The study was conducted following the guidelines established in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)(16). The study populations included Salmonella isolates from production animals such as cattle, poultry, pigs, meat products from these animals, and rodents. The primary outcome of interest was the reported frequencies of phenotypic and genotypic resistance to FQ: non-susceptibility to nalidixic acid (Nal ns), non-susceptibility to ciprofloxacin (Cip ns), non-susceptibility to enrofloxacin (Enr ns), frequency of mutations in QRDR genes, and the presence or absence of PMQR genes. The secondary outcomes included MDR, reported serotypes and sequence types (ST), amino acid substitutions in mutated genes, and the phenotypic and genotypic techniques used. MDR was defined as resistance to three or more drugs.

A literature search was conducted in English, Spanish, Portuguese, and French using Boolean logic tools with the operators “AND” and “OR” to search for relevant articles in the PubMed, Scopus, SciELO, and Latindex databases. The search aimed to identify pertinent articles published from January 1, 2020, to August 24, 2024. The search string that allowed for the identification of most studies was as follows: Salmonella AND (quinolone OR fluoroquinolone OR ciprofloxacin OR nalidixic acid OR enrofloxacin) AND (livestock OR cattle OR swine OR pig OR poultry OR rodent OR rat OR beef OR chicken OR pork OR meat) AND (Peru OR Brazil OR Colombia OR Ecuador OR Chile OR Venezuela OR Argentina OR Uruguay OR Bolivia OR Guyana OR Paraguay OR French Guiana OR Suriname). The search was conducted on August 24, 2024. Additional articles were also included, manually located in the Scopus, PubMed, SciELO, and Latindex databases. Additional articles were also included through manual searches of reference lists from selected studies and relevant journals.

The study selection was carried out by two independent reviewers (SBV and MIG), and the references were exported to the Rayyan online application software for screening and selection. In the first phase of review (screening), titles and abstracts were evaluated to identify studies related to the primary outcomes of interest. At this stage, exclusion criteria were applied to discard studies whose titles and/or abstracts were not relevant. Discrepancies between reviewers were resolved through discussion or, if necessary, by consulting a third reviewer.

In the second phase (eligibility), a full-text review of the selected articles was conducted, with detailed assessment based on the following eligibility criteria: (i) publication in English, Spanish, Portuguese, or French; (ii) inclusion of phenotypic and/or genotypic determinants of fluoroquinolone resistance; (iii) isolation of Salmonella from production animals, meat products, or rodents; (iv) exclusion of incomplete or unclear studies; (v) exclusion of studies conducted outside South America; and (vi) inclusion of studies published before August 24, 2024.

Duplicate references were identified and removed using EndNote software prior to the screening process. Additionally, data extraction was performed by one reviewer and independently validated by a second reviewer to minimize errors or inconsistencies. To assess the risk of bias and methodological quality of the included studies, we used the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for prevalence studies. Two independent reviewers (SBV and MIG) performed the quality assessment, and disagreements were resolved by consensus or by involving a third reviewer. Studies were not excluded based on quality, but the appraisal results were considered when interpreting the findings.

The following data were considered and extracted: (i) study identifier: Title, authors, year of publication, country, species (production animals, meat animals, or rodents); (ii) Methods: sample type, sample size, identification method, antimicrobial susceptibility testing (fluoroquinolones or quinolones), breakpoint/interpretive standard level, and gene detection (phenotype-based/genotype-based); (iii) Results: number of isolates, number of isolates tested for susceptibility, number of MDR strains, number of strains resistant to nalidixic acid (Nal ns), number of strains resistant to ciprofloxacin (Cip ns), number of strains resistant to enrofloxacin (Enr ns), number of strains examined for mutation detection (gyrA, gyrB, parC y parE), number of mutants, mutation positions, substituted amino acids, number of strains examined for plasmid-mediated quinolone resistance genes (PMQR) (qnrA, qnrB, qnrC, qnrD, qnrS, aac(6′)-Ib-cr, qepA, oqxA/B) and number of strains with PMQR genes.

Data were analyzed using descriptive statistics in Microsoft Excel (Microsoft 365®). Additionally, GIS software (QGIS 3.16.15) was used to generate maps illustrating the distribution of resistance patterns across South America.

Inferential statistical analyses were not applied in this study due to the high heterogeneity among the included articles in terms of study design, sample sizes, animal species, sampling matrices, and laboratory methodologies. As a result, quantitative synthesis through meta-analysis was not feasible. A descriptive approach was used instead, in line with the exploratory nature of this review.

Of the 27 studies, 88% (n = 24/27) employed phenotypic methods to detect fluoroquinolone resistance in Salmonella isolates the most commonly method used to assess AMR was disk diffusion (17), in 75% (n = 18/24) of the studies, followed by broth microdilution (18) in 25% (n = 6/24) of the studies, according to CLSI guidelines. The most studied fluoroquinolone was ciprofloxacin, present in 91.6% (n = 22/24) of the studies, with an overall resistance of 32.53%. It was followed by nalidixic acid, studied in 15 of the 24 studies, which showed an overall resistance of 60.6%. In contrast, enrofloxacin was the least evaluated, being analyzed in 10 of the 24 studies, with an overall resistance of 23.74%. MDR was assessed in 17 of the 24 studies (n = 17/24), with an overall percentage of 62%.

Of the 27 studies reviewed, 44% (n = 12/27) employed genotypic techniques to detect resistance genes in Salmonella. The most commonly used methodology was WGS, with the Illumina MiSeq platform being the most widely used, employed in 71% (n = 5/7) of the studies that applied Whole Genome Sequencing (WGS), followed by Illumina HiSeq. In comparison, the PCR technique was used in a smaller number of studies, as shown below. In all of the studies that used these techniques, mutations were found in gyrA 50% (n = 6/12), parC 58% (n = 7/12), gyrB 8% (n = 1/12), qnr genes 75% (n = 9/12), and aac(6′)-Ib-cr 8% (n = 1/12).

Phenotypic resistance to quinolones and fluoroquinolones in Salmonella from poultry and swine in South America reveals an alarming trend of antimicrobial resistance that varies between countries and production systems. This is particularly concerning, as fluoroquinolones are classified as “highest priority” by the World Health Organization (WHO) (12), they play a crucial role in the treatment of serious bacterial infections in humans and animals. Their prioritization is due to the fact that, in many cases, they are the only or few therapies available to treat severe non-human-origin infections (24).

Phenotypic resistance in Salmonella from poultry fecal samples is widely documented. In Brazil, resistance to nalidixic acid was reported at 90% in cloacal swabs and 28% in other samples, while ciprofloxacin resistance ranged from 0 to 16% (25, 26). In Colombia, resistance to ciprofloxacin (92.3%), levofloxacin (57%), and enrofloxacin (48.7%) was observed in poultry feces (24). Serotypes such as S. Paratyphi B, S. Minnesota, and S. Heidelberg have been identified, with the latter two showing multidrug resistance (MDR) rates of 64–66% (24–26).

In Chile, Salmonella isolates from chicken meat show 97% resistance to nalidixic acid, with 94% classified as MDR (27). Similar trends are observed in Brazil, where resistance to nalidixic acid ranges from 94.9 to 100%, likely due to the selective pressure from extensive antibiotic use in poultry production (28, 29). In Brazil, chicken carcass isolates also exhibit high FQ resistence, with ciprofloxacin resistance reaching 81.82% and MDR rates at 63.64% (30). However, other studies report significantly lower ciprofloxacin resistance (0.93–2.08%), suggesting possible methodological differences or evolving resistance patterns over time (31, 32).

In Colombia, older studies reported low resistance in poultry farms to levofloxacin, 2.3% in Cundinamarca, 0% in Santander (33), but more recent data show a sharp increase in resistance, likely due to continued antibiotic use (24). Genetic studies highlight Salmonella Heidelberg (ST15) as a major concern due to its high morbidity, resistance, and outbreak potential, posing a significant public health risk (34–37).

Genotypic studies identified key mutations in gyrA and parC, along with the qnrB19 gene, as major contributors to fluoroquinolone resistance (34, 35). Many ciprofloxacin-resistant strains carried both gyrA mutations and the qnrB gene, enhancing resistance (38). A study on S. Heidelberg from Brazilian poultry meat imported to the Netherlands found parC mutations in all 122 isolates, with 96.7% also carrying gyrA mutations (39). Similarly, qnrB19 was detected in S. Minnesota from Brazilian poultry meat (35, 37). The aac(6′)-Ib-cr gene, linked to fluoroquinolone resistance, was found in one S. Heidelberg isolate from Colombian broiler farms, which showed resistance to ciprofloxacin and levofloxacin (24, 40).

Antimicrobial resistance in Salmonella from swine varies across countries, influenced by production practices, antibiotic use, and regulations. In Brazil, de Quadros et al. (63) found that only 16% of Salmonella strains were fully susceptible, yet enrofloxacin inhibited 100% of them. This is notable, as enrofloxacin was widely used before 2017. However, stricter fluoroquinolone regulations in pork production appear to have reduced resistance, indicating a positive impact of recent policies (41).

However, Kich et al. (64) reported high nalidixic acid resistance (45.9%) in carcasses, while studies on mesenteric lymph nodes found resistance rates of 40.6–65.5% and MDR rates of 70.3–79% (42, 43). In Argentina, Vico et al. (66) observed 52% resistance to fluoroquinolones in mesenteric lymph nodes and 86% MDR. Resistance in S. Typhimurium and S. Derby, a key serotype in swine and pork products worldwide, is particularly concerning (44–46). The role of S. Derby, which is one of the most frequently reported serotypes in swine and pork products, both in Brazil and other regions such as the European Union and China (46, 47). In Colombia, Vidal et al. (67) reported 44% MDR in Salmonella from swine feces, with 55% ciprofloxacin resistance, emphasizing the impact of antibiotic overuse in pig farming.

Recent studies reveal distinct resistance patterns between poultry and swine. Pigs show higher resistance to azithromycin, ampicillin, and chloramphenicol, while poultry exhibit greater resistance to quinolones and sulfonamides, reflecting differences in antibiotic use between both industries (48).

Antimicrobial resistance data for cattle is limited. In Uruguay, high non-susceptibility to FQ has been reported, with 77.3% of isolates non-susceptible to ciprofloxacin and 6.6% to enrofloxacin (41). Additionally, 56% of isolates exhibited MDR particularly in serotypes like S. Typhimurium, S. Newport, and S. Anatum are notably prevalent in intensive cattle production, causing both enteric and invasive diseases such as septicemia (49).

Genotypic studies identify ST19 (S. Typhimurium) and its monophasic variant as key serovars in swine, commonly detected in Brazil and the European Union (50–54). S. Typhimurium is the second most common cause of salmonellosis outbreaks in the EU, with pork as the main source in 2014 (55).

The Ser83Phe mutation is frequently reported in pigs strains with reduced susceptibility to CIP, which has been detected in clinical cases in Peru (56). PMQR genes of the qnr alleles, such as the qnrB19 gene, are associated with the spread of Salmonella strains resistant to CIP in the United States (57). The Thr57Ser mutation has been described as relevant for reducing susceptibility to ciprofloxacin (58). The frequent detection of gyrA (position 83) and parC (position 57) mutations highlights their critical role in fluoroquinolone resistance. These findings underscore the need for targeted surveillance and intervention strategies to mitigate the spread of resistant strains.

The antimicrobial resistance observed to quinolones and fluoroquinolones reflects the selective pressure generated by the use of antibiotics in intensive animal production systems. This resistance can vary considerably between different matrices, such as meat, feces and carcass swabs, highlighting the importance of sampling multiple sources within the production system. Although antibiotics remain a crucial tool for treating bacterial diseases, the increase in antimicrobial resistance has reduced their effectiveness (59). Resistance to fluoroquinolones is particularly concerning, as they are commonly used to treat severe human infections, such as systemic salmonellosis in immunocompromised individuals (60).

These findings underscore the urgency of adopting a coordinated One Health approach that integrates surveillance, policy, and intervention strategies across human, animal, and environmental sectors to effectively mitigate the spread of antimicrobial resistance in South America.

This study has several limitations that should be considered when interpreting the results. First, not all South American countries published research on quinolone- and fluoroquinolone-resistant Salmonella between 2020 and 2024, which limits the representativeness of the data for the entire region. In addition, there was considerable variability in the types of samples or matrices used in the included studies—ranging from feces, tissues, food, and in some cases pooled samples from different animal species—making it difficult to isolate Salmonella-specific information.

Although the study initially sought to include data on fluoroquinolone-resistant Salmonella from rodents, no eligible studies were found from South America during the selected period. However, in veterinary medicine, rodents are recognized as important reservoirs and amplifiers of zoonotic infections, including Salmonella. Their absence in the published literature highlights a relevant knowledge gap that should be addressed in future research.

Finally, although a predefined protocol was used to guide the review process, it was not formally registered in a database such as PROSPERO. This omission is acknowledged as a limitation in the transparency of the study methodology.

Future studies should aim to include a broader range of countries and matrices, and explore the role of rodents in the dissemination of antimicrobial-resistant Salmonella.