Antimicrobial Susceptibility Profiles and Molecular Typing of Campylobacter jejuni and Campylobacter coli Isolates from Ducks in South Korea

Abstract

Campylobacter is a food-borne zoonotic pathogen that causes human gastroenteritis worldwide. Campylobacter bacteria are commensal in the intestines of many food production animals, including ducks and chickens. The objective of the study was to determine the prevalence of Campylobacter species in domestic ducks, and the agar dilution method was used to determine resistance of the isolates to eight antibiotics. In addition, multilocus sequence typing (MLST) was performed to determine the sequence types (STs) of selected Campylobacter isolates. Between May and September 2012, 58 duck farms were analyzed, and 56 (96.6%) were positive for Campylobacter. Among the isolates, 82.1% were Campylobacter jejuni, 16.1% were C. coli, and one was unidentified by PCR. Of the 46 C. jejuni isolates, 87.0%, 10.9%, and 21.7% were resistant to ciprofloxacin, erythromycin, and azithromycin, respectively. Among the C. coli isolates, all 9 strains were resistant to ampicillin, and 77.8% and 33.3% were resistant to ciprofloxacin and azithromycin, respectively. The majority of the Campylobacter isolates were classified as multidrug resistant. Twenty-eight STs were identified, including 20 STs for C. jejuni and 8 STs for C. coli. The most common clonal complexes in C. jejuni were the ST-21 complex and the ST-45 complex, while the ST-828 complex predominated in C. coli. The majority of isolates were of STs noted in ducks and humans from earlier studies, along with seven STs previously associated only with human disease. These STs overlapped between duck and human isolates, indicating that Campylobacter isolates from ducks should be considered potential sources of human infection.

INTRODUCTION

Campylobacters are the most commonly isolated bacterial enteric pathogens in developed and developing countries (1). Campylobacter jejuni is the predominant cause of campylobacteriosis, which is responsible for 93.4% of confirmed cases, whereas C. coli accounts for only 2.3% of outbreaks (2). Acute diarrheal illness is the main clinical sign, but more severe complications, such as Guillain-Barré syndrome, reactive arthritis, and a range of extraintestinal infections, occur in some serious cases (3, 4). Macrolides and fluoroquinolones are normally considered for treatment of Campylobacter enteritis; however, the prevalence of fluoroquinolone-resistant Campylobacter has increased dramatically in many countries since the last century, and such strains continue to exist with high-level resistance (5–7).

Poultry is the major source of human infection; chickens constitute the major reservoir of Campylobacter, highlighting the potential public health threat (8, 9). Ducks are the second largest poultry meat reservoir with a potential public health risk, but they have remained largely uncharacterized in comparison to chickens. There is some evidence of an increased risk of Campylobacter infection in humans associated with ducks (10). A study in the United Kingdom found that duck meat was implicated in 2% of campylobacteriosis outbreaks (11), and a large outbreak of campylobacteriosis was derived from duck products in the United Kingdom (12). Previous surveys indicated high prevalences of Campylobacter in domestic ducks and duck meat (11, 13). Therefore, there is reason to suspect that ducks are an important source of human campylobacteriosis that may be underestimated. An increase in Campylobacter resistance, particularly to fluoroquinolones and macrolides, has also been found in ducks (14–16). These findings indicate the potential risk of antibiotic-resistant Campylobacter subsequently being transferred to humans by ducks.

The proportion of global duck meat production occurring in Asia increased from 79.0% to 82.9% between 2002 and 2011 (17). In South Korea, duck production represents an important sector of animal husbandry, and consumption of duck meat has increased significantly, at an average rate of 13.4%, since 2001, which is faster than the case for any other meat source (18). However, little information is available on the prevalence and antimicrobial resistance of Campylobacter strains from ducks. Therefore, we investigated the occurrence and antimicrobial resistance of Campylobacter in domestic ducks in South Korea. In addition, we characterized antimicrobial resistance and correlated it with genotype.

MATERIALS AND METHODS

Isolation and identification of Campylobacter.

From May to September 2012, 58 farms were selected from five provinces. Five, 6, 12, 15, and 20 farms in Gyonggi, Chungnam, Chungbuk, Chonnam, and Chonbuk Provinces (Fig. 1), respectively, were sampled, which included >90% of the duck population in the five provinces (19). The sizes of the different farms varied from 5,000 to 20,000 ducks; the ducks were reared indoors with low curtain walls but not a closed production system until 6 to 7 weeks of age. Among the 58 farms, 9, 12, 19, and 18 farms had ducks of less than 1 week of age, less than 2 weeks but more than 1 week of age, less than 3 weeks but more than 2 weeks of age, and more than 3 weeks of age, respectively. Five pooled cloacal swabs from each farm were preenriched in Bolton broth (Oxoid Ltd., Basingstoke, England) and supplemented with cefoperazone, vancomycin, trimethoprim, and cycloheximide (Oxoid) for 24 h at 42°C in a microaerobic atmosphere of 10% CO₂, 5% O₂, and 85% N₂. The bacteria were streaked onto modified charcoal-cefoperazone-desoxycholate agar (Oxoid) containing an antibiotic supplement (cefoperazone and amphotericin; Oxoid) and incubated for 48 h at 42°C under microaerobic conditions. Presumptive Campylobacter colonies on the plate were further cultivated on 5% sheep blood agar plates (Komed, Seongnam, South Korea) for 24 to 48 h at 42°C under microaerobic conditions. One suspected colony was isolated from each farm.

FIG 1.

Locations of provinces in South Korea and prevalences of Campylobacter isolated in different provinces. The numbers represent the numbers of positive farms/total numbers of farms.

DNA templates were prepared using freshly grown Campylobacter colonies on blood agar by adding 500 μl sterile distilled water and boiling the samples in a heater block at 100°C for 15 min. Template DNA was stored at −20°C until it was used for PCR. The isolates were identified to the genus level by amplifying the 16S rRNA gene specific for Campylobacter species (20). We used multiplex PCR with specific primers to identify Campylobacter species, the hipO gene specific for C. jejuni, and the glyA gene specific for C. coli (21). After identifying each isolate, the Campylobacter isolates were initially stored in 20% glycerol in brain heart infusion broth (Oxoid) at −70°C.

Antimicrobial susceptibility testing.

The agar dilution method was used to determine susceptibility of the Campylobacter isolates to the following eight antimicrobial agents: ampicillin (Amp), azithromycin (Azi), ciprofloxacin (Cip), clindamycin (Cli), erythromycin (Ery), gentamicin (Gen), nalidixic acid (Nal), and tetracycline (Tet). Mueller-Hinton agar (Oxoid) plates supplemented with 5% lysed sheep blood (Oxoid) were used as culture medium for 2-fold serial dilutions of antibiotics ranging from 0.06 to 256 μg/ml. Fresh bacterial colonies taken directly from the agar plates and incubated for 24 h were resuspended in sterile tryptic soy broth (Merck Inc., Rahway, NJ) to obtain a suspension with a 0.5 McFarland turbidity. The plates were incubated under microaerobic conditions at 42°C for 24 h. MIC values were defined as the lowest concentrations producing no visible growth, and antimicrobial-free agar plates were included as a control for normal growth. The breakpoints were determined according to the National Antimicrobial Resistance Monitoring System criteria (22). Since there were no ampicillin breakpoints for Campylobacter, we used the breakpoints for Enterobacteriaceae from the Clinical and Laboratory Standards Institute criteria (23). C. jejuni ATCC 33560 was used as the reference quality control strain.

MLST.

All C. jejuni and C. coli isolates were characterized by multilocus sequence typing (MLST) based on primers for seven gene targets of each isolate, including aspA (encoding aspartase A), glnA (encoding glutamine synthase), gltA (encoding citrate synthase), glyA (encoding serine hydroxymethyl transferase), pgm (encoding phosphoglucomutase), tkt (encoding transketolase), and uncA (encoding the ATP synthase alpha subunit), using previously described conditions (24). Allele numbers, sequence types (STs), and clonal complexes (CCs) were assigned using the PubMLST database (http://pubmlst.org/campylobacter/) (25), and novel STs were submitted to the MLST database and assigned new numbers.

Statistical analysis.

The chi-square test was used for comparisons in order to determine if there were statistically significant differences at the 95% level in the prevalence of Campylobacter samples between different provinces. Differences were considered significant if the P values were <0.05.

RESULTS

Isolation and identification of Campylobacter.

Of the 58 farms, 56 (96.6%) had Campylobacter. Among the 56 isolates, 82.1% (46/56 isolates), 16.1% (9/56 isolates), and 1.8% (1/56 isolates) were identified as C. jejuni, C. coli, and unidentified Campylobacter species, respectively. The isolation rates were 95.0% (19/20 farms) in Chonbuk, 93.3% (14/15 farms) in Chonnam, 100.0% (12/12 farms) in Chungbuk, 100.0% (6/6 farms) in Chungnam, and 100.0% (5/5 farms) in Gyonggi (Fig. 1). All five provinces had high rates of Campylobacter isolation, with no significant differences (P > 0.05). All 9 C. coli isolates were isolated from Chungbuk Province, with a higher prevalence (75.0% [9/12 farms]) than that of C. jejuni (25.0% [3/12 farms]). Except for one farm with ducks of <1 week of age and one farm with ducks of >3 weeks of age which were negative for Campylobacter, Campylobacter was successfully isolated from ducks of the other age groups.

Antimicrobial susceptibility.

The antimicrobial susceptibilities of the 46 C. jejuni and 9 C. coli strains were determined against eight antimicrobial agents, and the results are presented in Table 1. Among the C. jejuni isolates, resistance to ciprofloxacin was the most common (40/46 isolates [87.0%]), followed by resistance to tetracycline (39/46 isolates [84.8%]) and nalidixic acid (39/46 isolates [84.8%]). All C. coli strains were resistant to ampicillin, and 88.9% (8/9 isolates) of the isolates were resistant to tetracycline. Bacterial resistance to macrolides (azithromycin and erythromycin) was also determined and presented resistance rates of 21.7% and 33.3%, respectively, for C. jejuni and 33.3% and 33.3%, respectively, for C. coli.

TABLE 1.

Distribution of MICs of eight antimicrobial agents for C. jejuni and C. coli isolates

Antimicrobial agent	Species	No. of isolates for each concn of antimicrobial (μg/ml)													Breakpoint (μg/ml)	MIC50/MIC90 (μg/ml)	Resistance rate (%)
		≤0.06	0.13	0.25	0.5	1	2	4	8	16	32	64	128	≥256
Ampicillin	C. jejuni	0	0	0	0	0	3	2	4	7	8	3	2	16	32	32/≥256	64.4
	C. coli	0	0	0	0	0	0	0	0	0	5	0	1	4	32	32/≥256	100.0
Azithromycin	C. jejuni	11	7	5	5	4	0	3	3	2	1	4			8	0.25/32	22.2
	C. coli	3	0	1	3	0	0	0	0	2	0	1			8	0.5/16	30.0
Ciprofloxacin	C. jejuni	2	1	1	0	0	2	1	2	7	20	9			4	32/64	86.7
	C. coli	0	0	1	0	1	0	1	0	5	2	0			4	16/32	80.0
Clindamycin	C. jejuni	2	8	10	12	4	3	3	1	2	0	0			8	0.5/4	6.7
	C. coli	0	1	1	3	1	3	0	0	1	0	0			8	0.5/2	10.0
Erythromycin	C. jejuni	2	8	8	8	5	4	4	1	0	0	1	4		32	0.5/64	11.1
	C. coli	0	2	1	3	0	0	1	0	0	0	1	2		32	0.5/128	30.0
Gentamicin	C. jejuni	5	4	10	11	10	0	1	1	0	0	0	3		8	0.5/4	8.9
	C. coli	0	1	0	6	2	0	0	0	0	0	0	1		8	0.5/1	10.0
Nalidixic acid	C. jejuni					0	0	3	1	1	2	2	19	17	64	128/≥256	84.4
	C. coli					0	0	1	0	1	1	1	4	2	64	128/≥256	80.0
Tetracycline	C. jejuni					1	4	2	0	1	6	10	4	17	16	64/≥256	84.4
	C. coli					0	0	0	1	0	3	0	0	6	16	256/≥256	90.0

Except for 1 strain of C. jejuni which was susceptible to all eight antimicrobial agents, the other 55 strains of Campylobacter were resistant to at least one antimicrobial agent (Table 2). The results indicate that 92.7% (51/55 isolates) of the strains were resistant to at least two antimicrobial agents, and the most frequent multidrug resistance pattern was resistance to ampicillin, ciprofloxacin, nalidixic acid, and tetracycline. All C. coli isolates were multidrug resistant. One isolate of C. jejuni and one isolate of C. coli showed resistance to all eight antimicrobial agents tested in this study.

TABLE 2.

Antimicrobial resistance patterns of C. jejuni and C. coli isolates

Campylobacter species	No. of agents to which isolate(s) is resistant	Antimicrobial resistance profile	No. of isolates	Rate (%)
C. jejuni	1	Cip	1	2.2
		Nal	1	2.2
		Tet	1	2.2
	2	Amp Tet	1	2.2
		Cip Nal	1	2.2
	3	Amp Azi Tet	1	2.2
		Amp Cip Nal	2	4.4
		Amp Cip Tet	1	2.2
		Amp EryTet	1	2.2
		Cip Nal Tet	10	22.2
	4	Amp Azi Cip Nal	1	2.2
		Amp Cip Nal Tet	13	28.9
	5	Amp Azi Cip Nal Tet	3	6.7
		Amp Cip Gen Nal Tet	1	2.2
	6	Amp Azi Cip Ery Nal Tet	2	4.4
		Amp Azi Cip Gen Nal Tet	1	2.2
		Amp Cip Cli Gen Nal Tet	1	2.2
		Azi Cip Cli Ery Nal Tet	1	2.2
	8	Amp Azi Cip Cli Ery Gen Nal Tet	1	2.2
C. coli	2	Amp Cip	1	10.0
		Amp Tet	1	10.0
	4	Amp Azi Ery Tet	1	10.0
		Amp Cip Nal Tet	5	50.0
	6	Amp Azi Cip Ery Nal Tet	1	10.0
	8	Amp Azi Cip Cli Ery Gen Nal Tet	1	10.0

MLST.

Twenty-eight different STs were identified among the samples: 20 STs for C. jejuni and 8 STs for C. coli (Table 3). Three complexes, namely, the ST-21 complex, the ST-45 complex, and the ST-828 complex, predominated and accounted for 60% of all isolates. The most common of these was the ST-21 complex, with 17 isolates divided into 3 STs, followed by the ST-45 complex, represented by 8 isolates in 2 STs, and the ST-828 complex, with 8 isolates and 7 STs. Forty-four of the isolates grouped into seven previously characterized CCs, whereas the remaining nine isolates had STs that were unassigned. Three new STs were identified in this study, but no new allelic sequences were found for any of the housekeeping genes, and all of the new STs resulted from new combinations of previously described alleles. A total of 73 alleles were identified across all seven loci, ranging from 7 alleles of glyA to 15 alleles of gltA.

TABLE 3.

Sources and STs of Campylobacter duck isolates from South Korea

Province	Campylobacter species	CCa	STb	No. of isolates	MLST allelic profile
					aspA	glnA	gltA	glyA	pgm	tkt	uncA
Chonnam	C. jejuni	ST-21	21	4	2	1	1	3	2	1	5
	C. jejuni		4253	1	2	17	52	3	2	1	5
	C. jejuni	ST-45	45	3	4	7	10	4	1	7	1
	C. jejuni	ST-692	4512	1	37	52	16	28	74	29	23
	C. jejuni	ST-1034	3266	1	8	61	4	64	74	57	23
	C. jejuni	ST-1332	5154	2	2	1	57	28	58	29	58
	C. jejuni	U	5857	1	8	2	80	28	74	24	23
	C. jejuni		7015	1	350	410	271	82	58	25	23
Chonbuk	C. jejuni	ST-21	21	4	2	1	1	3	2	1	5
	C. jejuni		50	1	2	1	12	3	2	1	5
	C. jejuni		4253	1	2	17	52	3	2	1	5
	C. jejuni	ST-45	45	2	4	7	10	4	1	7	1
	C. jejuni	ST-354	3784	1	8	315	2	2	11	12	6
	C. jejuni	ST-692	4536	1	37	52	4	64	129	29	23
	C. jejuni		7016	1	37	52	4	64	127	99	23
	C. jejuni	ST-1034	1034	1	2	61	4	64	74	25	23
	C. jejuni		5231	1	8	61	4	28	74	25	23
	C. jejuni	U	5096	1	22	8	4	28	74	99	35
	C. jejuni		5917	2	37	364	359	64	127	47	23
	C. jejuni		5857	2	8	2	80	28	74	24	23
Chungnam	C. jejuni	ST-21	21	2	2	1	1	3	2	1	5
	C. jejuni		50	1	2	1	12	3	2	1	5
	C. jejuni	ST-45	45	1	4	7	10	4	1	7	1
	C. jejuni		4852	1	37	7	10	4	1	7	1
	C. jejuni	ST-354	3784	1	8	315	2	2	11	12	6
Chungbuk	C. jejuni	ST-21	21	1	2	1	1	3	2	1	5
	C. jejuni	ST-45	4852	1	37	7	10	4	1	7	1
	C. jejuni	U	6149	1	37	364	4	28	74	99	23
	C. coli	ST-828	889	2	33	39	30	82	113	47	41
	C. coli		1593	1	32	39	30	82	104	43	17
	C. coli		2075	1	124	39	30	79	104	35	17
	C. coli		2397	1	184	39	30	82	113	43	17
	C. coli		4291	1	33	327	30	79	113	43	17
	C. coli		6582	1	33	39	402	82	113	43	17
	C. coli		7014	1	32	39	262	79	104	43	17
	C. coli	U	6148	1	33	38	30	79	113	47	17
Gyonggi	C. jejuni	ST-21	21	1	2	1	1	3	2	1	5
	C. jejuni		50	1	2	1	12	3	2	1	5
	C. jejuni	ST-1332	696	2	2	1	4	28	58	25	58
	C. jejuni		1332	1	2	1	29	28	58	25	58

^aU, no clonal complex assigned yet.

^bNew STs are shown in bold.

DISCUSSION

Ducks and their products are commonly consumed in the modern Asian diet, but little information is available about Campylobacter species from ducks. Therefore, the objective of this study was to determine the prevalence of Campylobacter infection in ducks from South Korea. Campylobacter was isolated from 96.6% of the duck cloacal samples, and such a high isolation rate of Campylobacter from ducks has been reported previously (13, 16). The present isolation rate was much higher than that from chicken flocks in China and Japan, with isolation rates of 77.7% and 47.2%, respectively (26, 27). Differences in avian species, temperature, moisture, and feed model may have influenced the prevalence of Campylobacter (28, 29).

C. jejuni is typically identified as the most prevalent species on commercial duck farms among the scarce Campylobacter studies on domestic ducks (13, 14), and our results were consistent with these findings, as 82.1% of the isolates were identified as C. jejuni. We found a high prevalence (88.9%) of Campylobacter in ducks of <1 week of age. A similar result has been reported previously, with 100% of ducks being contaminated with Campylobacter at 8 days of age (30). This result was different from the case for chickens, in which Campylobacter is almost never detected in animals of <2 weeks of age (31). The reason for the early Campylobacter colonization in ducks is unclear. While evidence exists for chickens (32), maternal antibody resistance to Campylobacter colonization is unknown for ducks. Other studies have reported that different life cycles and high levels of environmental contamination may explain the early colonization (13, 33), or ducks may be a prime host compared to chickens (34).

Antibiotic resistance, particularly multidrug resistance, is a public health problem. In our study, 44 (91.4%) C. jejuni strains and all C. coli strains were multidrug resistant. Ducks may play a role in transmitting multidrug-resistant Campylobacter to humans along the food chain, like the case for chickens (8). In addition, one strain of C. jejuni and one C. coli strain were resistant to all eight antimicrobials isolated from ducks. These observations highlight the need for rigorous surveillance of antibiotics used in ducks to control further emergence of antibiotic-resistant Campylobacter.

Our results indicate high-level resistance of the microorganisms from ducks to ciprofloxacin and nalidixic acid. The prevalence of fluoroquinolone-resistant Campylobacter varies greatly among different countries. No fluoroquinolone-resistant C. jejuni is found in Tanzania (35), whereas higher resistances (17.4 to 76%) to ciprofloxacin have been reported in other areas (14, 16, 36). Over 90% C. jejuni resistance to nalidixic acid was reported recently in Vietnam (15). Similarly, C. coli strains with high resistance to fluoroquinolones have been found in Iran, the United Kingdom, Malaysia, and Vietnam, with resistance rates ranging from less than 20% to 100% (11, 14, 15, 36). The high fluoroquinolone resistance rate for Campylobacter in our study may be attributed to the widespread use of fluoroquinolones in poultry production in South Korea before July 2010 (37). Additionally, one study showed that fluoroquinolone-resistant Campylobacter continues to persist even after removal of the selection pressure (38). Moreover, fluoroquinolone-resistant strains enhance the fitness of susceptible strains with no antibiotic selection pressure (39). An unexpected result was that two strains of C. jejuni and one strain of C. coli were resistant to ciprofloxacin but susceptible to nalidixic acid. The same pattern has been reported for C. jejuni and C. coli isolates from swine (40), but the mechanism of resistance to ciprofloxacin and susceptibility to nalidixic acid is unknown. It is well established that point mutations in gyrA can confer resistance to both ciprofloxacin and nalidixic acid or to nalidixic acid alone (41). Other reports show that ciprofloxacin resistance is not inevitable with the gyrA mutation (42). Further investigation is required to fully define the mechanisms involved in ciprofloxacin and nalidixic acid resistance in Campylobacter.

MLST is an important tool for elucidating the diversity and transmission routes of Campylobacter isolates for humans. In the present study, ST-21 and ST-45, assigned to CC-21 and CC-45, respectively, were more prevalent than other STs in C. jejuni isolates from South Korean ducks. These two STs are the most common STs in humans (43, 44). Our results are consistent with those of a previous report from the United Kingdom showing that the most common ST from domestic ducks was ST-45 (13). ST-21 from our study was the first to be found largely in ducks compared with previous studies and the MLST database (13, 15, 25). ST-21 has a wide host range in the MLST database, including chickens, cattle, wild birds, and the environment, but only one ST-21 isolate was from ducks. The reasons for the high prevalence (9/46 isolates [19.6%]) of ST-21 isolated from ducks in our study may be that ducks are an ST-21 host and that no large investigations have compared data for ducks with those for other poultry, such as chickens and turkeys, from the MLST database. The increasing fitness of certain STs may have induced the high prevalence in ducks, as poultry may mediate the microevolution of Campylobacter to become widespread (45). Ducks maintained in an open air system can easily exchange pathogens with the environment, wild animals, or other domestic animals near the duck farm, which is consistent with the case for other livestock, such as chickens (46).

Ducks are an important reservoir of Campylobacter and may threaten human health (11). In the only report of MLST data from human clinical samples in a restricted area in South Korea, only 2 common STs (ST-21 and ST-45) isolated from ducks overlapped with human isolates (47). However, 53.5% (15/28 STs) of the STs found in ducks were previously associated with human infections. In particular, three STs of C. jejuni (ST-3784, ST-3266, and ST-5154) and four STs of C. coli (ST-1593, ST-2075, ST-2397, and ST-4291) were particularly associated with human infections with ducks as the only source. Therefore, duck-associated Campylobacter poses a plausible risk to human health.

An infection with an antimicrobial-resistant Campylobacter strain may lead to a suboptimal outcome or treatment failure (48). No significant differences in antimicrobial resistance were observed between the STs in our study. All STs showed multidrug resistance, and the majority of the clonal complexes (ST-21, ST-45, and ST-828) were multidrug resistant and fluoroquinolone resistant. In particular, the STs that overlapped with humans showed multidrug resistance (Table 4). The high risk to humans is that ST-889 and ST-3784 were resistant to all eight antimicrobials tested. In addition to direct transmission to humans, ducks may also act as vehicles of disease in contact with other domestic animals, such as occurs with other human-pathogenic etiologies (49, 50). The rapid and widespread dissemination of multidrug-resistant Campylobacter clonal groups highlights the need to develop effective infection control measures for these groups in animal reservoir populations.

TABLE 4.

Distribution of MLST profiles and antimicrobial resistance patterns among C. jejuni and C. coli isolates from ducks

Species	CCa	STb	No. of isolates	Antimicrobial resistance pattern
C. jejuni	ST-21	21	1	Susceptible
	ST-21	21	1	Amp Azi Cip Gen Nal Tet
	ST-21	21	2	Amp Azi Cip Nal Tet
	ST-21	21	1	Amp Cip Nal
	ST-21	21	3	Amp Cip Nal Tet
	ST-21	21	1	Amp Ery Tet
	ST-21	21	2	Cip Nal Tet
	ST-21	21	1	Nal
	ST-21	50	1	Amp Azi Cip Ery Nal Tet
	ST-21	50	1	Amp Azi Cip Nal Tet
	ST-21	50	1	Cip Nal Tet
	ST-21	4253	1	Amp Cip Cli Gen Nal Tet
	ST-21	4253	1	Cip Nal Tet
	ST-45	45	1	Amp Azi Tet
	ST-45	45	2	Amp Cip Nal Tet
	ST-45	45	1	Amp Tet
	ST-45	45	1	Cip Nal
	ST-45	45	1	Cip Nal Tet
	ST-45	4852	1	Amp Cip Tet
	ST-45	4852	1	Cip Nal Tet
	ST-354	3784	1	Amp Azi Cip Cli Ery Gen Nal Tet
	ST-354	3784	1	Amp Cip Nal
	ST-692	4512	1	Cip Nal Tet
	ST-692	4536	1	Azi Cip Cli Ery Nal Tet
	ST-692	7016	1	Cip Nal Tet
	ST-1034	1034	1	Amp Azi Cip Nal
	ST-1034	3266	1	Amp Cip Nal Tet
	ST-1034	5231	1	Amp Cip Nal Tet
	ST-1332	696	1	Amp Cip Nal Tet
	ST-1332	696	1	Cip Nal Tet
	ST-1332	1332	1	Cip
	ST-1332	5154	1	Amp Cip Nal Tet
	ST-1332	5154	1	Tet
	U	5096	1	Cip Nal Tet
	U	5857	1	Amp Cip Gen Nal Tet
	U	5857	2	Amp Cip Nal Tet
	U	5917	2	Amp Cip Nal Tet
	U	6149	1	Amp Cip Nal Tet
	U	7015	1	Amp Azi Cip Ery Nal Tet
C. coli	ST-828	889	1	Amp Azi Cip Cli Ery Gen Nal Tet
	ST-828	889	1	Amp Cip Nal Tet
	ST-828	1593	1	Amp Azi Ery Tet
	ST-828	2075	1	Amp Cip Nal Tet
	ST-828	2397	1	Amp Cip
	ST-828	4291	1	Amp Cip Nal Tet
	ST-828	6582	1	Amp Tet
	ST-828	7014	1	Amp Azi Cip Ery Nal Tet
	U	6148	1	Amp Cip Nal Tet

^aU, no clonal complex assigned yet.

^bNew STs are shown in bold.

In conclusion, our results highlight the high level of contamination by Campylobacter in South Korean duck farms and the high resistance to antimicrobials in the fluoroquinolone family and show that South Korean ducks are a potentially important source of human infection. These results indicate the need to strengthen implementation of specific control procedures to decrease Campylobacter contamination of ducks.