Antimicrobial resistance among Campylobacter spp. strains isolated from different poultry production systems at slaughterhouse level

Abstract

The aim of the current work was to evaluate the prevalence and antimicrobial susceptibility of Campylobacter spp. isolated from different chicken production systems at the slaughterhouse level. Chicken sampling at slaughterhouse was performed for cecum, carcass, and breast meat from flocks of organic (n = 6), extensive indoor (n = 14), and intensive production (n = 14), totaling 34 ceca pools, 64 neck skin pools, and 132 breasts, representing 96,386 chickens. A collection of 167 strains were identified as Campylobacter coli (n = 85) and Campylobacter jejuni (n = 82) and were tested for susceptibility to 11 antimicrobial agents by the disk diffusion method. The frequency of Campylobacter in chicken samples from different production systems was between 79 and 100%. Campylobacter isolated from all origins were resistant to the fluoroquinolones studied (80–98%). However, for ciprofloxacin and ofloxacin, the Campylobacter isolates from extensive indoor chicken were significantly (P < 0.05) less resistant (77 and 58%) than that from organic (97 and 91%) and intensive production (96 and 95%). A high probability of tetracycline resistance occurrence was also found for the Campylobacter spp. tested (58% for C. jejuni and 76% for C. coli). A more frequent profile of multidrug resistance was noticed for isolates from intensive and organic production than for extensive indoor production. These results reinforce the need of efficient strategy implementation to control and reduce Campylobacter in chickens at production and slaughter levels, and the necessity to reduce the use of antimicrobials in poultry sector.

INTRODUCTION

Campylobacteriosis remains one of the most relevant causes of food-borne disease in humans, with a continuous increase in reported cases over the last 5 yr (World Organization for Animal Health, 2008; Food and Agriculture Organization of the United Nations and World Health Organization, 2009; Centers for Disease Control and Prevention, 2013; European Food Safety Authority, 2013). It is frequently associated with poultry products, namely the consumption of undercooked poultry and the handling of raw poultry (Gblossi Bernadette et al., 2012; Niederer et al., 2012). Although the infection is self-limited in most cases, people with febrile diarrheal illness can benefit from antibiotic treatments, especially when administered early in the course of the illness (Ternhag et al., 2007).

This disease poses additional health risks besides acute gastro-enteritis, including chronic illness such as polyarthropathies or neuropathies (Guillain-Barré syndrome and reactive arthritis). Antimicrobial therapy is imperative when long-lasting or systemic Campylobacter infections occur or when infection arises in immunosuppressed patients (Aarestrup and Engberg, 2001; Allos, 2001; World Organization for Animal Health, 2008; European Food Safety Authority, 2013; Iovine, 2013). When necessary, 2 major classes of antibiotics are currently used because of their large spectra activity on enteric pathogens (Gblossi Bernadette et al., 2012; Iovine, 2013; Xia et al., 2013) and are taken under consideration regarding the treatment of Campylobacter-induced infections, specifically macrolides (erythromycin) and fluoroquinolones (e.g., ciprofloxacin).

Broiler meat has become one of the most consumed meats, and a great diversity of products with different quality claims exist. Apart from intensive broiler production, other production systems have grown following consumer preferences, such as extensive and organic productions. The consumers’ perception of rearing practices is that organic or extensive chickens are of better quality or safer than intensively reared chickens; this opinion influences their choice of purchase (Thibodeau et al., 2011). However, according to Rosenquist et al. (2013), the production system requirements and the chicken poultry involved, particularly those of free-range and outdoors systems, can contribute as risk factors for an increased Campylobacter prevalence. The highest prevalence of Campylobacter among organic broilers is believed to be a result of the birds’ exposure to the outdoor environment, as well as their older slaughter age. The extensive indoor system is characterized by the availability of more space per bird (the stocking rate per meter-squared of floor space does not exceed 15 birds/m² in the case of chickens 15 birds) than the intensive system, along with a heavier bird variety and a slaughter age of 56 d or later. According to European legal requirements for organic broiler production, EC 834/2007 and EC 889/2009 (European Union, 2007, 2009), antibiotics may be used where necessary and under strict conditions when the use of phytotherapeutic, homeopathic, and other products is inappropriate. Recent surveys indicated that antimicrobial resistance among Campylobacter spp. is growing (Gblossi Bernadette et al., 2012; Iovine, 2013). Though most studies carried out have been associated with strains of Campylobacter isolated either from humans or from intensively reared chickens, the information comparing isolates from other chicken-production systems is scarce.

In this context, the current work was conducted to evaluate the prevalence of Campylobacter spp. at the slaughterhouse level from chickens of different production systems (intensive, extensive indoor, and organic) in Portugal, and to assess the antimicrobial resistance of the Campylobacter strains isolated. The incidence of antimicrobial, multiresistant Campylobacter from different poultry production systems was determined.

MATERIALS AND METHODS

Sampling and Isolation of Campylobacter

Different chicken flocks from organic (n = 6), extensive indoor (n = 14), and intensive production (n = 14), representing a total of 96,386 birds, were sampled during the slaughter and deboning process from December 2007 to October 2008, according to ISO 17604:2003 (International Organization for Standardization, 2003) and the guidelines of The National Veterinary Institute (2010). Sampling was performed for cecum, carcass (neck skin), and deboned poultry breast pieces, maintaining each flock’s traceability. For each chicken flock, a pool of 10 ceca was randomly sampled, and the neck skins of 5 carcasses, after the inside-outside shower and before rapid cooling, were collected twice constituting 2 pool samples. A total of 34 ceca pools and 64 neck skin pool samples were collected. Four carcasses per flock were marked and, after deboning (performed according to industrial practices), their breasts were sampled (n = 132) and delivered to the laboratory in a commercial refrigerated vehicle.

Campylobacter isolation was performed according to ISO 10272–1:2006 (International Organization for Standardization, 2006). Bacterial strains were grown and purified on Columbia Blood agar (BioMerieux, Marcy l’Etoile, France), supplemented with 5% horse blood and incubated at 42°C for 48 h under microaerobic conditions (5% O₂, 10% CO₂, and 85% N₂) in gas-tight containers. Identification was performed by hippurate hydrolysis allowing an initial differentiation between Campylobacter jejuni and non-C. jejuni strains. Campylobacter jejuni and Campylobacter coli identification was performed after DNA extraction with a Qiagen DNA-extraction Kit (Qiagen, Valencia, CA) by multiplex-PCR according to Denis et al. (1999). Strains were cryogenically preserved at −80°C (ThermoFisher Scientific, Waltham, MA) in brain heart infusion broth (Scharlau, Barcelona, Spain) with 15% glycerol (Scharlau).

Reference Strains

The reference strains C. jejuni NCTC11168 were kindly provided by Naoaki Misawa from the University of Miyazaki, Miyazaki, Japan, and C. coli ZIM 140 were kindly provided by Sonja Možina from the University of Ljubljana, Ljubljana, Slovenia. Also, Staphylococcus aureus ATCC 25923 and Enterococcus faecalis ATCC 29212 were used as controls.

Disc Diffusion Method

Disc diffusion susceptibility testing was performed according to the Comite de L’antibiogramme de la Société Française de Microbiologie (2012) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST, 2013) for 11 antibiotics of 7 different antibiotic classes: ampicillin, 10μg (AM); erythromicin,15 μg (ERI); tetracycline, 30 μg (TET); chloramphenicol, 30 μg (CHL); gentamicin, 10 μg (GEN); ciprofloxacin, 5 μg (CIP); nalidixic acid, 10 μg (NA); norfloxacin, 5 μg (NOR); ofloxacin, 5 μg (OFX); amoxacillin + clavulanic acid, 20 and 10 μg, respectively (AMC); and trimethoprim + sulfamethoxazole, 25 μg (STX; BioMérieux). Quality control strains Staph. aureus ATCC 25923, E. faecalis ATCC 29212, and C. jejuni NCTC11168, for which quality control ranges have been established, were included in the tests (according to Clinical and Laboratory Standards Institute, 2008). All bacterial isolates were grown on Mueller-Hinton agar (Scharlau) supplemented with 5% horse or sheep blood (BioMérieux). Campylobacter isolates were incubated at 42°C for 48 h under microaerobic conditions (5% O₂, 10% CO₂, and 85% N₂) in gas-tight containers, whereas quality control strains were incubated at 37°C for 24 h. Strains susceptibility categorization was carried out according to Comite de L’antibiogramme de la Société Française de Microbiologie (2012).

Statistical Analysis

The frequency of Campylobacter detection in the samples was calculated. Binary data of isolated strains’ resistance to several antibiotics (except to CHL and to GEN) were analyzed using Proc GLIMMIX of SAS 9.3 (SAS Institute Inc., Cary, NC) with the model using logit as a link function and the effect of production system and Campylobacter spp. as a fixed effect to model the probability of the occurrence of antibiotic resistance. Multiple comparisons of means were conducted with a Tukey adjustment and significance was declared at P < 0.05. The ILINK option of the LSMEANS statement of Proc GLIMMIX was used to back transform the logits into probabilities. No Campylobacter isolates from the intensive production system were resistant to CHL or GEN, resulting in a complete separation of data points that prevented the use of the previous model. For these variables, the effect of Campylobacter spp. was analyzed using Proc GLIMMIX with a simplified model that included only Campylobacter spp. as a fixed effect. The effects of the production system were analyzed using Proc LOGISTIC with the EXACT statement and the unbiased median estimates.

RESULTS

Prevalence of Campylobacter in Different Poultry-Production Systems

The prevalence of Campylobacter detection in flocks of chicken (n = 6, representing 2,634 birds on total) from organic production systems was high (100%), with predominance of C. coli isolated from ceca pool samples (Table 1). However, from the process of slaughter, the presence of both C. jejuni and C. coli was revealed on carcasses collected from different flocks (83%). A similar result was seen after the process of carcass deboning, with the presence of C. jejuni and C. coli on breasts collected from the 6 flocks.

Table 1.

Frequency of Campylobacter isolation on different samples collected from poultry of organic production system of different flocks and producers

				Sample
Item	Flock	No. of birds		Cecum		Neck		Breast
Producer
A	1	504		Campylobacter coli		Campylobacter jejuniand C. coli		C. jejuni
A	2	400		C. coli		C. coli		C. jejuniand C. coli
B	3	449		C. coli		C. jejuniand C. coli		C. jejuniand C. coli
C	4	520		C. coli		C. jejuniand C. coli		C. jejuni
D	5	257		C. coli		C. jejuniand C. coli		C. jejuniand C. coli
E	6	504		C. coli		ND1		C. jejuniand C. coli
Total	6	2,634		6		12		24
Detection (%)				100		83		100

¹ND = not detected.

The extensive production system of chicken flocks sampled represents 20,365 birds slaughtered (Table 2), and the presence of Campylobacter was detected in 79% of the ceca pool samples analyzed. Campylobacter coli were more frequently found in these flocks, but C. jejuni also coexists in part of the flocks analyzed. The frequency of Campylobacter spp. detected on neck samples of chicken carcass and breast samples of different flocks was high (86 and 100%, respectively).

Table 2.

Frequency of Campylobacter isolation on different samples collected from poultry of extensive production system of different flocks and producers

				Sample
Item	Flock	No. of birds		Cecum		Neck		Breast
Producer
A	1	941		Campylobacter coli		C. coli		Campylobacter jejuni
B	2	560		ND1		C. coli		NA1
C	3	1,524		C. jejuniand C. coli		C. jejuniand C. coli		C. jejuni
C	4	1,498		C. jejuniand C. coli		C. jejuniand C. coli		C. jejuni
C	5	160		C. jejuniand C. coli		C. jejuniand C. coli		C. jejuniand C. coli
D	6	2,344		C. jejuniand C. coli		C. jejuni		C. jejuniand C. coli
E	7	3,036		C. jejuniand C. coli		C. jejuniand C. coli		C. jejuniand C. coli
F	8	1,504		ND		C. jejuniand C. coli		C. jejuni
G	9	1,064		C. coli		C. coli		C. jejuni
H	10	1,512		C. coli		ND		C. jejuniand C. coli
E	11	1,080		C. coli		C. jejuni		C. jejuni
F	12	978		ND		C. coli		C. jejuniand C. coli
G	13	1,500		C. coli		ND		C. jejuniand C. coli
H	14	2,664		C. coli		C. coli		C. jejuniand C. coli
Total	14	20,365		14		28		52
Detection (%)				79		86		100

¹ND = not detected; NA = not analyzed.

Table 3 presents the frequency of Campylobacter isolation in chicken flocks of intensive production systems, representing 73,387 birds slaughtered. Seventy-nine percent of the ceca pooled from broiler flocks were positive for Campylobacter spp. The frequency of these pathogens on sampled flock carcasses analyzed was 100%, but after their deboning the breasts from different flocks were 93% positive for C. jejuni or C. coli.

Table 3.

Frequency of Campylobacter isolation on different samples collected from poultry of intensive production system of different flocks and producers

				Sample
Item	Flock	No. of birds		Cecum		Neck		Breast
Producer
1	1	5,616		Campylobacter coli		C. coli		C. coli
2	2	2,016		Campylobacter jejuniand C. coli		C. jejuniand C. coli		ND1
3	3	4,030		C. coli		C. jejuniand C. coli		C. jejuniand C. coli
4	4	5,200		C. jejuniand C. coli		C. jejuni		C. jejuniand C. coli
4	5	6,000		C. coli		C. jejuni		C. jejuniand C. coli
5	6	5,076		C. jejuni		C. jejuni/C. coli		C. jejuniand C. coli
5	7	7,600		ND		C. jejuni		C. jejuni
6	8	6,000		ND		C. coli		C. coli
5	9	4,710		C. coli		C. coli		C. jejuniand C. coli
7	10	5,616		C. coli		C. jejuni		C. jejuniand C. coli
8	11	6,048		ND		NA1		C. coli
9	12	4,665		C. coli		NA		C. jejuniand C. coli
10	13	5,520		C. jejuniand C. coli		C. coli		C. jejuniand C. coli
11	14	5,290		C. jejuni		C. jejuniand C. coli		C. jejuni
Total	14	73,387		14		24		56
Detection (%)				79		100		93

¹ND = not detected; NA = not analyzed.

In 76% of breast samples analyzed from different production systems (n = 132), Campylobacter spp. were present in 25 g, with C. jejuni being the main species isolated. Forty-nine percent of the breast samples were positive for C. jejuni and 27% were positive for C. coli. It was observed that 10% of samples were simultaneously positive for C. jejuni and C. coli.

Antimicrobial Resistance of Campylobacter Isolates from Different Poultry-Production Systems

The antimicrobial resistance probability occurrence for C. jejuni and C. coli isolated on chicken poultry samples from different production systems is presented in Table 4. It was noticed that C. coli typically tends to be more resistant to the antimicrobials tested than C. jejuni. However, except for the antimicrobials ERI and NOR, no significant difference was observed between the isolates of the species C. jejuni and C. coli analyzed. The highest probability of antimicrobial resistance occurrence on C. jejuni and C. coli were noticed for quinolones class tested, such as NA (92 and 98%), CIP (90 and 96%), NOR (80 and 95%; P < 0.05), OFX (81 and 91%).

Table 4.

Antimicrobial resistance probability of occurrence for Campylobacter jejuni and Campylobacter coli isolates from poultry samples of different production systems

	C. jejuni(n = 82)			C. coli(n = 85)
Antimicrobial (%)	Mean	SEM		Mean	SEM	P-value
Ampicillin	67.2	7.70		81.8	5.51	0.07
Amoxacillin + clavulanic acid	15.9	7.09		32.6	10.70	0.10
Ciprofloxacin	89.6	4.59		95.5	2.62	0.20
Norfloxacin	80.1	6.86		95.2	2.74	0.02
Ofloxacin	80.5	6.20		91.2	3.80	0.10
Nalidixic acid	91.5	5.30		98.0	1.70	0.10
Erythomicin	35.4	6.01		13.3	10.66	0.01
Tetracycline	57.8	9.85		76.4	7.39	0.06
Trimethoprim + sulfamethoxazole	34.2	6.32		47.1	6.73	0.14
Chloramphenicol1	17.9	4.39		11.9	3.58	0.29
Gentamicin1	21.0	4.58		8.5	3.12	0.03

¹Median unbiased estimate obtained from the exact conditional analysis.

From the penicillin group, the AM resistance prevalence for C. jejuni and C. coli was high (67 and 82%), whereas the penicillin combination of AMC substantially decreased the resistance occurrence of these pathogens (16 and 33%). Also, a high probability of TET resistance occurrence was revealed for the Campylobacter spp. tested (58 and 76%).

The STX tested on C. jejuni and C. coli had a moderate resistance prevalence (34 and 47%). Whereas for the macrolide ERI tested on C. jejuni presented a higher (P < 0.01) resistance probability occurrence (35%) than C. coli (13%). The Campylobacter spp. developed resistance to CHL (18 and 12%) and GEN (21 and 9%).

From Table 5, the Campylobacter isolates from organic chicken production revealed resistance prevalence against the antimicrobial under study, similar to that showed by intensive production isolates. In general, the isolates from extensive indoor chicken production were less resistant to antimicrobials than the other production systems; nevertheless, resistance occurrence among that group always greater than 40%.

Table 5.

Antimicrobial resistance probability of occurrence for Campylobacter isolates from poultry samples from different production systems

	Production system (n = 167)
	Organic (n = 43)			Extensive indoor (n = 41)			Intensive (n = 83)
Antimicrobial (%)	Mean	SEM		Mean	SEM		Mean	SEM	P-value
Ampicillin	77.8	10.29		74.8	9.69		72.9	7.97	0.93
Amoxacillin + clavulanic acid	43.0b	19.66		56.7b	15.70		2.7a	2.12	0.01
Ciprofloxacin	96.5ab	3.25		77.4a	9.59		96.4b	2.46	0.05
Norfloxacin	91.3	6.32		74.0	10.74		95.9	2.70	0.07
Ofloxacin	90.5b	5.86		58.3a	11.31		95.4b	2.72	0.01
Nalidixic acid	97.3	3.54		87.7	9.51		97.9	2.12	0.34
Erythromicin	26.8	17.12		16.2	10.40		25.6	11.73	0.81
Tetracycline	73.6ab	13.49		40.8a	13.81		82.9b	7.36	0.05
Trimethoprim + sulfamethoxazole	26.4	8.54		48.0	9.47		48.6	7.22	0.18
Chloramphenicol1	47.0b	—		7.1b	—		0.8a	—	0.001
Gentamicin1	13.2b	—		44.2b	—		0.8a	—	0.001

^a,bMean values on the same row not sharing a common superscript were significantly different.

¹Median unbiased estimate obtained from the exact conditional analysis.

The prevalence of resistance of Campylobacter isolated from all origins was 80 to 90% for the fluoroquinolones studied. However, for CIP and OFX, the Campylobacter isolates from chicken raised in the extensive indoors system were significantly less resistant (P < 0.05) than those from organic and intensive production.

Most of Campylobacter isolates were resistant to AM independent of its origin (72 to 77%), with an observed decrease of resistance from 56 and 42% with AMC on isolates from extensive and organic poultry production. The isolates from intensive chicken production were very susceptible to this penicillin combination (AMC), with a significant reduction in resistance incidence (3.0%; P < 0.01).

Tetracycline resistance was present in organic (73%) and intensive (82%) Campylobacter, with significantly less probability of occurrence (P < 0.05) for those from extensive indoor production (40%). All Campylobacter from different origins had a minor resistance probability occurrence for ERI (26–16%), whereas the values for STX were between 26 and 48%. The Campylobacter from intensive production were not resistant to CHL and GEN, but 44% of those from extensive indoor production presented resistance for GEN and 47% from organic production were resistant to CHL.

Antimicrobial Multiresistance of Campylobacter

The occurrence of a simultaneous resistance to various antibacterials for C. jejuni and C. coli was frequent (Table 6), illustrating a frequency higher than 20% of resistance to 3, 4, and 5 classes of key antimicrobials tested, particularly to C. coli. Campylobacter jejuni and C. coli isolates were 72 and 87% resistant for all fluoroquinolones tested, respectively. The resistance presented by C. jejuni and C. coli for 2 classes of antimicrobials were higher when TET (52 and 73%, respectively) was included than with ERI (18 and 35%). The profile of resistance including 3, 4 and 5 classes of antimicrobials for C. coli is higher than 20%.

Table 6.

Distribution of Campylobacter jejuni and Campylobacter coli according to their antimicrobial resistance profile

		C. jejuni			C. coli
Antimicrobial resistance profile (%)	Antimicrobial class	No. (n = 82)	%		No. (n = 85)	%
Nalidixic acid (NA), ciprofloxacin (CIP)	1	68	82.9		79	92.9
NA, CIP, norfloxacin (NOR)	1	62	75.6		77	90.6
NA, CIP, NOR, ofloxacin (OFX)	1	59	72.0		74	87.1
NA, CIP, NOR, OFX, tetracycline (TET)	2	43	52.4		62	72.9
NA, CIP, NOR, OFX, erythromycin (ERI)	2	15	18.3		32	37.6
NA, CIP, NOR, OFX, ERI, ampicillin (AM)	3	15	18.3		30	35.3
NA, CIP, NOR, OFX, ERI, TET	3	13	15.9		29	34.1
NA, CIP, NOR, OFX, ERI, TET, AM	4	13	15.9		27	31.8
NA, CIP, NOR, OFX, ERI, TET, trimethoprim + sulfamethoxazole (STX)	4	6	7.3		17	20.0
NA, CIP, NOR, OFX, ERI, TET, AM, STX	5	6	7.3		17	20.0
NA, CIP, NOR, OFX, ERI, TET, amoxicillin + clavulanic acid (AMC)	4	5	6.1		6	7.1
NA, CIP, NOR, OFX, ERI, TET, AMC, AM, STX	6	3	3.7		4	4.7
All antimicrobials tested	7	0	0.0		0	0.0

Campylobacter jejuni presented 16% of isolates with a profile of antimicrobial resistance to NA, CIP, NOR, OFX, ERI, TET, and AM. The addition of STX to this last profile described induces a decrease on isolates multidrug resistance (7%).

The distribution of Campylobacter according to their antimicrobial profile on resistance is different for the various production systems (Table 7). Intensive production Campylobacter isolates were highly resistant to all fluoroquinolones tested (94%), followed by those from the organic production system (81%). Only 48% of extensive production isolates were resistant to all fluoroquinolones, presenting lower rates (less than 5%) of multidrug resistance for 3 or more classes of antimicrobials.

Table 7.

Distribution of Campylobacter spp. strains from different production systems according to their antimicrobial resistance profile

		Organic			Extensive indoor			Intensive
Antimicrobial resistance profile (%)	Antimicrobial class	No. (n = 43)	%		No. (n = 41)	%		No. (n = 83)	%
Nalidixic acid (NA), ciprofloxacin (CIP)	1	37	86.0		30	73.2		80	96.4
NA, CIP, norfloxacin (NOR)	1	36	83.7		24	58.5		79	95.2
NA, CIP, NOR, ofloxacin (OFX)	1	35	81.4		20	48.8		78	94.0
NA, CIP, NOR, OFX, tetracycline (TET)	2	28	65.1		12	29.3		65	78.3
NA, CIP, NOR, OFX, erythromycin (ERI)	2	11	25.6		6	14.6		30	36.1
NA, CIP, NOR, OFX, ERI, ampicillin (AM)	3	11	25.6		6	14.6		28	33.7
NA, CIP, NOR, OFX, ERI, TET	3	10	23.3		2	4.9		30	36.1
NA, CIP, NOR, OFX, ERI, TET, AM	4	10	23.3		2	4.9		28	33.7
NA, CIP, NOR, OFX, ERI, TET, trimethoprim  + sulfamethoxazole (STX)	4	5	11.6		1	2.4		17	20.5
NA, CIP, NOR, OFX, ERI, TET, AM, STX	5	5	11.6		1	2.4		17	20.5
NA, CIP, NOR, OFX, ERI, TET, amoxicillin  + clavulanic acid (AMC)	4	8	18.6		2	4.9		2	2.4
NA, CIP, NOR, OFX, ERI, TET, AMC, AM, STX	6	4	9.3		1	2.4		2	2.4
All antimicrobials tested	7	0	0.0		0	0.0		0	0.0

The profile of antimicrobial multiresistance to NA, CIP, NOR, OFX, ERI, TET, AM, and STX was presented by 21% of the Campylobacter from intensive production and 12% from those of organic production. In general, Campylobacter isolates presented a profile of multidrug resistance with an incidence of 21 to 36% for intensive poultry-production systems and 12 to 27% for organic poultry-production systems.

DISCUSSION

Prevalence of Campylobacter in Different Poultry-Production Systems

The actual number of estimated Campylobacteriosis cases from the ones reported in the European Union is believed to be around 9 million each year, with a high cost related to public health systems and loss of productivity (European Food Safety Authority, 2013). The European Food Safety Authority (2013) reported that chickens and chicken meat may directly account for 50 to 80% of human cases. In the current study it was noticed that the frequency of Campylobacter in samples collected at the slaughterhouse level was extremely high independent of the type of poultry-production system. In the recent past (2008–2010), a high frequency of Campylobacter has also been noted in flock-based data from Spain, Slovenia, Poland, the United Kingdom, and Portugal (European Food Safety Authority, 2013; Carreira et al., 2012).

The slaughter line for chickens from organic and extensive production differs from that used to slaughter chickens from intensive production; in spite of that fact the ceca samples from the majority of the flocks were positive. These results suggest that the consumption of organic or extensive production chicken might not decrease consumer risk of exposure to Campylobacter spp. The entry point to the slaughterhouse of flocks positive for Campylobacter is the main reason for the pathogen dissemination all over the line’s plant (Reich et al., 2008; Hue et al., 2010).

At the slaughterhouse, the recovery of C. jejuni from samples increased as we progressed along the slaughter line; therefore, in more-processed samples (breast samples), the probability of isolation was higher than in less-processed samples (carcass neck samples). The cross contamination of these pathogens from one step to another on the slaughter line (scalding, defeathering, evisceration, carcass chilling, deboning), along with the hazard analysis critical control point plan implemented in the plant, is important (Hue et al., 2010). The persistence of C. jejuni in the slaughter environment is possibly due to their capacity for resistance even under oxidative stress or low temperatures and improved adaptation to biofilm formation (Gundogdu et al., 2011; Sulaeman et al., 2012). It was observed on breast samples that the dominant species was C. jejuni cohabitant with C. coli in 10% of the samples analyzed.

From the results obtained it seems that the urgent application of biosecurity measures, such as phage therapy, vaccination, or bacteriocines, used in feed or water at the production level are required to decrease Campylobacter prevalence and the risk level of Campylobacteriosis (European Food Safety Authority, 2013). This type of intervention could have particular importance in controlling Campylobacter on outdoor broiler production, such as organic production systems (Rosenquist et al., 2013).

Antimicrobial Resistance of Campylobacter Isolates from Different Poultry-Production Systems

The resistance prevalence for NOR and NA was high but similar when considering Campylobacter isolates from organic, extensive indoor, or intensive production. The level of resistance to most fluoroquinolones (CIP and OFX) was higher for organic or intensive production than for extensive indoor production. The very large prevalence of fluoroquinolones resistance for C. jejuni and C. coli was higher than the resistance level reported by European Food Safety Authority (2013) for CIP and NA. Fluoroquinolone-resistant Campylobacter spp. rapidly emerged as a consequence of the use of enrofloxacin in poultry production, which has probably exerted a selective pressure in animal reservoirs (Soonthornchaikul et al., 2006; Gblossi Bernadette et al., 2012). In Portugal, the selective pressure seems to be high once quinolones are used in veterinary medicine. This fact poses a major problem because Campylobacter isolates from humans were reported to be highly resistant to fluoroquinolones (Vicente et al., 2008).

The resistance prevalence for ERI was not significantly different when considering Campylobacter isolates from organic, extensive indoor, or intensive production. The current study reported a minor resistance to ERI similar to levels found in Italy (23% for C. coli; Parisi et al., 2007).

In general, C. coli tend to be more resistant to the antimicrobials tested. However, a significantly higher resistance to macrolides was found in C. jejuni, contrary to other reports (Alfredson and Korolik, 2007; Parisi et al., 2007). In spite of the fact that pathogen resistance to ERI has been increasing due to selective pressure at the production level (Gibreel and Taylor, 2006), the incidence of resistance in human Campylobacter isolates is still relatively low, and ERI could be regarded as the drug of choice in Campylobacteriosis treatment (Xia et al., 2013).

The high resistance prevalence for AM was similar when considering Campylobacter isolates from different production systems, suggesting that it might be used at production level. The AMC decreased Campylobacter resistance prevalence substantially (15 and 32%), with relevance for those from intensive production. Campylobacter strains are resistant to a large number of β-lactam antimicrobial agents, and clavulanic acid brought the susceptibility levels of β-lactamase-positive strains down to those of β-lactamase negative strains (Lachance et al., 1991; Alfredson and Korolik, 2007).

The prevalence of TET resistance for Campylobacter spp. was particularly high on intensive production, higher than in other studies where rates ranged from 50 to 68% (Van Looveren et al., 2001; Guévremont et al., 2006). This resistance to TET may be mediated by a self-transmissible plasmid encoding the gene tet(O) (Iovine, 2013), but the emergence and spread of TET resistance in Campylobacter from intensive production is likely driven by the selection pressure from antibiotic usage.

The prevalence of sulfametoxazole resistance of Campylobacter was lower than reported by Nobile et al. (2013). This resistance can be also related to the acquisition of horizontally transferred dfrgenes found on the chromosome in transposons or integrons, as has been reported by Lucey et al. (2000) and Gibreel and Skold (2000).

Campylobacter isolates from extensive indoor production presented resistance to GEN, whereas those from organic production were resistant to CHL. Despite the prohibition of CHL application in food animals, the antibiotic resistance in outdoor production could result from a selective pressure induced by the use of a synthetic amphenicol (florfenicol), causing cross-resistance or mobile genes to become widespread in nature owing to horizontal gene transfer (Tao et al., 2012).

Chen et al. (2013) reported an increased aminoglycoside resistance in Campylobacter, which accounted for 11.3% of human isolates and 12.5% of retail meat isolates in 2010. Gentamicin-resistant gene clusters on transmissible plasmids in C. jejuni and a genomic island in C. coli confer resistance to multiple aminoglycoside antibiotics (Qin et al., 2012).

Antimicrobial Multiresistance of Campylobacter

A more frequent profile of multidrug resistance was noted for isolates from intensive and organic production than from extensive indoor production. The high density used in intensive broiler production and the rearing conditions on outdoor organic production increase the risk of illness for both populations (Rosenquist et al., 2013), resulting in a subsequent increase of antibiotic use with a therapeutic objective. Therefore, the selection pressure in these animal reservoirs explains the high frequency of multidrug resistance presented by Campylobacter isolates from intensive and organic production. The rearing conditions could also give more exposure of Campylobacter to other gram-positive or -negative bacterial species resistant to antibiotics, with a potential increase of horizontal transmission of resistance genes carried on mobile elements, such as plasmids, phages, or other genetic mobile elements, being one cause of acquired resistance and contributing to the spread of antibiotic resistance determinants (Luangtongkum et al., 2009).

High levels of multiresistance to 3 antibiotics (ERI, CIP, and NA) in Campylobacter isolates on chicken meat at the retail level, from intensive and organic rearing systems, were also found by Soonthornchaikul et al. (2006). It is important to highlight the survivability and the spread of multiresistant Campylobacter strains, which mostly originated from intensive and organic production systems, and its implications on human infection and disease treatment.

The current study reported a high level of contamination in poultry at the slaughterhouse level by Campylobacter and also provided evidence that antimicrobial resistance is common among strains isolated from organic and intensive poultry production systems in Portugal, particularly between fluoroquinolones. These results reinforce the need to develop strategies to reduce Campylobacter in chickens, and the necessity to reduce the use of antimicrobials in the poultry sector and implement specific control procedures to decrease the level of contamination in poultry meat by Campylobacter strains.

The authors wish to thank the Campylobacter and E. coli—A network project, CampEc-NET (SAFEFOODERA 006465, ERA-FOOD/0001/2006, European Union) for its financial support and the Centro de Investigação Interdisciplinar em Sanidade Animal for logistic support.