sign in sign up
<<
Home , Listeria

735

Journal of Protection, Vol. 62, No. 7, 1999, Pages 735±740 Copyright ᮊ, International Association of Milk, Food and Environmental Sanitarians

Incidence of , , Campylobacter coli, and in Poultry Carcasses and Different Types of Poultry Products for Sale on the Belgian Retail Market

M. UYTTENDAELE,1* P. DE TROY,2 AND J. DEBEVERE1 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 1Laboratory of and , Department of and Nutrition, University of Ghent, Coupure Links 653, 9000 Gent, Belgium; and 2Delhaize±The Lion, Bettegemlaan, 1730 Zellik, Belgium

MS 98-219: Received 26 August 1998/Accepted 21 December 1998

ABSTRACT

From January 1997 to May 1998, 772 samples of poultry carcasses and poultry products for sale on the retail market in Belgium were analyzed for the presence of Salmonella spp., Salmonella Enteritidis, Campylobacter jejuni, C. coli, and Listeria monocytogenes per 100 cm2 or 25 g. Poultry samples were contaminated with Salmonella (36.5%), C. jejuni and C. coli (28.5%), and L. monocytogenes (38.2%). In about 12.3% of the poultry samples, the L. monocytogenes contamination level exceeded 1 CFU per g or cm2. Signi®cant differences in pathogen contamination rates of poultry products were noticed between the poultry products originating from Belgian, French, and U.K. abattoirs. Poultry products derived from broiler chickens running free in pine woods until slaughtering age (12 to 13 weeks) had a signi®cantly (P Ͻ 0.05) lower contamination rate of Salmonella than poultry products from enclosed broilers slaughtered at the age of 6 to 8 weeks. A signi®cantly (P Ͻ 0.05) lower pathogen contamination rate was noted for Salmonella, C. jejuni, and C. coli for poultry cuts without skin compared to poultry cuts with skin on. An increase in pathogen contamination rate was noticed during cutting and further processing. To diminish C. jejuni, C. coli, Salmonella, and L.monocytogenes contamination rates, hygienic rules of slaughter and meat processing must be rigorously observed. At the moment, zero tolerance for these pathogens is not feasible, and there is a need to establish criteria allowing these pathogens to be present at reasonable levels in the examined poultry samples.

Risk assessment is a tool used to evaluate the safety occur as a result of manipulation of raw poultry in the of food. Among the classes of hazards associated with food, kitchen. microbial contamination of the food supply is among the Information on contamination rates of poultry prod- most important. Risk assessment, however, is limited by a ucts for the above-mentioned pathogens is needed to char- lack of data on the incidence of pathogens in (13). acterize exposure in health risk assessment. This study was Poultry and poultry products are a common vehicle of conducted to estimate the incidence of Salmonella, C. je- . Microbial risks associated with raw poul- juni, C. coli, and L. monocytogenes on poultry carcasses try products include Salmonella spp., Campylobacter jeju- and poultry products distributed on the Belgian retail mar- ni, C. coli, and Listeria monocytogenes. Outbreaks involv- ket. ing large numbers of people are usually caused by Salmo- MATERIALS AND METHODS nella. Campylobacteriosis occurs as sporadic cases of ill- ness instead of outbreaks. Campylobacter are Sample collection. From January 1997 to May 1998, 772 increasing in Europe and other parts of the world (22, 27) samples of poultry carcasses and poultry products were analyzed and are the most common cause of human diarrhea in Eng- for the presence of Salmonella, Salmonella Enteritidis, C. jejuni, land and Wales (24, 28) and the United States (29). Eating C. coli, and L. monocytogenes. Samples were taken monthly at undercooked chicken is one risk factor associated with spo- the depot of a large chain of supermarkets in Belgium. At the central depot, chilled poultry carcasses and products were arriving radic campylobacteriosis (22). Worldwide, salmonellae and from several abattoirs in Belgium (10 sites), France (16 sites), and campylobacters are by far the most important pathogens occasionally the (2 sites), The Netherlands (1 associated with poultry products (7). Case±control studies, site), or Italy (1 site). From the central depot on, the chilled car- however, suggest that undercooking raw poultry is involved casses and products were distributed to individual supermarkets in human among individuals susceptible to the of the chain serving Belgian consumers. Neither at the central condition (14). Although suf®cient heating will eliminate depot nor at the individual supermarkets was there any processing the above-mentioned pathogens, cross-contamination can of the poultry products. Samples were collected at the central depot because this enabled gathering of samples from a large num- * Author for correspondence. Tel: ϩ32 9 264 61 78; Fax: ϩ32 9 225 55 ber of abattoirs at one point. In addition, the sampling plan rep- 10; E-mail: [email protected]. resented what is offered for consumption in the Belgian super- 736 UYTTENDAELE ET AL. J. Food Prot., Vol. 62, No. 7 market. Samples collected in the screening included (i) poultry instructions (Organon Teknika, Turnhout, Belgium). Samples de- carcasses of broiler chicken (industrial breeding, 6 to 8 weeks termined to be positive by the ELISA were con®rmed by streaking old), free-range chicken (12 to 13 weeks old), broiling hen (Յ2 M broth (which was kept at 4ЊC during the ELISA) on brilliant years old), spring chicken (4 to 6 weeks old), and guinea fowl green agar (OxoõÈd) and salmonella-shigella agar (OxoõÈd). After (Յ13 weeks old); (ii) poultry cuts, including leg, wing, breast, incubation for 24 h at 37ЊC, plates were searched for suspected and ®llets of broiler chickens and turkeys; (iii) processed poultry colonies (pink-red colonies on brilliant green agar and white col- in general products for barbecue, such as sausages and hamburgers onies with an occasional black precipitate on salmonella-shigella made of chicken or turkey meat, sliced poultry meat mounted on agar). These colonies (maximum ®ve) were subcultured to nutrient a skewer, and chicken wings coated with spices. Processed poultry agar (OxoõÈd) for purity and were con®rmed as Salmonella by in- was investigated only in the summer (May to September). oculation of a triple iron agar (OxoõÈd) tube, urea broth (Ox- Samples were taken in an aseptic manner (using sterilized oõÈd), and ®nally BBL CrystalID E/NF (Becton Dickinson, Cock- utensils under a laminar ¯ow hood). For poultry carcasses and eysville, Md., USA). Isolated Salmonella strains were sent to the cuts with skin on (e.g., chicken wings, chicken legs, and turkey National Institute for Veterinary Research (Brussels, Belgium) to legs), 300 cm2 (approximately 75 g) of skin was excised from two identify Salmonella Enteritidis, the serotype most frequently as- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 chicken carcasses or multiple chicken cuts from the same lot. sociated with Salmonella outbreaks (11). Samples of poultry carcasses consisted of skin of the wing, leg, breast, and part of the cloacal region. For poultry samples without Detection of C. jejuni and C. coli. The microbial technique skin (e.g., ®llets) and processed poultry, 75 g of meat was taken. for the investigation of C. jejuni and C. coli consisted of incu- The samples were put in a 500-ml sterile bottle and made up to bation in a jar under microaerophilic atmosphere in nutrient broth 300 ml with a saline and peptone solution consisting of 8.4 g of (OxoõÈd) supplemented with growth and selective supplement NaCl per liter (Vel, Leuven, Belgium) and 3 g of tryptone per (Preston medium, OxoõÈd) but without blood (31). After incubation liter (OxoõÈd, Hampshire, Basingstoke, UK). Samples were taken for 24 and 48 h, Campylobacter blood-free selective medium (Ox- at the central depot of the chain of supermarkets and transported oõÈd) was inoculated by streaking. The plates were searched for under refrigeration to the laboratory for analyses. typical colonies (small gray droplike or gray slimy colonies) after 48 h of incubation. Typical colonies were con®rmed as C. jejuni Microbiological analyses. In the laboratory, the 300-ml sa- or C. coli using microscopic analyses, a test, oxidase test, line and peptone sample suspension was transferred under a lam- and sensitivity to cephalotine and nalidixic acid (31). inar ¯ow hood to a sterile stomacher bag and homogenized using a Colworth stomacher (Seward Laboratory, London, UK). Detection of L. monocytogenes. Detection of L. monocyto- Next, three times 100 ml of the saline and peptone food genes was performed as described elsewhere (1). In short, demi- homogenate was transferred to a new stomacher bag and supple- Fraser broth was incubated for 24 h at 30ЊC and streaked on Ox- mented with 100 ml each of double-concentrated buffered peptone ford medium. Next, 0.1 ml demi-Fraser broth was transferred to water, double-concentrated Preston medium, or double-concen- 10 ml of Fraser broth (OxoõÈd), incubated for 24 and 48 h at 37ЊC, trated demi-Fraser medium (all from OxoõÈd) to look for Salmo- and streaked on Oxford medium (OxoõÈd). The Oxford medium 2 nella, C. jejuni, C. coli, and L. monocytogenes per 100 cm or 25 was incubated for 48 h, and typical colonies (maximum ®ve) were 2 g. For L. monocytogenes, detection per cm or per g was also puri®ed on tryptone soy agar (OxoõÈd) and identi®ed by motility performed. In Europe, a microbiological criterion for L. monocy- testing, fermentation of rhamnose, xylose, and mannitol, ␤-he- Ͻ 2 togenes of 100 CFU per g or per cm on the sell-buy is generally molysis, and a CAMP test (1). accepted (15). At this low level of contamination, (heating to a core temperature of 71ЊC) is suf®cient to inactivate Statistical analysis. Data were subjected to statistical anal- L. monocytogenes (18). In view of a objective of Ͻ100 ysis (logistic regression model) using SPSS for Windows, release 2 CFU per cm or per g at the ultimate date of consumption and 7.5 (SPSS, Inc., Chicago, Ill.). taking into account a maximum 2-log outgrowth of the pathogen under refrigeration temperatures during the storage period of poul- RESULTS try, L. monocytogenes should be absent per square centimeter or gram on the day of distribution to the retail market (day of sam- Poultry products for sale in the retail market in Bel- pling in the present study). Detection of L. monocytogenes per gium were contaminated with Salmonella (36.5%), C. je- square centimeter was performed (if poultry carcasses or poultry juni or C. coli (28.5%), and L. monocytogenes (38.2%) (Ta- samples with skin on were sampled) by means of transfer of 1 ml ble 1). Table 1 shows that Salmonella Enteritidis was not of the 300-ml saline and peptone food homogenate to 9 ml of the predominant Salmonella serotype in the poultry sam- demi-Fraser broth. Detection per gram (if poultry cuts without ples. Salmonella Enteritidis was isolated from 5.4% of the skin or processed poultry was sampled) was done by incubation poultry samples, whereas 36.5% of the samples were con- of 8 ml of the ®nal 200 ml of demi-Fraser broth food homogenate. taminated with Salmonella spp. Salmonella, C. jejuni, and Detection of Salmonella spp. and Salmonella Enteritidis. C. coli are well-known pathogens associated with poultry. The methodology applied for detection of Salmonella spp. con- L. monocytogenes is rarely cited as an important pathogen sisted of a nonselective enrichment phase in buffered peptone wa- for poultry products, although the results from the present Њ ter (18 to 24 h, 37 C). Next, 0.1 ml was transferred to 10 ml of study showed that incidence of L. monocytogenes in poultry Rappaport Vassiliadis broth (OxoõÈd) and incubated for 18 to 24 h products was as high as for Salmonella. In about 12.3% of at 42ЊC. One milliliter of Rappaport Vassiliadis broth was trans- the poultry samples, the contamination level of L. mono- ferred to 10 ml of M broth (OxoõÈd) plus novobiocine (Merck, 2 Darmstadt, Germany) (10 ␮g/ml) and incubated for 4 to 6 h at cytogenes exceeded 1 CFU per g or per cm . Because L. 37ЊC. One milliliter of M broth was boiled for 20 min, and 100 monocytogenes is a psychrotrophic pathogen, having the ␮l was analyzed by Salmonella-Tek, an enzyme-linked immuno- ability to multiply during storage at refrigeration tempera- sorbent assay (ELISA) procedure, according to the manufacturer's tures, there is a risk that in these samples the pathogen J. Food Prot., Vol. 62, No. 7 SALMONELLA, C. JEJUNI, C. COLI, AND L. MONOCYTOGENES IN POULTRY 737

TABLE 1. Incidence of Salmonella, Campylobacter jejuni, C. coli, and Listeria monocytogenes in poultry products for sale on the Belgian retail market according to country of origina Salmonella L. monocytogenes (Ͼ1 CFU/100 cm2 or 25 g) Campylobacter (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 Country of origin Enteritidis Other spp. or 25 g) or 25 g) (Ͼ1 CFU/cm2 or g)

Belgium 32/279 (11.5) Ab 121/279 (43.4) AD 54/247 (21.9) AE 44/279 (15.7) A 111/279 (39.8) AD France 3/434 (0.7) B 143/434 (32.9) BD 129/427 (30.2) BD 45/434 (10.4) B 156/434 (35.9) AD Italy 0/13 (0.0) 4/13 (30.8) 2/13 (15.4) 0/13 (0.0) 0/13 (0.0) The Netherlands 0/2 (0.0) 0/2 (0.0) 0/2 (0.0) 0/2 (0.0) 0/2 (0.0) United Kingdom 7/44 (15.9) A 14/44 (31.8) BD 24/44 (54.5) CE 6/44 (13.6) A 28/44 (63.6) BE Total 42/772 (5.4) 282/772 (36.5) D 209/733 (28.5) E 95/772 (12.3) 295/772 (38.2) E

a Values are numbers of samples per total number of samples. Values in parentheses are percentages of positive samples. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 b Incidence levels within columns followed by different letters (A, B,orC) differ signi®cantly (P Ͻ 0.05). (Because of the low number of samples from Italy and The Netherlands, these were not included in the statistical analyses.) Incidence levels within rows followed by different letters (D or E) differ signi®cantly (P Ͻ 0.05). (Only contamination levels of Salmonella spp., C. jejuni, C. coli, and L. monocytogenes [Ͼ1 CFU/100 cm2 or 25 g] were compared.) exceeds the accepted (European) level of contamination be- to poultry cuts with skin. However, contamination rates of fore the end of shelf life. Salmonella and C. jejuni and C. coli for poultry products Signi®cant (P Ͻ 0.05) differences in the pathogen con- without skin are still high (34.6 and 25.0%, respectively). tamination rates of poultry products were noticed between L. monocytogenes contamination (Ͼ1 CFU per 100 cm2 or the different countries of origin (Table 1). Contamination per 25 g), on the contrary, is signi®cantly (P Ͻ 0.05) higher with Salmonella was highest in products from Belgian ab- on poultry cuts without skin, indicating contamination from attoirs (10 sites), whereas C. jejuni and C. coli contami- the environment during processing (utensils, equipment, or nation was best controlled at these sites. Products from the personnel). United Kingdom (2 sites) had signi®cantly (P Ͻ 0.05) high- Table 4 shows that Salmonella Enteritidis was predom- er contamination with C. jejuni, C. coli, and L. monocyto- inantly associated with boiling hen carcasses and never or genes. Salmonella Enteritidis was only rarely isolated from rarely isolated from turkey products. products originating from French abattoirs (0.7%) (16 sites) An increase in the pathogen contamination rate was but was regularly isolated from Belgian (11.5%) and U.K. noticed during cutting and further processing of the poultry. (15.9%) poultry products. Chicken carcasses had contamination rates of 29.3 and In France, a label is assigned to poultry products de- 41.3% for Salmonella and L. monocytogenes, respectively; rived from free-range broiler chickens, which run free in chicken cuts, 44.0 and 46.7%, respectively; and processed pine woods until slaughtering age (12 to 13 weeks). Non- chicken products, 68.3 and 61.0%, respectively. Although labeled poultry products are derived from broilers raised in contamination rates for C. jejuni and C. coli similar to those enclosed pens and slaughtered at the age of 6 to 8 weeks. cited for the two above-mentioned pathogens were ob- These labeled poultry products have a signi®cantly (P Ͻ served for chicken carcasses (25.6%) and cuts (40.0%), the 0.05) lower Salmonella contamination rate for than nonla- C. jejuni and C. coli contamination rate observed in pro- beled products (Table 2). Although there was also a trend cessed chicken products was low (6.4%). Campylobacter is for lower contamination rates of C. jejuni, C. coli, and L. a fragile organism and dies off in the presence of air; thus, monocytogenes (Ͼ1 CFU per g or per cm2), this decrease contamination will be lower the longer the carcass is held was statistically not signi®cant (P Ͻ 0.05). after processing. Bostan et al. (6) reported a decrease of the Table 3 shows a signi®cantly (P Ͻ 0.05) lower path- number of C. jejuni from 2.8 to 4.3 ϫ 105 CFU per g in ogen contamination, except for L. monocytogenes (Ͼ1 CFU the beginning to 1.1 ϫ 104 CFU per g in refrigerated per cm2 or per g), for poultry cuts without skin compared minced meat, 3.8 ϫ 103 CFU per g in cubed meat, and

TABLE 2. Incidence of Salmonella, Campylobacter jejuni, C. coli, and Listeria monocytogenes in label and nonlabeled poultry products originating from France for sale on the Belgian retail marketa L. monocytogenes Salmonella spp. Campylobacter (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 Poultry sample or 25 g) or 25 g) (Ͼ1 CFU/cm2 or g) or 25 g)

Nonlabeled 40/87 (46.0) Ab 27/87 (31.0) A 13/78 (16.7) A 34/78 (43.6) A Label 12/82 (14.6) B 19/82 (23.2) A 6/82 (7.3) A 41/82 (50.0) A a Values are the number of positive samples per total number of samples. Values in parentheses are percentages of positive samples. b Incidence levels within columns followed by different letters differ signi®cantly (P Ͻ 0.05). 738 UYTTENDAELE ET AL. J. Food Prot., Vol. 62, No. 7

TABLE 3. Incidence of Salmonella, Campylobacter jejuni, C. coli, and Listeria monocytogenes in poultry cuts with and without skin for sale on the Belgian retail marketa L. monocytogenes Salmonella spp. Campylobacter (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 Poultry cut or 25 g) or 25 g) (Ͼ1 CFU/cm2 or g) or 25 g)

With skin 86/183 (47.0) Ab 71/183 (38.8) A 31/183 (16.9) A 91/183 (6.9) A Without skin 63/182 (34.6) B 45/180 (25.0) B 3/182 (1.6) B 32/182 (17.6) B a Values are the number of positive samples per total number of samples. Values in parentheses are percentages of positive samples. b Incidence levels within columns followed by different letters differ signi®cantly (P Ͻ 0.05).

Ͻ10 CFU per g in meat ball. Also, the C. jejuni and C. of samples were found to be positive for C. jejuni and C. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 coli contamination rate for processed turkey products was coli (32), whereas in the present study, 40.0% were found higher, although not signi®cantly (P Ͻ 0.05), than that for to be positive. For L. monocytogenes, 25.7% of poultry processed chicken products, whereas the opposite was ob- carcasses derived from one processing plant were reported served for Salmonella and L. monocytogenes. In general, to be positive (9). Other reports mention 20% positive poul- turkey products were contaminated with Salmonella spp. try meat samples (26) or 59% positive poultry products and L. monocytogenes at lower rates than chicken products (16). The contamination rate for L. monocytogenes deter- (Table 4). Finally, boiling hen carcasses had a signi®cantly mined in the present study (38.2%) was in this range. The (P Ͻ 0.05) higher contamination rate for L. monocytogenes: incidence of Salmonella in poultry in this study was 36.5%. 28% of boiling hen carcasses had a high level of L. mon- The percentage of Salmonella-positive samples reported for ocytogenes contamination (Ͼ1 CFU per g or per cm2). poultry products varies enormously (12.8 to 79%) (10). It is dif®cult to compare the incidence of these path- DISCUSSION ogens reported by different authors because of differences Contamination rates for C. jejuni and C. coli in broiler in time period, origin and relative age of the samples, sam- chickens of 36 and 27.9% have been reported in Denmark pling plan, sampling techniques, and methodology applied. (23) and Germany (3), respectively, compared to the con- In the present study, a large number of samples were ana- tamination rate of 25.6% found in the present study. For lyzed (n ϭ 772), and samples were obtained from abattoirs chicken cuts, contamination rates for C. jejuni and C. coli in different regions (and different countries) supplying of 61% have been reported in The Netherlands (12). In poultry products (carcasses, cuts, and meat of predomi- Northern Ireland, 38% of chicken pieces were reported to nantly broiler chickens and turkeys) to the Belgian retail be contaminated (19). In an earlier study in Belgium, 57.5% market. Individual abattoirs will have different pathogen

TABLE 4. Incidence of Salmonella, Campylobacter jejuni, C. coli, and Listeria monocytogenes in poultry carcasses, poultry cuts, and processed poultry for sale on the Belgian retail marketa Salmonella L. monocytogenes (Ͼ1 CFU/100 cm2 or 25 g) Campylobacter (Ͼ1 CFU/100 cm2 (Ͼ1 CFU/100 cm2 Poultry sample Enteritidis Other spp. or 25 g) (Ͼ1 CFU/cm2 or g) or 25 g)

Carcassesb Chicken 6/133 (4.5) A 39/133 (29.3) A 34/133 (25.6) A 15/133 (11.3) A 55/133 (41.3) A Boiling hen 5/32 (15.6) B 13/32 (40.6) A 7/32 (21.9) A 9/32 (28.1) B 26/32 (81.2) B Spring chicken 2/48 (4.2) A 13/48 (27.1) A 18/48 (37.5) A 8/48 (16.7) A 22/48 (45.8) A Guinea fowl 2/32 (6.2) A 6/32 (18.8) A 11/32 (34.4) A 2/32 (6.2) A 11/32 (34.4) A Cutsc Chicken 16/225 (7.1) C 99/225 (44.0) C 90/225 (40.0) C 36/225 (16.0) C 105/225 (46.7) C Turkey 0/164 (0.1) D 60/164 (36.6) C 32/162 (19.8) D 1/164 (0.6) D 30/164 (18.3) D Spring chicken 3/28 (10.7) C 8/28 (28.6) C 10/28 (35.7) C 8/28 (28.6) C 12/28 (42.8) C Guinea fowl 0/3 (0.0) 0/3 (0.0) 0/3 (0.0) 0/3 (0.0) 0/3 (0.0) Processedd Chicken 6/41 (14.6) E 28/41 (68.3) E 2/31 (6.4) E 14/41 (34.1) E 25/41 (61.0) E Turkey 2/66 (3.0) F 16/66 (24.2) F 5/39 (12.8) E 2/66 (3.0) F 9/66 (13.6) F Total 41/772 (5.4) 282/772 (36.5) 209/733 (28.5) 95/772 (12.3) 295/772 (38.2) a Values are the number of positive samples per total number of samples. Values in parentheses are percentages of positive samples. b Incidence levels within columns (carcasses sample) followed by different letters (A or B) differ signi®cantly (P Ͻ 0.05). c Incidence levels within columns (cuts sample) followed by different letters (C or D) differ signi®cantly (P Ͻ 0.05). (Guinea fowl not included in statistical analysis.) d Incidence levels within columns (processed sample) followed by different letters (E or F) differ signi®cantly (P Ͻ 0.05). J. Food Prot., Vol. 62, No. 7 SALMONELLA, C. JEJUNI, C. COLI, AND L. MONOCYTOGENES IN POULTRY 739 contamination rates, and signi®cant (P Ͻ 0.05) differences To lower C. jejuni, C. coli, Salmonella, and L. mon- were observed between pathogen contamination rates of ocytogenes contamination rates, hygienic rules of slaughter products originating from Belgium, France, and the United and meat processing must be rigorously observed. Im- Kingdom. However, a supermarket selling poultry carcasses proved hygiene control has been found to lower the number and products to consumers obtains poultry from different of Campylobacter on packaged poultry (21). The incidence abattoirs and different countries (international trade). By of L. monocytogenes especially may be reduced by im- sampling retail packs, a representative ®gure for the con- proving abattoir hygiene. For Salmonella, C. jejuni, and C. tamination rates of Salmonella spp., C. jejuni, C. coli, and coli, great emphasis should be put on prevention and con- L. monocytogenes in poultry products offered to the con- trol in poultry breeding. However, to succeed, a multifac- sumer for consumption is obtained. This relates immediate- eted approach must be applied (30). Preharvest food safety ly to the exposure of the consumer to these pathogens. programs implementing the rules of the Hazard Analysis The contamination rate for Salmonella reported in this and (HACCP) concept at farm level study (36.5%) was close to those observed in The Neth- from breeding to the slaughter house gate should be added Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 erlands (33.0 to 34.8%) (2). The incidence of C. jejuni and to existing HACCP programs from the slaughter house to C. coli in this study (28.5%) was close to that reported in the retailer (5). At the present moment, zero tolerance for Germany (27.9%) (3). These similarities indicate that the any of the three pathogens cannot realistically be achieved; data gathered in this study can be used for risk assessment thus, there is a need to establish criteria allowing these in countries other than Belgium, especially neighboring pathogens to be present at reasonable levels in poultry. countries. Consequently, implementation of good cooking techniques The trend of lower contamination rates in labeled and good kitchen and personal hygiene during preparation poultry products (free-range broilers, slaughtered at 12 to are necessary. Also, education of food handlers and infor- 13 weeks of age) compared to nonlabeled products (broilers mation to consumers about microbial risks associated with raised in enclosed pens, slaughtered at 6 to 8 weeks) sug- the consumption of poultry products and how to control gests that industrial food production and younger relative them (e.g., storage at the proper temperature and adequate age at slaughter have a negative effect on the incidence of cooking) are needed. pathogens. In industrialized production, environmental and hygienic conditions during breeding can be well controlled, ACKNOWLEDGMENT but, for example, providing Salmonella-contaminated feed M. Uyttendaele is indebted to the Fund of Scienti®c Research±Flan- or insuf®cient cleaning and disinfection of poultry houses ders (F.W.O.±Flanders) for a position as a postdoctoral research assistant. has a great impact on the entire ¯ock (10, 20). Also, in general, young hosts are more susceptible to salmonellae REFERENCES than older ones. This fact may be due to insuf®cient for- 1. Association FrancËaise de Normalisation. 1993. Normalisation fran- mation of stable intestinal ¯ora (4). cËaise, Syllabus eÂditeÂe et diffuseÂe par l'association francËaise de nor- Because of the way chickens are reared and trans- malisation (AFNOR). Paris. ported, they bring Salmonella into the processing plant on 2. Anonymous. 1996. Onderzoek van kip en kipproducten. Nieuwsbrief their skins (14). Salmonella cells are initially entrapped in Voedselveiligheid 1:6. 3. Atanassova, V., J. Altemeier, K. P. Kruse, and B. Dolzinski. 1998. a water ®lm on the skin and then migrate to the skin, where The detection of Salmonella and C. jejuni/coli in fresh poultry. they are entrapped in ridges and crevices that become more Fleischwirtschaft 78:364±366. pronounced in the skin after immersion in water (17). This 4. Bauer, J., and S. HoÈrmansdorfer. 1996. in farm ani- could explain the higher Salmonella contamination rates of mals. Fleischwirtschaft 76:726±728. poultry cuts with skin. Data obtained from a study moni- 5. Blaha, T. 1998. Pre-harvest food safety (epidemiological approach to complement traditional food security concepts for further reduc- toring the different processing steps in one plant showed tion of foodborne health risks to man, p. 521±525. In Proceedings that L. monocytogenes infrequently enters the processing of the 4th World Congress on Foodborne Infections and Intoxica- plant on live broiler hens (9). The organism appears to take tions, 7±12 June, Berlin. up residence in the plant, leading to cross-contamination 6. Bostan, K., H. Aksu, O. Ozgen, and M. Ugur. 1997. Survival of during processing (25). This is also indicated by the higher Campylobacter jejuni in some meat and meat products, p. 192. In Proceedings of the World Congress on Food Hygiene (WAVFH), incidence of L. monocytogenes on poultry cuts without skin 24±29 August, The Hague, Netherlands. (Ͼ1 CFU per 100 cm2 or per 25 g). 7. Bryan, F. L., and M. P. Doyle. 1995. Health risks and consequences The increase in the incidence of Salmonella during of Salmonella and Campylobacter jejuni in raw poultry. J. Food Prot. cutting and further processing was expected. It has been 58:326±344. reported that Salmonella incidence rates increased from 8. Carraminana, J. J., J. Yanguela, D. Blanco, C. Rota, A. I. Agustin, A. Arino, and A. Herrera. 1997. Salmonella incidence and distri- 30% in fecal material collected from incoming birds to 60% bution of serotypes throughout processing in a Spanish poultry in air-chilled carcasses, indicating that cross-contamination slaughterhouse. J. Food Prot. 60:1312±1317. occurred (8). Cross-contamination is also responsible for an 9. Cox, N. A., J. S. Bailey, and M. E. Berrang. 1997. The presence of increase in L. monocytogenes, C. jejuni, and C. coli con- Listeria monocytogenes in the integrated poultry industry. J. Appl. tamination in cuts. The low incidence of C. jejuni and C. Poultry Res. 6:116±119. 10. D'Aoust, J. Y. 1989. Salmonella, p. 327±446. In M. P. Doyle (ed.), coli in processed poultry may be attributed to a combina- Foodborne bacterial pathogens. Marcel Dekker, Inc., New York. tion of exposure to oxygen, decreased humidity, and sup- 11. Desenclos, J., P. Bouvet, V. Pierre, A. Brisabois, S. Fremy, C. Lah- pression of the organism by competitive micro¯ora. ellec, G. Grimont, and P. A. D. Grimont. 1996. Epidemiologie des 740 UYTTENDAELE ET AL. J. Food Prot., Vol. 62, No. 7

infections a Salmonella: tendances recentes en France et en Europe. 22. Neimann, J., J. Engberg, K. Molbak, and H. C. Wegener. 1998. Risk Bull. Soc. Fr. Microbiol. 121:209±215. factors associated with sporadic campylobacteriosis in Denmark, p. 12. Hahne, M., and E. de Boer. 1990. Cross contamination with Sal- 298±303. In Proceedings of the 4th World Congress on Foodborne monella spp. and C. jejuni/coli jejuni in raw poultry products. Voe- Infections and Intoxications, 7±12 June, Berlin. dingsmiddelentechnologie 23:11±13. 23. Nielsen, E. M., J. Engberg, and M. Madsen. 1997. Distribution of 13. Henry, C. J. 1997. Risk assessment, risk evaluation, and risk man- serotypes of C. jejuni and C. coli from Danish patients, poultry, agement, p. 283±300. In J. deVries (ed.), Food safety and toxicity. cattle and swine. FEMS Immunol. Med. Microbiol. 19:47±56. CRC Press, Boca Raton, Fla. 24. Nylen, G. 1998. Campylobacter seasonality in Europe, p. 293±297. 14. International Commission on Microbiological Speci®cations for In Proceedings of the 4th World Congress on Foodborne Infections Foods. 1998. Micro-organisms in foods 6: microbial ecology of food and Intoxications, 7±12 June, Berlin. commodities. Blackie Academic and Professional, London. 25. Ojeniyi, B., H. C. Wegener, N. E. Jensen, and M. Bisgaard. 1996. 15. Jouve, J. 1996. La qualite microbiologique des aliments, maõÃtrise et Listeria monocytogenes in poultry and poultry products: epidemio- criteÁres. Polytehnica, Paris. logical investigations in seven Danish abattoirs. J. Appl. Bacteriol. 16. Lawrence, L. M., and A. Gilmour. 1994. Incidence of Listeria spp. 80:395±401. and Listeria monocytogenes in a poultry processing environment and 26. Ozari, R., and F. A. Stolle. 1990. The occurrence of Listeria mon-

in poultry products and their rapid con®rmation by multiplex PCR. ocytogenes in meat, meat products and poultry. Arch. Lebensmittel- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/62/7/735/1671534/0362-028x-62_7_735.pdf by guest on 24 September 2021 Appl. Environ. Microbiol. 60:4600±4604. hyg. 41:47±50. 17. Lillard, H. S. 1989. Factors affecting the persistence of Salmonella 27. Schmidt, K. 1998. Situation of foodborne diseases in Europe, 1992± during the processing of poultry. J. Food Prot. 52:829±832. 1996, p. 262±266. In Proceedings of the 4th World Congress on 18. Mackey, B. M., C. Pritchet, A. Norris, and G. C. Mead. 1990. Heat Foodborne Infections and Intoxications, 7±12 June, Berlin. resistance of Listeria: strain differences and effects of meat type and 28. Smith, H. R. 1998. Serotyping and phage typing of enteric . curing . Lett. Appl. Microbiol. 10:251±255. De Ware(n)-Chemicus 28:27±29. 19. Madden, R. H., L. Moran, and P. Scates. 1998. Frequency of occur- 29. Smith, J. L. 1998. Foodborne illness in the elderly. J. Food Prot. 61: rence of C. jejuni/coli spp. in red meats and poultry in Northern 1229±1239. Ireland and their subsequent subtyping using polymerase chain re- 30. Stern, J., N. A. Cox, J. S. Bailey, S. E. Craven, J. E. Line, and R. action±restriction fragment length polymorphism and the random J. Meinersmann. 1997. Salmonella and Campylobacter epidemiology ampli®ed polymorphic DNA method. J. Appl. Microbiol. 84:703± 708. and control during broiler production, p. 29±35. In Proceedings of 20. Mead, G. C. 1996. Does the use of competitive exclusion and pro- the World Congress on Food Hygiene (WAVFH), 24±29 August, The biotics in poultry in¯uence the microbial status of poultry carcasses?, Hague, Netherlands. p. 33±40. In Factors affecting the microbial quality of meat, 1: dis- 31. Uyttendaele, M., and J. Debevere. 1996. Evaluation of Preston me- ease status, production methods and transportation of the live animal. dium for detection of C. jejuni/coli jejuni in vitro and in arti®cially Concerted action CT94-1456. University of Bristol Press, Bristol, and naturally contaminated poultry products. Food Microbiol. 13: UK. 115±122. 21. Mead, G. C., W. R. Hudson, and M. H. Hinton. 1995. Effect of 32. Uyttendaele, M., R. Schukkink, B. van Gemen, and J. Debevere. changes in processing to improve hygiene control on contamination 1996. Comparison of the nucleic acid ampli®cation system NASBA of poultry carcasses with campylobacter. Epidemiol. Infect. 115: and agar isolation for detection of pathogenic campylobacters in nat- 495±500. urally contaminated poultry. J. Food Prot. 59:683±687.