Occurrence of Campylobacter spp. in Cecal Contents Among Commercial Broilers in Iceland

Transcription

1 FOODBORNE PATHOGENS AND DISEASE Volume 2, Number 1, 2005 Mary Ann Liebert, Inc. Occurrence of Campylobacter spp. in Cecal Contents Among Commercial Broilers in Iceland N.J. STERN, 1 J. REIERSEN, 2 R. LOWMAN, 3 J.-R. BISAILLON, 3 V. FRIDRIKSDOTTIR, 4 E. GUNNARSSON, 4 K.L. HIETT, 1 and THE CAMPY-ON-ICE CONSORTIUM 5 ABSTRACT Campylobacter jejuni remains the most frequently reported bacterial cause of human gastroenteritis in Nordic countries. The primary source of transmission to humans is suggested as mishandled raw poultry or consuming improperly prepared chicken. The focus of this report was to characterize the prevalence and cell numbers of the organism within the commercial Icelandic poultry industry. Commercial broiler flocks were sampled from May 2001 through 2003 in a total population study. At the slaughter plant, 40 randomly selected ceca were obtained from each flock, pooled into four samples containing 10 ceca each, and analyzed. Cell numbers and prevalence of Campylobacter spp. were estimated by direct plating of dilutions onto Campy-Cefex agar and incubating the plates at 42 C under microaerobic atmosphere; colonies were confirmed as Campylobacter spp. by microscopy and latex agglutination to provide quantification of cell numbers per gm of cecal material. A total of 15.4% of the flocks carried the organism at at a maximum cell number of cfu/g, having a mean raw count of colonized birds at cfu/g (geometric mean of ). During the 3 years of sampling, the prevalence ranged from 17.6% to 17.3% to 12.7% for slaughter years 2001, 2002, and 2003, respectively. Isolation rates varied with numbers of catch lots (groups of birds taken for slaughter)/flock; with one catch lot/flock, the prevalence was 13.7%, with two 17.5%, and with three 33.3%. With increased flock size, isolation rates also increased; flocks of greater than 5,000 birds had a prevalence of 12.0% positive, 14.0% of flocks with 5,000 10,000 birds were positive, and 25.5% of flocks with more than 10,000 birds were positive for Campylobacter spp. Isolation rates varied with the processing lines: M was positive at 17.3%, B was positive at 10.1%, and G at 17.2%. Flocks were more frequently colonized in the warmer months, and younger birds were less frequently colonized than were older slaughtered birds. This study provides descriptive microbiology pertaining to Iceland broilers in a total population study. 1 USDA Agricultural Research Service, Poultry Microbiological Safety Research Unit, Athens, Georgia. 2 Chief Veterinary Office, Reykjavik, Iceland. 3 Canadian Food Inspection Agency, Ottawa, Ontario, Canada. 4 Institute of Experimental Pathology, Reykjavik, Iceland. 5 Campy-on-Ice Consortium: Iceland: Haraldur Briem, 11 Vala Fridriksdóttir, 8 Franklín Georgsson, 9 Eggert Gunnarsson, 8 Hjördís Harðardóttir, 10 Karl Kristinsson, 10 Jarle Reiersen. 7 Sweden: Eva Berndtson. 12 Canada: Jean-Robert Bisaillon, 16 Victoria Edge, 13 Aamir Fazil, 14 Pascal Michel, 14 Greg Paoli, 15 Ruff Lowman. 16 USA: Ken Callicott, 6 Kelli Hiett, 6 Norman Stern. 6 6 USDA Agricultural Research Service, Poultry Microbiological Safety Research Unit, Athens, Georgia. 7 Chief Veterinary Office, Reykjavik, Iceland. 8 Institute of Experimental Pathology, Reykjavik, Iceland. 9 Food Laboratory, Environmental and Food Agency of Iceland, Reykjavik, Iceland. 10 National University Hospital of Iceland, Reykjavik, Iceland. 11 Directorate of Health, Reykjavik, Iceland. 12 Swe-Chick, Kristianstad, Sweden. 13 Health Canada Center for Infectious Disease Prevention and Control, Guelph, Ontario, Canada. 14 Health Canada Laboratory for Foodborne Zoonoses, Guelph, Ontario, Canada. 15 Decisionalysis Risk Consultants, Inc., Ottawa, Ontario, Canada. 16 Canadian Food Inspection Agency, Ottawa, Ontario, Canada. 82

2 CAMPYLOBACTER IN ICELAND POULTRY PRODUCTION 83 INTRODUCTION CAMPYLOBACTERIOSIS is the most frequently reported bacterial cause of human gastroenteritis in many Nordic countries (Rautelin and Hanninen, 2000; Wedderkopp et al., 2000). Stern et al. (2003) reported that the prevalence of Campylobacter spp. enteritis in Iceland peaked at 116 domestically acquired cases per 100,000 persons in In the United States, approximately 2 million cases of human campylobacteriosis occur each year (Tauxe, 1992). Researchers have long suggested that a high percentage of human campylobacteriosis is associated with poultry (Norkrans and Svedhem, 1982; Hopkins and Scott, 1983; Oosterom et al., 1984; Harris et al., 1986; Deming et al., 1987; Kapperud et al., 1992; Tauxe, 1992; Frost, 2001). This relationship especially holds true in Iceland, where Hiett et al. (2001) reported that 85% of human isolates in 1999 had indistinguishable phylogenetic patterns of Campylobacter spp. when compared with isolates from poultry. Because of this strong association of human disease and poultry, we initiated an extensive study to determine what causal relationships were involved in the transmission of Campylobacter jejuni. The closed community and limited scale of well-defined broiler poultry operations in Iceland enabled us to conduct a total population-based study. We began a comprehensive cooperative program with the Iceland poultry industry and the Iceland Federal government to assess the presence of the organism in numerous potential sources on the island. We wished to characterize the prevalence and cell numbers of Campylobacter spp. colonizing broiler chickens to better understand transmission of the organism to humans. This article describes our findings related to carriage of Campylobacter spp. among broiler flocks produced in Iceland, from 2001 through MATERIALS AND METHODS Samples were taken over a 3-year period beginning in May of 2001 and continuing through January Three commercial companies M, B, and G were in operation at the beginning of the project and are included in the reported sampling. The centralized processing plants were located within 100 km of Reykjavik, Iceland. A fourth company, located on the north coast of Iceland, initiated operations during the course of the study and accounted for 5 10% of total Icelandic broiler production. Icelandic control program sampling results and isolates from this small production were shared with the project, but this small and remote production was excluded from the intensive project study, which focused on the main broiler production in the south of Iceland. Broilers were produced in Iceland consistent with modern commercial chicken production practices: Day of hatch chicks were delivered from the hatchery directly to broiler farm houses. Production houses were thoroughly cleaned, ceiling to concrete floor, and disinfected after each flock was transported. The majority of houses were populated with 5,000 10,000 birds, with newer houses having flocks sizes up to 24,000. As compared with many larger scale operations, sequential catch lots, most often on successive days, provided adequate amounts of poultry product to meet consumer demand. Birds grown an average age of 37 days (range, days) were generally transported less than 100 km to the processing plant in clean coops. At the processing plants, birds were stunned, exsanguinated, scalded, defeathered, and carcasses were eviscerated. From each catch, 40 randomly selected cecal samples (weighing 10 g each) were aseptically excised (with sterile instruments and clean plastic gloves) from the birds and pooled into four sterile plastic bags to create each of four sample lots per flock and transported on ice to the laboratory. Consistent with the official Nordic method of analysis (Berndtson, 1998), serial dilutions (in sterile Butterfield s buffer) of cecal contents were enumerated on Campy-Cefex agar (Stern et al., 1992) and incubated under microaerobic conditions at 42 C for 48 h. Isolated colonies having transluscent appearance consistent with typical Campylobacter spp. were assessed under phasecontrast microscopy for typical spiral-shaped cellular morphology and corkscrew movement. Latex agglutination was used for serological confirmation of Campylobacter spp. (Pan- Bio, Inc., Baltimore, MD; Latex-Campy).

3 84 Confirmed colonies were enumerated, and dilutions were calculated to provide cell numbers of the organism per gram of cecal material. Minimum sensitivity for enumeration and detection was 100 cfu/g. More Campylobacter-positive flocks may have been detected had an enrichment procedure been used. Microbiological testing results and related epidemiological information were recorded using spreadsheets (Microsoft Excel 2000), to be later transferred into a relational database (Microsoft Access 2002) for storage and editing purposes, and to establish links between the different sources of information in preparation of the querying process. Microsoft Access and Excel data processing and descriptive statistical functions were used to query the dataset to extract and summarize the descriptive information presented in the tables. RESULTS Table 1 provides the qualitative results for the analyses of 1,091 flocks that were tested in this study. In total, 15.4% (168) of the flocks STERN ET AL. manifested positive Campylobacter spp. colonization. This table also contains the quantitative data (cfu/g) from these same positive flocks. Owing to contemporary discussion on the subject, we analyzed the data by considering the mean raw counts, the geometric mean, and the mean of the log counts. The former provides greater emphasis on samples having higher cell numbers of Campylobacter spp. than does the latter analytical approach. The mean raw counts from the positive flocks was calculated at /g, the geometric mean at , while the mean of the positive log counts was calculated at /g of ceca. Table 1 also represents the data per slaughter year over the course of the study. Data gathering began in early May of 2001, which coincided with the upsurge in Campylobacter presence in Iceland. Although there was some variation in the numbers of flocks tested over the years, the prevalence of positive cecal samples was relatively consistent, with only limited variation, ranging from 12.7% to 17.6%, among the slaughter years. There was, however, almost one log difference between the 2001 and the 2002 data when comparing the TABLE 1. PREVALENCE AND LEVEL OF Campylobacter jejuni/coli WITHIN CECA OF COMMERCIALLY HATCHED OR SLAUGHTERED AND PROCESSED ICELAND BROILERS PER YEAR a From May Slaughter year to Jan 2004 Totals Isolation Percentage 17.62% 17.32% 12.74% 15.40% No. positive flocks No. tested flocks Raw counts from flocks Mean Median Standard deviation Minimum Maximum Log counts from flocks Mean log, geometric mean 6.59, , , , Standard error Median Minimum Maximum Confidence level (95%) a The geometric mean is the antilog of the mean log. It is often referred as the log mean, and is used to summarize data that have a few high or low values that would skew the arithmetic mean presented above. The geometric mean differs from the arithmetic mean in that data are transformed log values before averaging ( Student_Presentations/Preparation/Poster/Statistics/log_mean.html). Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated as cfu/g.

4 CAMPYLOBACTER IN ICELAND POULTRY PRODUCTION 85 means of the log counts in 2001 versus 2002 ( vs cfu/g). These values are, however, within the 95% confidence interval of each other. Table 2 provides data indicating the numbers of flocks tested with different numbers of catch lots. As the number of catch lots increased, there was a general increase in Campylobacterpositive flocks. The positive percentages were 13.7%, 17.5%, and 33.3% for flocks having 1, 2, or 3, catch lots respectively. Only three flocks had four catch lots, and the zero prevalence could have been an artifact of such a low sample size. The mean cell numbers of Campylobacter spp. per gram of cecal materials were very constant regardless of the numbers of catch lots. Table 3 addresses the question of whether flock size influences the prevalence of carriage of Campylobacter spp. Among all flocks, flocks with fewer than 5,000 birds were positive at a rate of 12.0%, flocks with 5,000 10,000 were positive at 14%, and flocks over 10,000 birds were positive at 25.5%. The cell numbers of Campylobacter spp., as measured by the mean of the log count, also trended upward with increased flock size, but not markedly. Table 4 represents the data considering differences in prevalence of Campylobacter-positive flock detection as related to the individual processing lines. Line B had overall rates of Campylobacter-positive flocks of 10.1%, whereas lines M and G had rates of 17.3% and 17.2%, respectively. The cell numbers of Campylobacter spp. per gram of ceca were quite constant among each of the participating lines, regardless of the statistical measure employed. Figure 1 indicates data on the prevalence of flock contamination by processing week in the year. For the 3 years of sampling, clusters of Campylobacter spp. colonization generally clustered within the warmer periods of the year. There were sporadic events of flock colonization outside of this time interval, but these were relatively infrequent. Even during the peak cluster time interval, the sporadic presence of the organism in the flocks ranged from 0% to almost 90%. The sporadic prevalence observed might well be explained by the limited numbers of flocks processed (average of 7.8 flocks per week) throughout all of Iceland in a given week. Figure 2 indicates the data describing the relationship of bird age to the prevalence of Campylobacter spp. recovery per flock. As the flock age increased, the prevalence in finding the organism also increased. Birds younger than 33 days of age did not yield the organism, TABLE 2. PREVALENCE AND LEVEL OF Campylobacter jejuni/coli WITHIN CECA OF COMMERCIALLY PROCESSED ICELAND BROILERS PER FLOCK PER NUMBER OF CATCH LOTS No. of catch lots Isolation Average flock size 5,967 10,158 13,573 13,309 Percentage flocks 13.70% 17.52% 33.33% 0 No. positive flocks No. tested flocks Raw counts from flocks Mean Median Standard deviation Minimum ,333 Maximum Log counts from flocks Mean log and geometric mean 6.03, , , Standard error Median Minimum Maximum Confidence level (95%) Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42 C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated as cfu/g.

5 86 STERN ET AL. TABLE 3. PREVALENCE AND LEVEL OF Campylobacter jejuni/coli CELL NUMBERS AND PRESENCE WITHIN CECA OF COMMERCIALLY PROCESSED, ICELAND BROILERS PER FLOCK SIZE 5,000 10,000 Flock size 5,000 birds birds 10,000 birds Isolation Percentage flocks 12.03% 13.97% 25.51% No. positive flocks No. tested flocks Raw counts from flocks Mean Median Standard deviation Minimum Maximum Log counts from flocks Mean log, geometric mean 5.91, , , Standard error Median Standard deviation Minimum Maximum Confidence level (95%) Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42 C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated as cfu/g. although this included only five flocks. The practical implications for processing birds of this age are clear in terms of total tonnage produced, although younger broiler chickens may be preferred for consumption. DISCUSSION Iceland has an international reputation for the effectiveness of its human disease surveillance. In part, this is because the Iceland Vet- TABLE 4. PREVALENCE AND LEVEL OF Campylobacter jejuni/coli CELL NUMBERS AND PRESENCE WITHIN CECA OF COMMERCIALLY PROCESSED, ICELAND BROILERS PER FLOCK PER LINE Line M B G Isolation Percentage flocks 17.29% 10.14% 17.24% No. positive flocks No. tested flocks Raw counts from flocks Mean Median Standard deviation Minimum Maximum Log counts from flocks Mean log and geometric mean 6.08, , , Standard error Median Minimum Maximum Confidence level (95%) Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42 C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated as cfu/g.

6 CAMPYLOBACTER IN ICELAND POULTRY PRODUCTION 87 FIG. 1. Prevalence of Campylobacter jejuni/coli within ceca of commercially processed Iceland broilers, per week of the year (May 2001 to January 2004). Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42 C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated. erinary Office has close ties with their public health authorities and insists upon maintaining the highest possible standards for their poultry production. In addition, the Iceland poultry industry is committed to addressing the Campylobacter spp. public health issue to reduce human disease and to enhance profitability. Because Campylobacter status is strictly monitored at the end of broiler production, flocks that are detected as positive at a pre-slaughter sampling of 10 fecal droppings at 2 4 days prior to slaughter are processed either at the end of the processing day and/or at the end of the processing week. These carcasses are then frozen before being released for commerce. FIG. 2. Prevalence of Campylobacter jejuni/coli colonizing ceca of commercially processed, Iceland broilers by age at slaughter. Direct plating of cecal suspension dilutions were made onto Campy-Cefex agar, plates incubated microaerobically at 42 C, suspect colonies examined microscopically, confirmed by latex agglutination test, and enumerated.

7 88 Freezing carcasses has been shown to reduce cell numbers of the organism by 1 2 logs, which reduces the public exposure via poultry in the same proportion (Stern et al., 1985). However, the profit derived by the poultry industry is eliminated when Campylobacter-positive flocks are frozen. Consequently, the industry is highly motivated to identify cost-effective interventions leading to the control of this human pathogen. The Iceland poultry industry is highly supportive of this ongoing study. The prevalence of Campylobacter-colonized flocks in Iceland was low when compared to either the 87.5% prevalence in the United States (Stern et al., 2001) or in other Scandinavian countries (Aho and Hirn, 1988; Berndtson et al., 1996). Different sensitivities of Campylobacter detection will vary with different protocols used. The frequencies in the Scandinavian countries referenced above appear to be lower than what has been found in the United States. This may be related to a lower concentration of environmental reservoirs of Campylobacter surrounding the poultry production houses, due to climatic factors, or perhaps due to Icelandic and general Scandinavian emphasis on biosecurity, construction quality of poultry houses, and sanitation between flocks to maintain control of Salmonella in poultry flocks and to prevent Campylobacter contamination of flocks. Even with relatively extraordinary measures taken to assure poultry isolation, Campylobacter spp. colonization occurs fairly routinely during summer production. Generally, the highest prevalence of colonization occurs at the same time when demand for fresh product is at its peak. Because of this spike in colonization during the warmest months, both consumers and producers have consequence; poultry producers suffer economic loss because they must freeze the contaminated products, and the consumers because their demand is not met by an adequate supply. Iceland has reduced the extraordinarily high cell numbers of disease related to public exposure to Campylobacter spp. through poultry that they had experienced in 1999 (Stern et al., 2003). This was most likely accomplished by a combination of enormous efforts directed at biosecurity, freezing of carcasses from known positive flocks, and perhaps some change in yet to be identified environmental factors. CONCLUSION We conducted a 3-year, 1,091 broiler flock survey for Campylobacter spp. in Iceland from May 2001 to January Flocks were colonized at a prevalence of 15.4% by cecal sampling. Rates of Campylobacter spp. colonization were relatively uniform throughout the period of sampling, and cell numbers were consistently at 10 7 cfu/g of cecal material among colonized flocks. Increased prevalence of colonization was related to the numbers of catch lots per flock, flock size, and age at time of slaughter. Variation was observed in prevalence of colonization related to commercial production lines. Higher prevalence of flock colonization by Campylobacter spp. was associated with the warmer months of production. ACKNOWLEDGMENTS Acknowledgement is extended to USDA Cooperative State Research, Education and Extension Service for providing financial assistance for project REFERENCES STERN ET AL. Aho, M., and J. Hirn Prevalence of campylobacteria in the Finnish broiler chicken chain from the producer to the consumer. Acta Vet. Scand. 29: Berndtson, E Campylobacter jejuni/coli Detection in Foods, 3rd ed. NMKL No Berndtson, E., U. Emanuelson, A. Engvall, et al A 1-year epidemiological study of campylobacters in 18 Swedish chicken farms. Prev. Vet. Med. 26: Deming, M.S., R.V. Tauxe, P.A. Blake, et al Campylobacter enteritis at a university from eating chickens and from cats. Am. J. Epidemiol. 126: Frost, J.A Current epidemiological issues in human campylobacteriosis. J. Appl. Microbiol. 90:85S 95S. Harris, N.V., N.S. Weiss, and C.M. Nolan The role of poultry and meats in the etiology of Campylobacter jejuni/coli enteritis. Am. J. Publ. Hlth. 76: Hiett, K.L., N.J. Stern, G. Alfredsson, et al Phylogenetic relationships of Campylobacter isolated in Iceland.

8 CAMPYLOBACTER IN ICELAND POULTRY PRODUCTION 89 Presented at the 11th International Workshop on Campylobacter, Helicobacter, and Related Organisms. Hopkins, R.S., and A.S. Scott Handling raw chicken as a source for sporadic Campylobacter jejuni infections. J. Infect. Dis. 148:770[letter]. Kapperud, G., Skjerve, E., Bean, N.H., et al Risk factors for sporadic Campylobacter infections: results of a case-control study in southeastern Norway. J. Clin. Microbiol. 30: Norkrans, G., and Å. Svedhem Epidemiologic aspects of Campylobacter jejuni enteritis. J. Hyg. Camb. 89: Oosterom, J., C.H. den Uyl, J.R.J. Bänffer, et al Epidemiologic investigations on Campylobacter jejuni in households with primary infection. J. Hyg. Camb. 92: Rautelin, H., and Hanninen, M.L Campylobacters: the most common bacterial enteropathogens in the Nordic countries. Ann. Med. 32: Stern, N.J., K.L. Hiett, G.A. Alfredsson, et al Campylobacter spp. in Icelandic poultry operations and human disease. Epidemiol. Infect. 130: Stern, N.J., P. Fedorka-Cray, J.S. Bailey, et al Distribution of Campylobacter spp. in selected United States poultry production and processing operations. J. Food Prot. 64: Stern, N.J., B. Wojton, and K. Kwiatek A differential-selective medium and dry ice-generated atmosphere for recovery of Campylobacter jejuni. J. Food Prot. 55: Stern, N.J., P.J. Rothenberg, and J.M. Stone Enumeration and reduction of Campylobacter jejuni in poultry and red meats. J. Food Prot. 48: Tauxe, R.V Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Campylobacter jejuni: Current Status and Future Trends. Nachamkin, I., Blaser, M., and Tompkins (eds.), American Society for Microbiology, Washington, D.C., pp Wedderkopp, A., E. Rattenborg, and M. Madsen National surveillance of Campylobacter in broilers at slaughter in Denmark in Av. Dis. 44: Address reprint requests to: Dr. N.J. Stern 950 College Station Rd. Athens, GA nstern@saa.ars.usda.gov