Salmonella Prevalence in Crops of Ontario and Quebec Broiler Chickens at Slaughter 1

Transcription

1 ENVIRONMENT AND HEALTH Salmonella Prevalence in Crops of Ontario and Quebec Broiler Chickens at Slaughter 1 J. R. CHAMBERS,*, J.-R. BISAILLON, Y. LABBÉ, C. POPPE, and C. F. LANGFORD*, *Southern Crop Protection and Food Research Centre, Research Branch, Agriculture and Agri-Food Canada, Box 3650, 95 Stone Road West, Guelph, Ontario, Canada, N1H 8J7; Food Inspection Directorate, Canadian Food Inspection Agency, 59 Camelot Court, Nepean, Ontario, Canada, K1A 0Y9; Health of Animals Laboratory, Health Canada, 110 Stone Road West, Guelph, Ontario, Canada, N1G 3W4; and P.O. Box 68, Campbell s Bay, Quebec, Canada, J0X 1K0 ABSTRACT Swabs of crop contents of 635 broiler chickens were obtained from 9 Ontario and 1 Quebec processing plants and cultured for Salmonella to determine prevalence in broiler crops. Serotypes of positive cultures were determined to evaluate the serotype profile. The overall prevalence of contamination was low (4.3%). Prevalence was higher in broilers sampled in Quebec (5.8%) than in those sampled in Ontario (.%). In Quebec, there were differences in prevalence among the groups of broilers sampled at the various plants. These differences were believed to be attributable to differences in Salmonella prevalence among groups of flocks delivered to the plants due to the limited exposure of the chickens to the plant. The serotype profile of Salmonella isolated from the crops of broilers in this study was similar in several respects to profiles Received for publication April 8, Accepted for publication May 7, Contribution number 46 of the Centre for Food and Animal Research, Agriculture and Agri-Food Canada. To whom correspondence should be addressed: chambersj@em. agr.ca obtained from other surveys of Canadian broiler flocks using either environmental samples or cloacal swabs. Similarities included: predominance of Salmonella hadar and Salmonella heidelberg; several other common serotypes at a low prevalence; little Salmonella enteritidis isolated in other studies, and no S. enteritidis isolated in this study. Results of this field survey of Salmonella in crops of broilers are similar to those of Canadian studies of other internal and environmental sites of broilers. The similarity indicates that monitoring of Salmonella environments of flocks of live broiler chickens should define profiles of Salmonella contamination of the carcasses and would also aid in determination of Salmonella contamination status of broiler flocks. Such information would assist efforts to reduce Salmonella contamination in broiler chickens. (Key words: Salmonella, prevalence, carcass contamination, broiler, crop) INTRODUCTION Salmonella isolations from humans in the U.S. have risen steadily since 1955 and a fourfold increase had accrued by the 1990s (Centers for Disease Control, 199). Globally, salmonellosis has remained one of the three most common meat-associated diseases in humans (Cooper, 1994). Infected chickens represent the greatest potential Salmonella hazard to public health (Suphabphant et al., 1983). Salmonella is often present in the intestinal tracts of mammals and birds, is readily acquired from feed and environmental sources, and contaminates body 1998 Poultry Science 77: parts of fowl on the farm (Bryan and Doyle, 1995). Canada is not exempt from this problem. Surveys of registered Canadian broiler flocks between December 1989 to May 1990 revealed that 6 of 300 (76.9%) randomly sampled flocks were contaminated with at least one of 50 different Salmonella serovars (Poppe et al., 1991a). Poultry carcasses sampled from federally inspected abattoirs across Canada between 1983 and 1986 revealed Salmonella contamination for 60.9% of broiler and 69.1% of turkey carcasses (Lammerding et al., 1988). In the live chicken, the ceca have been reported to be the primary predilection site of Salmonella colonization by several research teams (Brownell et al., 1969; Fanelli et al., 1971; Barrow et al., 1988; Xu et al., 1988). The cloaca (Snoeyenbos et al., 198; Xu et al., 1988) and the crop Abbreviation Key: BPW = Buffered Peptone Water; HACCP = Hazard Analysis Critical Control Points. 1497

2 1498 (Snoeyenbos et al., 198; Barrow et al., 1988; Impey and Mead, 1989) are also often commonly colonized; however, colonization of the crop is highly transient (Snoeyenbos et al., 198) or less persistent (Barrow et al., 1988). On the other hand, rates of Salmonella contamination were either equal or much greater for crop vs cecal samples of 150 to 00 broilers from each of three commercial broiler flocks (Hargis et al., 1995). After contents of the intestines are voided, these enteric bacteria contribute to the surface contamination of carcasses. Carcass rinses prior to scalding indicate that Salmonella are firmly attached to contaminated chickens on arrival at the processing plant (Lillard, 1989). Rupture of contaminated crops during carcass evisceration also contributes to carcass contamination. Salmonella cross-contamination occurs among birds during transit to market (Bryan and Doyle, 1995). Transportation of broilers from the farm to the processing plant has been associated with increases in frequency and level of broiler crop contamination by Salmonella (Hargis et al., 1995) and levels of carcass surface contamination by Campylobacter (Stern et al., 1995). This increase in Salmonella frequency may be attributed to preslaughter feed restriction. Humphrey et al. (1993) reported increased frequency of recovery of Salmonella enteritidis from crops of chickens that had been deprived of feed for 4 h. Moreover, crops were reported to be 80 times more likely than ceca to rupture during evisceration (Hargis et al., 1995). The crop appears to be an important Critical Control Point in the prevention of Salmonella contamination of carcasses. From a Hazard Analysis Critical Control Point (HACCP) approach, elimination of Salmonella from chickens prior to their arrival at the plant would prevent much of the subsequent carcass contamination with Salmonella due either to surface contamination of the live chicken or to crop colonization. In situations in which only some flocks of chickens are contaminated, scheduling the slaughter of negative flocks ahead of positive flocks on a daily basis would reduce Salmonella contamination of carcasses. To achieve this status, a means of evaluating Salmonella contamination status of the live chicken is required. Cecal contents sampling of chickens is feasible during slaughter; however, there is no convenient method for sampling cecal contents of live chickens. Therefore, the crop of live chickens appears to be a choice site for monitoring Salmonella contamination through swab sampling. A Canadian survey was carried out to determine: 1) the average rate of crop contamination by Salmonella in broiler flocks at processing plants in Ontario and in Quebec, and ) whether differences in rate of crop contamination of flocks existed between provinces and among broiler groups sampled at processing plants 3Starswab No. S131, Starplex Scientific, Etobicoke, ON, Canada, M9W 6Y3. CHAMBERS ET AL. within provinces. In addition, positive samples were serotyped to assess the diversity of the Salmonella serotypes involved in broiler crop contaminations. Sampling MATERIALS AND METHODS Crop swabs were obtained from 635 broilers being processed at 9 Ontario and 1 Quebec processing plants. Plants contributed from 5 to 8 swabs per sample day with one plant providing 11 swabs on one date. Dates of sampling included 3 consecutive d in March and 1 d between May 1 and June 1. Crops were sampled from most of the processing plants on four of these dates; however, crops were sampled at two plants on three dates, two on five dates, and three on six dates. Swab Procedure At the processing plant, swabs were collected after slaughter and plucking but prior to evisceration. At 1-h intervals, the 10th carcass subsequent to a carcass noted on the processing line was selected for crop swabbing. This carcass was removed from the line and had the crop removed manually. The esophagus at either end of the crop was tied closed to avoid escape of and carcass contamination by the contents. The crop was cut open using a sterile scalpel and the sterile swab (Microorganism Collection and Transport System, Starplex Scientific)3 was inserted into the contents of the crop and stirred to pick up material. Following removal from the crop, the swab was returned to the sterile test tube with protective medium and the stopper was replaced. The charcoal base of the protective medium (Microorganism Collection and Transport System) was expected to neutralize acidity of the crop contents and enhance the survival of enteric bacteria during shipment of the samples to the laboratory. Swab Storage and Transport Following collection, tubes with swabs for a specific day were placed in refrigerated storage containers for shipment to a central laboratory for culturing to detect and isolate Salmonella. Culturing Culture of samples commenced within d of collection with the exception of collections from three processing plants on 1 d. The latter samples had culture initiated within 3 d of collection. On arrival at the laboratory, each swab was transferred to a tube containing 10 ml 1.0% Buffered Peptone Water (BPW) and the swab stem was cut with flamed scissors before vortexing the swab in the BPW. The swab in the tube was then incubated for 4 h at 35 C (pre-enrichment) before 0.1 ml of the incubated BPW was transferred to a

3 SALMONELLA PREVALENCE IN CROPS OF BROILERS 1499 TABLE 1. Presence of Salmonella in crops of broilers at processing plants in Ontario and Quebec 1 Salmonella Ontario Quebec Total Positive Negative Total Percentage positive Chi-square test of difference between provinces = x 1 = ; (P < 0.05). modified semi-solid Rappaport-Vassiliadis (De Smedt and Bolderdijk, 1987) agar plate and incubated for 18 to 7 h at 4 C. Plates were examined after 18 to 4 h of incubation and at daily intervals thereafter to determine whether a zone of migration was visible. If no zones had developed after 7 h, the sample was considered negative and discarded. If zones of migration of at least 0 mm from the site of inoculation were detected, a loopful of the growth from the outer edge of migration was streaked onto a plate of MacConkey agar and incubated for 18 to 4 h at 35 C. Up to five isolated, colorless colonies from each plate were examined for slide-agglutination using the Bacto Salmonella O antiserum poly A-I and Vi.4 Each colony was then transferred to a half plate of MacConkey agar and incubated for 18 to 4 h at 35 C before refrigerated storage to await biochemical classification using the Enterotube system. Confirmed Salmonella colonies were transferred to nutrient agar slants and incubated for 18 to 4 h at 35 C before shipment to another laboratory for serotyping. Serotyping Procedures to serotype Salmonella have been described previously (Shipp and Rowe, 1980; Difco Laboratories, 1984; Popoff and Le Minor, 199). Statistical Analysis Results were analyzed by chi-square procedures (Snedecor and Cochran, 1967) to compare frequencies of Salmonella isolation between provinces and among plants within provinces. Binomial tables with only two levels of a factor (e.g., provinces) had chi-square values corrected for continuity. RESULTS AND DISCUSSION Salmonella Contamination of Crops Twenty-seven (4.3%) of the 635 broiler crop samples tested positive for Salmonella spp. (Table 1). These rates of 4Catalog no *, Difco Laboratories, Detroit, MI contamination are low in relation to crop Salmonella contamination rates (4.3% vs 16.0, 6.0 and 86.7%) of three commercial broiler flocks sampled in Texas (Hargis et al., 1995). These differences are difficult to explain unless climatic differences are involved. Unfortunately, there are no known other Canadian crop samples tested for Salmonella spp. on a contaminated chicken basis for comparison; however, the rates of contamination of Canadian broiler flocks, based on 1 pooled litter samples and 3 water samples per flock, were high (76.9%; Renwick et al., 199). These results, evaluated on a flock basis, should yield higher values than would be expected on an individual sample or chicken basis. Moreover, these flocks varied in age from near hatch to slaughter, whereas, in the current study, all flocks were at slaughter age, normally 5 wk or older. Salmonella colonization rates and counts tend to decline with age not only for broilers (Sadler et al., 1969) but also for litter samples after flocks reach 4 wk of age (Renwick et al., 199). These low levels of contamination of broiler crops by Salmonella suggest that this source of contamination should not be a major threat to poultry meat. On the other hand, Hargis et al. (1995) stated that crops are more than 80 times more likely to rupture during evisceration than ceca. Hence, crop contamination, even at these low levels, represents a serious Salmonella threat to broiler meat. The frequency of positive samples was higher (P < 0.05; x 1df = 4.1) in Quebec, 1 of 36 (5.8%), than in Ontario, six of 73 (.%) (Table ). Differences in Salmonella contamination of broiler litter samples have been observed among regions, which included Quebec and Ontario (Renwick et al., 199). Renwick et al. (199) reported lower contamination rates for litter and water samples from broiler flocks in Quebec than from those in Ontario. There were no differences in Salmonella frequency among groups of broilers sampled at plants within Ontario; however, broiler groups sampled from plants differed (P< 0.05; x 11 df = 0.03) in frequency in the province of Quebec. Broilers would be infected before arrival at the plant. Hence, these differences reflect differences among the groups of flocks supplied to each plant rather than differences in processing operations in the Quebec plants. Salmonella Serotypes Nine different Salmonella serotypes were represented in the 7 positive crop swab samples. Distribution of the serotypes is presented in Table 3. Salmonella hadar was the most common serotype, similar to observations by Irwin et al. (1989) and Poppe et al. (1991a). Salmonella heidelberg was the second most frequently isolated serotype, also as reported by Irwin et al. (1989). The remaining serotypes were common among the 9 other serotypes isolated from Ontario broilers by Irwin et al. (1989) and the top 15 serotypes found by Poppe et al. (1991a). Salmonella infantis, S. anatum, S. enteritidis, S. ohio, S. senftenberg, and S.

4 1500 CHAMBERS ET AL. TABLE. Presence of Salmonella in crops of broilers by processing plants in Ontario and Quebec Ontario 1 Quebec Plant No. positive No. tested No. positive No. tested Total Negative plants 4/9 or 44.4% 4/1 or 33.3% 1Chi-square test of differences among Ontario broiler groups = x 8 = (P = 0.531). Chi-square test of differences among Quebec broiler groups = x 11 = (P = 0.045). montevideo were reported by Poppe et al. (1991a); however, they were not observed in this study. The absence of S. enteritidis in this survey is consistent with the low frequencies reported in broilers and in layers (Poppe et al., 1991a,b) and the low frequency (1.5%) of S. enteritidis among all human Salmonella isolates in 1991 in Canada (Poppe, 1994). In contrast, high frequencies of contamination have been reported for S. enteritidis in other countries following a dramatic increase in incidence during the 1980s (Rodrigue et al., 1990). In many countries, this increase was due largely to emergence of S. enteritidis phage type 4 (Rodrigue et al., 1990). In Canada, only 18.3% of S. enteritidis isolates from humans were phage type 4 and many of these infections were probably acquired outside Canada (Lior and Khakhria, 199). The similar relative frequencies of Salmonella serotypes isolated from broilers sampled from across Canada or from parts of Canada by testing either environmental samples (Poppe et al., 1991a, 1994), cloacal swabs (Irwin et al., 1989), or crop swabs (current study), indicate that there is good general agreement. Multiple isolations of Salmonella serotypes from broiler crops were examined for evidence of clustering. Lack of TABLE 3. Profile of Salmonella serotypes from crops of Ontario and Quebec broilers clustering implies that Salmonella serotypes are generally available. There were six cases of multiple isolations of a serotype at different plants on the same date. The five serotypes involved were S. kentucky, S. thompson, S. heidelberg, S. hadar, and S. typhimurium. The duplicate isolation of S. hadar occurred twice, one pair of isolations in Quebec and another pair in Ontario. Multiple isolations of a serotype in the same plant on the same day involved one new serotype, S. mbandaka (a double isolation), and S. hadar (a quintuple isolation, four from the same flock). There were single occurrences of three serotypes, S. agona, S. berta, and S. schwarzengrund, plus other single test date occurrences of S. hadar (on four dates) and S. heidelberg (on one date), which had multiple occurrences on other test dates. Unfortunately, higher numbers of Salmonellacontaminated samples are required to make inferences about specific sources or vectors; however, evidence obtained suggests there is little clustering with the possible exception of S. hadar in one flock and that Salmonella serotypes appear to be widely distributed. ACKNOWLEDGMENTS The authors appreciate the capable technical assistance of Joanne Elsaesser with plating of the swabs and the assistance of the FP&I staff with collection of the swabs at the processing plants. Serotype Number S. hadar 13 S. heidelberg 3 S. mbandaka S. kentucky S. thompson S. typhimurium S. agona 1 S. berta 1 S. schwarzengrund 1 Total 7 REFERENCES Barrow, P. A., J. M. Simpson, and M. A. Lovell, Intestinal colonization in the chicken by food-poisoning Salmonella serotypes: microbial characteristics associated with faecal excretion. Avian Pathol. 17: Brownell, J. R., W. W. Sadler, and M. J. Fanelli, Factors influencing the intestinal infection of chickens with Salmonella typhimurium. Avian Dis. 13:

5 SALMONELLA PREVALENCE IN CROPS OF BROILERS 1501 Bryan, F. L., and M. P. Doyle, Health risks and consequences of Salmonella and Campylobacter jejuni in raw poultry. J. Food Prot. 58: Centers for Disease Control, 199. Summary of notifiable diseases, United States, Morbid. Mortal. Weekly Rep. 40(53):41. Cooper, G. L., Salmonellosis-infections in man and the chicken: pathogenesis and the development of live vaccines a review. Vet. Bull. 64: De Smedt, J. M., and R. F. Bolderdijk, Dynamics of Salmonella isolation with modified semi-solid Rappaport- Vassiliadis medium. J. Food Prot. 50: Difco Laboratories, Salmonella serology. Pages in: Difco Manual. 5th ed. Difco Laboratories, Detroit, MI. Fanelli, M. J., W. W. Sadler, C. E. Franti, and J. R. Brownell, Localization of salmonellae within the intestinal tract of chickens. Avian Dis. 15: Hargis, B. M., D. J. Caldwell, R. L. Brewer, D. E. Corrier, and J. R. Deloach, Evaluation of the chicken crop as a source of Salmonella contamination for broiler carcasses. Poultry Sci. 74: Humphrey, T. J., A. Baskerville, A. Whitehead, B. Rowe, and A. Henley, Influence of feeding patterns on the artificial infection of laying hens with Salmonella enteritidis phage type 4. Vet. Rec. 13: Impey, C. S., and G. C. Mead, Fate of salmonellas in the alimentary tract of chicks pre-treated with a mature caecal microflora to increase colonization resistance. J. Appl. Bacteriol. 66: Irwin, R. J., S. A. McEwen, R. C. Clarke, and A. H. Meek, The prevalence of verocytotoxin-producing Escherichia coli and antimicrobial resistance patterns of nonverocytotoxinproducing Escherichia coli and Salmonella in Ontario broiler chickens. Can. J. Vet. Res. 53: Lammerding, A. M., M. M. Garcia, E. D. Mann, Y. Robinson, W. J. Dorward, R. B. Truscott, and F. Tittiger, Prevalence of Salmonella and thermophilic Campylobacter in fresh pork, beef, veal and poultry in Canada. J. Food Prot. 51:47 5. Lillard, H. S., Incidence and recovery of Salmonellae and other bacteria from commercially processed poultry carcasses at selected pre- and post-evisceration steps. J. Food Prot. 5: Lior, H., and R. Khakhria, 199. Canada s most common Salmonella serotypes and Salmonella enteritidis phagetypes ( ). Safety Watch 7:3. Popoff, M. Y., and L. Le Minor, 199. Antigenic formulas of the Salmonella serovars. WHO Collaborating Centre for Reference and Research on Salmonella, Paris, France. Poppe, C., Salmonella enteritidis in Canada. Int. J. Food Microbiol. 1:1 5. Poppe, C., R. J. Irwin, S. Messier, G. G. Finley, and J. Oggel, 1991a. The prevalence of Salmonella enteritidis and other Salmonella spp. among Canadian registered commercial chicken broiler flocks. Epidemiol. Infect. 107: Poppe, C., R. J. Irwin, C. M. Forsberg, R. C. Clarke, and J. Oggel, 1991b. The prevalence of Salmonella enteritidis and other Salmonella spp. among Canadian registered commercial layer flocks. Epidemiol. Infect. 106: Renwick, S. A., R. J. Irwin, R. C. Clarke, W. B. McNab, C. Poppe, and S. A. McEwen, 199. Epidemiological associations between characteristics of registered broiler chicken flocks in Canada and the Salmonella culture status of floor litter and drinking water. Can. Vet. J. 33: Rodrigue, D. C., R. V. Tauxe, and B. Rowe, International increase in Salmonella enteritidis: A new pandemic? Epidemiol. Infect. 105:1 7. Sadler, W. W., J. R. Brownell, and M. J. Fanelli, Influence of age and inoculum level on shed patterns of Salmonella typhimurium in chickens. Avian Dis. 13: Shipp, C. R., and B. Rowe, A mechanised microtechnique for Salmonella serotyping. J. Clin. Pathol. 33: Snedecor, G. W., and W. G. Cochran, Statistical Methods. 6th ed. The Iowa State University Press, Ames, IA. Snoeyenbos, G. H., A. S. Soerjadi, and O. M. Weinack, 198. Gastrointestinal colonization by salmonellae and pathogenic Escherichia coli in monoxenic and holoxenic chicks and poults. Avian Dis. 6: Stern, N. J., M.R.S. Clavero, J. S. Bailey, N. A. Cox, and M. C. Robach, Campylobacter spp. in broilers on the farm and after transport. Poultry Sci. 74: Suphabphant, W., M. D. York, and B. S. Pomeroy, Use of two vaccines (live G30D or killed RW16) in the prevention of Salmonella typhimurium infections in chickens. Avian Dis. 7: Xu, Y. M., G. R. Pearson, and M. Hinton, The colonization of the alimentary tract and visceral organs of chicks with salmonellas following challenge via the feed: bacteriological findings. Br. Vet. J. 144: