RESEARCH NOTES Salmonella Contamination in Commercial Eggs and an Egg Production Facility FRANK T. JONES,1-2 DAVID V. RIVES,1 and JOHN B. CAREY3 Poultry Science Department, North Carolina State University, Raleigh, North Carolina 27695 and Poultry Science Department, Texas A&M University, College Station, Texas 77843-2472 ABSTRACT Egg samples were collected from various stages of an egg processing operation and from the attached production facility. Salmonella was isolated from 72.0% of all samples collected from the laying house environment. Recovery of Salmonella from flush water, ventilation fan, egg belt, and egg collector samples were (positive samples/total samples collected): 2/2,4/4,16/ 22, and 14/22, respectively. Salmonella was found on 7 of the 90 eggshells sampled before processing and 1 of 90 eggshells sampled after processing, but Salmonella was not found in the 180 eggs analyzed for internal contamination following processing. The one eggshell found positive for Salmonella following processing was detected when the pH of wash water samples was lowest (10.19). The 60 isolates from production facilities included the following Salmonella serotypes: S. agona, S. typhimurium, S. infantis, S. derby, S. heidelberg, S. California, S. montevideo, S. mbandaka, and untypable. The 22 isolates obtained from eggshells prior to processing were serotyped as S. heidelberg and S. montevideo. All five isolates obtained from eggshells after processing were serotyped as S. heidelberg. These data suggest that although the shells of about 1% of commercial eggs are contaminated with Salmonella, contamination of the internal contents of eggs with Salmonella is a rare event. {Key words: Salmonella, eggs, egg production) 1995 Poultry Science 74:753-757
INTRODUCTION Approximately 83% of all eggs consumed by the general population in the United States are transported to the purchaser in the shell (Anderson, 1993). Prior to 1970, eggs that were handled in an unsanitary fashion were often associated with foodborne illness (Woodward, 1989), but with the passage of the Egg Product Sanitation Act in 1970, illness due to contaminated eggs decreased dramatically
Received for publication September 1, 1994. Accepted for publication December 2, 1994. Worth Carolina State University. 2 To whom correspondence should be addressed. 3 Texas A&M University.
(Bryan, 1980). A low incidence of Salmonella was found in intact, correctly processed eggs (Baker et al, 1980b) and attempts to produce contaminated eggs through oral or intravenous infection of hens with Salmonella were unsuccessful (Baker et al, 1980a). The interior content of eggs was reported to be free of Salmonella at the time of lay (Mayes and Takeballi, 1983). In 1986, food poisoning outbreaks involving Salmonella enteritidis were associated with the consumption of Grade A, sanitized, and presumably uncracked eggs (Potter, 1987). Because several earlier studies had demonstrated that S. enteritidis was among the serotypes likely to cause ovarian infection (Faddoul and Fellows,
753
754
JONES ET AL.
1966; Snoeyenbos et al, 1969), it was hypothesized that these eggs came from hens with infected ovaries (St. Louis et al., 1988). Experimental infection of hens with S. enteritidis was also demonstrated to produce contaminated eggs (Gast and Beard, 1990). Recently, public health officials have also speculated that egg consumption may be related to Salmonella typhimurium infections in humans (Potter, 1989). Such events since 1986 have raised questions about the frequency of Salmonella contamination of both shell eggs (particularly the internal contents) and egg production facilities (Potter, 1987). In view of this situation, this study was undertaken in order to obtain a clearer understanding of the degree to which eggs, egg contents, and egg production facilities are contaminated with Salmonella.
MATERIALS AND METHODS A commercial egg production complex containing six environmentally controlled houses was sampled for Salmonella contamination. The complex had an on-site egg processing facility and each house was equipped with belts for egg collection as well as a flush system for manure removal. Manure was collected within a single lagoon on the complex. Environmental samples were collected from two houses within the complex. Because no differences were seen in either the distribution or rate of contamination of the houses sampled, the data were pooled. Within each house, 11 samples were collected from egg belts, 11 from egg collectors, 2 from fans, and 1 from flush waters. Egg belts and egg collectors were sampled by wetting a 5 x 5 cm sterile gauze pad with Tryptic Soy Broth (TSB)*
4
Difco, Detroit, MI 48232. 5Staalkat B. V., 7120 AB Aalten, The Netherlands. 6 Egg Turg, Oxford Chemical Division of Diversey Corp., Livonia, MI 48150. ^Fisher Acumet model 730 pH meter, Fisher Scientific, Pittsburgh, PA 15219-4785. 8 National Laboratories, Montvale, NJ 07645. 9 Nasco, Ft. Atkinson, WI 53538.
and rubbing a 5 x 10 cm area before returning it to the medium for incubation at 37 C for 18 to 24 h. Samples from fans were collected by rubbing an area of about 930 cm2 on the fan louvers with a sterile gauze pad as described above. Flush waters were sampled by wetting a sterile gauze pad in the waters and returning the pad to TSB for incubation. Gauze pads were incubated in 125 mL of TSB for analysis. Unused disposable latex gloves were worn during the collection of each sample. The processing facility washed eggs using a 180 case per hour machine5 with a single 473-L recirculating wash water tank. Water in the wash tank was maintained at ~ 46 C and a commercial egg washing detergent6 was added at .3% each time the tank was filled. Water was discarded after 4 h of processing and replaced with fresh well water. Eggs were collected for Salmonella analysis on 2 separate d in the morning, at noon, and in the afternoon. At each sampling, 30 eggs were randomly and aseptically collected both before processing at the accumulator and after processing and packaging. Wash water samples were collected each time egg samples were taken by using sterile 474-mL (16-oz.) glass jars to collect samples washer overflow. Wash water pH was measured in duplicate using a pH meter7 immediately following sample collection. Half the eggs collected were sampled for external salmonellae using the method of Gentry and Quarles (1972), and the liquid contents of the other half was assayed for contamination. Internal egg contamination was determined by soaking the egg in a 1% solution of Amphyl8 (a phenol disinfectant) for 30 s, blotting the egg dry on a clean paper towel, cracking the egg with the blunt end of flamed forceps, and allowing the contents of the egg to fall directly into a sterile whirl pak® bag.9 One hundred twenty-five milliliters of TSB was added to the contents of each bag prior to incubation. Samples were analyzed for the presence of Salmonella by a modified standard assay procedure, which involved pre-enrichment in TSB, enrichment in two broths [Tetrathionate
755
RESEARCH NOTE TABLE 1. Salmonella isolations from the laying house environment Positive samples
Location
Number of samples
Number
Percentage
Egg belts Egg collectors Ventilation fans Flush water All samples
22 22 4 2 50
16 14 4 2 36
72.7 63.6 100.0 100.0 72.0
Broth* (TB) and Selenite Cystine Broth* (SCB)], and incubation in M-Broth4 (MB) prior to testing for the presence of Salmonella using an immunoassay system.io One loop (3 mm) of MB, TB, and SCB from immunoassay positive samples was streaked on MacConkey Agar,4 Hektoen Enteric Agar,4 and Bismuth Sulfite Agar,4 and suspect colonies were tested biochemically using a commercial test kit.11 Isolates identified as Salmonella were sent to the National Veterinary Services Laboratory, Ames, LA 50010, for serotyping. Up to five isolates from each positive sample were serotyped. Details of the methods used are described by Jones et al. (1991). RESULTS Salmonella was isolated from 72.0% of the samples collected from the laying house environment (Table 1). The following salmonellae contamination rates were found in flush water, ventilation fan, egg belt, and egg collector samples, respectively (positive samples/total samples collected): 2/2, 4/4, 16/22, and 14/22. Salmonella was detected in the internal contents of none of the 180 eggs analyzed (data not shown). It should be noted that no attempt was made to detect Salmonella that may have been associated with the shell membranes. Contamination was detected on 7 of the 90 eggshells sampled before processing and 1 of 90 eggshells sampled after processing (Table 2).
10 Bio-EnzaBead™/ Organon Teknika, Durham, NC 27713. 1 Micro-Id, Organon-Teknika, Durham, NC 27713.
No S. enteritidis isolates were detected in either the production facilities or the eggs examined (Table 3). The 60 isolates from production facilities included the following Salmonella serotypes: S. agona, S. typhimurium, S. infantis, S. derby, S. Heidelberg, S. California, S. montevideo, S. mbandaka, and untypable. The 22 isolates obtained from eggshells prior to processing were serotyped as S. Heidelberg and S. montevideo (Table 3). All five isolates obtained from eggshells after processing were serotyped as S. Heidelberg. DISCUSSION Salmonella was found in 72.0% of the samples collected from the laying house environment (Table 1). Although a relatively small number of samples was collected, the contamination percentage is higher than that reported elsewhere. News reports have suggested that Salmonella was found in 53% of the Canadian egg operations examined (Elliott, 1990), but no details were given as to the conditions under which these samples were collected. Potter (1987) reported that Salmonella was isolated from 50% of the samples collected from one egg operation and 30% of the samples collected from another, but again no details regarding sampling were given. Although there are discrepancies as to the percentage of samples contaminated, these collective reports imply that Salmonella is a common inhabitant of egg production facilities. The data obtained from eggshells before washing (Table 2) are similar to those reported by Cox et al. (1973), who found that 7.6, 9.5, and 6.3% of the eggshells were contaminated when hens were artificially inoculated with Salmonella senften-
756
JONES ET AL. TABLE 2. Salmonella isolations from eggshells before and after washing Positive samples1
Sampling day
Time of day
Egg wash water2 pH
1 1 1 2 2 2 All
Morning Noon Evening Morning Noon Evening
10.60 10.25 10.55 10.58 10.19 10.34
J
After wash
Before wash Number
Percentage
Number
Percentage
1 3 1 0 1 1 7
6.7 20.0 6.7 0 6.7 6.7 7.8
0 0 0 0 1 0 1
0 0 0 0 6.7 0 1.1
A total of 15 eggs were sampled before and after washing at each time of day. One sample of egg wash water was collected as egg samples were collected.
2
TABLE 3. Salmonella serotypes isolated from egg production facilities and eggshells1 Salmonella Serotype S. agona S. California S. derby S. heidelberg S. infantis S. mbandaka S. montevideo S. typhimurium Untypable
E
gg belts 10 0 0 1 6 0 0 5 0
Egg
collectors
9 2 2 0 5 0 0 8 0
Ventilation fans 2 0 0 2 0 0 0 3 0
Flush water
• (Number of isolates\ ) 0 0 0 0 0 1 0 0 1
Before wash
After wash
Total
0 0 0 17 0 0 5 0 0
0 0 0 5 0 0 0 0 0
21 2 2 25 11 1 5 16 1
1
Where possible, five isolates were serotyped from each positive sample.
berg, Salmonella thompson, and S. typhimu-their liquid contents contaminated with rium, respectively. Although the data salmonellae. obtained from eggshells after wash are not conclusive, it is noteworthy that the one REFERENCES egg found positive for Salmonella followK. E., 1993. Egg quality: Future trends. ing processing was detected when the pH Anderson, Pages 9-11 in: Proceedings of the Florida of wash water samples was lowest. Poultry Institute, Tampa, FL. The lack of internal contamination Baker, R. C, J. P. Goff, and E. J. Mulnix, 1980a. Salmonellae recovery following oral and infound in this survey is similar to reports travenous inoculation of laying hens. Poultry by Baker et al. (1980b) and Chapman et al. Sci. 59:1067-1072. (1988) neither of whom isolated Salmonella Baker, R. C, J. P. Goff, and J. F. Timoney, 1980b. Prevalence of salmonellae on eggs from poultry from the interior of 1,400 and 1,000 eggs, farms in New York state. Poultry Sci. 59: respectively. However, because infected 289-292. hens have been shown to produce 1 Bryan, F. L., 1980. Foodborne diseases in the United States associated with meat and poultry. J. Food contaminated egg out of 200 laid (Morris, Prot. 43:140-150. 1990), a greater number of eggs must be Chapman, P. A., P. Rhodes, and W. Rylands, 1988. examined to determine the true percentSalmonella typhimurium phage type 141 infections in Sheffield during 1984 and 1985: associaage of eggs contaminated by Salmonella. tion with hen's eggs. Epidem. Inf. 101:75-82. Nonetheless, these data suggest that an Cox, N. A., B. H. Davis, A. B. Watts, and A. R. extremely small percentage of eggs have Colmer, 1973. Salmonella in the laying hen. 1.
RESEARCH NOTE
757
Salmonella recovery from viscera, feces and eggs Morris, G. K., 1990. Salmonella enteritidis and eggs: following oral inoculation. Poultry Sci. 52: assessment of risk. Dairy Food Environ. Sanit. 661-666. 10:279-281. Elliott, I., 1990. Salmonella found in half of Canada's Potter, M., 1987. Poultry egg related Salmonella poultry barns. Feedstuffs 62(7):1,4. enteritidis infections. Zootec. Int. 12:44-46. Faddoul, G. P., and G. W. Fellows, 1966. A five-year Potter, M., 1989. Public health significance of poultry survey of the incidence of salmonellae in avian infections by Salmonella, Campylobacter, and species. Avian Dis. 10:296-304. Listeria. Pages 1-12 in: Proceeding of the Avian Gast, R. K., and C. W. Beard, 1990. Production of Enteric Disease Symposium, American AssociaSalmonella enteritidis contaminated eggs by extion of Avian Pathologists/American Veterinary perimentally infected hens. Avian Dis. 34: Medical Association Annual Meeting, July 16, 438-446. 1989, Orlando, FL. Gentry, R. F., and C. L. Quarles, 1972. The measureSt. Louis, M. E., D. L. Morse, M. E. Potter, T. V. ment of bacterial contamination on egg shells. DeMelfi, J. J. Guzewich, R. V. Tauxe, and P. A. Poultry Sci. 51:930-933. Blake, 1988. The emergence of grade A eggs as a Jones, F. X, R. C. Axtell, D. V. Rives, S. E. Scheideler, major source of Salmonella enteritidis infections. F. R Tarver, Jr., R. L. Walker, and M. J. JAMA 259(14):2103-2107. Wineland, 1991. A survey of Salmonella contamination in modern broiler production. J. Food Snoeyenbos, G. H., C. F. Smyser, and F. Van Roekel, 1969. Salmonella infections of the ovary and Prot. 54:502-507. peritoneum of chickens. Avian Dis. 13:668-670. Mayes, F. J., and M. A. Takeballi, 1983. Microbial contamination of the hen's egg: A review. J. Woodward, S., 1989. Salmonella in eggs—The good Food Prot. 46:1092-1098. news and the bad news. Zootech. Int. 3:51-52.