Twenty-eight poultry farms producing broilers for local slaughterhouses in Austria voluntarily participated in the study between May 2019 and August 2020, some of which had private (n = 11) and others public (n = 17) water supplies. The fattening period at the participating poultry farms in Austria was five to six weeks. The poultry farms were divided into two distinct groups based on whether the farms employed solely chemical (T1) or a combination of chemical and mechanical (T2) water line treatment methods. An overview of the poultry farms included in the study is presented in Figure 1. Cleaning and water line treatment at the poultry farms was performed by the farmer. Since the participation of poultry farms in the study was voluntary, poultry farms 6, 8, 9, 12, and 13 withdrew their participation after T1 and were substituted by the poultry farms 15–19 during T2. The study was conducted in collaboration with a private laboratory (HYGIENICUM GmbH, Graz, Austria), which provided training on the water line cleaning procedures to be implemented at the poultry farms to the participating farmers.

During T1 water line treatment, water lines were drained and filled with a commercially-available solution of which the main disinfecting component contained 4.0 ppm active chlorine dioxide (ClO₂) solution (Calgonit CD-K1/K2, Calvatis GmbH, Ladenburg, Germany). The commercial solution was retained in the water lines for 24 h. Measurements of free ClO₂ inside the waterlines were not obtained. Subsequently, the water lines were washed with the supply water by continuous flushing for 10 min. Under normal operating conditions. The T2 water line was performed by continuous pumping of acidic cleaner containing 3.0% peroxyacetic acid (PAA) and hydrogen peroxide (Calgonit DS 625, Calvatis GmbH, Ladenburg, Germany) continuously for 30 min using highpressure air pump. The water lines were then washed with the supply water and purged using a highpressure air pump until no inorganic and organic debris were visible in the water. Subsequently, the water line disinfection was performed using a commercial disinfection solution containing 4.0 ppm active ClO₂ solution (Calgonit CD-K1/K2) which was retained in the water lines for 24 h. Subsequently, the water lines were washed with supply water by flushing for 10 min. Under normal operating conditions.

Water samples were collected by employees from the private laboratory, samples were taken from the end nipple of the drinking water line inside the vacant poultry house (HYGIENICUM GmbH, Graz, Austria). One water line was sampled at four and five poultry farms, while two water lines (line 1 and 2) were sampled at ten and nine poultry farms during T1 and T2 water line treatments (Figure 1). Two sampling timepoints were chosen, namely before treatment (BT) at the end of fattening period of 5–6 weeks, and after the water line treatment (AT) before restocking of the subsequent production cycle. As shown in Figure 1 and Supplementary Table S1, in six poultry farms during the T1 and T2 water line treatment, water samples were collected at two different sampling intervals, while other poultry farms were sampled only once. Additionally, at some poultry farms from some water lines the duplicate samples were collected, while from other poultry farms only a single sample was collected. Therefore, in total 36 (T1) and 33 (T2) BT and corresponding AT samples were collected for the microbial analysis in the present study. The water samples were collected in sterile 500 mL bottles by the private laboratory and immediately transported to the laboratory at 4°C for microbial analysis.

Prior to analysis, 500 mL of water samples were centrifuged at 8000 rpm for 30 min at 4°C (Thermo Scientific, Sorvall Lynx 4000 centrifuge). All but 10 mL of the supernatant was discarded, the remainder was then resuspended using a serological 10 mL pipette (Greiner Bio One, Frickenhausen, Germany) and vortexed for 30 s.

Campylobacter selective enrichment and isolation were performed according to the ISO 10272-1:2006 standard for the detection of Campylobacter spp. in foodstuff (27). Five milliliters of the supernatant were transferred to 45 mL of Bolton broth (Thermo Fisher Scientific Ltd., Hampshire, United Kingdom) supplemented with 5% hemolyzed horse blood (Oxoid Ltd., Hampshire, United Kingdom). The Bolton broth enrichment was incubated for up to 48 h at 42°C under microaerobic conditions (10% CO₂, 3% O₂, 87% N₂). After incubation modified charcoal cefoperazone deoxycholate agar (mCCDA) (Oxoid Ltd) was inoculated by fractionated loop inoculation (10 μL) and incubated at 42°C for 48 h under microaerobic conditions. Quantification of aerobic mesophilic count (AMC), Enterobacteriaceae (EB), and Pseudomonadaceae (PS) counts was carried out according to ISO reference methods (28, 29). For enumeration of AMC, EB, and PS, 5 mL of the re-suspended supernatant was transferred to 45 mL buffered peptone water (BPW) (Biokar Solabia diagnostics, Pantin Cedex, France). Subsequently, serial ten-fold dilutions were prepared up to dilution 10⁻⁵ in BPW (Biokar Solabia diagnostics, Pantin Cedex, France). The AMC were enumerated on trypto-caseine soy agar with 0.6% yeast extract (TSAYE) (Biokar Solabia diagnostics), while EB and PS were enumerated on red bile glucose agar (VRBG) (Merck KGaA, Darmstadt, Germany). Each dilution step (100 μL) was plated on selective agar media for the enumeration of AMC, EB, and PS counts. For dilution 10⁻¹ the volume of 1 mL was divided (333 μL) on three agar plates per selective medium. Agar plates were incubated at 30°C (AMC) and 37°C (EB, PS) aerobically for up to 48 h. The EB and PS counts on VRGB agar were differentiated by their ability to ferment glucose, leading to pink colonies with or without precipitation and pale colonies for PS. Presumptive EB and PS isolates were confirmed using oxidase reaction (BioMerieux, Marcy I’Etoile, France). The minimum and maximum limits for the determination of the AMC, EB, and PS in the samples ranged between 10 and 300 CFU.

Microbial quality of water samples before (BT) and after (AT) sanitation were categorized according to AMC, EB, and PS load in two contamination levels, <4.0 log₁₀ CFU/ml and ≥4.0 log₁₀ CFU/ml based on existing studies (4, 7, 12).

The predominant bacterial colony morphologies were collected from each water sample for further confirmation. Specifically, 1–5 colonies were selected from TSAYE (n = 224), VRBG (n = 206) and mCCDA agar (n = 41) and then subcultured on the respective medium. The isolate list is provided in the Supplementary Table S1. The purified colonies, comprising isolates from T1 BT samples (n = 123), T1 AT samples (n = 113), T2 BT samples (n = 139), T2 AT samples (n = 96) were stored at – 80°C in brain heart infusion broth (Biokar Solabia diagnostics) supplemented with 25% (v/v) glycerol (Merck KgaA).

For DNA extraction of Campylobacter spp. isolates 10 μL loop of bacterial material was resuspended in 100 μL of 0.1 M Tris–HCl buffer pH 7 (Sigma Aldrich, St. Louis, MO, United States) and mixed with 400 μL Chelex® 100-Resin (BioRad, Hercules, CA, United States) (30). The bacterial Chelex® 100Resin suspension was heated at 100°C for 10 min on a block heater (Thermo Fischer Scientific Inc.), followed by short centrifugation step at 15,000 ×g (Eppendorf Centrifuge 5,425) for 5 s. The supernatant (100 μL) was transferred to a maximum recovery tube (Corning Incorporated Life Sciences, Reynosa, Mexico) and stored at −20°C until analysis. Campylobacter spp. were identified using multiplex PCR targeting genes including the conserved genus-specific 23S rRNA gene, the Campylobacter jejuni hippuricase gene (hipO) and the Campylobacter coli serine hydroxymethyltransferase (glyA) gene, as previously described (31). Briefly, a single reaction mixture (20 μL) contained diethylpyrocarbonate (DEPC) treated water (Sigma Aldrich), 1× PCR buffer, 2 mM MgCl₂, 500 nm hipO forward and reverse primer, 1,000 glyA forward and reverse primer, 200 nm 23S forward and reverse primer, 200 μM dNTP mix, 1.5 U of Platinum Taq DNA polymerase (Platinum™ Taq DNA Polymerase, DNAfree, Thermo Fisher Scientific Inc., Waltham, MA, United States), and 5 μL template genomic DNA. The amplification was performed in T100™ Thermal Cycler (Bio-Rad, Hercules, CA, United States). The PCR cycling conditions included initial denaturation at 94°C for 2 min, 30 cycles of denaturation (94°C for 30 s), primer annealing (59°C for 30 s), elongation (72°C for 30 s) and final elongation (72°C for 7 min). The gel electrophoresis of PCRamplicons was performed in a 1.5% agarose gel containing 0.5× TrisBorateEDTA (TBE) buffer (Sigma Aldrich, St. Louis, MO, United States) and 3.5 μL peqGREEN DNA gel stain (VWR International, Radnor, United States), at 120 V for 30 min. The DNA standard Thermo Scientific™ GeneRuler™ 100 bp (Thermo Fisher Scientific Inc., Waltham, United States) was applied for fragment length comparison. We utilized the following control isolates for the DNA extraction and multiplex PCR: C. jejuni strain DSM 4688 and C. coli strain DSM 4689, obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ), Braunschweig, Germany.

For DNA extraction of isolates from TSAYE and VRBG, bacterial cells were lysed by boiling the suspension. A 10 μL loop of bacterial material was re-suspended in 100 μL 0.1 M Tris–HCl pH 7 buffer (Sigma Aldrich, St. Louis, MO, United States), briefly vortexed and heated at 100°C for 15 min (Thermo Scientific™ block heater, Thermo Fischer Scientific Inc.). The suspension was then centrifuged for 5 s at 15,000 ×g (Eppendorf Centrifuge 5,425, Hamburg, Germany) and the supernatant (70 μL) was transferred into maximum recovery tubes (Corning Incorporated Life Sciences, Reynosa, Mexico) and stored at −20°C until analysis. For identification of bacteria isolates (n = 471) the partial amplification of 16S rRNA gene was performed following the methods of (32, 33), using universal primer pairs 616F (5’AGAGTTTGATYMTGGCTC3′) and 1492R (5’GGYTACCT TGTTACGACTT3′) (both Microsynth AG, Blagach, Switzerland). A single PCR reaction (45 μL) contained 1× PCR buffer, 2 mM MgCl₂, 200 nM forward and reverse primer, 250 μM dNTP mix, 2 U of Platinum Taq DNA polymerase (Platinum™ Taq DNA Polymerase, DNAfree, Thermo Fisher Scientific Inc.) and 5 μL template genomic DNA. The DNA amplification was performed in T100™ Thermal Cycler (BioRad, Hercules, CA, United States). The PCR cycling conditions included initial denaturation at 95°C for 5 min, 35 cycles of denaturation (94°C for 30 s), primer annealing (52°C for 30 s), elongation (72°C for 60 s) and final elongation (72°C for 7 min). Subsequently, the PCR amplicons were sent for purification and sanger sequencing to LGC Genomics (LGC Genomics GmbH, Berlin, Germany). The gel electrophoresis of PCR-amplicons was performed in a 1.5% agarose gel containing 0.5× TrisBorateEDTA (TBE) buffer (Sigma Aldrich, St. Louis, MO, United States) and 3.5 μL peqGREEN DNA gel stain (VWR International, Radnor, United States), at 120 V for 30 min. The DNA standard Thermo Scientific™ GeneRuler™ 100 bp (Thermo Fisher Scientific Inc., Waltham, United States) was applied for fragment length comparison. The PCR amplicons were sequenced using a 1492R (5’GGYTACCTTGTTACGACTT3′) primer. The nucleotide sequences were qualityevaluated by using Finch TV 1.4.0 (34) and MEGA X (35). The bacterial nucleotide Basic Local Alignment Search Tool (BLAST) algorithm from the National Centre for Biotechnology Information (NCBI)¹ was used for taxonomy assignment. Sequences were assigned to genus or species level according to best matches and highest similarities (1,040 to 1,120 bp fragment length, similarity cutoff ≥97.0%). The partial rRNA gene sequence data from the isolates were deposited in the GenBank database under accession numbers MZ642358 to MZ643011.² Subsequent identification of opportunistic pathogens among identified isolates was performed using the bacterial metadata base BacDive (36) and List of Prokaryotic names with Standing in Nomenclature (LPSN) (37).

Opportunistic pathogens with clinical relevance isolated from water samples during T1 and T2 water line treatment were subjected to antimicrobial susceptibility testing (AST). The set of isolates included most frequently isolated OP, such as Pseudomonas spp. (n = 17), Ochrobactrum spp. (n = 4), Stenotrophomonas spp. (n = 3), and human relevant opportunistic pathogens including Citrobacter spp. (n = 2), Enterobacter spp. (n = 2), Klebsiella spp. (n = 1), and Aeromonas spp. (n = 1).

AST was performed for a total of 30 bacterial isolates using Sensititre™ Avian AVIAN1F Vet AST Plate (ThermoFischer Scientific Inc., Waltham, MA, United States), according to the manufacturer’s instructions. Briefly, single colonies were picked from fresh cultures grown on TSAYE for 24 h at 30°C, suspended in in sterile water to an optical density of a 0.5 McFarland standard (~ 10⁸ CFU/mL). 50 μl volumes of the bacterial suspension were transferred to wells containing different concentrations of lyophilized antimicrobials. Plates were sealed and incubated at 30°C for 24 to 48 h, after which minimum inhibitory concentrations (MIC) were read visually and defined as the lowest concentration of a given antibiotic at which no growth of the test organism was observed. E. coli strain ATCC 25922 was used as the internal quality control isolate. The minimum inhibitory concentration (MIC) breakpoints and definitions for multi-drug resistance (MDR; resistance to two or more antibiotic classes) (38) were determined following the standards provided by the Clinical and Laboratory Standards Institute (CLSI) manuals (39–41).

A descriptive analysis was carried out (mean, median, and standard deviation) for AMC, EB, and PS counts. The normal distribution of each data set (T1 and T2) was investigated using the Shapiro–Wilks test. Due to nonnormal distribution of data, the median values of AMC, EB, and PS counts were calculated. The Wilcoxon–Mann–Whitney rank sum test performed as a twosided test was applied to identify whether there was a significant difference between median AMC, EB and PS counts of BT and AT samples. Median AMC, EB, and PS counts in AT samples were compared for different water supplies (public vs. private), water line treatments (T1 vs. T2), following log₁₀ transformation, using Wilcoxon–Mann–Whitney rank sum test. Values of p < 0.05 were considered as statistically significant. Statistical analyses were carried out using the R software package for statistical computing.³

Ninety-nine BT samples and their corresponding AT water samples were microbiologically assessed, with a maximum acceptable microbial limit of 4.0 log₁₀ CFU/ml for AMC, EB, and PS counts (Table 1). Due to non-normal distribution of the data, we used the Wilcoxon–Mann–Whitney twosided rank sum test to assess the median values for AMC, EB, and PS counts. No significant differences (p ≥ 0.05) were observed between the median AMC, EB, and PS counts of the BT and AT samples after T1 water line treatment (Table 1). Furthermore, we did not observe any significant difference between median AMC, EB, and PS counts in poultry farms with private and public water supply. Among the water samples, the highest median AMC counts were observed in BT (5.9 ± 1.02 log₁₀ CFU/ml, median ± MAD; MAD: median absolute deviation) and AT (6.0 ± 1.17 log₁₀ CFU/ml) samples. Higher median AMC counts in BT and AT samples were observed in poultry farms with a private well than those with a public water supply (Table 1). The lowest median counts were observed for EB in both BT (3.6 ± 2.13 log₁₀ CFU/ml) and AT (2.3 ± 1.52 log₁₀ CFU/ml) samples. In AT samples higher median EB counts were observed in poultry farms with public water supply. The PS resulted in the second highest median counts, which remained unchanged in BT (4.7 ± 1.44 log₁₀ CFU/ml) and AT (4.7 ± 2.48 log₁₀ CFU/ml) samples. Higher median PS counts were detected in both BT and AT samples in poultry farms with public water supply.

After T1 water line treatment, high (>4.0 log₁₀ CFU/ml) AMC, EB, and PS counts from BT samples decreased below the maximum acceptable microbial limit in 8/36, 7/36, and 9/36 AT samples, respectively (Supplementary Table S2). The AMC, EB, and PS below the microbial limit were observed in 1/36, 18/36, and 7/36 BT and AT samples, respectively. The AMC, EB, and PS counts above the maximum acceptable microbial limit were observed in 27/36, 11/36, and 20/36 AT samples, respectively, after T1 treatment.

During T2 water line sampling, no significant differences (p ≥ 0.05) were observed in the median AMC, EB, and PS counts between the BT and AT samples (Table 1). No significant difference was observed between median AMC, EB, and PS count in poultry farms with private and public water supply. The highest median counts were for AMC counts in both BT (4.6 ± 1.55 log₁₀ CFU/ml) and AT (4.7 ± 1.85 log₁₀ CFU/ml) samples, followed by the PS counts in BT (3.5 ± 1.62 log₁₀ CFU/ml) and AT (3.1 ± 2.05 log₁₀ CFU/ml) samples The lowest counts were observed in the median EB counts of BT (2.4 ± 1.63 log₁₀ CFU/ml) and AT (1.6 ± 0.42 log₁₀ CFU/ml) samples. Higher median AMC, EB, and PS counts were detected in AT samples in poultry farms with public water supply.

After T2 water line treatment, high (>4.0 log₁₀ CFU/ml) AMC, EB, and PS counts from BT samples decreased below the maximum acceptable microbial limit in 8/33, 5/33, and 14/33 AT samples, respectively (Supplementary Table S3). The AMC, EB, and PS counts below the microbial limit were detected in 4/33, 25/33, and 10/33 samples in both BT and AT, respectively. The AMC, EB, and PS counts remained above the maximum acceptable microbial limit in 21/33, 3/33, and 9/33 AT samples, respectively, after T2 water line treatment.

The impact of T1 and T2 water line treatment on private and public water supply was evaluated by calculating the log₁₀ ratio from CFU log₁₀ counts detected in BT and AT water samples (Table 1). No significant differences (p ≥ 0.05) in log₁₀ ratios were observed for AMC, EB, and PS counts after T1 and T2 water line treatment. The log₁₀ ratio was not significantly different (p ≥ 0.05) between private and public supplied poultry farms after T1 and T2 water line treatment. The median AMC, EB, and PS ratios after T1 waterline treatment were −0.2 ± 2.13, −0.6 ± 1.79, and 0.0 ± 2.26, respectively. The analysis of log₁₀ ratios after T2 waterline treatment resulted in median values of −1.1 ± 2.13 for AMC, 0.0 ± 2.94 for EB, and 0.0 ± 3.12 for PS counts. Although log₁₀ ratios between poultry farms with private and public water supplies were not significantly different, we observed higher median log₁₀ reduction of AMC, EB, and PS counts at poultry farms with private water supply. During T2 water line treatment higher median log₁₀ reduction was observed for AMC and EB counts at poultry farms with public water supply, while higher median log₁₀ reduction for PS counts was observed in poultry farms with private water supply.

Out of the 14 poultry farms assessed, five farms exhibited microbial counts below the acceptable microbial limit (<4.0 log₁₀ CFU/ml) subsequent to the T1 water line treatment (Figures 2A–C). Among these farms, three had a private water supply, while the remaining two had public water supplies. Notably, poultry farm 7, which had a public water supply, exhibited an AMC count below the maximum acceptable microbial limit in both BT and corresponding AT water sample. Furthermore, 11 poultry farms exhibited EB counts below the maximum acceptable microbial limit. Of these, nine poultry farms demonstrated EB counts below the microbial limit in both BT and corresponding AT samples. Additionally, among 14 poultry farms examined, a total of eight poultry farms exhibited PS counts below the microbial limit. Out of these, four poultry farms demonstrated PS counts below the microbial limit in both BT and corresponding AT samples. Among the poultry farms that underwent two samplings, poultry farms 12 and 13 exhibited AMC and PS counts exceeding the microbial limit in one of the sampling events. Furthermore, poultry farm 12 demonstrated EB counts above the microbial limit on one of two sampling occasions.

During T2 waterline treatment AMC counts below the microbial limit were observed in six out of 14 poultry farms (Figures 3A–C). Of these, two poultry farms demonstrated AMC count below the microbial limit in both BT and corresponding AT samples (Figure 3B). EB counts below the microbial limit were observed in 12 out of 14 poultry farms, and among them, nine poultry farms had EB counts below the microbial limit in both the BT and corresponding AT samples. Similarly, PS counts below the microbial limit were observed in ten from 14 poultry farms, and among them, three poultry farms demonstrated PS counts below the microbial limit in BT and corresponding AT samples. Among the poultry farms subjected to two samplings, poultry farm 18 demonstrated AMC, PS and EB counts below the microbial limit during one of the sampling occasions. However, after second sampling, the AMC load in water samples exceeded the microbial limit. Notably, the PS and EB counts remained below the microbial limit during both sampling occasions.

Isolate taxonomic assignment was performed using partial sequencing of 16S rRNA gene. In the present study, isolate sequences showed ≥97.0% similarity to the reference sequence in the NCBI database. In BT samples, 123 isolates corresponded to 24 genera and 55 species, while in AT samples, the 113 isolates corresponded to 22 genera and 40 species. Further analysis of bacterial isolates revealed that in BT and AT samples, 43.1% (n = 41 isolates) and 36.3% (n = 53 isolates) of sequenced isolates were assigned to OP, found in 29/36 BT and 17/36 AT samples (Table 2). The isolates from BT samples contained OP represented by 16 genera and 19 species, while isolates from AT samples contained OP represented by 12 genera and 12 species OP. Furthermore, C. jejuni was detected using multiplex PCR in a single BT water sample from a poultry farm with a public water supply.

During the T2 water line treatment, 139 isolates in the BT corresponded to 26 genera and 46 species, whereas 96 isolates in AT samples corresponded to 21 genera and 33 species (Table 2). Among the sequenced isolates, 30.9% (n = 43 isolates) and 33.3% (n = 33 isolates) corresponded to OP, isolated from 20/33 BT and 14/33 AT samples, respectively. The OP in the BT samples comprised 10 genera, and 14 species, while the OP in the AT samples comprised 11 genera and 14 species. No Campylobacter spp. were detected in poultry drinking water samples during the T2 water line treatment.

Figures 4A,B represents the taxonomic classification of assigned isolate sequences at phylum, and genus level. The predominant phyla in BT and AT samples were Pseudomonadota, followed by Bacillota, Actinomycetota, and Bacteroidota (Figure 4A). The frequently isolated genera during both T1 and T2 water line treatment in BT and AT samples were Aeromonas, Bacillus, Citrobacter, Enterobacter, Pseudomonas and Stenotrophomonas (Figure 4B). Among these, Pseudomonas (BT 38.2%; AT 32.7%) and Bacillus (BT, 13.0%, AT, 11.5%) were most commonly observed genera during T1 water line treatment. Similarly, during T2 water line treatment, Pseudomonas (BT, 31.7%; AT, 33.3%) and Bacillus (BT, 10.1%; AT, 11.5%) were predominant genera in BT and AT samples. The Figure 4B depicts the percentage identification of other observed genera during T1 and T2 water line treatments. The majority of sequenced isolates classified as OP in BT and AT samples during T1 and T2 water belonged to the Pseudomonas spp., followed by Stenotrophomonas spp., Citrobacter spp., Ochrobactrum spp., and Acinetobacter spp. (Figure 4C). Furthermore, isolates of Enterobacter spp. and Klebsiella spp. genera were isolated during T1 and T2 sampling. The Pseudomonas spp. isolates identified as OP were most frequently observed bacteria sequences during both T1 (BT, 22/123 isolates; AT, 10/113 isolates) and T2 (BT, 13/139 isolates; AT, 10/96 isolates) sampling.

Before and after the T1 water line treatment, Pseudomonas spp. was isolated from BT and AT samples in 12/14 and 9/14 poultry farms, respectively (Table 3). Isolate sequences of OPs were detected in BT samples of 11 out of 14 poultry farms and in AT samples of 9 out of 14 poultry farms. Among the frequently observed genera before and after T2 treatment, the genus Pseudomonas was isolated from the BT and AT samples in 12 out of 14 poultry farms and 9 out of 14 poultry farms, respectively (Table 4). The OP were observed in 10 out of 14 poultry farms in BT samples and in 9 out of 14 poultry farms in AT samples after T2 water line treatment.

The susceptibility of bacterial isolates recovered from BT (n = 14) and AT (n = 16) water samples during T1 and T2 water line treatments to 18 antibiotic agents commonly used in poultry production was evaluated using Avian AVIAN1F Vet AST susceptibility plates (Table 5). The goal was to investigate AMR in the most frequently isolated OP isolates, including isolates belonging to Pseudomonas spp., Stenotrophomonas spp., Ochrobactrum spp., as well as AMR in specific waterborne OP important to human health, such as Aeromonas spp., Citrobacter spp., Enterobacter spp., and Klebsiella spp.

The highest level of AMR was observed against spectinomycin and sulfadimethoxin (90.0%; 27/30 isolates each), followed by ceftiofur (83.3%; 25/30 isolates), florfenicol (66.6%; 20/30 isolates), and neomycin (53.5%, 16/30 isolates). Further, some isolates were resistant to enrofloxacin (23.3%; 13/30 isolates), trimethoprim-sulfamethoxazole (23.1%; 3/13 isolates), sulfathiazole (20.0%; 6/30 isolates), streptomycin (16.7%, 5/30 isolates), gentamicin (13.3%; 4/30 isolates), and trimethoprim-sulfamethoxazole (10.0%, 3/30 isolates).

The MDR was exhibited among the isolates of Pseudomonas spp., (17/17 isolates), and Stenotrophomonas spp. (1/3 isolates), Ochrobactrum spp. (4/4 isolates), Citrobacter spp. (2/2 isolates), and Enterobacter spp. (1/2 isolates). All Pseudomonas spp. isolates showed resistance patterns exhibiting resistance to a minimum of four and a maximum of eight antibiotics. Tested Stenotrophomonas spp. isolates also demonstrated resistance patterns to a minimum of four and a maximum of six antibiotics. All tested Ochrobactrum spp. were resistant to four antibiotics. The isolates of Citrobacter spp. were resistant to six antimicrobial classes and nine different antibiotics. The isolates of Enterobacter spp. showed resistance patterns to a minimum of two and a maximum of four antibiotics. The isolates of Klebsiella spp. were resistant to five antibiotics, while Aeromonas spp. isolate was susceptible to all tested antibiotic agents.