Food Microbiology 28 (2011) 1117e1123

Contents lists available at ScienceDirect

Food Microbiology
journal homepage: www.elsevier.com/locate/fm

Evaluation of ISO 10272:2006 standard versus alternative enrichment and plating combinations for enumeration and detection of Campylobacter in chicken meat
Ihab Habib a, c, *, Mieke Uyttendaele a, Lieven De Zutter b
a

Laboratory of Food Microbiology and Food Preservation, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, B-9000 Ghent, Belgium Department of Veterinary Public Health and Food Safety, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, B-9820 Merelbeke, Belgium c Food Hygiene and Control Division, High Institute of Public Health (HIPH), Alexandria University, 165 El-Horrya Avenue, Alexandria, Egypt
b

a r t i c l e i n f o
Article history: Received 8 October 2010 Received in revised form 2 March 2011 Accepted 2 March 2011 Available online 10 March 2011 Keywords: Campylobacter Chicken meat Enrichment Enumeration Methods performance

a b s t r a c t
In the present study, we evaluate the recommended ISO 10272:2006 versus alternative procedures for Campylobacter enumeration and enrichment in naturally contaminated chicken meat samples (n 49). Three enrichment media were evaluated; Bolton broth, Preston broth and CampyFood broth (bioMrieux SA, Marcy lEtoile, France). In addition, three selective plating agars were compared; modied charcoal cefoperazone deoxycholate agar (mCCDA), CampyFood agar (CFA; bioMrieux SA) and Brilliance CampyCount agar (BCC; Oxoid, Basingstoke, England). Direct plating on CFA provided the highest number of Campylobacter positive samples (17/49); however this was not statistically different (P > 0.05) from numbers of positive samples recovered by direct plating on mCCDA (15/49) or BCC agars (14/49). Also, there was no signicant difference between Campylobacter counts on the three compared media (P > 0.05). The coloured colonies of Campylobacter on CFA and BCC were easier to record and count than those on mCCDA. Enrichment of chicken meat samples in Bolton broth for 48 h and subsequent plating on CFA provided signicantly higher (P < 0.05) Campylobacter detection compared to the other brothagar combinations. Enrichment in Preston broth for 24 h followed by plating on mCCDA gave a higher number of positive samples (20/49) than 48 h enrichment in Bolton broth and plating on mCCDA (15/49). Enrichment in Bolton broth for 48 h followed by plating on CFA recovered 35% of samples below the limit for quantications (<10 CFU/g, n 34), as identied by direct plating on mCCDA. Compared to the current ISO method, some alternative combinations of enrichment and agar media could provide signicantly better detection and enumeration of Campylobacter in chicken meat. 2011 Elsevier Ltd. All rights reserved.

1. Introduction Campylobacter jejuni and Campylobacter coli are among the most common bacterial causes of human gastroenteritis worldwide. Infected humans exhibit a range of clinical symptoms varying from mild, watery to severe inammatory diarrhoea (Humphrey et al., 2007). In addition, C. jejuni has been identied as an important infectious trigger for Guillain-Barr syndrome, the most common cause of acute accid paralysis in polio-free regions (Godschalk et al., 2004). Another issue of concern with regard to Campylobacter is the increase in antimicrobial resistance appearing in various regions around the world (Humphrey et al., 2007). Infection with antimicrobial-resistant Campylobacter may lead to suboptimal outcomes of antimicrobial treatments or even treatment failure

* Corresponding author. Laboratory of Food Microbiology and Food Preservation, Faculty of Bioscience Engineering, Ghent University, Coupure links 653, B-9000 Ghent, Belgium. Tel.: 32 09 264 60 85; fax: 32 09 225 55 10. E-mail address: ihab.habib@ugent.be (I. Habib). 0740-0020/$ e see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.fm.2011.03.001

(Engberg et al., 2004). The majority of human campylobacteriosis cases are sporadic, and consumption or mishandling of contaminated raw or undercooked poultry meat is believed to be an important vehicle of infection (EFSA, 2009). Source-attribution studies, outbreak investigations and case-control reports all incriminate chicken meat as a major source of the foodborne transmission of Campylobacter infection (Butzler, 2004; Friedman et al., 2004; Humphrey et al., 2007; Mullner et al., 2009). Effective monitoring and risk assessment of Campylobacter contamination in chicken meat is largely dependent on the availability of reliable detection and enumeration methods. For Campylobacter enumeration, direct spread plating on modied charcoal cefoperazone deoxycholate agar (mCCDA) was found to be a reliable alternative to the most probable number method (Scherer et al., 2006). In addition, Rosenquist et al. (2007) concluded, through a collaborative study, that direct plating on mCCDA is an acceptable protocol for enumeration of thermotolerant Campylobacter in chicken meat. Currently, mCCDA is the recommended medium by the International Organization for Standardization

1118

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123

(ISO) for enumeration of thermotolerant Campylobacter in foods, based on spread plate count after incubation for 48 h at 41.5  C (ISO, 2006a). On the other hand, detection of Campylobacter after enrichment is recommended for recovery of sublethally damaged cells from food and environmental samples. The formulations of Campylobacter enrichment broth media have been modied extensively over the last two decades, mostly to avoid inhibitory effects on Campylobacter spp. by components in the broth formulae (Corry et al., 1995). In 2006, the ISO standard method for detection of thermotolerant Campylobacter in food recommended the use of Bolton broth, instead of Preston broth, for selective enrichment. Enrichment procedures are recommended to start at 37  C for 4e6 h followed by 44 h at 41.5  C. After enrichment, plating is carried on mCCDA and a second solid medium left to the choice of the researcher (ISO, 2006b). Campylobacters are fastidious bacteria with demanding growth requirements. Thus, special handling and careful testing procedures are required in order to detect them in food samples. Adding to that, recent reports indicate that there might be considerable false-negative ndings associated with campylobacters testing in food (Habib et al., 2008; Edson et al., 2009; EFSA, 2010). Hence, the use of an appropriate enrichment broth and selective plating medium is important for the certain recovery of C. jejuni. In particular, stressed and heat-injured cells might require a longer cultivation time in a suitable enrichment broth. In the present study, we evaluate the recommended ISO 10272:2006 procedures for Campylobacter enrichment and enumeration versus alternative combinations of enrichment and plating media. The evaluation was performed on naturally contaminated chicken meat samples and included evaluation of two new selective agars and a new enrichment broth. 2. Materials and methods 2.1. Media and samples The evaluated procedures involved testing the performances of three enrichment media; Bolton broth ((CM983 with supplement

SR183, Oxoid, Basingstoke, England), recommended in the current ISO standard method), Preston broth ((CM689 with supplement SR117, Oxoid), recommended in the former version of the ISO standard method) and a new broth medium, CampyFood broth (CFB; bioMrieux SA, Marcy lEtoile, France). CFB is commercialized as a liquid in ready-to-use plastic bags, and its patented composition is currently condential. In addition, three selective plating agars were evaluate; modied charcoal cefoperazone deoxycholate agar (mCCDA, Oxoid), which is recommended in the current ISO standard method, the chromogenic-like agar CampyFood agar (CFA, bioMrieux SA) and the chromogenic agar Brilliance CampyCount agar (BCC, Oxoid). The last two agar media are commercialized in the format of ready-to-use plates. The detailed compositions of these two new patented agars are currently condential. In total, 49 samples of chicken meat were included in the present study. Samples were collected from February to June 2010, from local supermarkets and poultry processing companies. These samples were 31 fresh chicken meat items (chicken carcasses, llets, legs and wings) and 18 chicken meat preparations (minced and portioned chicken meat to which herbs, sauces or marinades were added). For chicken meat and meat preparations made from whole pieces of meat, the test-portion was taken, as far as possible, from the surface of the meat; starting with the skin, if present, but scraping away any sauce or nonmeat components as the presence of seasonings and marinades may interfere with the analysis. For samples made from minced chicken meat, a representative testportion was taken from throughout the sample. For broiler carcasses, the test-portion was taken from neck skin. 2.2. Microbiological testing A schematic representation of the analysis protocol and different enrichment/plating combinations evaluated in this study is given in Fig. 1. For Campylobacter detection after enrichment, a representative 10 g test-portion was homogenized in 90 ml 0.1% peptone water (Oxoid). From this homogenate, 10 ml were transferred to 90 ml of each of the enrichment broths that were compared. Incubation of the enrichment cultures was under

10 g test-portion + 90 ml 0.1% Peptone water Campylobacter detection and enumeration Enrichment:* 10 ml + 90 ml BB 10 ml + 90 ml PB 10 ml + 90 ml CFB
4-6 h @37C + 40-48 h @41.5C 48 h @41.5C 48 h @41.5C

Subsequent plating: After 24 h, and After 48 h enrichment mCCDA + CFA+ BCC mCCDA + CFA+ BCC mCCDA + CFA

Identification to genus and species levels: - ISO 10272 - Multiplex-PCR

Enumeration after direct plating: On mCCDA + CFA + BCC (in duplicates) Enumeration of background bacteria Aerobic bacterial count Enterobacteriaceae E. coli Pseudomonas Plate count agar (PCA; 30C/ 48 h) Violet red bile glucose agar (VRBG; 37C/ 24 h) Tryptone Bile X-Glucuronide (TBX; 44C/ 24 h) Cephaloridine-fucidin-cetrimide agar (CFC; 25C/ 48 h)

* BB, Bolton broth; PB, Preston broth; CFB, CampyFood broth; mCCDA, modified charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; and BCC, Brilliance CampyCount agar.
Fig. 1. Schematic summary of the comparison procedures of the ISO standard versus alternative protocols for Campylobacter detection and enumeration in chicken meat samples.

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123

1119

microaerobic conditions generated by evacuating 80% of the normal atmosphere and introducing a gas mixture consisting of 8% CO2, 8% H2, and 84% N2 into stainless-steel jars (Don Whitley Scientic, Shipley, UK). After 24 h, and again after 48 h, 10 mL of the enrichment culture was streaked onto the surface of a plate of each agar medium, which was then incubated at 41.5  C for 48 h under microaerobic conditions. Conrmation of presumptive colonies to the genus level was conducted as described in the ISO 102721:2006 method. For Campylobacter enumeration, serial decimal dilutions were prepared from the same sub-sample homogenate (Fig. 1). One ml from the 101 dilution was spread plated over 4 agar plates (distributed as 0.3 ml, 0.3 ml, 0.3 ml, and 0.1 ml), and then from further serial dilutions with plating of 0.1 volumes. Microaerobic incubation was as described above, and conrmation of presumptive colonies to the genus level was conducted as described in ISO 10272-2:2006. From presumptive positive colonies, recovered after enrichment or after direct plating, up to ve well dened colonies were restreaked for purity on mCCDA and incubated micro-aerobically at 41.5  C for 24 h. DNA lysates were prepared by suspending a 1 ml loop-full of bacteria from each colony in an Eppendorf tube containing 1 ml Peptone Buffer Saline (PBS, SigmaeAldrich, Bornem, Belgium). The Eppendorf tube was centrifuged at 14,000 rpm for 5 min, the supernatant was discarded and the pellet was re-suspended in 100 ml Tris-EDTA (TE) buffer 10:1 (Promega, Madison, USA). The suspension was heated at 95  C for 15e20 min and then transferred directly to ice. The lysed DNA was diluted 10 fold in TE 10:1, and stored at 20  C. Multiplex-PCR was used for further conrmation and differentiation of C. jejuni and C. coli according to the procedures described by Vandamme et al. (1997). The initial background ora of the tested chicken meat samples (aerobic bacterial count, Enterobacteriaceae, E. coli and Pseudomonas) were enumerated by spiral plating (Eddy Jet, IUL Instruments, Barcelona, Spain) from each homogenate as indicated in Fig. 1. In an attempt to screen for E. coli with possible resistance to the antimicrobial supplements used in Bolton and Preston broths, plating was performed on Tryptone Bile X-Glucuronide agar (TBX; CM0945, Oxoid) to which the antimicrobial supplement for one or other of the enrichment broths was added, and growth of E. coli on these plates was monitored. 2.3. Data analysis Data were compiled in Excel sheet and imported to the computer program STATA version 8 (StataCorp, 2007) for statistical analyses. The signicant differences (P-value  0.05) between Campylobacter counts, as a function of agar media, were tested using generalized linear models starting with Poisson regression analysis. In cases of extra-Poisson dispersion, the analyses were tted by negative binomial regression analysis. Results for detection (Negative/positive) using different enrichment/plating combinations were cross tabulated. Logistic-regression analysis was used to evaluate the impact of the following variables on the Campylobacter detection rate: (1) the combination of enrichment/ plating media, (2) effect of the length of incubation time (24 versus 48 h) in enrichment broths, and (3) the interaction between enrichment/plating combination and incubation time. The reference combination category in this regression model was the ISO standard procedure, based on enrichment in Bolton broth and subsequent plating on mCCDA. Numerical results were transformed to base-10 log values for descriptive analysis. Campylobacter counts on the two new plating agars (CFA and BCC) were compared versus the ISO recommended agar medium (mCCDA) using the concordance correlation coefcient (CCC). This coefcient indicates the precision of result (how far each count deviates from the line of t

to the data), and also estimates the bias correction factor (BCC), a parameter showing the accuracy of result (how far the line of t deviates from the 45 line of origin) (Lin, 1989). 3. Results Table 1 summarizes the performance of the three selective agars used for Campylobacter enumeration in chicken meat samples. Direct plating on CFA provided the highest number of Campylobacter positive samples (17/49); however this was not statistically different (P > 0.05) from numbers of positive samples recovered by direct plating on mCCDA or BCC agars. Also, there was no signicant difference between the overall means of Campylobacter counts obtained with the three media (P > 0.05). Twenty samples were positive for Campylobacter by direct plating on one or more of the three selective agars. The positive detection agreement between the three agars was only in 10 of these 20 samples. For Campylobacter recovery by direct plating, false-negative ndings varied between the three agars. The rate of false-negatives was signicantly less (P < 0.05) on CFA (3/10) compared to mCCDA (5/10) and BCC (6/10). There was substantial concordance between Campylobacter counts on CFA and mCCDA (Fig. 2); this was indicated by a CCC value of 0.90 (95% CI: 0.795, 1.008). This result is also supported by a value of 0.981 for the estimated BCC (measure of accuracy). A lower CCC (0.802) and BCC (0.958) were estimated for the correlation between Campylobacter counts on mCCDA and BCC agar. Table 2 indicates that 31 samples (63.3%) were positive for Campylobacter, based on different combinations of enrichmentplating procedures. However, all combinations of enrichment and subsequent plating could lead to false-negative results. This varied from as low as 5/49 samples, using 48 h enrichment in Bolton broth and subsequent plating on CFA, to 18/49 samples when using 48 h enrichment in CFB and subsequent plating on mCCDA. Enrichment of chicken meat samples in Bolton broth for 48 h followed by plating on CFA provided signicantly higher (P < 0.05) Campylobacter detection compared to other combinations. The fastest Campylobacter enrichment recovery was obtained using Bolton broth and subsequent plating on CFA (Fig. 3). However, Preston broth gave a consistent enrichment performance, regardless of the subsequent plating agar; as presented in Fig. 3, number of Campylobacter positive samples after enrichment in Preston broth for just 24 h was signicantly higher (P < 0.05) than the number of positive samples detected after the same period of incubation in Bolton broth or CFB. Added to that, extending the

Table 1 Campylobacter recovery from chicken meat samples (n 49) after direct plating on the three agar media.a mCCDA CFA BCC

Mean (log10 CFU/g) SDb: 1.92 0.77 Results concordancec: 1.92 0.88 2.15 0.69 Total 29 10 3 3 2 1 1

a mCCDA, modied charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; BCC, Brilliance CampyCount. b Mean and standard deviation (SD) was calculated for countable results. c () indicates a countable result, and () indicates a result below quantication limit.

1120

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123

Fig. 2. Concordance correlation coefcient (CCC) plot of agreement between Campylobacter counts (log10 CFU/g) on modied charcoal cefoperazone deoxycholate agar (mCCDA) versus CampyFood agar (CFA) (I), and mCCDA versus Brilliance CampyCount agar (BCC) (II).

enrichment period to 48 h in Preston broth was not associated with a signicant increase (P > 0.05) in the number of Campylobacter positive samples, compared to 24 h enrichment. Enrichment in Preston broth for 24 h followed by plating on mCCDA gave a higher number of positive samples than 48 h enrichment in Bolton broth and plating on mCCDA (Fig. 3). To evaluate the potential of enrichment protocols in recovering low levels of Campylobacter contamination, samples (n 34) identied as below the limit for quantications (<10 CFU/g) by direct plating on mCCDA were further examined for the correspondent results with enrichment (Fig. 4). Enrichment in Bolton broth for 48 h followed by plating on CFA recovered Campylobacter from 35% (12/34) of the samples in which Campylobacter was below limit for quantication; while enrichment in Preston broth allowed recovery of Campylobacter from 29% (9/34) of the samples regardless of the subsequent plating agar. On TBX plates with Preston supplement there was no growth of E. coli colonies from any sample homogenate. However, on TBX agar with Bolton supplement there was growth of E. coli colonies with homogenates from 16 samples (32.6%). The mean count for E. coli
Table 2 Variation between enrichment broths-subsequent plating mediaa combinations of the number of samples in which Campylobacter was detected after 48 h enrichment. No. of samples Bolton broth mCCDA 18 6 4 3 3 2 2 2 1 1 1 1 1 1 1 1 1 BCC CFA Preston broth mCCDA BCC CFA CampyFood broth mCCDA CFA

growing on TBX agar with Bolton supplement was 1.82 log10 CFU/g with a standard deviation of 2.05 log10 CFU/g Table 3 indicates that among these 16 samples with E. coli resistant to antimicrobials in Bolton supplement, 5 samples were Campylobacter-negative after enrichment in all three broths, and these 5 samples were also below the limit for quantication by direct plating. Two of those 16 samples were positive for Campylobacter by all enrichment protocols. Campylobacter counts in these two samples were 2.5 and 2.6 log10 CFU/g, which is higher than the mean Campylobacter count for the tested chicken meat samples. The remaining samples showed results for Campylobacter, depending on the combination of enrichment broth and plating agar (Table 3). 4. Discussion The development of selective agar media and enrichment broths for isolation of Campylobacter from food and environmental samples is an area of active research. Among the plethora of selective agars, mCCDA is the most widely used in food laboratories (Corry et al., 1995). However, from a practical point of view, the black colour of mCCDA and the grey colour and variable shapes of Campylobacter colonies on it sometimes makes distinguishing between typical and atypical colonies difcult. Results from the present study indicate that the performances of CFA and BCC media are comparable to that of mCCDA for the purpose of enumeration of Campylobacter from naturally contaminated chicken meat. The ease of discerning (coloured colonies on a transparent agar background) is important for reducing errors in counting. In recent years, there has been a growing demand for quantitative data to describe the occurrence and dynamics of Campylobacter in the broiler meat chain, especially to support quantitative risk assessment modelling (Uyttendaele et al., 2006; Nauta et al., 2007). Given the high level of correlation between counts on mCCDA and CFA or BCC media, the easier discernment of presumptive colonies on the new chromogenic agars, than on the ISO recommended mCCDA, is advantageous. Although various enrichment broths have been developed and introduced, Bolton broth is currently the medium recommended by the U.S. Food and Drug Administration (Hunt et al., 2001), the International Standard Organisation (ISO, 2006a) and the Nordic Committee of Food Analysis (Rosenquist et al., 2007). Nevertheless, some comparisons of the performances of Bolton broth and other enrichment media are available. Baylis et al. (2000) compared Bolton Broth, Preston broth and Campylobacter Enrichment broth.

a mCCDA, modied charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; BCC, Brilliance CampyCount.

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123

1121

30 27 24

No. of positive samples

21 18 15 12 9 6 3 0 mCCDA/ 24 h mCCDA/ 48 h mCCDA/ 24 h mCCDA/ 48 h mCCDA/ 24 h

11.76

Bolton broth

Preston broth

CampyFood broth

Fig. 3. Detection of Campylobacter from chicken meat samples (n 49) in relation to enrichment broth, duration of incubation and subsequent plating agars (mCCDA, modied charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; and BCC, Brilliance CampyCount agar).

In this comparison, Bolton broth showed a good compromise between recovery of Campylobacter and inhibition of competitors. The same study also found that E. coli and Pseudomonas spp. were frequently isolated from BB. In line with those ndings, we found that E. coli from chicken meat could be resistant to the antimicrobial supplements of Bolton broth. Jasson et al. (2009) showed that Bolton broth supported the growth of extended-spectrum-betalactamase (ESBL) E. coli present in poultry meat; and that these ESBL E. coli could crowd mCCDA plates and masked Campylobacter colonies, leading to false-negative results. These previous ndings, and ours, raise concerns about the selectivity of Bolton broth. Campylobacter spp. has a slower growth rate than many other bacteria found on meat and competes poorly outside of its intestinal niche (Musgrove et al., 2001). Hence, the growth of competing
40
35.29

bacteria during enrichment, among other factors, might adversely affect the detection of Campylobacter, especially if it is present in small number. In our study, the combination of enrichment in Bolton broth and subsequent plating on CFA provided a higher recovery. This might be due to the ease of identication of Campylobacter colonies on CFA, but might also be due to the specic antimicrobials included in the medium. Thus, CFA can be recommended as a second medium to be included in parallel with mCCDA for plating enrichment cultures of chicken meat samples. Another intriguing nding of this study was that the former ISO approach of enrichment in Preston broth and subsequent plating on mCCDA gave signicantly higher recovery of Campylobacter than the current ISO approach of enrichment in Bolton broth and subsequent plating on mCCDA. Unlike the Bolton-mCCDA

Samples giving Campylobacter colonies (%)

35 30 25 20 15 10 5 0 Bolton broth + CFA Preston broth + mCCDA Preston broth + BCC Preston broth + CFA Bolton broth + mCCDA Bolton broth + BCC CampyFood CampyFood broth + broth + CFA mCCDA
29.41 29.41 29.41

17.65 14.71 14.71

Fig. 4. Recovery of Campylobacter from samples (n 34) with count less than 10 CFU/g after incubation in enrichment broth for 48 h. Number above each bar represent the percentage of recovery for each brotheagar combination. Subsequent plating agars are; mCCDA, modied charcoal cefoperazone deoxycholate agar; CFA, CampyFood agar; and BCC, Brilliance CampyCount agar.

mCCDA/ 48 h

CFA/ 24 h

CFA/ 48 h

BBC/ 24 h

BBC/ 24 h

CFA/ 48 h

CFA/ 24 h

CFA/ 48 h

CFA/ 24 h

BBC/48 h

BBC/48 h

1122

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123

Table 3 Campylobacter detection in sixteen chicken meat samples that contained Escherichia coli resistant to Bolton broth supplement. a Campylobacter Bolton broth count (log10 CFU/g)b Preston broth CampyFood broth

mCCDA BCC CFA mCCDA BCC CFA mCCDA CFA <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 1.0 1.5 1.7 2.6 2.5

a None of these sixteen samples contained E. coli resistant to Preston broth supplement. b Count as obtained by direct plating on mCCDA, with limit for quantication of 1 log10 CFU/g.

combination (in which cefoperazone is included in both the broth and the agar), the combination Preston-mCCDA is composed of a selective agar with an antimicrobial supplement (based on cefoperazone) that differs from that in the enrichment broth (which is based on rifampicin and polymxin). Theoretically, the later combination would be the best for isolating Campylobacter. Nevertheless, all of the evaluated combinations of enrichment broth and subsequent plating agars could give false-negative results. The false-negative issue was highlighted in a recent study by Edson et al. (2009), who presented results from prociency testing of Campylobacter spp. detection in U.S. food laboratories over the period 1999 to 2007. The cumulative 9 years false-negative rate was 13.6%, which is higher than that for other foodborne bacteria (Edson et al., 2009). Our ndings show that careful choice of enrichment broth and subsequent plating agar can play a role in reducing false-negative results for Campylobacter in chicken meat samples. Even so, recovery of Campylobacter was enhanced by prolonged enrichment in Bolton broth for 48 h. However, this benet did not arise with enrichment in Preston broth or CFB. This is in line with a nding by Uyttendaele and Debevere (1996), as incubation of samples in Preston broth for 24 h was sufcient for enrichment of Campylobacter in articially inoculated chicken meat. Preston broth provided fast and consistent enrichment performance, regardless of the subsequent plating medium. Bolton broth contains meat peptone (10 g L1) as a main source of carbohydrates, while Preston broth contains both peptone (10 g L1) and Lab-Lemco powder (10 g L1). The later material is known for its excellent growthpromoting qualities (Atlas, 2006), so it may enhance Campylobacter recovery and growth in Preston broth. Compared to Bolton broth and Preston broth, the newly released CFB medium showed lower recovery of Campylobacter from chicken meat samples. No reason for that can be suggested, as the medium formulation is currently under patent right. However, CFB medium is commercialized in a liquid format, in ready-to-use mini bags. This format is not common among food microbiology media, which mostly are provided in dehydrated powdered form. Whether the unusual format of CFB medium has an impact on its efcacy (e.g. the broth might be more exposed to light and susceptible to temperature uctuations during delivery to and storage in the laboratory) needs to be evaluated further by the manufacturer.

In the present study, sample enrichment gave higher rates of Campylobacter recovery than direct plating. The number of Campylobacter in some samples may not have been large enough to allow for the recovery by direct plating. If this were the case, then enrichment would increase the rate of Campylobacter isolation from such samples, especially if the bacterial cells are injured (Gharst et al., 2006). Enrichment in Bolton broth and subsequent plating on CFA provided the highest recovery of Campylobacter from samples with low Campylobacter contamination levels (<10 CFU/g), with Campylobacter being recovered from almost twice as many samples as by the present ISO enrichment protocol. Proper choice of enrichment broths, depending on sample ecology and Campylobacter injury status, is important for the rapid and efcient enrichment of Campylobacter spp. in chicken meat samples (Kim et al., 2009). In conclusion, the present study adds to the current knowledge on performance evaluation of detection methods for Campylobacter in chicken meat. New selective agars (CFA and BCC) showed an attractive performance for easy and precise Campylobacter enumeration. Successful recovery of Campylobacter after enrichment in Bolton broth is very dependent on the choice of subsequent plating agar, and our ndings show that CFA performed signicantly better than mCCDA medium when coupled with Bolton broth. On the other hand, enrichment in Preston broth provides fast and consistent recovery, regardless of the subsequent plating media. Acknowledgements Ihab Habib is indebted to the Fund for Scientic ResearchFlanders (FWO-Vlaanderen, project G.024.09N) for a position grant as postdoctoral fellow. The authors would like to thank Justine Platteau, Martine Boonaert and Carine Van Lancker for technical assistance with this study. References
Atlas, R.M., 2006. Handbook of Microbiological Media for the Examination of Food, second ed. CRC Press/ITPS. Baylis, C.L., MacPhee, S., Martin, K.W., Humphrey, T.J., Betts, R.P., 2000. Comparison of three enrichment media for the isolation of Campylobacter spp. from foods. J. Appl. Microbiol. 89, 884e891. Butzler, J.P., 2004. Campylobacter, from obscurity to celebrity. Clin. Microbiol. Infec 10, 868e876. Corry, J.E., Post, D.E., Colin, P., Laisney, M.J., 1995. Culture media for the isolation of Campylobacter. Int. J. Food Microbiol. 26, 43e76. Edson, D.C., Empson, S., Massey, L.D., 2009. Pathogen detection in food microbiology laboratories: an analysis of qualitative prociency test data, 1999e2007. J. Food Saf. 29, 521e530. Engberg, J., Neimann, J., Nielsen, E.M., Aerestrup, F.M., Fussing, V., 2004. Quinoloneresistant Campylobacter infections: risk factors and clinical consequences. Emerg. Infect. Dis. 10, 1056e1063. European Food Safety Authority (EFSA), 2009. EFSAs 12th Scientic Colloquium e Assessing health benets of controlling Campylobacter in the food chain. http:// www.efsa.europa.eu/en/colloquiacampylobacter/publication/ colloquiacampylobacter.pdf. European Food Safety Authority (EFSA), 2010. Analysis of the Baseline Survey on the Prevalence of Campylobacter in Broiler Batches and of Campylobacter and Salmonella on Broiler Carcasses in the EU, 2008-Part A: Campylobacter and Salmonella Prevalence Estimates. http://www.efsa.europa.eu/en/scdocs/doc/1503.pdf. Friedman, C.R., Hoekstra, R.M., Samuel, M., Marcus, R., Bender, J., Shiferaw, B., Reddy, S., Ahuja, S.D., Helfrick, D.L., Hardnett, F., Carter, M., Anderson, B., Tauxe, R.V., 2004. Risk factors for sporadic Campylobacter infection in the United States: a casecontrol study in foodnet sites. Clin. Infect. Dis. 38 (Suppl. 3), S285eeS296. Gharst, G., Hanson, D., Kathariou, S., 2006. Effect of direct culture versus selective enrichment on the isolation of thermophilic Campylobacter from feces of mature cattle at harvest. J. Food Prot. 69, 1024e1027. Godschalk, P.C., Heikema, A.P., Gilbert, M., Komagamine, T., Ang, C.W., Glerum, J., Brochu, D., Li, J., Yuki, N., Jacobs, B.C., van Belkum, A., Endtz, H.P., 2004. The crucial role of Campylobacter jejuni genes in anti-ganglioside antibody induction in Guillain-Barre syndrome. J. Clin. Invest. 114, 1659e1665. Habib, I., Sampers, I., Uyttendaele, M., Berkvens, D., De Zutter, L., 2008. Baseline data from a Belgium-wide survey of Campylobacter species contamination in chicken meat preparations and considerations for a reliable monitoring program. Appl. Environ. Microbiol. 74, 5483e5489.

I. Habib et al. / Food Microbiology 28 (2011) 1117e1123 Humphrey, T., OBrien, S., Madsen, M., 2007. Campylobacters as zoonotic pathogens: a food production perspective. Int. J. Food Microbiol. 117, 237e257. Hunt, J. M., Abeyta, C., Tran, T., 2001. Campylobacter, F.D.A. Bacteriological Analytical Manual, eighth ed., Revision A. http://www.fda.gov/food/scienceresearch/ laboratorymethods/bacteriologicalanalyticalmanualbam/ucm072616#authors. (accessed 25.01.11). International Organisation for Standardization (ISO), 2006a. ISO 10272-1 Microbiology of Food and Animal Feeding StuffsdHorizontal Method for Detection and Enumeration of Campylobacter sppdPart 2: Enumeration Method. ISO, Geneva, Switzerland. International Organisation for Standardization (ISO), 2006b. ISO 10272-1 Microbiology of Food and Animal Feeding StuffsdHorizontal Method for Detection and Enumeration of Campylobacter sppdPart 1: Enrichment Method. ISO, Geneva, Switzerland. Jasson, V., Sampers, I., Botteldoorn, N., Lopez-Galvez, F., Baert, L., Denayer, S., Rajkovic, A., Habib, I., De Zutter, L., Debevere, J., Uyttendaele, M., 2009. Characterization of Escherichia coli from raw poultry in Belgium and impact on the detection of Campylobacter jejuni using Bolton broth. Int. J. Food Microbiol. 135, 248e253. Kim, S.A., Lee, Y.M., Hwang, I.G., Kang, D.H., Woo, G.J., Rhee, M.S., 2009. Eight enrichment broths for the isolation of Campylobacter jejuni from inoculated suspensions and ground pork. Lett. Appl. Microbiol. 49, 620e626. Lin, L.I., 1989. A concordance correlation coefcient to evaluate reproducibility. Biometrics 45, 255e268. Mullner, P., Jones, G., Noble, A., Spencer, S.E., Hathaway, S., French, N.P., 2009. Source attribution of food-borne zoonoses in New Zealand: a modied Hald model. Risk Anal. 29, 970e984.

1123

Musgrove, M.T., Berrang, M.E., Byrd, J.A., Stern, J., Cox, A., 2001. Detection of Campylobacter spp. in ceca and crops with and without enrichment. Poult. Sci. 80, 825e828. Nauta, M.J., Jacobs-Reitsma, W.F., Havelaar, A.H., 2007. A risk assessment model for Campylobacter in broiler meat. Risk Anal. 27, 845e861. Rosenquist, H., Bengtsson, A., Hansen, T.B., 2007. A collaborative study on a Nordic standard protocol for detection and enumeration of thermotolerant Campylobacter in food (NMKL 119, 3. Ed., 2007). Int. J. Food Microbiol. 118, 201e213. Scherer, K., Bartelt, E., Sommerfeld, C., Hildebrandt, G., 2006. Comparison of different sampling techniques and enumeration methods for the isolation and quantication of Campylobacter spp. in raw retail chicken legs. Int. J. Food Microbiol. 15, 115e119. StataCorp, 2007. Stata Statistical Software: Release 10. StataCorp LP, College Station, TX. Uyttendaele, M., Baert, K., Ghar, Y., Daube, G., De Zutter, L., Herman, L., Dierick, K., Pierard, D., Dubois, J.J., Horion, B., Debevere, J., 2006. Quantitative risk assessment of Campylobacter spp. in poultry based meat preparations as one of the factors to support the development of risk-based microbiological criteria in Belgium. Int. J. Food Microbiol. 111, 149e163. Uyttendaele, M., Debevere, J., 1996. Evaluation of Preston medium for detection of Campylobacter jejuni in vitro and in articially and naturally contaminated poultry products. Food Microbiol. 13, 115e122. Vandamme, P., Van Doorn, L.J., al Rashid, S.T., Quint, W.G., van der Plas, J., Chan, V.L., On, S.L., 1997. Campylobacter hyoilei Alderton et al., 1995 and Campylobacter coli Veron and Chatelain 1973 are subjective synonyms. Int. J. Syst. Bacteriol. 47, 1055e1060.