Use of MIDI Fatty Acid Methyl Ester Analysis To Monitor the Transmission of Campylobacter during Commercial Poultry Processing

Transcription

1 1610 Journal of Food Protection, Vol. 67, No. 8, 2004, Pages Copyright, International Association for Food Protection Use of MIDI Fatty Acid Methyl Ester Analysis To Monitor the Transmission of Campylobacter during Commercial Poultry Processing ARTHUR HINTON, JR., 1 * J. A. CASON, 1 MICHAEL E. HUME, 2 A KIMBERLY D. INGRAM 1 1 Poultry Processing and Meat Quality Research Unit, Agricultural Research Service, U.S. Department of Agriculture, 950 College Station Road, Russell Research Center, Athens, Georgia 30605; and 2 Food and Feed Safety Research Unit, Agricultural Research Service, U.S. Department of Agriculture, 2881 F&B Road, Southern Plains Research Center, College Station, Texas 77845, USA MS : Received 19 December 2003/Accepted 16 March 2004 ABSTRACT The presence of Campylobacter spp. on broiler carcasses and in scald water taken from a commercial poultry processing facility was monitored on a monthly basis from January through June. Campylobacter agar, Blaser, was used to enumerate Campylobacter in water samples from a multiple-tank scalder; on prescalded, picked, eviscerated, and chilled carcasses; and on processed carcasses stored at 4 C for 7 or 14 days. The MIDI Sherlock microbial identification system was used to identify Campylobacter-like isolates based on the fatty acid methyl ester profile of the bacteria. The dendrogram program of the Sherlock microbial identification system was used to compare the fatty acid methyl ester profiles of the bacteria and determine the degree of relatedness between the isolates. Findings indicated that no Campylobacter were recovered from carcasses or scald tank water samples collected in January or February, but the pathogen was recovered from samples collected in March, April, May, and June. Processing generally produced a significant (P 0.05) decrease in the number of Campylobacter recovered from broiler carcasses, and the number of Campylobacter recovered from refrigerated carcasses generally decreased during storage. Significantly (P 0.05) fewer Campylobacter were recovered from the final tank of the multiple-tank scald system than from the first tank. MIDI similarity index values ranged from to based on MIDI fatty acid methyl ester analysis of Campylobacter jejuni and Campylobacter coli isolates. Dendrograms of the fatty acid methyl ester profile of the isolates indicated that poultry flocks may introduce several strains of C. jejuni and C. coli into processing plants. Different populations of the pathogen may be carried into the processing plant by successive broiler flocks, and the same Campylobacter strain may be recovered from different poultry processing operations. However, Campylobacter apparently is unable to colonize equipment in the processing facility and contaminate broilers from flocks processed at later dates in the facility. Campylobacter is the major cause of human foodborne illnesses in developed countries, and contaminated poultry products are commonly cited as vehicles of outbreaks of campylobacteriosis (14, 26). Campylobacter jejuni and Campylobacter coli are routinely isolated from broiler rearing houses (17), and the bacteria are frequently found on the feathers, on the skin, and in the intestinal contents of live broiler chickens (2, 9, 26). Consequently, live broilers may transfer the enteropathogen from the farm to the poultry processing facility, where the bacteria can be spread to processing equipment, water, and personnel (1, 4, 15, 18). Commercial broiler processing consists of several operations including scalding, defeathering (picking), evisceration, and chilling (4). Although processing generally reduces microbial contamination of broiler carcasses (10), some processing steps may actually increase the level of carcass contamination (4, 28). Even though some microorganisms that are carried on the feathers and skin of broilers are removed by scalding and picking operations, other microbes may be spread to neighboring carcasses by scalder water and the mechanical picker (13, 15, 23, 28, 29). Sim- * Author for correspondence. Tel: ; Fax: ; ahinton@saa.ars.usda.gov. ilarly, evisceration removes enteric bacteria carried in the alimentary tract of live broilers, but evisceration and other processing operations may cause leakage of contaminated digestive contents from the cloaca and ruptured portions of the alimentary tract (24). Immersion chilling also leads to cross-contamination by bacteria that are not killed by antimicrobials in the chill water (29). Some microorganisms carried into the processing facility by live broilers may colonize processing equipment and contaminate broilers that are processed on subsequent days in the same facility (12, 13). The MIDI Sherlock microbial identification system (MIS) is a computer-operated automated system that identifies microorganisms based on the composition of their cellular fatty acids (22, 25). The MIS software also contains a dendrogram program that compares the similarity of the fatty acid profile of the cells identified by the system and establishes the degree of relatedness of microorganisms at the strain, biotype, or genus level. Since organisms linked at the strain level are considered to have originated from the same source, the MIDI Sherlock MIS may be used to track the movement of bacteria between different locations (11, 19, 21). The purpose of the present study was to enumerate

2 J. Food Prot., Vol. 67, No. 8 MIDI-FAME ANALYSIS OF CAMPYLOBACTER IN POULTRY PROCESSING 1611 Campylobacter spp. associated with various commercial poultry processing operations from January through June and to use the Sherlock MIS to identify Campylobacter isolates and to monitor the movement of Campylobacter strains through the processing line and on refrigerated carcasses. These findings may provide information that will be useful in identifying critical control points in processing that should be monitored to decrease contamination of processed poultry by Campylobacter. MATERIALS A METHODS Processing facility samples. Broiler carcasses and water samples from scald tanks were collected from a commercial poultry processing plant at monthly intervals from January through June. Carcasses that had undergone various stages of processing were taken from selected sites while the facility was in operation. During each visit, 6 prescalded (PS), 6 picked (PP), 6 eviscerated (EV), and 18 chilled (CH) carcasses were collected; placed in individual, sterile plastic bags (Fisher Scientific, Pittsburgh, Pa.); and transported to the laboratory for immediate analysis or refrigerated storage (13). Water samples were collected from each tank of the multiple-tank counterflow scald tank system as carcasses were passed through water in the tanks. The scald tank system is composed of three adjacent tanks that are maintained at a constant water level and temperature during scalding of carcasses, and all carcasses are passed successively through each tank during the scalding operation (5). A sterilized Wheaton stainless steel bomb sampler (Fisher Scientific) was used to collect water samples. The bomb sampler was lowered beneath the surface of the scald water and filled with water that was transferred to sterile plastic containers. Water samples were also transported to the laboratory for analysis. Enumeration and identification of Campylobacter. After samples were collected each month, all prescalded, picked, and eviscerated carcasses and water samples were analyzed immediately upon arrival at the laboratory. Six of the chilled carcasses were also immediately analyzed, and the remaining 12 chilled carcasses were placed in a 4 C refrigerator for storage. Six of the stored carcasses were analyzed after 7 days (R7) of refrigeration, and the remaining six carcasses were analyzed after 14 days (R14) of storage. The whole carcass rinse procedure was used to recover bacteria from the broiler carcasses, using previously described methods (6, 7, 11). Serial dilutions of the carcass rinsates and scald tank water were prepared in 0.1% Bacto Peptone (Difco, Becton Dickinson, Detroit, Mich.) and plated on Bacto Campylobacter agar, Blaser (Difco). Inoculated plates of the selective medium were incubated for 48 h at 42 C in a BBL GasPak jar system (Becton Dickinson, Sparks, Md.) containing an activated BBL CampyPak plus gas generator envelope. After incubation, plates were examined for the presence of Campylobacter-like colonies that appear as nonhemolytic, mucoid, gray, colonies on the media. Campylobacter-like colonies were counted, and the number of presumptive Campylobacter recovered from each sample was recorded. Direct plating procedures were unable to detect the bacterium in samples containing less than 1.00 log CFU of Campylobacter per ml. Two or three colonies from each sample were selected from plates containing isolated colonies. Colonies were streaked on Bacto Campylobacter agar, Blaser, and incubated for 48hat42 C in a BBL GasPak jar system with a microaerophilic gas generator envelope. Each isolate was then subjected to the latex-campy(jcl) Campylobacter culture confirmation test (Integrated Diagnostics, Inc., Baltimore, Md.). Colonies that were positive for the latex agglutination tests were subjected to further analysis. Presumptive Campylobacter isolates that were positive for the latex agglutination test were identified using the MIDI Sherlock MIS (MIDI, Inc., Newark, Del.). First, colonies were transferred to Remel blood agar (Remel Inc., Lenexa, Kans.) and incubated at 35 C for 48 h in a BBL GasPak jar system with a microaerophilic gas generator envelope. After three consecutive transfers and incubation on blood agar, cultures were harvested and fatty acids were extracted from the cellular membranes using the four-step extraction procedure recommended by the manufacturer (11, 22). The first step of the extraction procedure consisted of saponification by heating the cells at 100 C for 30 min in a methanolic sodium hydroxide solution composed of 15% (wt/vol) NaOH (Spectrum Chemical Manufacturing Corp., Gardena, Calif.) dissolved in 50% (vol/vol) methanol (Sigma-Aldrich, Inc., St. Louis, Mo.). Saponification lysed the bacteria and released cellular fatty acids. Methylation of released fatty acids was accomplished by mixing the resultant suspension in a solution of 3.25 N hydrochloric acid (Ricca Chemical Co., Arlington, Tex.) in methanol (46%, vol/vol) at 80 C for 10 min. The fatty acid methyl esters (FAMEs) were extracted from the aqueous phase to the organic phase of the mixture by a 10-min liquid-liquid extraction in a 1: 1 solution of hexane (Sigma-Aldrich, Inc.) and methyl-tert-butyl ether (Sigma-Aldrich, Inc.). A 5-min base wash of the FAME extracts in a sodium hydroxide solution (1.08%, wt/vol) was performed to remove free fatty acids and residue from the organic extract. FAME extracts were transferred to sample vials for identification and quantification by an Agilent 6890 gas chromatograph (GC) (Agilent Technologies, Palo Alto, Calif.). The GC was operated by a computer containing ChemStation (Agilent Technologies) and MIDI Sherlock MIS, version 4.5, software. ChemStation software set the GC parameters of operation and controlled GC functions throughout FAME analysis. The MIS software identified and quantified FAMEs of bacterial isolates and compared the FAME profiles with the library of known isolates contained in the CLIN 40 MIS computer library. Isolates were identified as Campylobacter when their FAME profiles matched the profile of Campylobacter isolates contained in the MIS library. The degree of the match between the unknown isolates and the data for the isolate entered in the MIS library is indicated by the similarity index value assigned to the unknown. Similarity index values may range from 0.00 (no match) to 1.00 (identical match). Dendrograms. The dendrogram program of the MIDI Sherlock MIS was used to determine the degree of relatedness (21) between Campylobacter isolates recovered from carcass rinsates and scald water. The MIDI dendrogram program determines the degree of relatedness of microorganisms based on the similarity of the FAME profiles generated during identification of the isolates. Isolates linked at a Euclidean distance of 10 generally belong to the same microbial species; isolates linked at a Euclidean distance of 6 generally belong to the same subspecies; and isolates linked at a Euclidean distance of 2.5 generally are considered to be the same strain (19, 21). Statistical analysis. Group means of data for the number of Campylobacter recovered were compared to determine significant differences in the number of bacteria recovered on each medium from carcasses taken from different stages of processing or refrigerated for different lengths of time and from scald tank water samples. A value of 0.99 log CFU/ml was assigned to samples from which no Campylobacter were recovered. Data were analyzed using GraphPad InStat version 3.05, 32 bit for Windows 95-

3 1612 HINTON ET AL. J. Food Prot., Vol. 67, No. 8 NT (GraphPad Software, San Diego, Calif.) to perform one-way analysis of variance. When the analysis of variance detected significant differences in group means, the Tukey-Kramer multiple comparisons test was used to determine which treatment groups differed significantly. All significant differences were determined at P RESULTS A DISCUSSION Population of Campylobacter on processed carcasses and in scald tank water. No Campylobacter were recovered from broilers processed at the facility in the months of January or February, but the bacterium was recovered from carcasses processed in March, April, May, and June (Table 1). The level of contamination of broiler chickens by Campylobacter varies throughout the year (30). The number of Campylobacter recovered from broilers usually increases during the warmer months of the year, then declines as environmental temperatures decrease (27). The correlation between the time of year and Campylobacter colonization of broilers was probably the reason that no Campylobacter were recovered during the first 2 months of the study. Findings indicated that commercial poultry processing generally reduced contamination of broiler carcasses by Campylobacter; although individual processing steps did not always significantly decrease the number of Campylobacter recovered from broiler carcasses (Table 1). On broilers processed in March, neither scalding, picking, evisceration, nor chilling produced significant reductions in the number of Campylobacter recovered. Therefore, the number of Campylobacter was the same on broilers entering the processing line as on broilers exiting the processing line. The number of Campylobacter on processed carcasses did decrease during refrigerated storage, however. Various processing steps were effective in decreasing the number of Campylobacter recovered from carcasses processed at the facility during April, May, and June. Each processing step possesses characteristics that may reduce Campylobacter populations. Scalding may reduce the population of these bacteria because of the ability of the scald water to rinse away contaminated dirt and debris on feathers and skin of the broilers. Furthermore, the relative high temperatures of the water may be lethal to the bacteria (31). Scalding broiler carcasses with 58 or 60 C water has been reported to reduce the number of Campylobacter on the carcasses, but scalding at temperatures of 51.8 C did not always reduce the level of contamination (8, 24, 29). The combination of exposure to the 56.4 C water in the third scald tank and the progressive cleansing of the carcasses as they were passed through the counterflow currents of the tank could have been factors in reducing the number of Campylobacter recovered from the carcasses. There was usually a reduction in the number of Campylobacter recovered from scald tank water as the broilers were moved through each of the tanks (Table 2). The recovery of the bacterium from scald water indicated that the pathogen was being shed by carcasses as they were moved through the water. Significantly fewer Campylobacter were recovered from tank 2 than from tank 1 in March through June, and significantly fewer Campylo- TABLE 1. Number of Campylobacter recovered on Campylobacter Blaser agar from rinsates of whole carcass rinses taken during various stages of commercial processing No. of Campylobacter (log CFU/ml) by month a January February March b April c May b June b Broiler carcass (6/6) A (6/6) B (6/6) C (6/6) D (4/6) D 1.00 (0/6) D (6/6) A (4/6) B (6/6) B (3/6) B 1.00 (0/6) B (6/6) A (6/6) B (6/6) C (2/6) C 1.00 (0/6) D 1.00 (0/6) D (6/6) A (6/6) AB (6/6) AB (6/6) AB (5/6) BC (1/6) C d Prescalded Picked Eviscerated Chilled Refrigerated 7 days Refrigerated 14 days a Values are averages standard deviations with number of Campylobacter-positive samples/total number of samples in parentheses. Values of 1.00 indicate that no Campylobacter were recovered by direct plating. Different letters in the same column indicate significant differences (P 0.05) in the number of CFU per milliliter recovered from whole carcass rinsates of the carcasses. b Counts include species identified as C. jejuni only. c Counts include species identified as C. jejuni and C. coli. d, samples not analyzed for the presence of Campylobacter.

4 J. Food Prot., Vol. 67, No. 8 MIDI-FAME ANALYSIS OF CAMPYLOBACTER IN POULTRY PROCESSING 1613 TABLE 2. Number of Campylobacter recovered on Campylobacter Blaser agar taken from a commercial multiple-tank counterflow scald tank system Scald tank Temp ( C) No. of Campylobacter (log CFU/ml) by month a January February March b April c May b June b (6/6) A (6/6) B (6/6) C (6/6) A (2/6) B (3/6) B (6/6) A (5/6) B 1.00 (0/6) C (6/6) A (6/6) B (4/6) C a Values are averages standard deviations with number of Campylobacter-positive samples/total number of samples in parentheses. Values of 1.00 indicate that no Campylobacter were recovered by direct plating. Different letters in the same column indicate significant differences (P 0.05) in the number of CFU per milliliter recovered from whole carcass rinsates of the carcasses. b Counts include species identified as C. jejuni only. c Counts include species identified as C. jejuni and C. coli. bacter were recovered from tank 3 than from tank 2 in March, May, and June. The reduction in the number of Campylobacter recovered from successive tanks of the multiple-tank counterflow scald tank system was probably due to the temperature of water in the tanks and the removal of bacteria from the carcasses as the were passed to progressively cleaner water (5). The temperature of water in tanks 1, 2, and 3 was maintained at average temperatures of 45.0, 49.9, and 57.2 C, respectively. Feather removal by the mechanical picker may also remove bacteria that are associated with feathers and skin of the bird, although some reports have indicated that defeathering can increase carcass contamination (1, 24). Cross-contamination in mechanical defeatherers may occur by microorganisms that have colonized the machine (13) and by enteric bacteria, such as Campylobacter, that are spread onto the carcass as contaminated fecal matter is forced out of the cloaca and onto the carcass surface during the process (24). The combination of scalding and picking produced significant reductions in the number of Campylobacter recovered from carcasses processed in April, May, and June. The evisceration process produced significant reductions in the number of Campylobacter recovered from carcasses processed in April and June. Since fecal material is the primary source of Campylobacter contamination of processing facilities, evisceration may physically remove the bacteria from the carcasses. On the other hand, evisceration may also increase carcass contamination by Campylobacter, if contaminated intestinal contents are spilled onto the surface of the carcass during the process (1, 24). The chilling process also pro- FIGURE 1. MIDI Sherlock microbial identification system dendrogram of selected Campylobacter jejuni recovered from samples of scald tank water and rinsates of broiler carcasses collected in March. First segment of strain name indicates the month in which samples were taken; second segment indicates processing operation or carcass treatment (SC1 3, scald tank 1, 2, or 3 water samples; PS, prescalded carcass; PP, picked; EV, eviscerated; CH, chilled; R7, 7 days of storage at 4 C; R14, 14 days of storage at 4 C); third segment indicates sample or carcass number; and fourth segment indicates isolate number. a o Similar superscripts indicate that isolates belonged to the same strain.

5 1614 HINTON ET AL. J. Food Prot., Vol. 67, No. 8 FIGURE 2. MIDI Sherlock microbial identification system dendrogram of selected Campylobacter jejuni and Campylobacter coli recovered from samples of scald tank water and rinsates of broiler carcasses collected in April. First segment of strain name indicates the month in which samples were taken; second segment indicates processing operation or carcass treatment (SC1 3, scald tank 1, 2, or 3 water samples; PS, prescalded carcass; PP, picked; EV, eviscerated; CH, chilled; R7, 7 days of storage at 4 C; R14, 14 days of storage at 4 C); third segment indicates sample or carcass number; and fourth segment indicates isolate number. a h Similar superscripts indicate that isolates belonged to the same strain ( a c,e,g,h C. jejuni isolates and d,f C. coli isolates). duced significant reductions in the number of Campylobacter recovered from carcasses processed in April and June. Immersion chilling may also reduce contamination by Campylobacter by physically removing the bacterium from carcasses (16, 24), and antimicrobial chemicals, such as chlorine (20, 31), may be added to chill water to kill undesirable microorganisms. Although results from some studies have indicated that Campylobacter may survive on fresh poultry during refrigerated storage (3), findings from the present study indicated that the population of the bacterium significantly (P 0.05) decreases during storage at 4 C. The pathogen was not recovered from scald water temperatures maintained at 60 C during commercial processing (29). Additionally, the carcasses were moved to progressively cleaner water as they moved from tank to tank, so the volume of debris and microorganisms that were being added in the water as the carcasses moved from tank to tank was reduced (5). MIS identification and construction of dendrograms. Sherlock MIS FAME analysis identified numerous C. jejuni and C. coli isolates that were recovered from the processing samples (Figs. 1 through 4). C. jejuni and C. coli are the two Campylobacter spp. most frequently isolated from the fecal material of poultry (17), and they are the species most commonly associated with human gastroenteritis (26). C. coli was only recovered from samples collected in April, while C. jejuni was recovered from samples collected from March through June and was the only Campylobacter species recovered from scald water samples and carcasses collected in May and June. The Sherlock MIS assigned each isolate a similarity index value after comparing the FAME profile of the unknown isolate to profiles of known isolates in the MIS library. Campylobacter isolates recovered from the processing facilities were assigned similarity index values between and (data not shown) by the Sherlock MIS. Similarity index values between and indicated that the isolates were atypical in their relation to Campylobacter strains in the MIS library. Assigned values of less than indicated that a similar FAME profile was not FIGURE 3. MIDI Sherlock microbial identification system dendrogram of selected Campylobacter jejuni recovered from samples of scald tank water and rinsates of broiler carcasses collected in May. First segment of strain name indicates the month in which samples were taken; second segment indicates processing operation or carcass treatment (SC1 3, scald tank 1, 2, or 3 water samples; PS, prescalded carcass; PP, picked; EV, eviscerated; CH, chilled; R7, 7 days of storage at 4 C; R14, 14 days of storage at 4 C); third segment indicates sample or carcass number; and fourth segment indicates isolate number. a g Similar superscripts indicate that isolates belonged to the same strain.

6 J. Food Prot., Vol. 67, No. 8 MIDI-FAME ANALYSIS OF CAMPYLOBACTER IN POULTRY PROCESSING 1615 FIGURE 4. MIDI Sherlock microbial identification system dendrogram of selected Campylobacter jejuni recovered from samples of scald tank water and rinsates of broiler carcasses collected in June. First segment of strain name indicates the month in which samples were taken; second segment indicates processing operation or carcass treatment (SC1 3, scald tank 1, 2, or 3 water samples; PS, prescalded carcass; PP, picked; EV, eviscerated; CH, chilled; R7, 7 days of storage at 4 C; R14, 14 days of storage at 4 C); third segment indicates sample or carcass number; and fourth segment indicates isolate number. a i Similar superscripts indicate that isolates belonged to the same strain. in the current Sherlock MIS database; however, the isolates were identified as the species that it was most closely related to in the database (22). These findings indicate that although the Sherlock MIS can be used to identify Campylobacter recovered from poultry processing environments, researchers may encounter a significant number of the poultry isolates that are not closely related to isolates in the Sherlock MIS library. Researchers may customize their Sherlock MIS library databases by use of the MIDI library generation software (21). By using the library generation software, the FAME profiles of Campylobacter spp. may be entered to produce an updated, customized Sherlock MIS library. A customized library would produce identifications with higher similarity index values than Sherlock MIS identifications because they will have more poultryrelated Campylobacter entered in the custom database. An isolate may be included in a dendrogram regardless of the similarity index value assigned during the identification phase, and dendrograms of selected Campylobacter isolates indicated that the same strain of the bacterium may be isolated from scald tank water and from carcasses taken from different locations in the processing line on the same day. Recovery of the same Campylobacter strain from different processing locations indicated that a group of bacteria with the same source of origin had survived earlier processing steps and was recovered from a carcass that had undergone one or more processing steps and/or refrigerated storage. The most widespread degree of cross-contamination was detected between processing samples collected in March (Fig. 1). For example, analyses of the March samples indicated that the same C. jejuni strain could be isolated from a prescalded carcass and a carcass that had been refrigerated for 14 days and that another strain of the bacterium could be recovered from an eviscerated carcass and from a carcass refrigerated for 7 days. The indication of widespread cross-contamination by C. jejuni in the month of March may have been related to the fact that none of the processing steps reduced the population of the bacterium recovered from the carcasses. Indications of cross-contamination by C. jejuni in April (Fig. 2), May (Fig. 3), and June (Fig. 4) and by C. coli in April (Fig. 2) were also apparent. Spread of the same Campylobacter strain was most evident among prescalded, picked, and eviscerated carcasses. Unlike results from studies that examined the spread of psychrotrophic, spoilage bacteria and yeasts during commercial poultry processing (12, 13), findings from the present study indicate that the same Campylobacter strain could not be isolated from samples taken from the same processing facility in different months (data not shown). Apparently, unlike some spoilage bacteria and yeasts, Campylobacter may be unable to colonize processing equipment. Therefore, the processing facility does not serve as a reservoir, spreading bacterium to flocks that are processed in the facility on subsequent dates (24). Previous studies have indicated that different Campylobacter isolates are recovered from poultry during different times of the year (14). Campylobacter isolated from processed carcasses are usually of fecal origin and do not appear to be indigenous to the facility (24). Since live poultry carry different strains of the bacterium during different parts of the year, this may lead to different strains of the bacterium being isolated from the processing plant in different times of the year. Sanitation measures and the inability of the organisms to survive outside of host animals reduce the likelihood that the bacterium will survive for significant periods of time in processing plants (10). Therefore, the goal of reducing the contamination of processed poultry by Campylobacter may focus primarily on eliminating the pathogen in live animals and preventing fecal contamination during processing. ACKNOWLEDGMENTS The authors acknowledge the technical assistance of Jerrie Barnett, Charles Hernandez, Fredda J. Murray, and Kathy Orr. REFERENCES 1. Berrang, M. E., R. J. Buhr, J. A. Cason, and J. A Dickens Broiler carcass contamination with Campylobacter from feces during defeathering. J. Food Prot. 64:

7 1616 HINTON ET AL. J. Food Prot., Vol. 67, No Berrang, M. E., J. A. Dickens, and M. T. Musgrove Effects of hot water application after defeathering on the levels of Campylobacter, coliform bacteria, and Escherichia coli on broiler carcasses. Poult. Sci. 79: Blankenship, L. C., and S. E. Craven Campylobacter jejuni survival in chicken meat as a function of temperature. Appl. Environ. Microbiol. 44: Bryan, F. L., and M. P. Doyle Health risks and consequences of Salmonella and Campylobacter jejuni in raw poultry. J. Food Prot. 58: Cason, J. A., A. Hinton, Jr., and K. D. Ingram Coliform, Escherichia coli, and salmonellae concentrations in a multiple-tank, counterflow poultry scalder. J. Food Prot. 63: Cox, N. A., J. E. Thomson, and J. S. Bailey Sampling of broiler carcasses for Salmonella with low volume water rinse. Poult. Sci. 60: Dickens, J. A., N. A Cox, J. S. Bailey, and J. E. Thomson Automated microbiological sampling of broiler carcasses. Poult. Sci. 64: Doyle, M. P., and D. J. Roman Growth and survival of Campylobacter fetus subsp. jejuni as a function of temperature and ph. J. Food Prot. 44: Genigeorgis, C., M. Hassuneh, and P. Collins Campylobacter jejuni infection on poultry farms and its effect on poultry meat during slaughtering. J. Food Prot. 49: Geornaras, I., and A. von Holy Bacterial counts associated with poultry processing at different sampling times. J. Basic Microbiol. 40: Glucksman, A. M., H. D. Skipper, R. L. Brigmon, and J. W. Santo Domingo Use of MIDI-FAME technique to characterize groundwater communities. J. Appl. Microbiol. 88: Hinton, A., Jr., J. A. Cason, and K. D. Ingram Enumeration and identification of yeasts associated with commercial poultry processing and spoilage of refrigerated broiler carcasses. J. Food Prot. 65: Hinton, A., Jr., J. A. Cason, and K. D. Ingram Tracking spoilage bacteria in commercial poultry processing and refrigerated storage of poultry. Int. J. Food Microbiol. 91: Hudson, J. A., C. Nicol, J. Wright, R. Whyte, and S. K. Hasell Seasonal variation of Campylobacter types from human cases, veterinary cases, raw chicken, milk, and water. J. Appl. Microbiol. 87: Humphrey, T. J., and D. G. Lanning Salmonella and Campylobacter contamination of broiler chicken carcasses and scald tank water: the influence of water ph. J. Appl. Bacteriol. 63: Izat, A. L., F. A. Gardner, J. H. Denton, and F. A. Golan Incidence and level of Campylobacter jejuni in broiler processing. Poult. Sci. 67: Kazwala, R. R., J. D. Collins, J. Hannan, R. A. P. Crinion, and H. O Mahony Factors responsible for the introduction and spread of Campylobacter jejuni infection in commercial poultry production. Vet. Rec. 126: Kotula, K. L., and Y. Pandya Bacterial contamination of broiler chickens before scalding. J. Food Prot. 58: Leonard, R. B., J. Mayer, M. Sasser, M. L. Woods, B. R. Mooney, B. G. Brinton, P. L. Newcomb-Gayman, and K. C. Carroll Comparison of MIDI Sherlock system and pulsed-field gel electrophoresis in characterizing strains of methicillin-resistant Staphylococcus aureus from a recent hospital outbreak. J. Clin. Microbiol. 33: Mead, G. C., W. R. Hudson, and M. H. Hinton Effect of changes in processing to improve hygiene control on contamination of poultry carcasses with campylobacter. Epidemiol. Infect. 115: MIDI, Inc LGS user s manual, Sherlock library generation system, version 4.5. MIDI, Inc., Newark, Del. 22. MIDI, Inc MIS operating manual, Sherlock microbial identification system, version 4.5. MIDI, Inc., Newark, Del. 23. Mulder, R. W. A. W., L. W. J. Dorresteijn, and J. van der Broek Cross-contamination during the scalding and plucking of broilers. Br. Poult. Sci. 19: Oosterom, J., J. Notermans, H. Karman, and G. B. Engels Origin and prevalence of Campylobacter jejuni in poultry processing. J. Food Prot. 46: Peltroche-Llacsahuanga, H., S. Schmidt, M. Seibold, R. Lutticken, and G. Haase Differentiation between Candida dubliniensis and Candida albicans by fatty acid methyl ester analysis using gasliquid chromatography. J. Clin. Microbiol. 38: Saleha, A. A., G. C. Mead, and A. L. Ibrahim Campylobacter jejuni in poultry production and processing in relation to public health. Worlds Poult. Sci. J. 54: Stern, N. J., P. Fedorka-Cray, J. S. Bailey, N. A. Cox, S. E. Craven, K. L. Hiett, M. T. Musgrove, S. Ladely, D. Cosby, and G. C. Mead Distribution of Campylobacter spp. in selected U. S. poultry production and processing operation. J. Food Prot. 64: Thomas, C. J., and T. A. McMeekin Contamination of broiler carcass skin during commercial processing procedures: an electron microscopic study. Appl. Environ. Microbiol. 40: Wempe, J. M., C. A. Genigeorgis, T. B. Farver, and H. Yusufu Prevalence of Campylobacter jejuni in two California chicken processing plants. Appl. Environ. Microbiol. 45: Willis, W. L., and C. Murray Campylobacter jejuni seasonal recovery observations of retail market broilers. Poult. Sci. 76: Yang, H., Y. Li, M. G. Johnson Survival and death of Salmonella Typhimurium and Campylobacter jejuni in processing water and on chicken skin during poultry scalding and chilling. J. Food Prot. 64: