Antibiotic Resistance of Escherichia coli Strains Isolated From Pet Parrots in Hanoi, Vietnam

Birds are the third most popular pets, after dogs and cats, and are close companions to humans, and their numbers are rising worldwide and they play an important role in human life (Pomba et al., 2017). The majority of captive birds belong to two orders: Passeriformes, which includes canaries and sparrows, and Psittaciformes, which includes parrots, parakeets, and lovebirds (Boseret et al., 2013). However, many recent studies have shown that companion animals can carry and spread zoonotic bacteria, including E. coli (Day, 2016; Boroomand and Faryabi, 2020). E. coli is also one of the most common pathogens, responsible for a range of diseases such as colibacillosis, respiratory infections and septicemia in birds (Machado et al., 2018), as well as urinary tract infections, diarrhea and blood infections in humans (Karpman and Ståhl, 2014; Pakbin et al., 2021), and is considered a source of bacteria containing genes associated with antibiotic resistance (Machado et al., 2018; Prestinaci et al., 2015). E. coli can easily adapt to different environments, which helps it develop many drug resistance mechanisms, and even harmless E. coli can spread antibiotic resistance genes in bacteria (Szmolka and Nagy, 2013).

Antibiotic resistance is a One Health problem, affecting humans, animals, and the environment (Argudín et al., 2017). Antibiotic resistance keeps getting worse and, unfortunately, has serious health and economic effects around the world. Numerous studies have shown that drug-resistant bacteria and/or bacteria carrying antibiotic resistance genes can be transmitted from animals to humans and vice versa, through contaminated food, the environment, or through contact (Damborg et al., 2016; Argudín et al., 2017; Pompa et al., 2017). Close contact between people and their pet birds creates many chances for bacteria to spread, including of antibiotic-resistant bacteria (Machado et al., 2018; Rahman et al., 2020; Hosseinian, 2022). Furthermore, owners often treat sick birds themselves, without veterinary advice, which can promote antibiotic resistance (Giacopello et al., 2015).

In recent years, keeping pet birds, including parrots, has become quite popular in Vietnam. However, few studies have looked at diseases in pet birds or the risk of them spreading diseases to people. Specifically, to our knowledge, there are currently no studies have examined antibiotic resistance in bacteria from pet birds in Vietnam. Therefore, investigating the prevalence and antibiotic resistance of E. coli strains isolated from pet parrots in this study will provide additional data on the antibiotic resistance situation in livestock in Vietnam.

In this study, we collected 89 fecal samples from healthy parrots kept as pets in people’s homes in Hanoi, Vietnam, from June 2025 to April 2026. The samples were collected following Vietnam’s national technical regulation on animal diseases - General requirements for sample collection, storage and shipment of the Ministry of Agriculture and Rural Development (2011). Briefly, fresh fecal samples from the cages were carefully collected by using sterile spoons, each sample in its own sterile bag, labeled, stored in a cooler with ice, and taken immediately to the laboratory of Department of Veterinary microbiology and Infectious diseases, Faculty of Veterinary Medicine, Vietnam National University of Agriculture for analysis within 24 hours.

At the laboratory, about one gram of each fecal sample was mixed with buffered peptone water (BPW, Merck, Germany) at a 1:9 ratio. Next, a loopful of the mixed culture was streaked onto MacConkey agar (Merck, Germany) and incubated at 37°C for 24 hours. Then, the pink colonies (Figure 1) were cultured onto eosin methylene blue agar (EMB, Merck, Germany) and continuously incubated at 37°C for 24 hours. Only one colony with the typical metallic green sheen on EMB agar (Figure 2) were streaked into triple sugar iron agar (TSI, Merck, Germany) tube and incubated at 37°C for 24 hours. Colonies exhibiting a typical TSI profile, including glucose and lactose fermentation with gas production and no H2S (Figure 3), were identified as E. coli by Gram staining and biochemical tests, such as indole production, methyl red, Voges-Proskauer, and citrate utilisation reactions, with the expected IMViC result being “++--” (Figure 4). We stored all isolates in brain heart infusion broth with 50% glycerol at -20°C for subsequent experiments.

DNA was extracted using the TopPURE® Genomic DNA Extraction Kit (ABT, Vietnam) according to the manufacturer’s instructions. We used specific primers (Malinen et al., 2003) with the following sequences based on the 16S gene for E. coli (5’-GTTAATACCTTTGCTCATTGA-3’ and 5’-ACCAGGGTATCTAATCCTGTT-3’) with an expected PCR product of 340 bp were used (Figure 5). PCR cycling conditions were initial denaturation at 94 ºC for 5 min followed by 35 cycles of denaturation at 94 ºC for 30 sec, annealing at 60 ºC for 30 sec and extension at 72 ºC for 45 sec followed by final extension at 72 ºC for 10 min. The reaction components included 12.5 µl of GoTaq® Green Master Mix (Promega, USA), 1 µl each of the forward and reverse primers (10 µM), 8.5 µl of purified water, and 2 µl of template DNA. The PCR products were electrophoresed on 1.5% agarose gel supplemented with RedSafe™ nucleic acid staining solution (Intron, Korea).

Antibiotic susceptibility test was examined according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI, 2020). Agar diffusion method was performed on Mueller-Hinton agar (MHA, Merck, Germany) following Bauer et al. (1966) and 10 different antibiotic agents (Nam Khoa, Vietnam) belonging to seven classes were used (Figure 6), including penicillins (amoxicillin, 10 µg), carbapenems (meropenem, 10 µg), cephalosporins (cefotaxime, 30 µg), aminoglycosides (kanamycin, 30 µg; neomycin, 30 µg; streptomycin, 10 µg), quinolones (nalidixic acid, 30 µg), fluoroquinolones (levofloxacin, 5 µg; norfloxacin, 10 µg), sulfonamides (trimethoprim/sulfamethoxazole, 1.25/23.75 µg). The inhibition zone diameters of the ten antibiotics for E. coli are shown in Table 1. Escherichia coli ATCC 25922 strain was used for quality control. An isolate was determined to be antibiotic-resistant or multidrug-resistant (MDR) based on the definition of Magiorakos et al. (2012).

Table 1: CLSI breakpoints (inhibition zone diameters) for ten antibiotic tested.

The isolation and antibiotic resistance rates of E. coli strains were recorded and calculated using Microsoft Excel 2016. The 95% confidence intervals (95% CI) for the proportions were estimated using the Clopper-Pearson binomial distribution (Exact Binomial Test) with R software.

Overall, the rate of E. coli isolation from fecal samples was 59.6% (Table 2). Of these, the fecal samples from Red-breasted parakeet samples had the highest rate (75.0%), followed by Monk parakeet (70.0%), Macaw parrot (66.7%). Isolation rates for other species ranged from 50.0% to 58.8% in fecal samples from other parrot species such as Lovebirds, Sun parakeet, and African grey parrot.

The isolates showed resistance to each of the 10 antibiotics belonging to 7 different antibiotic classes, but at different rates (Table 3). Resistance was highest to amoxicillin (94.3%), followed by trimethoprim/sulfamethoxazole (50.9%) and streptomycin (47.2%). Resistance to norfloxacin, levofloxacin, and nalidixic acid was 35.8%, 41.5%, and 45.3%, respectively. Cefotaxime, kanamycin and neomycin were resistant at rates ranging from 30.2-34.0%. Notably, three (5.7%) strains were resistant to meropenem, a last-resort antibiotic for serious human infections.

Table 2: Isolation rates of E. coli from fecal samples collected from pet parrots (n = 89).

* Note: Other species include Alexandrine parakeet (Psittacula eupatria, n = 4); Amazon parrot (Amazona ochrocephala, n=5); Budgie (Melopsittacus undulatus, n=6); Cockatiel (Nymphicus hollandicus, n =4).

Table 3: Antibiotic susceptibility of E. coli strains isolated from pet parrot fecal samples (n = 53).

Table 4: Antibiotic resistance phenotypes of the isolated E. coli strains (n = 53).

AMX: Amoxicillin; CTX: Cefotaxime; KAN: Kanamycin; LEV: Levofloxacin; MEM: Meropenem; NEO: Neomycin; NAL: Nalidixic acid; NOR: Norfloxacin; STM: Streptomycin; SXT: Trimethoprim/Sulfamethoxazole.

Only one of the 53 isolates was susceptible to all the antibiotics tested (Table 4). The remaining 52 (98.1%) strains were resistant to at least one antibiotic and exhibited 31 different resistance phenotypes. The most common pattern was resistance to amoxicillin alone (10 strains, 18.9%), followed by AMX-CTX-KAN-NEO-STM-NAL-LEV-NOR-SXT resistance was found in 5 (9.43%) strains isolated from African grey parrots and Red-breasted parrots. The resistance rates for 1 and 2 antibiotic classes were 15.1% and 20.8%, respectively (Table 5). Meanwhile, the resistance rates for 3-6 antibiotic classes ranged from 13.2% to 17.0%. Only one strain (1.9%) was resistant to seven classes. Of these, 33 isolates (62.3%) were resistant to three or more antibiotic classes, meeting the definition of multidrug-resistant (MDR).

Table 5: Resistance to different antibiotic classes of isolated E. coli strains (n= 53).

*Note: “No resistance” means susceptible to all 10 antibiotics tested. The number of strains resistant to three or more antibiotic classes is 33 (62.3%).

Previous studies conducted in Bangladesh (Hasib et al., 2025) and Turkey (Diren Sigirci et al., 2020) reported isolation rates of 48.7% in 150 and 52.3% in 172 parrots and parakeets samples, respectively. However, lower isolation rates, ranging from 18.3% to 36.0%, were observed in studies conducted in Bangladesh (Nupur et al., 2023), Italy (Varriale et al., 2020), Brazil (Marques et al., 2024), and Egypt (Samir et al., 2025). These differences could be due to geography, sampling methods, or lab techniques. Additionally, sanitary conditions in the cages, along with the hot and humid climate in northern Vietnam, may be related to the high rate of E. coli isolates in fecal samples from parrot species.

We found that 52 of 53 isolates (98.1%) were resistant to at least one antibiotic, and the most common resistance was identified as amoxicillin (94.3%). High amoxicillin resistance (81-100%) has also been seen in other studies in Brazil (Pontes et al., 2018), Italy (Varriale et al., 2020) and Bangladesh (Nupur et al., 2023). Cefotaxime is a critically important antibiotic for human medicine (WHO, 2017), but resistance was observed in 30.2% of the E. coli strains in present study. One study in Egypt found even higher resistance (66.7%) in parrots with respiratory illness (Samir et al., 2025). Meropenem is a last-resort antibiotic for serious human infections (WHO, 2017). Similar studies conducted in Brazil (Marques et al., 2024), Egypt (Samir et al., 2025), and Bangladesh (Hasib et al., 2025) all showed that all isolated E. coli strains were susceptible to this antibiotic. Although only three strains (5.7%) were resistant to meropenem in this study, this is concerning because meropenem is not approved for use in animals.

Trimethoprim/sulfamethoxazole is an oral broad-spectrum antibiotic often used in birds, and it is particularly proper used for therapy of birds (Diren Sigirci et al., 2019). Resistance to trimethoprim/sulfamethoxazole were 44.4% and 46.0% in studies conducted in Egypt (Samir et al., 2025) and Turkey (Diren Sigirci et al., 2020). However, this antibiotic resistance rate reached 85.0% and 100% in studies from Bangladesh (Hasib et al., 2025) and Italy (Varriale et al., 2020). Notably, parrot species-derived E. coli strains exhibited trimethoprim/sulfamethoxazole resistance at rates ranging from 12.1-33.0% in a study conducted in Brazil (Machado et al., 2018; Pontes et al., 2018; Marques et al., 2024). The rates of resistance to aminoglycosides such as kanamycin, neomycin, and streptomycin were 30.2%, 34.0%, and 47.2%, respectively. Aminoglycosides are important therapeutic agents in clinical practice, often used to treat infections, especially severe infections caused by Gram-negative bacteria, but they are used less often in birds because they can be toxic (Flammer, 2006). However, some previous studies have documented aminoglycoside resistance in E. coli strains isolated from parrots, such as resistance rates of 25.0% and 34.0% for kanamycin and streptomycin in Turkey (Diren Sigirci et al., 2020); and resistance rates of 29.8% and 42.6% for kanamycin and neomycin in Bangladesh (Nupur et al., 2023). In Brazil, E. coli strains isolated from parrots showed streptomycin resistance ranging from 2.0% to 37.2% and 74.0% (Lopes et al., 2015; Machado et al., 2018; Pontes et al., 2018). Different lab methods and study populations may explain the varying resistance rates.

The rates of resistance to quinolone and fluoroquinolone classes such as norfloxacin, levofloxacin, and nalidixic acid ranged from 35.8% to 45.3%. Lower resistance rates for these antibiotics, from 13.0% to 17.0%, were observed in E. coli strains isolated from parrots in Turkey (Diren Sigirci et al., 2020). In Brazil, parrot-derived E. coli strains showed resistance to nalidixic acid at levels of 6.1% - 25.6% - 30.0% (Lopes et al., 2015; Machado et al., 2018; Pontes et al., 2018). Levofloxacin resistance of E. coli strains in a similar study conducted in Bangladesh was 19.1% (Nupur et al., 2023). Notably, parrot-derived E. coli strains with respiratory tract infections in Egypt showed resistance to norfloxacin at a high level of 51.9% (Samir et al., 2025). Differences in antibiotic resistance rates and phenotypes may be related to differences in antibiotic use practices, veterinary surveillance, and differences in hygiene and biosecurity measures between countries.

We found 31 different resistance patterns among the isolates, of which 33 (62.3%) were identified as multidrug-resistant. This variety may come from sampling different parrot species kept under different conditions of care, disease prevention, and treatment. Household pets are recognised as significant risks for zoonotic diseases and considered as a source of multidrug-resistant (MDR) bacteria (Diren Sigirci et al., 2020; Ohene Larbi et al., 2021), which can pose serious risks. The rates of MDR E. coli were consistent with the 59.0% and 67.0% observed in E. coli strains isolated from parrots in Brazil (Pontes et al., 2018) and Turkey (Diren Sigirci et al., 2020). However, in previous studies conducted in Italy (Varriale et al., 2020) and Brazil (Lopes et al., 2015), the rates of multidrug-resistant strains were identified as 40.1% and 43.2%, respectively. Notably, MDR rates were reported as high as 100% in studies in Bangladesh (Hasib et al., 2025). The rise of MDR bacteria calls for tighter control of antibiotic use to minimize the spread of MDR pathogens in livestock, including pet birds, and avoid significant risks to animal and community health (Szmolka and Nagy, 2013; Monteiro et al., 2025).

This study was only conducted with a small sample size on healthy parrots raised in Hanoi, and specific drug resistance genes such as ESBL, carbapenemase (bla­­NDM, blaKPC, blaOXA-48, blaIMP, blaVIM…) have yet to be identified. However, this study has provided basic data on the infection rates and antibiotic resistance of E. coli isolates from pet birds, including many introduced species. Therefore, larger-scale studies on more bird species are needed, as well as the use of molecular methods to detect antibiotic resistance genes, in order to determine the relationship between drug resistance genes, including plasmids, and multidrug-resistant phenotypes. In fact, finding out exactly how these bacteria become resistant would help us understand their role in the One Health context.

Overall, the E. coli strains we isolated were resistant to many antibiotics, and many were MDR. This suggests that pet birds could be reservoirs of antibiotic-resistant bacteria, which is a concern for both human and animal health. Although we did not study transmission to humans, the presence of MDR E. coli in pet birds is a potential public health concern that warrants further investigation. Future studies should examine ESBL and carbapenemase genes using PCR, and birds should be sampled at different times to see if resistance changes over time.

The authors would like to thank the students for transporting samples and the pet owners for their cooperation and assistance in the sampling process.

This preliminary study documented the presence of antibiotic-resistant E. coli strains in pet parrots in Hanoi, Vietnam. The study results suggest that pet parrots may be carriers of multidrug-resistant E. coli and contribute to the spread of antibiotic resistance in the environment. This underscores the need for monitoring and controlling antibiotic use in both animal husbandry and veterinary medicine to mitigate this problem in Vietnam.

CTTH: Study design, investigation and methodology, and manuscript writing. HTH, NTT, NMH, LBN, NTTM: Sample collection, experimentation and data analysis. TLO: Laboratory research design, sample testing supervision. THT: Conceptualization, supervision, writing, review and editing, project management.

The present study received no external funding. The authors contributed to this study independently.

The present study was conducted by collecting samples in accordance with the guidelines outlined in the animal welfare and safety procedures of the Committee on Animal Research and Ethics (CARE), Faculty of Veterinary Medicine, Vietnam National University of Agriculture, Vietnam (Approval No. CARE-2025/08).

AI-generated tools were not used to create any scientific content. Any AI assistance was limited to minor language editing, and all ideas, interpretations, and conclusions remain entirely the authors.

The authors have declared no conflict of interest.