationally based analysis of a resistant bacterial load for each food.In addition, the information presented was obtained more than a year ago, which makes it invalid for immediate surveillance purposes, but does not affect its usefulness as a resource for illustrating a problem that demands surveillance and control.Despite these limitations, a broad range of antibiotics were analyzed, including most classes used in human and veterinary treatment, and the number of isolates studied for each bacterial genus was suffi cient for making preliminary estimates of AMR prevalence in each case, although without claims as to their representativity.

CONCLUSIONS
Resistant phenotypes were identifi ed in more than half the bacteria isolated from foods, with a higher percentage found in animal products such as meat, dairy, eggs and foods made from these ingredients.Low percentages of AMR were found for antibiotics classifi ed as critical for human use.These results may serve as an alert to the dangers of acquiring foodborne antibiotic-resistant bacteria and demonstrate the need to establish a surveillance system and institute related control in Cuba.

Table 2 : Relation between antibiotic resistance of Escherichia coli, Salmonella and Staphylococcus and food type from which isolates were recovered (n = 1065). INHEM 2004-2018 Susceptibility
a Percentage refers to total number of isolates in category b Percentage refers to total number of foods analyzed per microorganism AMR: Antimicrobial resistance INHEM: National Institute of Hygiene, Epidemiology and Microbiology a Percentage refers to total number of isolates in category b Percentage refers to total number of foods analyzed per microorganism INHEM: National Institute of Hygiene, Epidemiology and Microbiology

Table 3 : Percentage of resistance by antibiotic and microorganism. INHEM 2004-2018 Antibiotic Salmonella n = 236 E. coli n = 220
INHEM: National Institute of Hygiene, Epidemiology and MicrobiologyOriginal ResearchPeer Reviewed ampicillin and sulfamethoxazole/trimethoprim (Table3).A low percentage (2.8%) of ESβL enzyme was detected in 97 E. coli isolates obtained from fresh meats.Geographical distribution of isolates (Table4) showed that the highest percentage, 52.7% of the total, was identifi ed in Havana Province at INHEM's laboratory.The percentage of isolates sent from provinces outside Havana was low.The highest percentage came from Santiago de Cuba (11.0%); the rest were less than 10.0%.

Table 4 : Isolates studied, by microorganism and province where identifi ed. INHEM 2004-2018
Percentage refers to total number of isolates for province, b Percentage refers to total number of isolates INHEM: National Institute of Hygiene, Epidemiology and Microbiology * Special Municipality a